Geotechnical Design Manual Final

Transcript

FEDERAL DEMOCRATIC REPUBLIC OF ETHIOPIA ETHIOPIAN ROADS AUTHORITY GEOTECHNICAL DESIGN MANUAL 2013 Geotechnical Design Manual - 2013 Foreword FOREWORD The road network in Ethiopia provides the dominant mode of freight and passenger transport and thus plays a vital role in the economy of the country. The network comprises a huge national asset that requires adherence to appropriate standards for design, construction and maintenance in order to provide a high level of service. As the length of the road network is increasing, appropriate choice of methods to preserve this investment becomes increasingly important. In 2002 the Ethiopian Roads Authority (ERA) published road design manuals to provide a standardized approach for the design, construction and maintenance of roads in the country. Due to technological development and change, these manuals require periodic updating and expanding. The new series of manuals has particular reference to the prevailing conditions in Ethiopia and reflects the experience gained through activities within the road sector during the last 10 years. The updating of existing manuals and preparation of new manuals was undertaken in close consultation with the federal and regional roads authorities and stakeholders in the road sector. Most importantly, a series of thematic peer review panels was established that comprised local experts from the public and private sector who provided guidance and review for the project team. The Geotechnical Design Manual is a new addition to the ERA series of manuals. The standards set out shall be adhered to unless otherwise directed by ERA. However, I should emphasize that careful consideration to sound engineering practice shall be observed in the use of the manual, and under no circumstances shall the manual waive professional judgment in applied engineering. For simplification in reference this manual may be cited as ERA’s Geotechnical Design Manual - 2013. On behalf of ERA I would like to thank DFID, Crown Agents and the AFCAP management team for their cooperation, contribution and support in the development of the manual. I would also like to extend my gratitude and appreciation to all of the industry stakeholders and participants who contributed their time, knowledge and effort during the development of the documents. Special thanks are extended to the members of the various Peer Review Panels, whose active support and involvement guided the authors of the manual and the process. It is my sincere hope that this manual will provide all users with a standard reference and a ready source of good practice for the geotechnical design of roads, and will assist in a cost effective operation, and environmentally sustainable development of our road network. I look forward to the practices contained in this manual being quickly adopted into our operations, thereby making a sustainable contribution to the improved infrastructure of our country. Comments and suggestions on all aspects from any concerned body, group or individual as feedback during its implementation is expected and will be highly appreciated. Addis Ababa, 2013 Zaid Wolde Gebriel Director General, Ethiopian Roads Authority Ethiopian Roads Authority Page i Preface Geotechnical Design Manual – 2013 PREFACE The Ethiopian Roads Authority is the custodian of the series of technical manuals, standard specifications and bidding documents that are written for the practicing engineer in Ethiopia. The series describes current and recommended practice and sets out the national standards for roads and bridges. The documents are based on national experience and international practice and are approved by the Director General of the Ethiopian Roads Authority. The Geotechnical Design Manual -2013 forms part of the Ethiopian Roads Authority series of and Roadbridges and Bridge Designisdocuments. all roads in Ethiopia, as follows: The complete series of documents, covering 1. 2. 3. 4. Route Selection Manual Site Investigation Manual Geotechnical Design Manual Geometric Design Manual 5. 6. 7. 8. Pavement Design Manual Volume I Flexible Pavements Pavement Design Manual Volume II Rigid Pavements Pavement Rehabilitation and Asphalt Overlay Design Manual Drainage Design Manual 9. Bridge Design Manual 10. Low Volume Roads Design Manual 11. Standard Environmental Procedures Manual 12. 13. 14. 15. Standard Technical Specifications. Standard Drawings Best Practice Manual for Thin Bituminous Surfacings Standard Bidding Documents for Road Work Contracts – A series of Bidding Documents covering a full range from large scale projects unlimited in value to minor works with an upper threshold of $300,000. The higher level documents have both Local Competitive Bidding and International Competitive Bidding versions. These documents are available to registered users through the ERA website: www.era.gov.et Manual Updates Significant changes to criteria, procedures or any other relevant issues related to new policies or revised of theshould land be or incorporated that are mandated by the from relevant Government Ministrylaws or Agency into the manual their Federal date of effectiveness. Other minor changes that will not significantly affect the whole nature of the manual may be accumulated and made periodically. When changes are made and approved, new page(s) incorporating the revision, together with the revision date, will be issued and inserted into the relevant chapter. Page ii Ethiopian Roads Authority Geotechnical Design Manual - 2013 Preface All suggestions to improve the manual should be made in accordance with the following procedures: 1. Users of the manual must register on the ERA website: www.era.gov.et 2. Proposed changes should be outlined on the Manual Change Form and forwarded with a covering letter of its need and purpose to the Director General of the Ethiopian Roads Authority. 3. Agreed changes will be approved by the Director General of the Ethiopian Roads Authority on recommendation from the Deputy Director General (Engineering Operations). 4. The release date will be notified to all registered users and authorities. Addis Ababa, 2013 Zaid Wolde Gebriel Director General, Ethiopian Roads Authority Ethiopian Roads Authority Page iii Preface Geotechnical Design Manual – 2013 ETHIOPIAN ROADS AUTHORITY CHANGE CONTROL DESIGN MANUAL This area to be completed by the ERA Director of Quality Assurance MANUAL CHANGE Manual Title:____________________________ _______________________________________ Section Table Figure Page Explanation CHANGE NO.___________________ SECTION NO.__________________ Suggested Modification Submitted by: Name:____________________________________Designation:_______________________________ Company/Organisation Address ____________________________________________________________________ _______________________________email:___________________________________________Date:___________ Manual Change Action Authority Date Signature Recommended Action Approval Registration Director Quality Assurance Deputy Director General Eng. Ops Approval / Provisional Approval / Rejection of Change: Director General ERA:__________________________________ Date: __________________ Page iv Ethiopian Roads Authority Geotechnical Design Manual - 2013 Acknowledgements ACKNOWLEDGEMENTS The Ethiopian Roads Authority (ERA) wishes to thank the UK Government’s Department for International Development (DFID) through the Africa Community Access Programme (AFCAP) for their support in developing this manual. It will be used by all authorities and organisations responsible for the provision of roads in Ethiopia. From the outset, the approach to the development of the manual was to include all sectors and stakeholders in Ethiopia. Our own extensive local experience and expertise was supplemented by inputs from an international team of experts and shared through review workshops to discuss and debate the contents of the draft manual. ERA wishes to thank all the individuals who gave their time to attend the workshops and provide valuable inputs to the compilation of the manual. In addition to the workshops, Peer Review Groups comprising specialists drawn from within the local industry were established to provide advice and comments in their respective areas of expertise. The contribution of the Peer Review Group participants is gratefully acknowledged. Finally, ERA would like to thank Crown Agents for their overall management of the project. As with the other manuals of this series, the intent was, where possible, and in the interests of uniformity, to use those tests and specifications included in the AASHTO and/or ASTM Materials references. Where no such reference exists for tests and specifications mentioned in this document, other references are used. List of Persons Contributing to Peer Group Review No. Name Organization 1 Abebe Asefa, Ato Ethiopian Roads Authority 2 Alemayehu Ayele, Ato Ethiopian Roads Authority 3 Asnake Haile, Ato OMEGA Consulting Engineers 4 Asrat Sewit, Ato Saba Engineering 5 Colin Gourley, Dr. ERA/DRID 6 Daniel Nebro, Ato Ethiopian Roads Authority 7 Efrem Degefu, Ato BEACON Consulting Engineers 8 Fikert Arega, W/ro Ethiopian Roads Authority 9 Muse Belew, Ato Ethiopian Roads Authority 10 Shimelis Tesfaye, Ato Spice Consult 11 Tewodros Alene, Ato Ethiopian Roads Authority 12 Zerihun Nuru, Ato Gondwana Engineering Ethiopian Roads Authority Page v Acknowledgements Geotechnical Design Manual – 2013 Project Team No. 1 Name Organization Role Bekele Negussie ERA AFCAP Coordinator for Ethiopia 2 Abdo Mohammed ERA Project Coordinator 3 Daniel Nebro ERA Project Coordinator 4 Frew Bekele ERA Project Coordinator 5 Lulseged Ayalew, Dr AFCAP Consultant Lead Author 6 Robert Geddes AFCAP/Crown Agents Technical Manager 7 Les Sampson AFCAP/Crown Agents Project Director 8 Gareth Hearn URS Technical Review 9 Tim Hunt URS Technical Review 10 Alemgena Araya, Dr Consultant Final Review Addis Ababa, 2013 Zaid Wolde Gebriel Director General, Ethiopian Roads Authority Page vi Ethiopian Roads Authority Geotechnical Design Manual - 2013 Table of Contents TABLE OF CONTENTS FOREWORD ............................................................................................................................. I PREFACE ................................................................................................................................ II ACKNOWLEDGEMENTS .........................................................................................................V TABLE OF CONTENTS .......................................................................................................... VII LIST OF ILLUSTRATIONS .......................................................................................................X LIST OF TABLES ................................................................................................................XIII GLOSSARY OF TERMS .........................................................................................................XV ABBREVIATIONS ................................................................................................................. XIX 1 1.1 1.2 2 INTRODUCTION ............................................................................................................ 1-1 Scope ...................................................................................................................... 1-1 Structure ................................................................................................................. 1-2 PAVEMENT SUBGRADE ................................................................................................ 2-1 2.1 2.2 General.................................................................................................................... 2-1 Geotechnical Design Considerations ...................................................................... 2-1 2.2.1 Strength......................................................................................................... 2-1 2.2.2 Stiffness ......................................................................................................... 2-2 2.2.3 Moisture and density .................................................................................... 2-5 2.3 Special Considerations ........................................................................................... 2-6 2.3.1 Expansive soils ............................................................................................. 2-6 2.3.2 Compressible soils ...................................................................................... 2-13 2.3.3 Collapsible soils ......................................................................................... 2-13 2.3.4 Dispersive soils ........................................................................................... 2-17 2.4 Subgrade Treatment .............................................................................................. 2-19 2.4.1 Moisture control ......................................................................................... 2-20 2.4.2 Removal and replacement .......................................................................... 2-21 2.4.3 Soil stabilization ......................................................................................... 2-22 3 ROAD EMBANKMENTS ................................................................................................. 3-1 3.1 Types of Embankments .......................................................................................... 3-1 3.1.1 Rock fill embankments .................................................................................. 3-1 3.1.2 Earth-fill embankments ................................................................................ 3-2 3.1.3 Embankments on soft ground ....................................................................... 3-2 3.2 Design Considerations ............................................................................................ 3-3 3.3 Settlement Analysis ................................................................................................ 3-5 3.3.1 Primary consolidation .................................................................................. 3-6 3.4 3.5 3.6 3.7 3.3.2 Time for settlement ..................................................................................... 3-10 3.3.3 Secondary compression .............................................................................. 3-11 Settlement Mitigation ........................................................................................... 3-13 3.4.1 Preloading and Surcharge.......................................................................... 3-13 3.4.2 Vertical drains ............................................................................................ 3-15 3.4.3 Removal and replacement .......................................................................... 3-16 Bridge Approach Embankments ........................................................................... 3-17 Stability Assessment ............................................................................................. 3-21 Fill Slope Stabilization ......................................................................................... 3-24 Ethiopian Roads Authority Page vii Table of Contents Geotechnical Design Manual – 2013 3.7.1 Staged construction .................................................................................... 3-24 3.7.2 Base reinforcement..................................................................................... 3-28 3.7.3 Ground improvement .................................................................................. 3-29 3.7.4 Lightweight fills .......................................................................................... 3-29 3.7.5 Removal and replacement .......................................................................... 3-29 3.7.6 Toe berms and shear keys .......................................................................... 3-29 3.8 Embankments in Hilly Areas ............................................................................... 3-31 3.9 Fill-slope Angles and Benches ............................................................................. 3-34 3.10 Wall-supported Embankments ............................................................................. 3-37 3.11 Mortared Masonry Walls ..................................................................................... 3-40 3.12 3.13 4 Gabion Supported Embankments ......................................................................... 3-40 Reinforced Embankments .................................................................................... 3-42 ROADSIDE SLOPES ...................................................................................................... 4-1 4.1. 4.2. 4.3. Natural Slopes ........................................................................................................ 4-2 Cut Slopes .............................................................................................................. 4-5 Landslides ............................................................................................................... 4-9 4.3.1. Types of landslides ..................................................................................... 4-10 4.3.2. Causes of landslides ................................................................................... 4-11 4.3.3. Roadside landslides .................................................................................... 4-15 4.3.4. Roadbed landslides .................................................................................... 4-18 4.3.5. Landslide stabilisation ............................................................................... 4-20 4.4. Soil slope cuts....................................................................................................... 4-21 4.4.1. Design considerations ................................................................................ 4-23 4.4.2. Cut slope angles, profiles and benches ...................................................... 4-24 4.4.3. Colluvium and talus slopes ........................................................................ 4-26 4.4.4. Residual soils.............................................................................................. 4-27 4.4.5. Collapsible soils ......................................................................................... 4-29 4.4.6. Latosols ...................................................................................................... 4-30 4.5. Rock Cut Slopes ................................................................................................... 4-31 4.5.1. The role of discontinuities .......................................................................... 4-31 4.5.2. The role of weathering ............................................................................... 4-35 4.5.3. The role of groundwater............................................................................. 4-36 4.5.4. Mode of failures ......................................................................................... 4-38 4.5.5. Design considerations ................................................................................ 4-40 4.5.6. Safe angles of cut and benches................................................................... 4-43 4.5.7. Methods of rock excavation ........................................................................ 4-45 4.6. Soil Slope Stability Analyses ............................................................................... 4-53 4.6.1. Limit equilibrium methods ......................................................................... 4-54 4.6.2. The use of computer programs ................................................................... 4-59 4.6.3. Determination of shear strength parameters ............................................. 4-59 Stability analysis procedures ..................................................................... 4-60 4.7. 4.6.4. Rock slope stability analyses ................................................................................ 4-61 4.7.1. Kinematic analysis ..................................................................................... 4-61 4.7.2. Limit equilibrium analyses ......................................................................... 4-62 5 5.1 5.2 5.3 5.4 GEOTECHNICAL REPORT AND CHECKLIST ................................................................. 5-1 Preliminary Level Geotechnical Report ................................................................. 5-1 Final Level Geotechnical Report ............................................................................ 5-2 General Geotechnical Report Outline .................................................................... 5-5 Checklist ................................................................................................................. 5-7 Page viii Ethiopian Roads Authority Geotechnical Design Manual - 2013 6 Table of Contents REFERENCES AND BIBLIOGRAPHY .............................................................................. 6-1 APPENDIX A - SOIL STABILIZATION ..................................................................................... 1 Ethiopian Roads Authority Page ix List of Illustrations Geotechnical Design Manual – 2013 LIST OF ILLUSTRATIONS Figure 2-1: Typical Resilient Modulus correlations to empirical soil properties and Classification categories. From NCHRP (2001) ................................................................ 2-5 Figure 2-2: Graphical illustration of the depth of desiccation and moisture fluctuation. Modified from BRAB (1978) ............................................................................................. 2-9 Figure 2-3: Mechanism of soil collapse in natural soils................................................... 2-14 Figure 2-4: Typical results of a one dimensional oedometer test. Modified from Univ Iowa (2013) ............................................................................................................................... 2-16 Figure 2-5: Dispersive potential determined based on percentage sodium and total dissolved solids. From US DOI Bureau of the Interior (1991) ........................................ 2-18 Figure 2-6: Sources of moisture in pavements. Modified from US DOT FHWA (2006) 2-20 Figure 3-1: Example of stress-strain incompatibility. From Abramson et al (2002) ......... 3-3 Figure 3-2: Typical consolidation curve for normally consolidated soil, (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress. From US DOT FHWA (2006B) .................................................................................................. 3-7 Figure 3-3: Typical consolidation curve for over-consolidated soil, (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress. From US DOT FHWA (2006B)......................................................................................................... 3-8 Figure 3-4: Typical consolidation curve for under-consolidated soils (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress. From US DOT FHWA (2006B)......................................................................................................... 3-9 Figure 3-5: Log t method of determining the coefficient of consolidation. From US DOT FHWA (2006B) ................................................................................................................ 3-12 Figure 3-6: Concept of pre-loading and its effect on magnitude and time of settlement . 3-13 Figure 3-7: Effect of surcharge on magnitude and time of settlement ............................. 3-14 Figure 3-8: Use of vertical drains to accelerate settlement. From NCHRP (1989) reproduced in US DOT FHWA (2006B) ......................................................................... 3-16 Figure 3-9: Removal and replacement beneath an embankment ..................................... 3-17 Figure 3-10: Elements of a bridge approach embankment. From Briaud et al (1997) .... 3-17 Figure 3-11: Settlement and down-drag in bridge abutments and piles. Modified from US DOT FHWA 2006B ......................................................................................................... 3-18 Figure 3-12: Suggested details of a bridge approach embankment. Modified from US DOT FHWA 2006B .................................................................................................................. 3-19 Figure 3-13: Modes of side slope failures in embankments. From IOWA State (2013) and US DOT FHWA (2006B) ................................................................................................ 3-21 Figure 3-14: Typical circular arc failure mechanism. From US DOT FHWA (2006B) .. 3-23 Figure 3-15: Effect of flooding and rapid-drawdown on embankment stability .............. 3-28 Figure 3-16: Concept of calculating the percent consolidation in staged construction. From Washington State DOT (2013) ......................................................................................... 3-28 Figure 3-17: Use of a counterweight berm (a) and a shear key (b). From US DOT FHWA (2006B)............................................................................................................................. 3-30 Figure 3-18: Typical construction of embankments in hilly areas. From FAO (1998) ... 3-31 Page x Ethiopian Roads Authority Geotechnical Design Manual - 2013 List of Illustrations Figure 3-19: Types of slope instability commonly seen in fills and the underlying hillslope. From MPWT (2008) .............................................................................................. 3-32 Figure 3-20: Typical side-slopes of a rock fill embankment ............................................ 3-35 Figure 3-21: Typical side-slopes of an earth fill embankment .........................................3-35 Figure 3-22: Fill embankment on a hill-side with outward dipping berms ...................... 3-36 Figure 3-23: Embankment on inward inclined hill-side benches. From Keller and Sherar (2011) ...............................................................................................................................3-36 Figure 3-24: Benched fill on a benched hill-side slope. From JKR (2010) .....................3-37 Figure 3-25: Typical use of retaining structures in road embankments ........................... 3-38 Figure 3-26: Gravity and semi-gravity retaining walls .................................................... 3-39 Figure 3-27: Terminology associated with semi-gravity retaining walls ......................... 3-39 Figure 3-28: Typical types of gabion walls ...................................................................... 3-41 Figure 3-29: Application of reinforced slopes in road construction. From New York State DOT (2007) ...................................................................................................................... 3-43 Figure 3-30: Typical construction of reinforced fills. From NY State DOT (2007) ........ 3-44 Figure 3-31: Failure modes for reinforced soil embankments. From US DOT FHWA (2001) ...............................................................................................................................3-45 Figure 3-32: Steps for design of reinforced soil slopes. From US DOT FHWA (2001).. 3-46 Figure 4-1: Terms used commonly to define a road and associated slopes .......................4-1 Figure 4-2: Natural and cut slopes adjacent to a road ........................................................ 4-2 Figure 4-3: Illustration of the terms used to describe stages of slope failure. From Cruden and Varnes (1996) ..............................................................................................................4-4 Figure 4-4: Graphic description of the evolution and extent of slope failure. From Cruden and Varnes (1996) ..............................................................................................................4-5 Figure 4-5: Types of cross-section design. Modified from Keller and Sherar (2011) ....... 4-7 Figure 4-6: Example of a box cut ....................................................................................... 4-8 Figure 4-7: Full cut cross-section ....................................................................................... 4-9 Figure 4-8: Schematic profile of cut slope benches ........................................................... 4-9 Figure 4-9: Adversely jointed rock mass that can fail if joints become undercut by excavation ......................................................................................................................... 4-15 Figure 4-10: Perched water table and the formation of landslides on road cuts .............. 4-15 Figure 4-11: Landslides above a road. From Hearn and Hunt (2011) .............................. 4-16 Figure 4-12: Retrogressive landslide developed on a natural slope ................................. 4-16 Figure 4-13: Landslide on a cut slope .............................................................................. 4-17 Figure 4-14: Landslide affecting both the cut and natural slope above............................ 4-17 Figure 4-15: Failure of a fill slope ....................................................................................4-18 Figure 4-16: Example of a failure affecting the entire road ............................................. 4-19 Figure 4-17: Stability condition of a clay cut slope over time From Bishop and Bjerrum (1960) reproduced in Abramson et al (2002) ................................................................... 4-22 Figure 4-18: Typical cut slope ratios in most soil types ................................................... 4-24 Figure 4-19: Types of cut-slope profiles .......................................................................... 4-26 Figure 4-20: Road cut on a colluvial slope ....................................................................... 4-27 Figure 4-21: Simplified illustration of the formation of residual soils ............................. 4-28 Ethiopian Roads Authority Page xi List of Illustrations Geotechnical Design Manual – 2013 Figure 4-22: Relatively steep 1:1 cut slope in residual clay ............................................ 4-29 Figure 4-23: Failure in black cotton soil due to infiltration of water ............................... 4-29 Figure 4-24: Cut through weathered rock and residual soils ........................................... 4-30 Figure 4-25: Description of different discontinuity parameters ....................................... 4-33 Figure 4-26: Graphical illustration of the rock quality designation (RQD) ..................... 4-35 Figure 4-27: Effect of discontinuity characteristics on groundwater level ...................... 4-37 Figure 4-28: Rock slide initiated by groundwater seepage out of the slope .................... 4-38 Figure 4-29: Modes of failure in rock slopes. Modified from GWP Consultants (2008) 4-39 Figure 4-30: Example of toppling failure ......................................................................... 4-41 Figure 4-31: General guide to select a method of rock excavation. From Pettifer and Fookes (1994) ................................................................................................................... 4-46 Figure 4-32: Blast design parameters. Modified from US DOI Bureau of Reclamation (1998) ............................................................................................................................... 4-47 Figure 4-33: Blasting in limestone ................................................................................... 4-49 Figure 4-34: Rock displacement as a result of uncontrolled blasting .............................. 4-50 Figure 4-35: Equilibrium conditions used to define the factor of safety ......................... 4-55 Page xii Ethiopian Roads Authority Geotechnical Design Manual - 2013 List of Tabless LIST OF TABLES Table 2-1: California Bearing Ratio (CBR). After US DOT FHWA (2006) ..................... 2-2 Table 2-2: Resilient modulus (M R). Modified from US DOT FHWA (2006) ................... 2-3 Table 2-3: Swelling characteristics of subgrade soils. Modified from Montana DOT (2008) ............................................................................................................................................ 2-7 Table 2-4: Collapsible soils. From US DOT FHWA (2006) ............................................ 2-15 Table 2-5: A summary of remedial measures to reduce the effect of dispersive soils. From Soil and water Management (2008) ................................................................................. 2-19 Table 3-1: Engineering properties and field and laboratory tests for embankment design. From Washington State DOT (2013) ................................................................................. 3-4 Table 3-2: Correlations between Cc & soil index parameters. Modified by US DOT FWHA (2006B) from Holtz & Kovacs (1981) ............................................................................... 3-7 Table 3-3: General considerations for specification of selected structural backfill. From US DOT FHWA 2006B .........................................................................................................3-20 Table 3-4: Suggested gradation for drainage aggregate. From United States DOT FHWA (2006B) ............................................................................................................................. 3-20 Table 3-5: Slope assessment guidelines for the design of embankments and cuts. From US DOT FHWA (2006B) ....................................................................................................... 3-25 Table 3-6: Design techniques useful for mitigating embankment failure. From US DOT FHWA (2006B) ................................................................................................................ 3-26 Table 3-7: Slope stabilization techniques for embankments on hill slopes. Modified from MPWT (2008) .................................................................................................................. 3-33 Table 3-8: Preliminary fill slope angles ...........................................................................3-34 Table 4-1: Slope stability problems associated with natural slopes. Modified from Hearn and Hunt (2011) .................................................................................................................. 4-3 Table 4-2: Simple classification of landslide types. Modified from Cruden and Varnes (1996) ...............................................................................................................................4-11 Table 4-3: Common landslide causal factors. Modified from Nettleton et al (2005) ...... 4-12 Table 4-4: Natural and artificial causes of landslides....................................................... 4-13 Table 4-5: Geological, topographical and hydrological causal factors ............................4-14 Table 4-6: Indications of slope instability at or above a road .......................................... 4-20 Table 4-7: Common landslide remedial measures. From Sassa and Canuti (2008) ......... 4-21 Table 4-8 Soil cut slope ratios (H:V) for preliminary design purposes ........................... 4-25 Table 4-9: Types of discontinuities and their characteristics. From GWP Consultants (2008) ...............................................................................................................................4-32 Table 4-10: Fracture density. Modified from US DOI Bureau of Reclamation (1998) ... 4-34 Table 4-11: Aspects to be considered during rock cut slope design. ............................... 4-42 Table 4-12: Indicative maximum cut slope angles (H:V) for rock slopes (without discontinuity control) ........................................................................................................ 4-44 Table 4-13: Factors controlling bench dimension on rock slopes. Modified from GWP Consultants (2008) ............................................................................................................ 4-45 Table 4-14: Controlled blasting techniques. From US DOT FHWA CFLHD (2011) .....4-51 Ethiopian Roads Authority Page xiii List of Tables Geotechnical Design Manual – 2013 Table 4-15: Parameters used for drilling in presplit, smooth and cushion blasting. From US DOT FHWA CFLHD (2011) ........................................................................................... 4-52 Table 4-16: Commonly used limit equilibrium methods ................................................. 4-57 Table 4-17: Advantages and limitations of conventional rock slope analyses. From Eberhardt (2003), modified from Coggan et al (1998) .................................................... 4-62 Table 5-1: Checklist of important information in geotechnical reports ............................. 5-8 Page xiv Ethiopian Roads Authority Geotechnical Design Manual - 2013 Glossary of Terms GLOSSARY OF TERMS Anchored Wall Walls that derive their capacity to resist lateral loads by their structural components being restrained by tension elements connected to anchors (and additionally by partial embedment of their structural components into existing ground). The anchors may be ground anchors (tiebacks), passive concrete anchors, passive pile anchors, or pile group anchors. Atterberg Limits A basic measure of the nature of a fine-grained soil. Depending on the water content of the soil, it may appear in four states: solid, semi-solid, plastic and liquid. In each state the consistency and behaviour of a soil is different and thus so are its engineering properties. Bearing capacity The capacity of soil to support the loads applied to the ground. The bearing capacity of soil is the maximum average contact pressure between the foundation and the soil which should not produce shear failure in the soil. Bench A near-horizontal step on a cut slope, usually to act as a buffer against surface erosion, or to permit the construction of a bench drain, or to act as an intermediate rock fall area, or a combination of all three. Also used to facilitate the keying in of a fill slope into an underlying natural slope. Berm A near-horizontal step on an embankment slope, usually to provide additional weight at the toe or to act as a buffer against surface erosion, or both. Black Cotton Soil A soil in which there is a high content of expansive clay known as smectite that forms deep cracks in drier seasons or years. Black cotton soils typically form from highly basic rocks, such as basalt, in climates that are seasonally humid or subject to erratic droughts and floods, or to impeded drainage. Depending on the parent material and the climate, they can range from grey or red to the more familiar deep black, particularly in Ethiopia. California Bearing Ratio (CBR) An indirect measurement of soil strength based on resistance to penetration. Collapse Potential The change in void ratio (∆e) upon wetting. Collapsible Soils Soils that appear to be strong and stable in their natural (dry) state, but which rapidly consolidate under wetting, generating large and often unexpected settlements. Compressible Soils Soil deposits which are susceptible to large settlements and deformations because of a relatively rapid decrease in void volume upon loading. Ethiopian Roads Authority Page xv Glossary of Terms Geotechnical Design Manual – 2013 Dispersive Soils Soils that, when placed in water, have repulsive forces between the clay particles that exceed the attractive forces. This results in the colloidal fraction going into suspension and, in still water, staying in suspension. In moving water, the dispersed particles are carried away or eroded. Earth fill Embankment Embankments typically built by compacting earthen materials such as natural soils. Expansive Soils Soils which exhibit significant volume changes in the presence of water. Gabion Wall Constructed from rectangular steel wire mesh baskets that are filled on site with stone or rock to form a gravity retaining structure. Geogrid Comprises a regular grid of plastic with large openings (called apertures) between the tensile elements. The function of the apertures is to allow the surrounding soil materials to interlock across the plane of the geogrid. Geomembrane Used to retard or prevent fluid from penetrating the soil and consists of continuous sheets of low permeability materials. Geosynthetic Often used to cover a wide range of different artificially manufactured materials, including geotextiles, geogrids, and geomembranes. Geotextile A permeable geo-synthetic comprised solely of textile, usually created from polymer, most commonly polypropylene, but also potentially including polyester, polyethylene, or nylon. A geotextile is usually classified as either woven or non-woven. Gravity Wall A wall that derives its capacity to resist lateral loads through its dead weight. Liquid Limit The moisture content at the point between the liquid and plastic limits of a clay as determined using a liquid limit device. Modulus of Elasticity The slope of the stress–strain curve in the elastic deformation region. Non- gravity Wall A wall that relies on structural components of the wall partially embedded in foundation passive resistance to resist lateral loads, or asmaterials a resulttoofmobilize their structural components being restrained by tension elements connected to anchors, or a combination of the two. Oedometer Laboratory apparatus for carrying out one-dimensional consolidation tests for the determination of settlement. Pavement The constructed portion of a road that includes the sub-base, base, Page xvi Ethiopian Roads Authority Geotechnical Design Manual - 2013 Glossary of Terms and surface layers. Plastic Limit The moisture content at the lower limit of the plastic state of clay. Plasticity Index A measure of the plasticity of a soil. The plasticity index is the size of the range of water contents where the soil exhibits plastic properties. The PI is the difference between the liquid limit and the plastic limit Preloading The process of compressing the subsoil prior to placing the Primary Consolidation permanent load. Settlement of a soil associated with the readjustment of soil particles due to the migration of water out of the voids Reinforced Embankment This is a form of mechanically stabilized earth that incorporates planar reinforcing elements in the constructed embankment with face inclinations of up to 70 degrees. Metallic strips, geosynthetics, and polymer and wire grids are used as reinforcing elements. Resilient Modulus The resilient modulus (MR) measures the amount of recoverable deformation at any stress level for a dynamically loaded test specimen and is an indication of the stiffness of the layer immediately under the pavement. Rip-rap Rock or other material used to armour streambeds, bridge abutments, and other structures against scour, or water erosion. It can be made from a variety of rock types, commonly granite or limestone, and often used on a waterway or water containment where there is potential for water erosion Rock fill Embankment Embankments usually containing more than 75% by volume of large fragments, 100 mm or greater in size. Secondary Compression Occurs when a soil continues to settle after the excess pore water pressures are dissipated to a negligible level. The occurrence of secondary compression is independent of the stress state. Semi-gravity Wall Similar to gravity walls, except that it relies on its structural components to mobilize the dead weight of fill to derive its capacity to resist lateral loads. Shrinkage Limit The moisture content where further loss of moisture will not result in any more volume reduction. Soil Stabilisation The use of mechanical or chemical modifiers to enhance the strength of soils and reduce the change in moisture. Subgrade The supporting ground beneath a pavement structure Surcharging The process of subjecting the ground to a higher pressure than that during the service life in order to achieve a higher initial rate of Ethiopian Roads Authority Page xvii Glossary of Terms Geotechnical Design Manual – 2013 settlement and thus reducing long term deformation. Swell Potential The swell of a laterally-confined specimen when it is surcharged and flooded. Triaxial Compression Test A compression (or shear) test where the stress applied in the vertical direction (along the axis of the cylindrical sample) can be different from the stresses applied in the horizontal directions perpendicular to the sides of the cylinder, i.e. the confining pressure). Vertical Drain Sometimes called a ‘wick drain’, and is a prefabricated drain consisting of a plastic core that is wrapped with geo-textile and is installed using a mandrel Void Ratio The ratio of the volume of voids in a soil to the volume of solids. Page xviii Ethiopian Roads Authority Geotechnical Design Manual - 2013 Abbreviations ABBREVIATIONS AASHTO American Association of State Highway and Transportation Officials Cα Coefficient of Secondary Compression CBR California Bearing Ratio (as described in AASHTO T 193 or ASTM D 1883) Cc Compression Index CD Consolidated Drained CPT Cv Cone Penetration Test Coefficient of Consolidation CU Consolidated Undrained DCP Dynamic Cone Penetrometer DI Design Index E Modulus of Elasticity eo Initial Voids Ratio ERA Ethiopian Roads Authority ESP Exchangeable Sodium Percentage FS Factor of Safety FWD Falling Weight Deflectometer FWHA Federal Highway Agency Gs Specific Gravity H:V Horizontal:Vertical KPa Kilopascals LL Liquid Limit LSPI Lime Slurry Pressure Injection MR Resilient Modulus Nc Bearing Capacity factor NCHRP National Cooperative Highway Research Program NHI National Highway Institute pH Potential Hydrogen PI Plasticity Index RF RQD Reduction Factor Rock Quality Designation SPT Standard Penetration Test TDS Total Dissolved Solids Tv Time Factor USCS Unified Soils Classification System UU Unconsolidated Undrained Ethiopian Roads Authority Page xix Chapter 1 Geotechnical Design Manual - 2013 1 Introduction INTRODUCTION This manual has been prepared to provide guidance on the standards of practice used for solving geotechnical problems relating to the design and construction of roads in Ethiopia. Its purpose is to present procedures and guidelines useful to design stable subgrades, embankments, and cut slopes. It also aims to achieve consistency in the way geotechnical problems and solutions are approached in the practice of road design for new road construction, upgrading, or road rehabilitation projects. The design and construction of roads requires good communication and coordination between all team members, and especially between the geotechnical engineer and other personnel. This interaction is important in the design of reliable and cost-effective structures. Close and effective communication between geotechnical, highway and structural engineers needs to take place at all times to ensure compatibility with the various design criteria. Much of the technical guidelines and design requirements provided in this manual have been taken from international references such as those produced by the United States Federal Highway Agency (FHWA), United States Army Corps of Engineers (US ACE), US Bureau of Reclamation and others. These sources are listed in the Bibliography at the end of the manual. Users are strongly recommended to use this manual in conjunction with the latest edition of the Site Investigation Manual of the Ethiopian Road Authority (ERA). The first step in performing any geotechnical analysis and design is a thorough review of any test data and engineering parameters available for the proposed project and any associated geotechnical investigation. Experience has shown that an inadequate geotechnical investigation can lead to excessive risk both in terms of schedule and cost. The level of geotechnical field investigation necessary for analysis and design depends on the type of the road project and the geotechnical issue under consideration. This information is given in the Site Investigation Manual which must be consulted in conjunction with this document. 1.1 Scope The subject of Geotechnical Engineering is very wide and for roads its application ranges from simple geotechnical investigation for route selection to foundation analysis and design of bridges. This manual focuses on geotechnical parameters required for subgrade design and construction, the determination of the geotechnical inputs needed for embankment design, the geotechnical aspects of cut-slope design and slope stability, the evaluation and selection of appropriate slope stabilization techniques, and the standard of geotechnical reporting. Each project presents unique considerations and requires an approach that involves a thorough knowledge of the ground conditions. This manual is not intended to serve as the sole reference of geotechnical services on individual projects. Instead, the scope of services for each project should be formulated using this manual and others as references. It must be emphasized that the manual is a guidance document, and that occasions may arise when the methods and recommendations given are inadequate or unsuitable. In such circumstances the use of alternative approaches and engineering judgement will be Ethiopian Roads Authority Page 1-1 Chapter 1 Introduction Geotechnical Design Manual – 2013 necessary. Each situation will require its own investigation and design, and the contents of this manual are intended to be for indicative purposes only. 1.2 Structure The manual is structured into four chapters and an appendix, with Chapter 1 being this Introduction. Chapter 2 discusses the characteristics of pavement subgrades. A subgrade is the in-situ material upon which the pavement structure is placed. The performance of a pavement often depends on the quality of its subgrade. For the pavement to resist deformation and other types of failures, the subgrade must be able to support loads transmitted from above. This load bearing capacity is often affected by the degree of compaction, moisture content, and soil type. Chapter 2 contains a detailed discussion on factors that affect the strength and stiffness characteristics of pavement subgrades, the methodologies used for evaluating and describing different subgrade soils, and the soil treatment and stabilization methods available. Chapter 3 addresses the geotechnical aspects important for the design of road embankments, including rock and earth embankments and lightweight fills, etc. The primary geotechnical issues that impact the performance of all these types of embankments are global stability, internal stability, consolidation and settlement. Chapter 3 presents a detailed discussion of these geotechnical issues with recommendations on material requirements, staged construction, settlement monitoring as well as details of other mitigation measures used to increase embankment performance. In addition, Chapter 3 includes a discussion of design procedures for reinforced embankment slopes. Reinforcement allows embankments to be constructed to greater heights and/or with steeper side slopes. In a typical application, the reinforcement is placed at the base of the embankment, and geotextiles or geo-grids are used as the reinforcing materials. As with many topics, the design procedures given in Chapter 3 are brief and users of this manual are advised to consult other references for more detailed outlines. Chapter 4 deals with the design of slopes as they relate to road construction. Some of the subjects covered are the geological aspects that control the nature of rocks and soils exposed in cut slopes and their stability, types of slope failures, earthworks design and excavation techniques, and stability analysis. As landslides and unstable slopes are common in many parts of Ethiopia, it is vital that engineers engaged in road design activities are aware of the basic principles of slope stability. They must understand how these principles are applied to the design of stable roads through various geological materials and landscapes. Chapter 4 also contains general guidelines and engineering practices needed to satisfy the overall stability requirements during road construction for the type of materials and topographic conditions encountered in Ethiopia. Chapter 5 summarises the requirements for geotechnical reporting. Appendix A covers the topic of soil stabilisation. Page 1-2 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 2 PAVEMENT SUBGRADE 2.1 General Pavement Subgrade The subgrade is the supporting ground beneath a pavement structure. It is located below the base and sub-base courses. It is usually investigated to such depth as may be important to structural design and pavement life, and it may consist of materials forming the natural ground surface or exposed in excavations. In a fill section, the subgrade is the upper part of the embankment. The strength, stiffness, and moisture characteristics of materials forming the subgrade can have a significant influence on pavement performance. The subgrade and base layers must be strong enough to resist shear failure and should have adequate stiffness to minimize vertical deflection. Stronger and stiffer materials provide a more effective foundation for the riding surface and will be more resistant to stresses from repeated loadings and environmental conditions. A critical component of pavement design is, therefore, the investigation and testing of the subgrade upon which the pavement structure will be constructed. In cases where the subgrade is of inadequate strength, methods of improvement must be considered. 2.2 Geotechnical Design Considerations The performance of a road pavement surface is significantly affected by the characteristics of the subgrade. Desirable properties that the subgrade should possess include high strength and stiffness, good drainage, ease of compaction, and low compressibility and swelling. 2.2.1 Strength The strength and stiffness properties control the ability of the subgrade to support loads transmitted from the pavement layers (load-bearing capacity). This ability is often influenced by the degree of compaction, moisture content, soil type, and history of consolidation. The California Bearing Ratio (CBR) is an indirect measure of the strength of the subgrade (Table 2-1). It is also the most widely used method for designing pavement structures. The higher the CBR value of a subgrade, the more strength it has to support the pavement. This means that a thinner pavement structure could be designed on a subgrade with higher CBR compared to a lower CBR value. However, it should be noted that although the CBR value is directly correlated with strength, the change in pavement thickness needed to carry a given traffic load is not directly proportional to the change in CBR value of the subgrade soil. For example, a one-unit change in CBR from 5 to 4 requires a greater increase in pavement thickness than does a change in CBR from 10 to 9. Generally, the strength and stiffness of fine-grained soils are low, especially when they are exposed to water. The fines reduce the overall strength of the subgrade because they reduce the particle-to-particle contact that provides friction to a soil matrix. Sands and gravels with high CBR values are often considered as the best subgrades for formation. In general, a subgrade having a CBR of 10 or greater can usually support heavy loads and repetitious loading without excessive deformation. Ethiopian Roads Authority Page 2-1 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 Table 2-1: California Bearing Ratio (CBR). After US DOT FHWA (2006) Description Uses in pavements Laboratory determination Field measurement Comment The California Bearing Ratio or CBR is an indirect measurement of soil strength based on resistance to penetration. • Direct input to some empirical pavement design methods. • Correlations with resilient modulus and other engineering properties. The laboratory test is determined based on AASTHO T 193 or ASTM D 1883. It is based on the resistance to penetration by a standardized piston moving at a standardized rate for a prescribed penetration distance. Typically, soaked conditions are used to simulate anticipated long term conditions in the field. The testfor is run on three identically compacted seriesThe of tests CBR is run a given relative compaction and samples. moisture Each content. geotechnical or materials engineer must specify the conditions (dry, at optimum moisture, after soaking, 95% relative compaction, etc.) under which each test should be performed. Field measurements are conducted based on ASTM D 4429. Test procedure is similar to that for laboratory determination. Most CBR testing is laboratory based; thus the results will be highly dependent on the representativeness of the samples tested. It is important that the testing conditions be clearly stated. For example, CBR values measured from as-compacted samples at optimum moisture and density conditions can be significantly greater than CBR values measured from similar samples after soaking. For field measurement, care should be taken to make certain that the deflection dial is anchored well outside the loaded area. Field measurement is made at the field moisture content while laboratory testing is typically performed for soaked conditions, so soil-specific correlations between field and laboratory CBR values are often required. 2.2.2 Stiffness The subgrade soil stiffness is measured by resilient modulus (Table 2-2). The resilient modulus (MR) of a material is an estimate of its modulus of elasticity (E). While the modulus of elasticity is stress divided by strain for a slowly applied load, the resilient modulus is stress divided by strain for rapidly applied loads, such as vehicle loads on pavements. MR measures the amount of recoverable deformation at any stress level for a dynamically loaded test specimen. It is defined simply as the ratio of the cyclic axial stress to resilient axial strain as given below. Δ  = Δ Where Δ is the repeated deviator stress and Δ is the recoverable resilient axial strain. AASHTO recommends the use of a resilient modulus (MR) value obtained from a repeated triaxial test for the design of pavements, especially for heavily trafficked pavements. As shown 2-2, thereInarethecurrently five test efforts protocols in been use for resilient modulus testing in in Table the laboratory. past, numerous have made for obtaining appropriate MR values that are representative of field conditions and seasonal moisture variations. The purpose of using seasonal modulus values, particularly for fine grained soils, is to qualify the relative damage of a pavement during each season of the year and include this as part of the overall design. Page 2-2 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade Table 2-2: Resilient modulus (M R). Modified from US DOT FHWA (2006) Description Uses in pavements Laboratory determination The resilient modulus (MR) measures the amount of recoverable deformation at any stress level for a dynamically loaded test specimen and is an indication of the stiffness of the layer immediately under the pavement • Characterization of subgrade stiffness for flexible and rigid pavements (AASHTO 1986/1993; National Cooperative Highway Research Program NCHRP 2004 1-37A, Mechanistic-Empirical). • Determination of structural layer coefficients in flexible pavements (AASHTO 1986/1993). • Characterization of unbound layer stiffness (NCHRP 2004 1-37A). There are currently five test protocols in use for resilient modulus testing in the laboratory: AASHTO T 292-91, AASHTO T 294-92, AASHTO T 307-99, AASHTO T P46-94, NCHRP 1997 1-28A. The harmonized protocol developed in NCHRP Project 1-28A attempts to combine the best features from all of the earlier test methods with a loading sequence that minimizes the potential for premature failure of the test specimen. All of the test procedures employ a closed loop electro-hydraulic testing machine to apply repeated cycles of load-pulse at 0.2 second loading time followed by a 0.8 second rest time for subgrade materials. Axial deformation is measured on the sample using clamps positioned one quarter and three quarters from the base of the test specimen. For very soft specimens, the displacement may be measured between the top and bottom plates. Different specimen sizes, compaction procedures, and loading conditions are usually recommended for granular base/sub-base materials, coarse-grained subgrades, and fine-grained subgrades. These different procedures reflect the different particle sizes of the materials, the state of stress specific to each layer in the pavement structure, and the mechanical behaviour of the material type. Field measurement Comment In-situ resilient modulus values can be estimated from back-calculation of falling weight deflectometer (FWD) test results or correlations with Dynamic Cone Penetrometer (DCP) values. No definitive studies have been conducted to date to provide guidance on differences between measured MR from the various laboratory test protocols. Field MR values determined from FWD back-calculation are often significantly higher than design MR values measured from laboratory tests because of differences in stress states. The 1993 AASHTO guide recommends for subgrade soils that field MR values be multiplied by a factor of up to 0.33 for flexible pavements and up to 0.25 for rigid pavements to adjust to design MR values. NCHRP 2004 1-37A recommends adjustment factors of 0.40 for subgrade soils and 0.67 for granular bases and sub-bases under flexible pavements. The 1993 AASHTO guide includes procedures for incorporating seasonal variations into an effective MR for the subgrade. Seasonal variations of soil properties are included directly in the NCHRP 2004 1-37A design methodology. The procedure in the 1993 AASHTO guide for incorporating seasonal variations into the effective subgrade (MR) can be briefly summarized as follows: • Determine an MR value for each time interval during a year. Typically, time intervals of two weeks or one month duration are used for this analysis. Methods for determining the MR value for each time interval include: Ethiopian Roads Authority Page 2-3 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 Laboratory measurement at the estimated in-situ water content for the time interval. Back calculation from falling weight deflectometer (FWD) tests performed o during each season. Estimate a relative damage X corresponding to each seasonal modulus value using the empirical relationship: o •  = 1.18 108 ( )−2.32 �as the sum of the relative damage values • Compute the average relative damage • for each season divided by the number of seasons. Determine the effective subgrade M R from using the inverse of the above equation:  = 3015(�)−0.431 However, the practice at present is to use the CBR value only during design. This is partly because of the difficulty to determine the MR values as these values depend on many factors. For example, the MR value of granular soils is significantly dependent on gradation, shape, angularity, surface texture, and moisture content. For fine-grained soils, plasticity index, clay content, the specific gravity, and soil consolidation affect the resilient modulus. AASHTO (1993) and NCHRP (2004) provide different methodologies to obtain MR values. As mentioned above, the AASHTO guide recommends the resilient modulus test be performed on laboratory samples using relevant stress levels and moisture conditions simulating the primary moisture seasons. In the laboratory, the MR test applies a repeated axial cyclic stress of fixed magnitude, and load and cycle duration to a cylindrical test specimen. While the specimen is subjected to this dynamic cyclic stress, it is also confined by a static stress provided by a triaxial pressure chamber. The test is essentially a cyclic version of a triaxial compression test with the cyclic load application thought to accurately simulate actual traffic loading. When test facilities are unavailable for performing the test, estimation of the resilient modulus can be made from standard CBR and soil index properties. Figure 2-1 provides a graphical chart suggested by NCHRP (2001) which is useful to correlate the CBR and M R values based on AASHTO and USCS soil groups. It should, however, be noted that the CBR value is a static property that cannot account for the actual response of the pavement under the dynamic loads of moving vehicles, and correlations should always use local experience. Page 2-4 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade Figure 2-1: Typical Resilient Modulus correlations to empirical soil properties and Classification categories. From NCHRP (2001) 2.2.3 Moisture and density Moisture is one of the main factors which affects the strength and stiffness of a subgrade. It also controls the ease of compaction (density) and the compressibility and swelling/shrinkage characteristics of subgrade soils. There are numerous ways through which water can percolate and affect the moisture content of the subgrade. For example seepage from higher ground, either along the pavement or within cuts can cause fluctuations in subgrade moisture conditions. In order to determine the possible consequences, sensitivityduring of theorsubgrade strength and stiffness to changes in moisture content should the be evaluated before design. Generally, in sandy and gravelly soils, small fluctuations in water content produce little change in their strength and stiffness. In silty soils, however, any fluctuation in water content can bring about a certain change in volume, and may also produce large changes in strength and stiffness. Typically these soils attract and retain water through capillary action, and do not drain well. The change in water content in clays often results in large variations in volume, and there may be large changes in strength and stiffness too, Ethiopian Roads Authority Page 2-5 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 particularly if the moisture content is near or above optimum. Typically these soils attract and retain water through matrix suction. Pavement performance also depends on subgrade density. Usually, the depth of influence for wheel loading varies between 1.5 and 3.0 m below the pavement surface. One of the factors which determine the depth of influence for wheel loading is the inherent characteristic of the subgrade and its natural density. If the density varies, then an adequate subgrade compaction is necessary for obtaining a high-quality travel surface. In-situ soils used as subgrades for the construction of road pavements are invariably compacted to improve their density. The purpose of compaction is generally to enhance the strength of a soil by increasing density. The evaluation of density reached as a result of the compactive efforts of compaction equipment is the most common quality-control measurement made at construction sites. Compaction also increases stiffness, decreases the sensitivity of the subgrade soil to changes in moisture content, minimizes long term settlement, and reduces the swelling potential of expansive soils. 2.3 Special Considerations Considering variables such as soil type and clay mineralogy along a length of a roadway, the geology (soil genesis), climate and topography make each project unique with respect to subgrade materials and conditions. For this reason, in addition to suitable foundation materials, several collapsible or highly compressible, expansive or swelling, and dispersive soils occur in many countries and in various regions of Ethiopia. Identification of these problematic subgrade soils is reviewed in the ERA Site Investigation Manual. The characteristics of these soils and the design alternatives to achieve adequate subgrades to support pavement structures are given below. Additional information can also be obtained in the ERA Pavement Design Manual. 2.3.1 Expansive soils Soils which exhibit significant volume changes in the presence of water are termed as expansive soils. These soils exhibit behaviour opposite to consolidation and compression. Expansive soils generally owe their expansive character to their constituent clay minerals, past and present loading history, and to their natural and imposed environments. Swelling may also be due to chemical processes acting on certain non-clay minerals which result in the formation of new minerals of lesser density. Typical damage to roads on expansive soils includes longitudinal unevenness and bumpiness, differential movement near culverts, and longitudinal cracking. The nature of volume change beneath pavements in the vertical direction often takes the form of a general upward movement beginning shortly after the start of construction and continuing until an equilibrium subgrade moisture condition is achieved. Cyclic expansion and contractions of subgrade soils usually occur at the perimeter of pavements and on shoulders exposed to rainfall and evaporation. Local expansion may also occur from poor drainage. In addition, cut sections may lead to local heaving due to removal of surcharge and subsequent increase of moisture content. Table 2-3 summarizes the general characteristics of swelling subgrade soils. Page 2-6 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade Table 2-3: Swelling characteristics of subgrade soils. Modified from Montana DOT (2008) Description Swelling is a large change in soil volume induced by changes in moisture content. Uses in pavements Swelling subgrade soils can have a seriously detrimental effect on pavement performance. Swelling soils must be identified so that they can be removed, stabilized, or treated in the subgrade preparation and pavement design. Soil composition and genesis, clay mineralogy, soil thickness, variation in Intrinsic properties thickness, diagenetic factors, depth below the ground surface, the presence of permeable layers, and dry density. External factors Climatic factors, seasonal moisture fluctuation, time, vegetation, surcharge, and poor drainage. Laboratory determination Swell potential is measured using either the AASHTO T 258 or ASTM D 4546 test protocols. The swell test is typically performed in a consolidation apparatus. The swell potential is determined by observing the swell of a laterally-confined specimen when it is surcharged and flooded. Alternatively, after the specimen is inundated, the height of the specimen is kept constant by adding loads. The vertical stress necessary to maintain zero volume change is the swelling pressure. Field measurements Indirect techniques; swell potential can be estimated using some soil physical and index properties. The swell test can be performed on undisturbed, remoulded, or compacted Comment specimens. If the soil structure is not confined such that to swelling may occur laterally and vertically, triaxial tests can be used determine three dimensional swell characteristics. The volume changes exhibited by expansive soils are related to the interactions of various intrinsic and external factors. The intrinsic factors are soil composition and thickness, dry density, soil fabric and moisture content while the external factors include climate and time. Laboratory related variables which influence the measurement of volume change are initial moisture content, initial dry density, soil fabric, surcharge load, solution characteristics, time allowed for swell, stress history (loading sequence), sample size and shape and temperature. 2.3.1.1 Intrinsic properties Soil composition involves the type and amount of clay mineral within a soil and the size and specific surface area of these minerals. Often, the potential for volume change is mainly determined by mineralogical composition of the clay content. Clays containing montmorillonite expand highly, compared with those composed of kaolinites and illites. Montmorillonite is formed from the weathering of volcanic ash or primary silicate minerals such as feldspars, pyroxenes or amphiboles under those conditions which result in the retention of bases and silica. These conditions are promoted by insufficient leaching of the soil by downward moving water. Long term physical and chemical weathering or alterations of clay soils as a result of changes in overburden conditions or groundwater environment are generally termed Ethiopian Roads Authority Page 2-7 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 diagenetic factors. The diagenetic factors are generally reflected in such phenomena as inter-particle bonding due to recrystallization of the contacts between clay minerals under high overburden stress conditions or by cementation of particles as a result of precipitation of cementing agents from the groundwater. In general, the differences in behaviour of expansive soils between the undisturbed and disturbed states are related to the presence and absence of diagenetic bonds. Apart from soil composition, the properties of the soil profile which influence volume change include the total layer thickness and its variation, depth below the ground surface, and the presence of lenses and layers of more permeable materials. Obviously, the thicker the layer of expansive soil, the greater is the total potential volume change, provided that a source of moisture is available throughout the layer. Variations in thickness of the layer will result in variations of the magnitudes of volume change, or more precisely, differential volume change. Differential expansion, just like differential settlement, is a major problem to pavement structures. The phenomenon of volume change in expansive soils is also the direct result of the availability and variation in the quantity and chemical property of water in the soil. The volume change of expansive soils is primarily due to the hydration of clay minerals. The degree of hydration is influenced by the amount and type of ions present in pore fluids. Pore fluids containing high concentrations of cations tend to reduce the magnitude of volume change of an expansive soil. The soil fabric refers to the orientation of the constituent particles. In the case of expansive soils, the fabric consists of the arrangements of the plate-like clay minerals with each other and with the non-clay components. The arrangement of clay minerals influences the amount and to some degree the direction (lateral or vertical) of volume change exhibited by an expansive soil. The dry density is an important factor in determining the magnitude of volume change. The swell or swelling pressure of an expansive soil increases with increasing dry density for constant moisture content. The reason is that higher densities result in closer particle spacing, therefore causing greater particle interaction. This particle interaction, or more precisely, double-layer water interaction, results in higher osmotic repulsive forces and a greater volume change. This holds true for both disturbed and undisturbed materials. Another important influence of dry density on volume change is its interrelationships with some of the other intrinsic factors. For example, the dry density of a material influences the soil fabric or inter-particle arrangement. Permeability also plays an important role in the time rate of volume change. The permeability of a soil is a function of the initial moisture content, dry density, and soil fabric. For compacted soils, the permeability is greater at low moisture contents and dry densities and decreases to some relatively constant value at about the optimum moisture content. Above optimum, the permeability is essentially constant. The obvious reason for this minimum permeability near the optimum moisture content and maximum dry density is that the voids available for moisture movement are at a minimum because of the close particle spacing. In many cases, however, permeability can normally be enhanced by fissures, fractures, and desiccation cracks. The depth of desiccation is important to the magnitude and rate of volume change and can be defined as the depth to which evaporation influences are reflected in the soil profile (Fig Page 2-8 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade 2-2). Generally, the depth of desiccation is high in warm climate with enough moisture in the soil. Changes in the overburden conditions and the proximity of the groundwater table have an important influence on the depth of desiccation. Figure 2-2: Graphical illustration of the depth of desiccation and moisture fluctuation. Modified from BRAB (1978) 2.3.1.2 External factors Both natural (climate and time) and man-made (surcharge, vegetation, surface drainage, etc) external factors play a role to determine the magnitude of volume change in expansive soils. For instance, the depth of seasonal moisture variation comprises some thickness of the surface material which is influenced by variations in climatic conditions. As would be expected, the larger depths of seasonal moisture change occur in areas where the seasonal climatic changes are greatest, i.e. long dry season followed by excessive rainfall. Ambient temperature conditions also influence the depth of seasonal variations. During rainfall and colder seasons, moisture accumulates closer to the surface and dissipates by evaporation when the climate is warm. Seasonal moisture variations in the central part of Ethiopia and in some northern and western haveareas beenwith reported to occur a depth of between 3 and In lowland highlands and semi-arid reduced sourcesupoftowater, variation is limited to a5m. depth of 2m. Seasonal moisture variations are relatively constant for given climatic conditions and identical soil profile since the general trend is toward accumulation or loss of total moisture content until the weather changes. The influence of time on volume change is another interrelated property which has its major impact on the rate at which expansion occurs. The time to the first occurrence of volume change and the rate of expansion are functions of the permeability of the soil and Ethiopian Roads Authority Page 2-9 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 the availability of water. Expansion occurs as soon as moisture is made available and continues until an equilibrium condition is reached with regard to the source of water. The application of surcharge to an expansive material reduces the amount of volume change that is likely to occur. Often, the presence of a layer of non-expansive overburden may reduce the effect of the underlying expansive material. It may be noted that confinement has its greatest influence on expansive soils in a stress-related sense. This means the greater the confinement, the greater the stress and the smaller would be the deformation. However, in road construction, the load applied by a pavement is often far less than that needed to control deformation. Vegetation such as trees, shrubs, and some grasses are conducive to moisture movement or depletion by transpiration. In areas where vegetation is removed, the moisture that was being used by plants will tend to accumulate beneath the pavement structure and enhance the volume change. Vegetation with large root systems located in close proximity to pavements will result in differential moisture conditions and thus differential volume change. Poor surface drainage leads to moisture accumulation or ponding which can provide a source of moisture for expansive subgrades. It is a frequent problem associated with highways on expansive soils. The problem may be eliminated or reduced by locating the pavement on an embankment constructed with non-expansive materials, locating the side ditches as far away as possible from the edge of the road, and ensuring sufficient lateral and longitudinal gradients so that surface water is removed away from the road with minimum surface infiltration. 2.3.1.3 Identification and testing The identification and testing of expansive soils are discussed in the ERA Site Investigation Manual. The purpose is to qualitatively and quantitatively describe the volume change behaviour of soils. The obvious need for qualitative identification is to inform the design engineer of the potential for expansion and to generally classify this potential on the basis of probable severity. Quantitative testing is necessary to obtain measurable properties for predicting or estimating the magnitude of volume change the soil will experience in order to ascertain approximate treatment or design alternatives. Generally, there are three ways of identification and testing. The first is an indirect technique in which one or more of the intrinsic properties are described or measured, and complemented with experience to determine potential volume change. The information may come from soil composition and genesis, chemical and physical properties of parent material, the climate of the area, index properties of soils, and soil classification systems. The most widely used indicators for identification of expansive soils are the index properties that are routinely determined by most road agencies. Experience has shown that the volume change behaviour correlates reasonably well with liquid limit, plasticity index, and shrinkage limit. In most cases, a combination of observed Atterberg limits and prior experience with materials within a given area could be the main identification methods used for expansive soils. For instance, if the liquid limit is above 70, then the material is often highly expansive and may not be used for fills. If the liquid limit is between 40 and 70, then some type of Page 2-10 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade treatment may be necessary to avoid distress. Similarly, the plasticity index is also useful in that if it is below 15, then minimal problems are anticipated. If the plasticity index is greater than 35, the material must be treated to minimize the problem or it should be discarded. Moreover, a measure of the activity (the ratio of the plasticity index to the clay fraction) is a good indication of the presence of smectites (expansive clay minerals). In addition to the use of Atterberg limits to determine swell behaviour, an engineering soil map is sometimes prepared during site investigation where pedological soil groups along route corridors are described on the basis of index properties. This map can be used as guide to locate expansive soils in the region. The second and direct technique involves actual measurement of volume change in a laboratory. This technique provides swell or swelling pressure values. Since subgrade loading conditions are minimal, usually it is the swell rather than the swelling pressure which is used by many road agencies. The swell defines deformation while the swelling pressure is related to stress generated by the volume change. Many laboratories use the odometer swell test with a minimal surcharge to determine volume change. In practice, however, both the swell and swelling pressure values are rarely taken directly into account in the design of pavements. Instead, they are used only to assess the magnitude of volume change and to assist in the selection of appropriate treatment techniques. The problem is that the laboratory tests are conservative because of the method by which water is made available to specimens. In nature, it is highly unlikely that in-situ subgrade soils would have a sufficient source of water for complete saturation. The third approach involves data from indirect and direct techniques that are correlated either directly or by statistical means to determine the comparative severity of the degree of expansion, its probable severity, and its lateral or regional distribution. In pavement design, the determination of the extent of volume change from indirect techniques and performing laboratory tests only on selected samples is often the most cost effective and faster way of dealing with expansive soils. For instance, it is known that the appearance of highly expansive soils such as Black Cotton soils is distinct after desiccation. The surfaces exhibit polygonal shrinkage cracks which reflect the percentage of clay and possibly the presence of expandable clay minerals. The size of polygons is also an indicator, in that small polygons are often the result of a high clay content. Examining these and other fissures in the field and considering the regional distribution of expansive soils from other sources could help the pavement design engineer to estimate the magnitude of volume change. However, it should also be noted that any clay-rich soil may exhibit polygonal cracking upon drying and it is important to distinguish this from an expansive soil. 2.3.1.4 Treatment options When expansive soils are encountered in areas where significant moisture fluctuations in the subgrade are expected, consideration should be given to the following measures to minimize future volume changes and damage to the pavement structure. • • • For relatively thin (up to say 600 mm) layers of expansive clays near the ground surface, remove and replace the expansive soil with select borrow materials. Extend the width of the bottom of the pavement layers on both sides of the road to reduce the change in subgrade moisture at the edges of the pavement. In addition, increase the roadway camber to reduce the possibility of surface infiltration. Scarify, stabilize, and re-compact the upper portion of the expansive clay subgrade. The amount of heave of expansive soils can also be reduced if compacted to low Ethiopian Roads Authority Page 2-11 Chapter 2 Pavement Subgrade • • • • • • • Geotechnical Design Manual – 2013 densities at moisture contents wet of optimum. However, if the soils are compacted below optimum, they may exhibit excellent immediate stability, but they may fail to satisfy specified density requirements. Upon saturation, the strength of this material could also significantly reduce. Loading expansive soils with a stress greater than the swelling pressure is also a way of preventing swelling. Although pavement loads are generally insufficient, using relatively thick bases and sub-bases (or capping layers) may also be useful to reduce the effect of swelling to a certain extent. In addition, pre-loading or placing 1.0m or more of permanent compacted fill on the existing ground surface prior to pavement construction will reduce the negative (suction) pore-water pressure and thereby decrease the potential for swell. With pre-loading, swelling tends to be more uniform. Pre-wetting the subgrade is one way of reducing the effect of swelling soils. The objective of pre-wetting is to allow desiccated swelling soils to reach equilibrium prior to placement of the pavement. The most commonly applied method for accelerating swelling by this technique is ponding. The time needed for ponding and the depth of moisture penetration in the subgrade is determined based on the characteristics of the soils and their swelling potential. Constructing the pavement during the time of the year when the moisture content of subgrade soils is close to the anticipated equilibrium value may reduce further moisture movement and hence expansion. In areas with deep cuts in over-consolidated expansive clay soils, completing the excavation to the design elevation, and allowing the subsurface soils to rebound prior to placing the pavement layers may also reduce further expansion. Since the change in moisture content is the main factor influencing the volume change of swellingbesoils, it is In obvious that if waterproof the soil is isolated water, volume change should reduced. this context, asphalticfrom membranes are sometimes used in different forms to limit the surface penetration of water. In addition, vertical moisture barriers placed adjacent to pavements or around the perimeter of foundations down to the maximum depth of moisture changes may also be an effective method in maintaining uniform soil moisture. Partial encapsulation along the edge of the pavement or full encapsulation can also be used to reduce change in subgrade moisture. Encapsulation involves maintaining the moisture content at desired constant level by wrapping the subgrade soils in waterproof membranes. Chemical stabilization has been used for altering the clay structure to prevent or minimize the swelling of expansive clays. Lime or cement stabilization are accepted methods for this purpose. Lime injected or mixed into expansive soil can reduce the potential for heave. Fissures should exist in situ to promote the penetration of lime. However, lime may be detrimental in soils containing sulphates. Potassium solutions injected into expansive soils can cause a base exchange, increase the soil permeability, and reduce the potential for swell. Sealed pavements (e.g. asphalt surfaced) will always be more successful in maintaining a more constant moisture content in the underlying subgrade soils, compared to gravel roads where surface water infiltration is impossible to prevent. The placement of geotextile between the subgrade and the pavement layers (subbase) to help distribute differential movements associated with heave more evenly. Page 2-12 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 2.3.2 Pavement Subgrade Compressible soils Compressible soils are soil deposits which are susceptible to large settlements and deformations because of a relatively rapid decrease in void volume upon loading. All soils compress when load is applied. Some, such as gravels and sands, compress very little. Others, such as those containing large amounts of silt and clay, compress significantly. When a load is applied, water molecules in silt or clays flow out of the soil mass. As the water flows out, the soil grains move closer. As this happens, the overall volume of the soil mass decreases, which is reflected as settlement or deformation on the pavement surface. This movement is largely unrecoverable. Compressible soils usually have very low density. Organic and peaty soils are also prone to compression. They are usually found in low lying areas that are prone to flooding. When a road is to be built in areas with thick deposits of highly compressible soils, specific index properties must be examined to estimate settlement. If compressible soils are not treated, large surface depressions with random cracking can develop. The surface depressions can allow water to pond on the road surface and readily infiltrate the pavement structure, potentially creating further problems. When existing subgrade soils do not meet minimum design requirements and are susceptible to large settlements over time, the following treatment options should be considered: • • • • • 2.3.3 Remove and process soil to attain the approximate optimum moisture content, and replace and compact. Remove and replace subgrade soil with suitable compacted borrow or select embankment materials. If soils are non-plastic, consider dynamic compaction of the soils from the surface to increase the dry density. If the soil is extremely wet or saturated and relatively permeable, consider dewatering using well points or deep horizontal drains. If horizontal drains cannot be daylighted, connection to storm drainage pipes or roadside ditches may be required. Consolidate deep deposits of very weak saturated soils with large fills (surcharge) prior to construction. After construction, the fills can either be left in-place or removed, depending on the final elevation. Consider wick drains to accelerate consolidation. Collapsible soils Collapsible soils are those that appear to be strong and stable in their natural (dry) state, but which rapidly consolidate under wetting, generating large and often unexpected settlements. When dry or at low moisture content, collapsible soils give the appearance of a stable deposit. Often, the loose structure of these soils is held together by small amounts of clay minerals or calcium carbonate (Figure 2-3). The introduction of water dissolves the bonds created by these cementing materials and allows the soil to take a denser packing under any type of compressive loading. The condition for collapse is that the soil mass must be in a partially saturated condition and then wetted up and loaded simultaneously, which can occur beneath pavement structures. A short description of collapsible soils is given in Table 2-4. Ethiopian Roads Authority Page 2-13 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 Figure 2-3: Mechanism of soil collapse in natural soils. Collapsible soils are found in areas where there are loess deposits (windblown silts). In Ethiopia, loess deposits are common in the southern part of the Omo River valley. Loess, unlike other non-cohesive soils, can stand on nearly vertical slopes until saturated. It has a low relative density, a low unit weight, and a high void ratio. Other types of collapsible deposits include alluvial soils formed in flood plains and semi-arid lowlands, residual soils formed as a result of the removal of organics by decomposition or the leaching of certain minerals (calcium carbonate), and residual soils formed by weathering of volcanic rocks such as welded tuff and ash. Collapsible soils may also be suspected in undeveloped areas that have young, accumulating sandy and silty soils. Volcanic ashes (andosols with allophane clay minerals), often found in Ethiopia, typically have low density and high porosity, and are easily crushed during compaction. These soils can pose significant problems if present in the subgrade or used as fill, as their structure is lost through compaction and they can have very high water contents, causing them to become difficult to manage and susceptible to flow failure. Page 2-14 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade Table 2-4: Collapsible soils. From US DOT FHWA (2006) Description Collapsible soils exhibit large decreases in strength at moisture contents approaching saturation, resulting in a collapse of the soil skeleton and large decreases in soil volume. Uses in pavements Collapsible subgrade soils can have a seriously detrimental effect on pavement performance. Collapsible soils must be identified so that they can be removed or stabilized. Laboratory determination Collapse potential is measured using the ASTM D 5333 test protocol. The collapse potential of suspected soils is determined by placing an undisturbed, compacted, or remoulded specimen in an oedometer. A load is applied and the soil is saturated to measure the magnitude of the vertical displacement. Field measurement Indirect techniques such as excavating and refilling of pits and measuring the depth below the srcinal ground surface. Comment The collapse during wetting occurs due to the destruction of clay binding, which provides the srcinal strength of these soils. Remoulding and compacting may also destroy the srcinal structure. On pavements, the result of collapse of the subgrade is mostly manifested by the development of a deeply rutted and often uneven road surface and significant deterioration of the riding quality of the road with or without cracking (Page Green 2008). However, such rutting and unevenness is more likely to be due to other factors such as poor compaction or the presence of compressible soils. Among the common artificial sources of wetting in the surroundings of pavements are: (a) irrigation of agricultural lands; (b) leakage from unlined drains; and (c) seepages from ponded areas such as retention basins. The severity of collapse depends on the extent of wetting, depth of the deposit and loading from the overburden weight and structure. Minor artificial and natural wetting is often confined to the top layers beneath the ground surface. Sustained, long term infiltration can lead to soil wetting at depth below the surface which in extreme circumstances can be quite serious and can lead to significant settlements and ground cracking. Unlike expansive soils, where the heave is restricted by the overlying load, collapsible soils can be subject to collapse over their entire thickness (Paige-Green 2008). 2.3.3.1 Identification The potential for collapse at a specific location is initially evaluated based on the geological and environmental setting. The subsurface conditions are then evaluated by ground investigation, for example using power auger borings, trial pits and open excavations. These may be supplemented by boreholes that can extend to depths sufficient to define the thickness of the collapsible soil. If a pit is excavated, then filling the hole with the same amount of material can indicate whether the soils are collapsible, especially if the difference in levels is high after tamping and levelling. In the laboratory, the potential degree of collapse is best determined using oedometer tests (various tests including the double oedometer, single oedometer and collapse potential test are used) in which the collapse potential under specific loads is determined. For subgrades under roads, the actual quantification is not that important. The primary objective is to recognize the problem and implement measures to disrupt the collapse structure as far as Ethiopian Roads Authority Page 2-15 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 possible and to produce a subgrade that is relatively uniform, so that differential settlement is minimized (Page-Green 2008). An example of a typical stress-strain curve obtained by a one dimensional oedometer test is shown in Fig. 2.4. The collapse potential is defined as the change in void ratio ( ∆e) upon wetting compared to eo + 1. Settlement estimates are generally made by taking the collapse potential over the potential depth of wetting. Clearly, the larger the collapse potential, the more collapsible the soil is considered to be. Collapse potential in the order of 1% is considered to be mild, while that above 10% is considered to be severe. Figure 2-4: Typical results of a one dimensional oedometer test. Modified from Univ Iowa (2013) Usually, the collapse potential of the soil depends on density, gradation, the initial water content, composition, and the extent of loading at the time of wetting. A good indication of the potential for high collapse potential is a very low density. Typical collapsible soils have densities of less than 1600 kg/m3 (mostly in the range of 1000 to 1585 kg/m 3). In addition, more than 60% of the mass of collapsible material lies in the 0.075 to 2 mm range and less than 20% is finer than 0.075 mm (Page-Green 2008). In addition, high collapse potentials are recorded for low initial moisture contents. For a given initial water content, the collapse potential decreases with increase in the energy of compaction. 2.3.3.2 Treatment options If pavements are to be constructed over collapsible soils, special remedial measures may be required to prevent large-scale cracking and differential settlement. The most obvious remedial measure is to preclude the presence of water. This is, however, impractical. Compaction of subgrades using conventional compaction plant has been shown to be only moderately effective in removing the collapse potential to any significant depth, even after the addition of water. However, modern high energy compaction techniques using large impact rollers, with or without the addition of water, have proved most effective in reducing the collapse potential to a significant depth. If this work is done in the wet season, a more economical and effective result is obtained as a result of the improved lubrication offered by water (Page-Green 2008). Page 2-16 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade In addition, various ground modification methods can be used to prevent or limit collapse from occurring, or cause the collapse to occur before construction. These methods include partial removal and replacement; densification of the collapsible soil in-place such as by compaction grouting; and pre-wetting of the collapsible soil followed by surcharge loading to cause settlement before construction; conventional compaction with heavy vibratory roller for shallow depths (within 0.3 to 0.6 m); and dynamic or vibratory compaction for deeper compressible soils of more than 0.5 m (combined with inundation with water). More options will generally be available to new construction compared to existing pavements where there are constraints to mitigation options. For depths of collapsible soils greater than 1.5 m, lime pressure and sodium silicate injections could also be helpful, though expensive. 2.3.4 Dispersive soils Dispersive soils are those soils that, when placed in water, have repulsive forces between the clay particles that exceed the attractive forces. This results in the colloidal fraction going into suspension and, in still water, staying in suspension. In moving water, the dispersed particles are carried away or eroded. This obviously has serious implications in earth dam engineering, but it is of less consequence in road (Page-Green 2008) construction, especially when compared to the effects of expansive and collapsible soils. In some places, however, the inclusion of dispersive soils in the subgrade or fill can lead to significant pavement failures through piping, tunnelling and the formation of cavities. It is, therefore, important to identify dispersive soils prior to design (Page-Green 2008). Dispersive soils have not been definitively associated with any specific geological srcin but predominantly have a high sodium cation content. They occur mostly as alluvial clays in the form of slope wash, lake bed sediments, loess deposits, and flood plain silts and clays. Early studies indicated that dispersive clays were associated only with soils formed in arid or semi-arid climates and in areas of alkaline soils. Recently, however, the same soils have been found in humid climates and in various geographic locations. In areas of sloping topography where dispersive soils exist, a characteristic pattern of surface erosion is evidenced by jagged, sinuous ridges and deep rapidly forming channels and tunnels. In gently rolling or flat areas there is frequently no surface evidence of dispersive clay, due to an overlying protective layer of silty sand or topsoil from which the dispersive soil particles have been removed. An absence of surface erosion patterns typical of dispersive soils does not necessarily indicate that dispersive soils are not present. Dispersive, slaking and erodible soils are similar in their field appearance (highly eroded, gullied and channelled exposures), but differ significantly in the mechanisms of their actions. Unlike dispersive soils, slaking soils disintegrate in water to silt, sand and gravel sized particles without going into dispersion. The cause of slaking is probably a combination of swelling of clay particles, the generation of high pore air pressures as water is drawn into the voids in the material, and softening of any cementation. Purely erodible soils will not necessarily disintegrate or go into dispersion in water. They tend to lose material as a result of the frictional drag of water flowing over the material exceeding the cohesive forces holding the material together (Page-Green 2008). It is not very important (nor even really possible) to quantify the actual potential effect of dispersive material, as the process is time related and, given enough time, all of the colloidal material could theoretically be dispersed and removed, leading to piping, internal erosion and eventually loss of material on a large scale. However, most studies reported in Ethiopian Roads Authority Page 2-17 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 the literature have shown that failures of pavements built on dispersive clay soils occurred on first wetting. Further damage was associated with the presence of water and cracking by shrinkage, differential settlement, or maintenance deficiencies. Hence it is important to identify the presence of dispersive soils so that the necessary precautions can be taken in an appropriate time. 2.3.4.1 Identification Identification of dispersive soils should start with field investigations to determine if there are any surface indications such as unusual erosional patterns with sub-soil pipes and deep gullies, concurrent excessive turbidity any nearby Areas of soils, poor crop production andwith stunted vegetation may inindicate highlybodies salineoforwater. carbonate-rich many of which are dispersive. However, dispersive soils can also occur in neutral or acidic soils and can support lush grass growth. Although surface evidence can give a strong indication of dispersive soils, lack of such evidence does not preclude the presence of dispersive clay at depth and further investigation should be undertaken. In many cases, dispersive clays are more related to pedogenic processes or depositional environments and are better identified on soil or agricultural maps if available at the required scale. Figure 2-5: Dispersive potential determined based on percentage sodium and total dissolved solids. From US DOI Bureau of the Interior (1991) Dispersive clays cannot be identified by the standard index tests such as visual classification, grain size analysis, specific gravity, or from Atterberg limits and therefore other laboratory tests have been devised for this purpose. The five laboratory tests most generally performed to identify dispersive clays are the crumb test, the double hydrometer test, the pinhole test, the dissolved salts in the pore water test, and exchangeable sodium percentage (ESP) based tests. Determination of the exchangeable sodium percentage (ESP) has been suggested as the best test, and has been implemented widely, with the addition of results from the pinhole test. Figure 2.3 presents a method of classifying dispersive soils using percent sodium and total dissolved solids (TDS). However, as discussed in the ERA Site Investigation Manual, the crumb test on undisturbed samples is the best first indication for pavement design purposes. Dispersive soils tend to produce a colloidal suspension or cloudiness over the crumb during the test, Page 2-18 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade without the material necessarily disintegrating fully. Disintegration of the crumb in slaking soils is very rapid and forms a heap of silt, sand and gravel. Erodible soils do not necessarily disintegrate in the crumb test as they require a frictional force of moving water to loosen the surface material. Clay soils should be routinely tested for dispersive characteristics during design studies for pavement and hydraulic structures where the clay may be subjected to potential erosion and piping. 2.3.4.2 Treatment options The remedial measures for avoiding dispersive soil damage to road pavements are summarized in Table 2-5 and from the relatively to the demanding costly. Precautions include, butrange are not limited to propersimple moisture andmore density control, and use of filters and filter drains, selective placement of materials, use of sand-gravel blankets or lime-modified soil on slopes, and chemical treatments of dispersive clays. Avoiding the use of dispersive soils in fills as far as possible and removing and replacing them in the subgrade is the preferred solution. It is important to manage water flows and drainage in the area well. Since the presence of sodium as an exchangeable cation in clays is a major problem, treatment with lime or gypsum will allow the calcium cations to replace sodium and reduce the potential for dispersion. It is also important that the material is compacted at 2 to 3% above optimum moisture content to as high a density as possible (Page Green 2008). Table 2-5: A summary of remedial measures to reduce the effect of dispersive soils. From Soil and water Management (2008) 1. Minimize disturbance to topsoil and vegetation. 2. Choose construction methods that minimize the need for excavation and subsoil exposure. 3. Avoid concentrating water flow over areas that have dispersive topsoil or sub-soils. If possible divert water to areas where the soil is not dispersive. 4. Immediately infill any trenches or holes to prevent collection and ponding of water on subsoil surfaces. 5. Always compact dispersive sub-soils that have been disturbed or excavated. Dispersive soils require above average compaction. Careful control of compaction and water content is important during construction. 6. Top dress the surface of potentially dispersive soils with up to 2% gypsum (if soil pH > 6.5) or up to 4% lime by dry mass of soil (if soil pH <5) or a mixture of both (if soil pH is within the range of 5 to 6.5). 7. Cover dispersive soils with a minimum 100 mm layer of non-dispersive soil prior to revegetation, or the placement of pavement layers. 2.4 Subgrade Treatment Proper treatment of subgrade soils and the preparation of the foundation are important to ensure a long-lasting pavement that does not require excessive maintenance. The subgrade should be treated to form a construction pad or a long-term subsurface layer capable of carrying pavement applied loads. Subgrade soils can be treated using a variety of methods or a combination of them. Techniques that can be used to improve the strength and stiffness of the subgrade and increase pavement performance include moisture control removal and replacement, soil stabilization (modification), and the use of geosynthetics. Ethiopian Roads Authority Page 2-19 Chapter 2 Pavement Subgrade 2.4.1 Geotechnical Design Manual – 2013 Moisture control It is known that excess moisture has a damaging effect on pavement structures. This also applies to expansive and to a lesser extent, dispersive soils. Moisture, in combination with other factors, can have a profound negative effect on both material properties of the subgrade and the overall performance of the pavement. As shown in Figure 2-6, moisture can enter the pavement from a variety of sources. It may seep downward from a higher ground, may infiltrate through the surface, or could flow laterally from the pavement edges and shoulder ditches. Capillary action and moisture-vapour movement are also important. Capillary effects are the result of surface tension and the attraction between water and soil. The movement of vapour is associated with fluctuating temperatures and other climatic conditions. Figure 2-6: Sources of moisture in pavements. Modified from US DOT FHWA (2006) Hence, knowledge of groundwater and its movement are critical to the performance of the pavement as well as stability of the subgrade, embankments, side slopes and cut sections. Groundwater can be especially problematic for pavement subgrades in low-lying areas where inundation is common. Moisture control is therefore often an essential part of pavement design. A major issue in geotechnical design of pavements is especially to prevent the subgrade from becoming saturated or even exposed to constant high moisture levels. The three main approaches for controlling or reducing the problems caused by moisture on pavements are: • • • Prevent moisture from entering the pavement. Use materials and design features that are insensitive to the effects of moisture. Quickly remove moisture that enters the pavement. The first two approaches involve the upper pavement layers, and are covered in the ERA Pavement Design Manual. The last approach (removal of moisture) normally needs a subsurface investigation and groundwater study to design the most appropriate drainage system. It is necessary to drain the ground properly and to allow exposed soils of the subgrade to dry out. Deep subsurface drains are usually installed to reduce groundwater levels. These drains intercept the lateral flow of subsurface water beneath the pavement structure, and remove the water that infiltrates the pavement surface. Various types of longitudinal roadside drains placed in trenches beneath shoulders at shallower depths can also be used to handle water infiltrating the pavement from above and the sides. Page 2-20 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade In some cases, subsurface drainage may remove water, but may not significantly reduce the moisture content of fine grained soils in the subgrade. Drying should then be accomplished through evaporation of soil moisture at the time of construction. Disking and tilling of the soil accelerates the drying process, by reducing the size of soil lumps, thereby increasing the surface area exposed to evaporation. These methods are generally effective only in the top 200 to 300 mm of the subgrade, and are highly dependent on weather and environmental conditions. 2.4.2 Removal and replacement Removal of naturally occurring soil andofreplacing it with problems. a suitable material the most obvious method of eliminating many the subgrade In some iscases this approach may be economical if the thickness of the layer to be removed is less than about 1 to 2 m and suitable replacement material is available. Unfortunately, this is generally not the case, and the excavation and replacement solution is extended only to a depth which will reduce the subgrade problem to a tolerable minimum. Hence the required depth of excavation depends upon the suitability of the subgrade soil and the anticipated characteristics of fill that is available nearby. Usually, the selection of appropriate material for replacing poor subgrade soils is a critical issue. Some road agencies use granular materials to replace unsuitable subgrade soils for structural and drainage reasons. Often a granular layer is used to provide uniformity and support as a construction platform. A thick mixed gravel/sand/silt layer may be used as an alternative to soil stabilization for subgrade improvement in areas with large quantities of readily accessible, good quality borrow material. Subgrade improvement is often the preferred way of dealing with weak and poorly drained soils compared to increasing the pavement layer thicknesses. The objectives and benefits of thick mixed gravel/sand/silt layers for subgrade improvement are to: • • • • • Increase the supporting capacity of weak, fine-grained subgrades. Provide a minimum bearing capacity for the design and construction of pavements. Provide uniform subgrade support over sections with highly variable soil conditions. Reduce the seasonal effects of moisture and temperature on subgrade support. Reduce subgrade rutting potential of flexible pavements. Several field and laboratory methods are used to characterize the strength and stiffness of granular materials, including the California Bearing Ratio (CBR) and resilient modulus (MR). In general, materials with CBR values of 20% or greater which corresponds to M R of approximately 120 MPa can be used (US DOT FHWA (2006A). These are typically sand or granular materials with or without limited fines, corresponding to AASHTO A-1 and A2 (GW, GP, SW and SP) soils. The type of granular material used is normally a function of material availability and cost. Pit-run gravel/sand/silt is the most common. Although the high shear strength of crushed stone may be more desirable, the use of crushed materials is unlikely to be economically feasible. The thicknesses of granular layers vary, depending upon their intended use. To increase the composite subgrade design values ( i.e. combination of granular layer over natural soil), it is usually necessary to place a minimum of 0.5 – 1.5 m of embankment Ethiopian Roads Authority Page 2-21 Chapter 2 Pavement Subgrade Geotechnical Design Manual – 2013 material, depending on the strength of the granular material relative to that of the underlying soil. In some instances, the use of a thick granular layer can allow the dimension of the overlying layers to be reduced without compromising the strength and serviceability of the entire pavement structure. The placement of a granular layer over a comparatively weak underlying soil forms a nonhomogeneous subgrade in the vertical direction. Pavement design requires a single subgrade design value, for example CBR or MR. This is generally determined through laboratory or field tests, when the soil mass in the zone of influence of vehicle loads is of the same type, or exhibits similar properties. In the case of a non-homogeneous subgrade, the composite reaction of the embankment and soil combination can vary from that of the natural soil to that of the granular layer. Most commonly, the composite reaction is a value somewhere between the two extremes, dependent upon the relative difference in moduli, and the thicknesses of the granular layer. The actual composite subgrade response is not known until the embankment (granular) layer is placed in the field, and it may also be different once the upper pavement layers are placed. To account for non-homogenous subgrades in pavement structural design, it is recommended to characterize the individual material properties by traditional means, such as resilient modulus or CBR testing, and to compare these results to field tests performed over the constructed embankment layers, as well as the completed pavement section. Some road agencies use in-situ plate load tests to verify that a minimum composite subgrade modulus has been achieved. Deflection devices, including the Falling Weight Deflectometer (FWD), can be used for testing the compacted embankment layer and the constructed pavement surface. It is advisable to use caution when selecting a design subgrade value for a nonhomogenous subgrade. Experience has shown that a good-quality embankment layer must be as high as 1 m or more, before the composite subgrade reaction begins to resemble that of the granular layer. This means that, for granular layers of up to 1 m in height, the composite reaction can be much less than that of the embankment layer itself. If too high a subgrade design value is selected, the pavement will be under-designed. Granular layers less than 0.5 m thick have minimal impact on the composite subgrade reaction, when loaded under the completed pavement section. Structural design charts found in the ERA Flexible Pavement Design Manual will usually provide the necessary guidance. 2.4.3 Soil stabilization Soil stabilization is a general term that involves the use of mechanical or chemical modifiers to enhance the strength of soils and reduce the change in moisture. The process is often called soil modification when the purpose is to change the physical properties and thereby improve the quality of the subgrade soil. Soil stabilization is usually performed for the following reasons: • As a construction platform to dry very wet soils and facilitate compaction of the upper layer. For this case, the stabilized soil is usually not considered as a structural layer in the pavement design process. • To strengthen a weak soil and restrict the volume change potential of a highly plastic (expansive) or compressible soil. For this case, the modified soil is usually given some structural value in the pavement design process. Page 2-22 Ethiopian Roads Authority Chapter 2 Geotechnical Design Manual - 2013 Pavement Subgrade For further information on soil stabilization, including the use of geotextiles and geogrids, see Appendix A. Ethiopian Roads Authority Page 2-23 Chapter 3 Geotechnical Design Manual - 2013 3 Road Embankments ROAD EMBANKMENTS In road construction, embankment design is treated separately from the sub-grade primarily because embankments are usually constructed of engineered fill imported from other locations. The engineered fill is normally compacted as it is placed. Compaction of the fill is monitored to confirm that it is constructed in accordance with the specification. 3.1 Types of Embankments A road embankment may be constructed from either rock-fill or earth-fill. Fills that comprise both coarse and fine grained materials and are known as earth-rock (composite) embankments. 3.1.1 Rock fill embankments In most cases, rock fill embankments contain more than 75% by volume of large fragments, 100 mm or greater in size. When these materials are placed and compacted, the embankment produced as a result primarily derives its stability from the mechanical interlock of coarser particles. Since standard laboratory compaction tests on coarse materials are of doubtful accuracy, a given amount of compactive effort is normally specified for rock-fill embankments. Special consideration should be given to the type of material that will be used in rock-fill embankments. In some areas, moderately weathered or very soft rocks may be encountered in cuts and used as embankment fills. The use of these materials can result in significant long term settlement and stability problems as the rock degrades. Such rocks should be checked by a slake durability test. If the rock is found to be non-durable, it should be physically broken down and compacted as earth embankment provided the material meets the specification for earth-fill. Shales and mudstones found in sedimentary sequences in northern and eastern Ethiopia are good examples of low durability rocks and special compaction techniques are required if they are to be used as fill. As with all rock-fill, any oversize materials should be removed. Generally, in order to maintain the long-term stability of a rock-fill embankment, 90% of the rock fragments with dimensions greater than 100 mm should have a Point Load strength of 2.0 MPa or greater. In addition, the maximum size of the coarse particles and lift thickness should be specified, as they may vary in accordance with the hardness of rocks in the region. However, in many instances, both the particle size and lift thicknesses are limited to 300 mm. The advantage of rock fills compared with earth embankments is that they can be constructed to steeper side slopes. As a consequence the fill volume is lower than that for a similar height earth embankment, particularly on sloping ground and they can therefore be constructed more quickly. Their disadvantage is that their unit volume is usually more costly than earth, particularly if the rock is not available locally within the project area. Ethiopian Roads Authority Page 3-1 Chapter 3 Road Embankments 3.1.2 Geotechnical Design Manual – 2013 Earth-fill embankments Earth-fill embankments are typically built by compacting earthen materials such as natural soils. Hence the compaction properties of the soil materials (optimum water content and maximum dry density) are very important to the long-term performance of the embankments. Compressibility and shear strength can also be useful measures to determine the stability of earth embankments. In addition, drainage is an important issue to prevent the loss of shear strength due to saturation. 3.1.3 Embankments on soft ground If soils underlying an embankment are predominately cohesive, then the primary design issues will be bearing capacity, side-slope stability during construction, and long-term settlement. These design issues will usually require the collection of undisturbed soil samples for laboratory strength and consolidation testing. It may also be desirable to collect in situ vane shear strength data and conduct DCP tests. The vane shear test can provide valuable in-situ strength data, particularly in soft clays. The DCP information can be used to identify locations for sampling and the occurrence of cohesionless layers that could increase the rate of consolidation. It will usually be necessary to perform triaxial compression tests in the laboratory to determine undrained strengths as well as total stress and effective stress parameters. Consolidation tests can be conducted to define the preconsolidation pressure, the compressibility index and the coefficient of consolidation. Cohesionless soils underlying an embankment are not usually a major geotechnical design concern for static loading. This is because most cohesionless soils exhibit good bearing capacity and low compressibility. Settlements will generally be small and will occur rapidly during placement of the fill. If the project area is in a seismically active zone then the liquefaction potential should be considered. In this case, it is necessary to determine the level of groundwater table and its fluctuation characteristics. Grain-size distribution data are also needed. Embankments sometimes have to be built on weak foundation materials or soft ground, such as soft clays and silts and organic materials, including peat. Settlement or piping may occur as a result, irrespective of the stability of the embankment. These can result in significant settlement of the embankment during construction which in turn will lead to an increase in the required fill quantity, long term total and differential settlements affecting the serviceability of the road, and instability of the embankment. Embankments on soft ground also have a tendency to spread laterally because of horizontal earth pressures acting within the embankment. These earth pressures cause horizontal shear stresses at the base of the embankment that must be resisted by the shear strengths of foundation soils. Soft soils do not have adequate shear strength and failure may occur as a result. Hence, embankments on soft soils are ideally designed so that the increase in stress is relatively small, i.e. shallow embankments. Embankment fills over soft ground are frequently stronger and stiffer than their foundations. This leads to the possibility that an embankment will deform as the foundation fails under the weight of the embankment and the possibility of progressive failure because of stress–strain incompatibility between the embankment and its foundation (Figure 3-1). Because of this problem, peak strengths of the embankment and the foundation soils do not mobilize simultaneously. Often, stability analyses performed using peak strengths of soils would overestimate the factor of safety. Page 3-2 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-1: Example of stress-strain incompatibility. From Abramson et al (2002) Where embankments are required to be built over soft ground, stability and settlement of the fill should be carefully evaluated. Some very soft and highly organic materials need to be tested in situ with vane shear apparatus, as retrieving undisturbed samples from these soils for laboratory testing is difficult. In addition, the level of groundwater should be determined and its fluctuation monitored. Attention should be given to the internal stability of the entire embankment foundation interface rather than the embankment or the foundation soils alone. Consideration should be given to removing the soft underlying layer and replacing it with free-draining material prior to the placement of the embankment. 3.2 Design Considerations Table 3-1 provides a summary of the engineering properties and field and laboratory tests needed for the design of embankments. The key geotechnical issues are stability and settlement characteristics of the foundation soils and the bearing capacity at the base. The impact of these considerations on stages of construction is also an issue that should be considered during design. Embankments provide adequate support for roadways if the additional stress from traffic loads and pavement structures does not exceed the shear strength of the soils. Overstressing the embankment may result in slope failure. In addition to issues related to side slope stability, any anticipated settlement problem must also be considered. This settlement can be short term created as a result of the addition of increased load on the soil beneath the embankment (bearing failure during placement of the fill) or long-term due to consolidation. The primary design issue is whether the existing foundation soil can support the new embankment loads without undergoing bearing failure, or excessive settlement. These conditions are most critical when soft cohesive soils are present below the embankment. Staged construction may need to be considered whereby the embankment is constructed in a number of stages and excess pore water pressures allowed to dissipate over a period of time before the next lift is constructed. Sometimes trial embankments are constructed at the beginning of the construction process to check assumptions made during the design. Ethiopian Roads Authority Page 3-3 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Table 3-1: Engineering properties and field and laboratory tests for embankment design. From Washington State DOT (2013) Engineering parameters Required information for analyses • Subsurface profile • Settlement (soil, ground water, (magnitude and rate) rock) • Bearing capacity • Compressibility • • • • • • • • • Slope Lateralstability pressure Internal stability Borrow source evaluation (available quantity and quality of borrow soil) Required reinforcement Liquefaction Delineation of soft soil deposits Potential for subsidence (karst, mining, etc.) Constructability parameters • Shear strength parameters • Unit weights • Time-rate consolidation parameters • Horizontal earth pressure coefficients • Interface friction parameters • Pullout resistance • Geologic mapping including orientation and characteristics of rock discontinuities • Shrink/swell and degradation of soils Field testing Laboratory testing • Nuclear density • Plate load test • Test fill (water • CPT content and pore pressure measurement) • SPT • PMT • Dilatometer • Vane shear • Rock coring (RQD) • Geophysical testing • Piezometers • Settlement plates • Slope inclinometers • One dimensional Oedometer • Triaxial tests • Unconfined compression • Direct shear tests • Grain size distribution • Atterberg limits • Specific gravity • Organic content • Moisture-density relationship • Hydraulic conductivity • Geosynthetic soil testing • Shrink/swell • Slake durability • Unit weight • Relative density The second type of load is the long-term operational loading. This loading occurs after the embankment is constructed to the final grade and excess pore-water pressures have dissipated. The long-term stability of embankment slopes should be analysed especially in fine grained soils. When foundation soils are cohesive and not heavily over-consolidated, potential settlement will be a design consideration. Consolidation settlement and secondary compression can continue for many years and, depending on the thickness of soils, the amount of settlement can be very high. Significant settlement can result in distress to the pavement at the top of the embankment, as well as differential settlement of the pavement at cut and fill transitions and bridge approaches. Another load that may occur in embankments is seismic loading. This load is usually a rare occurrence but may be very important in areas like the rift valley of Ethiopia where earthquakes are much more common. The primary geotechnical concern during the design earthquake is the potential for liquefaction in foundation soils beneath the embankment. Liquefaction could lead to bearing failures, side slope failures and post-liquefaction settlement. The potential for side slope failure without liquefaction is also a design consideration. The duration of loading during a seismic event is usually short; however, pore-water pressures can redistribute after an earthquake leading to liquefaction-related failures that may occur several minutes after the main earthquake event. Page 3-4 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Embankments under 5 m high in areas of stable ground and with slopes not greater than 1.5H:1V generally do not require a detailed geotechnical investigation and analysis. These embankments can be specified particularly when based on past experience in the same region and on engineering judgment. On the same basis, embankments over 5 m high and those constructed over soft ground will usually require a detailed geotechnical analysis. 3.3 Settlement Analysis Settlement is the amount of vertical deformation that occurs when a load is applied or an embankment is placed over compressible soils. The settlement may be due to the consolidation of the embankment itself as well as the underlying soils. The total settlement in an embankment is a summation of three potential components: immediate settlement of the fill or the foundation soil, primary consolidation of the foundation soil, and secondary compression controlled by the composition and structure of the foundation soil skeleton. Settlement caused by lateral deformation of the foundation soil at the edges of an embankment is not considered here. Immediate settlement of the fill and the foundation soil occur during construction and will not usually have any impact on the future pavement. Therefore, the analysis of settlement for embankment design focuses on primary consolidation and secondary compression of the foundation soil. Because primary consolidation and secondary compression can continue long after the embankment is constructed (post construction settlement), they represent the major settlement considerations for embankment design. Usually, post-construction settlement can damage structures located within the embankment, especially if these facilities are also supported by adjacent soils that do not settle appreciably, leading to differential settlements. Differential settlements that occur along the longitudinal axis of the embankment because of changes in consolidation properties of underlying clays can cause transverse cracking on the surface of the embankment. Post-construction settlement adjacent to bridges can also create deformation to the road surface, or down drag and lateral squeezing of the foundation. Generally, settlement in the range of 30 to 60 mm throughout the design life of the road is considered to be tolerable provided that it is uniform, occurs slowly, and does not take place adjacent to a pile supported bridge. One-dimensional consolidation tests are often used for the determination of settlement in the laboratory. The results of these one-dimensional consolidation tests are expressed in an e-log p (void ratio versus the log of pressure) or ε-log p (strain versus the log of pressure) plot, which is called a consolidation curve. Settlement due to consolidation can be estimated from the slope of this curve. This procedure is generally used in practice despite the fact that not all of the soil beneath the embankment undergoes one dimensional consolidation. As stated earlier, for cases where there is uncertainty in the settlement estimate using consolidation tests, it may be desirable to construct and monitor test fills. Information from test-fill monitoring can be used to develop better estimates of soil compressibility and the rate at which settlement will occur. Ethiopian Roads Authority Page 3-5 Chapter 3 Road Embankments 3.3.1 Geotechnical Design Manual – 2013 Primary consolidation The settlement associated with the readjustment of soil particles due to migration of water out of the voids is known as primary consolidation. The amount of primary consolidation depends on the initial void ratio of the soil. The greater the initial void ratio, the more water that can be squeezed out, and the greater the primary consolidation. The rate at which primary consolidation occurs is also dependent on the rate at which the water is squeezed out of the soil voids. The response of foundation soils to additional loads is also dependent upon their stress history. settlement happens when the weight of the embankment exceeds the previous Usually, stress history of the soils. Depending upon the magnitude of the existing effective pressure relative to the maximum past effective stress at a given depth, foundation soils can be classified as normally consolidated, over-consolidated, or under-consolidated. When the existing load is equal to the historical load, the soil is said to be normally consolidated. Pre-consolidation pressure in excess of the current vertical effective stress occurs in over-consolidated soils. Soils are considered to be under-consolidated when consolidation under the existing load is still occurring and will continue to occur until primary consolidation is complete, even if no additional load is applied. Settlement computation starts with the soil profile being divided into layers, with each layer reflecting changes in soils properties. Thick layers with similar properties are also subdivided to improve the analysis since the settlement calculations are based on the stress conditions at the midpoint of the layer (i.e. it is preferable to evaluate a 6 m thick layer as two 3 m thick layers). The total settlement is the sum of the settlement from each of the compressible layers. The settlement of an embankment resting on n layers of normally consolidated soils can be computed from Fig.3-2a and using the following Equation 3-1: S= n Cc H o i o ∑ 1+ e  pf    po  log10  where Cc is the compression index, eo is initial void ratio, Ho is layer thickness, po is initial (current) effective vertical stress and p f = po+ ∆p is final effective vertical stress at the centre of layer n, and p c is the maximum past effective vertical stress. The total settlement is the sum of the compressions in n soil layers. Common correlations for estimating Cc are given in Table 3-2. Page 3-6 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-2: Typical consolidation curve for normally consolidated soil, (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress. From US DOT FHWA (2006B) Sometimes the consolidation data is presented in terms of vertical strain ( εv) instead of void ratio. In this case the slope of the virgin portion of the consolidation curve is called the modified compression index and is denoted as Ccε as shown in Fig.3-2b, where: Ccε= Cc /(1+eo). As mentioned earlier, the total settlement is computed by summing the settlements computed from each subdivided compressible layer within the zone of influence. The assumption is made that the initial and final stress calculated at the centre of each sub-layer is representative of the average stress for the sub-layer, and the material properties are reasonably constant within the sub-layer. The sub-layers are typically 1.5 m to 3 m thick for pavement design applications. Table 3-2: Correlations between Cc & soil index parameters. Modified by US DOT FWHA (2006B) from Holtz & Kovacs (1981) Correlation Soil Cc= 0.156eo + 0.0107 Cc = 0.5Gs(PI/100) All clays Cc= 0.30(eo- 0.27) Cc = 0.0115wn Inorganic, silt, silty clay Organic soils, peat Cc = 0.75(e-o 0.50) Low plasticity clays Where eo is initial void ratio, PI is plasticity index, LL is liquid limit, and wn is water content If the water content of a clay layer below the water table is closer to the plastic limit than the liquid limit, the soil is likely to be over-consolidated. This means that in the past the clay was subjected to a greater stress than now exists. Over-consolidation can be due to the weight of a natural soil deposit that has since been eroded away, the weight of a previously placed fill that is removed, loads due to structures that have been demolished, or due to Ethiopian Roads Authority Page 3-7 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 desiccation. Normally consolidated soils undergo large settlements compared to overconsolidated soils. The curves shown in Figure 3-3 can be used to determine the settlement of an embankment resting on n layers of over-consolidated soils. As a result of pre-consolidation, the field state of stress will reside on the initially flat portion of the e-log p curve. Fig. 3-3a and 3-3b illustrate the case where a load increment, ∆p, is added so that the final stress, p f, is greater than the maximum past effective stress, pc. The settlements for n layers of overconsolidated soils will be computed from the following Equation 3-2 that corresponds to Fig.3-3a, where Cr is the recompression index: S = n Ho  P  P  Cr log10 c + Cc log10 f  P Pc  o o ∑1+ e i In cases where the foundation soils represent both normally and over-consolidated layers, the total settlement is computed by using a combination of the corresponding equations. Figure 3-3: Typical consolidation curve for over-consolidated soil, (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress . From US DOT FHWA (2006B) Under-consolidation is the term used to describe the effective stress state of a soil that has not been fully consolidated under an existing load. Consolidation settlement due to the existing load will continue to occur under that load until primary consolidation is complete, even if no additional load is applied. This condition is indicated in Fig. 3-4a and Fig. 3-4b by ∆po. As a result of under-consolidation, the field state of stress will reside entirely on the virgin portion of the consolidation curve. The settlements for the case of n layers of under-consolidated soils are computed by the following Equation 3-3, which corresponds to Fig. 3-4a: S= n H o Cc  P  P  log10 o + log10 f  P Po  o  c ∑ 1+ e i Page 3-8 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Several methods are available to estimate the stress distribution at any point in the embankment. Perhaps the simplest approach is the 2V:1H method. This empirical approach is based on the assumption that the area the load acts over increases geometrically with depth. Since the same vertical load is spread over a much larger area at depth, the unit stress decreases. Apart from the vertical stress, the parameters that control stress distribution at depth are the dimension of the embankment, the inclination of the embankment side slopes, depth below the ground surface, and horizontal distance from the centre of the load to the point in question. Figure 3-4: Typical consolidation curve for under-consolidated soils (a) void ratio versus vertical effective stress and (b) vertical strain versus vertical effective stress. From US DOT FHWA (2006B) The step-by-step procedures for determining the amount of and time for consolidation to occur for a single-stage construction of an embankment on soft ground is outlined below: 1. From laboratory consolidation test data determine the e-log p curve and estimate the change in void ratio that results from the added weight of the embankment. Create the virgin field consolidation curve by using standard guidelines. 2. Determine if the soil is normally consolidated, over-consolidated or underconsolidated. 3. Use Equations 3-1, 3-2 and 3-3 to compute the primary consolidation settlement for each sub-layer of the foundation soils. 4. The total settlement a soil layer is a sum of all sub-layer settlements The detailed methodology for the estimation of consolidation is provided in many text books. The computations are conducted manually or by using a spreadsheet. There are also many computer programs that can compute settlement. The advantage of computer programs is that multiple runs can be made quickly, and they include subroutines to estimate the increased vertical effective stress caused by the embankment or other loading conditions. Ethiopian Roads Authority Page 3-9 Chapter 3 Road Embankments 3.3.2 Geotechnical Design Manual – 2013 Time for settlement Many projects cannot accept the programming impact associated with waiting for primary consolidation to occur. Therefore, estimating the time of settlement during design is often as important as estimating the magnitude of settlement. The ability to quantify both the magnitude and time of settlement depends on field investigations, the quality of laboratory testing, the size of the embankment, and type and consistency of the foundation soils. As shown in the following Equation 3-4, the time for primary consolidation settlement depends on the coefficient of consolidation, the thickness of the layer and the time after loading: t= Tv H d2 Cv Where t is the time (days) needed, T v is a dimensionless time factor, Hd (m) is the longest distance to a drainage boundary, C v is the coefficient of consolidation measured in m 2/day. As shown in the equation above, the time of settlement is directly proportional to the distance to a drainage surface squared (Hd2); therefore, reducing the drainage thickness (H) by a factor of two results in a four-fold increase in settlement rate. Generally, determination of the average drainage path length is an important component of field exploration. This drainage path or the distance the pore water must flow through a compressible layer depends on the permeability of materials below and above it. If the clay deposit has a significant number of sand inter-layers, the rate of settlement increases significantly because of the reduced drainage path length. Generally, the longest drainage distance of a soilWhen confined by permeable layers on bothonsides equal one-half of the layer thickness. confined by a permeable layer one isside andtoan impermeable boundary on the other, the longest drainage distance is equal to the layer thickness itself. The value of the dimensionless time factor, Tv, for any average degree of consolidation can be taken from tables available in many text books. The average degree of consolidation at any time, t, can be defined as the ratio of the settlement at that time (S t) to the settlement at the end of primary consolidation (S f), when excess pore water pressures are zero throughout the consolidating layer (i.e. U= St/S f). At the end of primary consolidation all excess pore water pressures have dissipated and the average degree of consolidation (U) approaches 100%. The normalized time factor, Tv, is used to compute the settlement time for various percentages of settlement due to primary consolidation in order to develop a predicted settlement-time curve. Normally, experience has shown that the rate of settlement is faster than is estimated from calculations. A part of this faster rate can be attributed to the existence of thin drainage layers. The step by step procedures to determine rate of settlement are as follows: 1. Determine the coefficient of consolidation (C v) from laboratory consolidation test data. The two graphical procedures commonly used for this purpose are the logarithm-of-time (log t) and the square root of time(√) methods. Because both methods are different approximations of a theory, they do not give the same Page 3-10 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments answers. Often the √ method gives slightly greater values than the log t method. The log t method is shown in Figure 3-5. Plot the dial readings for sample deformation for a given load increment o against time on a semi-log paper. Plot two points, P and Q on the upper portion of the consolidation curve o which correspond to time t1 and t2, respectively. Note that t 2 = 4 t1. The difference of the dial readings between P and Q is equal to x. Locate o point R, which is at a distance x above point P. Draw the horizontal line RS. The dial reading corresponding to this line is o d0, which corresponds to 0% consolidation. o Project the straight-line portions of the primary consolidation and the flatter portion towards the end of the consolidation curve to intersect at T. The dial reading corresponding to T is d 100, i.e. 100% primary consolidation. The sample deformation beyond t100 is due to secondary compression. o Determine the point V on the consolidation curve which corresponds to a dial reading of (d0+d100)/2 = d50. The time corresponding to the point V is t50, i.e. 50% consolidation. 2. Determine the average drainage path length (H d) during field exploration. o Calculate Cv from Equation 3.4 for desired U. For example for U=50%, the value of Tv is 0.197 from standard tables. 3. Establish the time to achieve 90% - 95% primary consolidation. An alternative approach to hand calculation is the use of computerized methods. There are various programs which can calculate the time rate of settlement for various boundary conditions including the effects of staged construction and strip drains in addition to calculating the stresses and settlements. Some programs also allow for simulation of multiple layers undergoing simultaneous consolidation. 3.3.3 Secondary compression The end of primary consolidation is considered as the amount of compression that occurs during the period of time required for excess pore-water pressure to dissipate because of an increase in effective stress. Secondary compression occurs when the soil continues to settle after the excess pore water pressures are dissipated to a negligible level, i.e. primary consolidation is essentially completed. It can take years for primary settlement to complete and secondary compression continues for decades. The occurrence of secondary compression is independent of the stress state and theoretically is a function only of the secondary compression index and time. Ethiopian Roads Authority Page 3-11 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-5: Log t method of determining the coefficient of consolidation. From US DOT FHWA (2006B) Secondary compression is normally evident in the settlement-log time plot (Figure 3.5), when the specimen continues to consolidate beyond 100% of primary consolidation, i.e. beyond t 100. Secondary compression continues through the life of the embankment, though the normal assumption is that it decreases according to the logarithm of time as shown below: Sc = C α 1 + eo  t2    t1  H o log10  Where Sc is secondary compression, Cα is the coefficient of secondary compression, Ho is layer thickness, eo is the initial void ratio, t1 is the time when approximately 90% of primary consolidation has occurred for the actual clay layer being considered, and t 2 is the service life of the structure or any other time of interest. The values of C α can be determined from the dial reading vs. log time plots of one-dimensional consolidation test shown in Fig 3-5. For many soils, contribution from secondary settlement is small. However, in soft soils and particularly in soils with a high peat or organic content, it can be large and difficult to predict. Like the primary consolidation, detailed the amount of secondary compression are given in thenumerical literature.procedures Generally,for theestimating contributions from the individual soil layers are summed to estimate the total settlement. Since secondary compression is not a function of the stress state in the soil but rather how the soil breaks down over time, techniques such as surcharging to pre-induce settlement are sometimes only partially effective at mitigating the effect of secondary compression. Often the owner must accept the risks and maintenance costs associated with secondary compression if a cost/benefit analysis indicates that mitigation techniques such as using lightweight fills or over-excavating are too costly. Page 3-12 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 3.4 Road Embankments Settlement Mitigation Once the amount and rate of settlement are determined from laboratory tests and field measurements, the next stage would be to select appropriate techniques that can assist mitigating the short and long-term effects. Mitigation techniques are employed when it is observed that the extent of settlement is beyond the amount that can be tolerated. Often, there is an attempt to reduce potential settlement by compaction of the foundation soil, but this process is expensive and is rarely applied. Instead, the most commonly used methods to mitigate settlement include acceleration using surcharges and wick drains, lightweight fills and removal and replacement. 3.4.1 Preloading and Surcharge Preloading is the process of compressing the subsoil prior to placing the permanent load. This method involves the placement and removal of a fill similar to or greater than the permanent load. The concept of preloading and its effect on reducing settlement is shown in Figure 3-6. Figure 3-6: Concept of pre-loading and its effect on magnitude and time of settlement In preloading, the surcharge is removed after the settlement objectives have been met in order to avoid additional deformation. If the material cannot be moved to another part of the project site for use as site fill or another surcharge, it is often not economical to bring the extra fill to the site only to haul it away again. Also, when fill soils must be handled multiple times (such as with a “rolling” surcharge) for preloading, it is advantageous to use gravel borrow material to reduce workability issues during the rainy season and wet weather conditions. Surcharging is a process which subjects the ground to a higher pressure than that during the service life in order to achieve a higher initial rate of settlement thus reducing long Ethiopian Roads Authority Page 3-13 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 term deformation. Unlike preloading, a large proportion of the fill is left behind after the required settlement is achieved. Figure 3-7 illustrates the idea behind surcharging and the results achieved. In embankment construction, the surcharge is constructed to a predetermined height, usually between 300 mm and 3 m above the final grade elevation. The surcharge is maintained for a predetermined waiting period (typically 3 to 12 months) based on settlement-time calculations. Figure 3-7: Effect of surcharge on magnitude and time of settlement Depending upon the strength of the consolidating layer(s) the surcharge may have to be constructed in stages. The actual dimensions of the surcharge and the waiting period for each stage depend on the strength and drainage properties of the foundation soil as well as the initial height of the proposed embankment. The length of the waiting period can be estimated by using laboratory consolidation test data. The actual settlement occurring during embankment construction is then monitored with geotechnical instrumentation. When the settlement with surcharge equals the settlement srcinally estimated, then surcharging could be removed. In addition to decreasing the time to reach the target settlement, surcharges can be used to reduce the impact of settlement from secondary compression. The intent is to use the surcharge to pre-induce the settlement estimated to occur from primary consolidation and secondary compression due to the embankment load. For example, if the estimated primary consolidation under an embankment is 400 mm and secondary compression is estimated as an additional 200 mm over 25 years, then the surcharge would be designed to achieve 600 mm of settlement or greater under primary consolidation only. Using a surcharge typically may not completely eliminate secondary compression, but it has been successfully used elsewhere to reduce its magnitude. Page 3-14 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments However, for highly organic soils or peats where secondary compression is expected to be high, the success of a surcharge to reduce secondary compression may be quite limited. Other means are needed to address secondary compression in this case, such as removal and replacement. The significant design and construction consideration for using surcharges is the embankment stability. It is known that new fill embankments over soft soils can result in stability problems. Hence, the stability of a surcharged embankment must be checked as part of the embankment design to ensure that an adequate short term safety factor exists (see Section 3.6). The stability of the embankment can be monitored in the field using geotechnical instruments. 3.4.2 Vertical drains Primary consolidation of some highly plastic clays can take many years to be completed. Surcharging alone may not be effective in reducing settlement time sufficiently since the longest distance to a drainage boundary may be significant. In such cases, vertical drains or wick drains can be used to accelerate the settlement, either with or without surcharge treatment. The vertical drains accelerate the settlement rate by reducing the drainage path the water must travel to escape from the soil to half the horizontal distance between them, as illustrated in Figure 3-8. In most applications, a permeable sand blanket, 0.6 to 1 m thick, is placed on the ground to permit free movement of water away from the embankment and to create a working platform for installation of the drains. The drains are installed prior to placement of the embankment. The time for consolidation is proportional to the square of the length of the longest drainage path. Thus if the length of the drainage path is shortened by 50%, the consolidation time is reduced by a factor of four. Vertical drains and sand blankets should have high permeability to allow the water squeezed out of the subsoil to travel relatively quickly through them. Generally, wick drains are small prefabricated drains consisting of a plastic core that is wrapped with geotextile, which functions as a separator and a filter to keep holes in the plastic core from being plugged by the adjacent soil. The drains are usually 100 mm wide and about 6.25 mm thick, produced in rolls that can be fed into a mandrel. They are installed by pushing or vibrating a mandrel into the ground with the wick drain inside. Predrilling of dense soil deposits may be required in some cases to reach the design depth. Since vertical drains are generally expensive, the feasibility of a surcharge solution should always be considered first. Ethiopian Roads Authority Page 3-15 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-8: Use of vertical drains to accelerate settlement. From NCHRP (1989) reproduced in US DOT FHWA (2006B) 3.4.3 Removal and replacement Removal and replacement (over-excavation) refer to excavating soft compressible soils from below the embankment and replacing them with higher quality, less compressible soil (Figure 3-9). Where analyses indicate that more foundation settlement would occur than can be tolerated, partial or complete removal of compressible foundation material may be necessary. However, because of high costs associated with excavating and disposing of unsuitable soils and the difficulty of excavating below the water table, removal and replacement is only justified under certain conditions. Some of these conditions include the following: • • • • • The area requiring over-excavation is not wide; The unsuitable soils are near the ground surface and do not extend very deeply (removal of unsuitable material beyond the depth of 3 m is not normally economically feasible); Temporary dewatering is not required to support or facilitate the excavation; The unsuitable soils can be dumped on site or can be disposed of safely elsewhere close by; Suitable fill materials are readily available to replace the volume of unsuitable soils. Page 3-16 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-9: Removal and replacement beneath an embankment 3.5 Bridge Approach Embankments Usually, excess materials from a roadway excavation or a convenient borrow site are used to construct bridge approach embankments. However, because only a small settlement is desirable at abutments, requiring piles and drilled shafts, it is better to use select materials and increase compaction requirements to prevent differential settlement. Figure 3-10: Elements of a bridge approach embankment. From Briaud et al (1997) At bridge approaches, deformation can occur both in the vertical and lateral directions. Settlements of that vertical deformation. As approach embankments and may foundation soils settle, are the results result is materials surrounding deep foundation systems cause negative skin friction or a down-drag effect on individual piles or drilled shafts (Fig. 3-11). Past studies have indicated that this effect can occur with as little as a 10 mm settlement. If enough down-drag occurs, the axial capacities of piles or shafts may be exceeded. Ethiopian Roads Authority Page 3-17 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-11: Settlement and down-drag in bridge abutments and piles. Modified from US DOT FHWA 2006B Therefore it is important to evaluate potential approach embankment settlements close to the bridge structure and determine if down-drag will occur. Once it is determined that down-drag is possible, an estimate of the negative skin resistance should be performed. The use of a waiting period between the completion of the embankment and the installation of foundations permits settlement to occur prior to driving piles or installing shafts. Downdrag loads may be considered negligible if the settlement following completion of the waiting period is expected to be less than 10mm. In addition, any large settlement near a bridge structure can lead to the formation of differential settlements in the road surface. Often, bridge approach slabs are used to provide a transitional road surface between the pavement on the approach embankment and the actual structure of the bridge. Due to the deformation of the approach embankment fills, these slabs can settle and/or rotate, Depending on the configuration of the approach slab, e.g., how the slab is connected to the abutment and/or the wing walls, voids may develop under the slab as the approach fill settles. Such voids can then fill with water, and water pressure may act against structural elements or soften the soils with associated reduction in strength. Generally the problems in bridge approach embankments are classified as internal deformation within the embankment and external deformation in foundation soils. Internal deformation is a result of compression of fill materials, commonly related to drainage problems. Poor drainage can cause softening of the embankment soils, reduce the stability of soils near the slope, and potentially lead to migration of fill material and creation of voids or substantial vertical and lateral deformations. External deformation is due to the vertical and lateral deformation of the foundation soils on which the embankment is placed. Furthermore, deformation may include both primary consolidation and secondary compression depending on the type of foundation soils. Lateral squeeze of the foundation soils can occur if the soils are soft and if their thickness is less than the width of the end slope of the embankment. Internal deformation within embankments can be controlled by using fill materials that have the ability to resist the anticipated loads imposed on them. A well-constructed soil embankment will not excessively deform internally. A typical approach embankment cross section is shown in Figure 3-12. Special attention should be given to the interface area between the bridge structure and the approach embankment, as this is where the bump at the end of the bridge occurs. Usually, the reasons for the bump are poor compaction of Page 3-18 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments embankment material near the structure, migration of fine soils into drainage material, and, loss of embankment material due to poor drainage. Poor compaction is usually caused by restricted access of standard compaction equipment to that area. Proper compaction can be achieved by optimizing the soil gradation in the interface area to permit compaction to maximum density with minimum effort. Figure 3-12: Suggested details of a bridge approach embankment. Modified from US DOT FHWA 2006B Selected structural backfill is usually placed in relatively small quantities and in relatively confined areas to minimize differential settlements. Table 3-3 lists specification considerations for selected structural back fill to ensure the construction of a durable and dense backfill. In order to drain water from the embankment, under-drain filters can be constructed. The drainage aggregate used for properties under-drain should consist are of crushed sand,3-4. or screened gravel. Suggested forfilters drainage aggregate providedstone, in Table The soundness of the drainage aggregate should also be tested. Generally, the aggregate should not have a loss exceeding 20% by weight after four cycles of the magnesium sulphate soundness test. Ethiopian Roads Authority Page 3-19 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Table 3-3: General considerations for specification of selected structural backfill. From US DOT FHWA 2006B Consideration Lift thickness Largest particle size Gradation and % fines Plasticity index Durability T99 density control Compatibility Comment Limit to 150 mm to 200 mm, so compaction is possible with small equipment. Limit to less than ¾ of lift thickness. Use well graded soil for ease of compaction. Typical gradation: Sieve Size % passing (by weight) 100 mm 100 No. 40 (0.425 mm) 0 to 70 No. 200 (0.075 mm) 0 to 15 The limitation on percent fines (particles smaller than No. 200 sieve) is to prevent piping and allow gravity drainage. For rapid drainage, consideration may be given to limiting the percent fines to 5%. The PI should not exceed 10 to control long-term deformation. The material should be substantially free of shale or other soft, poor-durability particles. A material with a magnesium sulfate soundness loss exceeding 30 should be rejected. Small equipment cannot achieve AASHTO T180 densities. A minimum of 100% of standard Proctor maximum density is required. Particles should not move into voids of adjacent fill or drain material. In areas where selected materials are not available, the use of geosynthetic materials to reinforce the abutment backfill and approach area can reduce the differential settlement at the end of the bridge. Such reinforced fills can be designed using the principles of reinforced soil slopes. Table 3-4: Suggested gradation for drainage aggregate. From United States DOT FHWA (2006B) Sieve Size (mm) 25.4 12.7 6.3 (No 3) 2.00 (No 10) 0.85 (No 20) % passing (by weight) 100 30 – 100 0 – 30 0 – 10 0–5 In addition to overall and differential settlement problems, bridge approach embankments can become saturated as a result of inundation during the occurrence of floods. When the water level falls rapidly, excess pore pressures can develop and lead to instability of embankment slopes. Page 3-20 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 3.6 Road Embankments Stability Assessment In addition to settlement, the design of embankments should also consider side slope stability and bearing capacity. Stability problems in the form of rotational or sliding block failures (Figure 3-13) most often occur when the embankment is built over soft soils such as low strength clays and silts, or when the foundation soil is overstressed during or immediately after construction. Failure can also occur if the fill materials are not compacted to specification. In addition, failures can also occur when the fill materials at the toe of the embankment are eroded. Dynamic forces from earthquakes, blasting or pile driving can also trigger failures of embankment slopes. Usually, short-term stability of embankments on cohesive soil is more critical than long-term stability, because the foundation soil will gain shear strength as the pore pressures dissipate. Figure 3-13: Modes of side slope failures in embankments. From IOWA State (2013) and US DOT FHWA (2006B) Instability of embankments can generally be classified as internal or external. Internal stability generally results from the selection of poor quality embankment materials and/or improper placement of the embankment fills. Shallow, planar slope failures in embankment side slopes are examples of internal instability. Usually such kinds of failures are manifested as sloughing of the surface of the slope. On the other hand, deep rotational and planar failures that involve both the embankment and the foundation soils are considered to be external. In general, embankments that are 5 m or less in height with 1.5H:1V or flatter side slopes, may be designed based on past precedence and engineering judgment provided there are no known problem soil conditions such as organic and soft soils. Embankments over 5 m in height or any embankment on soft soils, in unstable areas, or those comprised of lightweight fill require more in-depth stability analyses, as do any embankments with side Ethiopian Roads Authority Page 3-21 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 slope inclinations steeper than 1.5H:1V. Moreover, any fill placed near or against a bridge abutment or foundation, or that can impact a nearby buried or above-ground structure, will likewise require stability analysis. Prior to commencing a stability analysis, the following key issues need to be addressed: • • • • Is the site underlain by soft silt, clay or peat? If so, a staged stability analysis may be required. Are there geometrical and site constraints which need embankment slopes steeper than 1.5H:1V? If so, a slope stability assessment may be needed to evaluate the various alternatives. Is the embankment temporary or permanent? Factors of safety for temporary fills may be lower than for permanent embankments, depending on the site conditions and the availability of materials. Does the new embankment have an impact on nearby structures or bridge abutments? If so, more elaborate sampling, testing and analysis are required. Are there potentially liquefiable soils at the site? If so, seismic analysis to evaluate this and ground improvement may be needed. Methodologies for analysing the stability of embankment slopes are available in many reference books. A discussion on slope stability is also given in Chapter 4. Generally, experience and observations of failures of embankments constructed over relatively deep deposits of soft soils have shown that when failure occurs, the embankment settles, the adjacent ground rises and the failure surface follows a circular arc of the type shown in Figure 3-14. A slip circle failure in embankments may be a base circle, a toe circle, or a slope circle. A base slip circle develops when there is a significant thickness of weak foundation soil. The base of the failure arc is tangent to the base of the weak layer, and the arc will have a significant portion of its length in the weak soil. A slip circle develops within the embankment and intersects with the slope. A toe slip circle develops in the embankment and intersects at the toe. This happens, sometimes, when the embankment material becomes saturated and failure occurs. In general, at failure, the driving and resisting forces act as follows: • • The force driving movement consists of the embankment weight. The driving moment is the product of the weight of the embankment acting through its centre of gravity times the horizontal distance from the centre of gravity to the centre of rotation (Lw). The resisting force is the total shear strength acting along the slip plane. The resisting moment is the product of the resisting force times the radius of the circle (LS). Page 3-22 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-14: Typical circular arc failure mechanism. From US DOT FHWA (2006B) The factor of safety (FS) against failure is equal to the ratio between the resisting and driving moments as given below. Failure takes place when the factor of safety is less than one.  =    ℎ ℎ ∗  =   ℎ ∗  A rule of thumb based on simplified bearing capacity theory can be used to make a preliminary estimate of the factor of safety (FS) against circular arc failure for an embankment built on a clay foundation without presence of free water. This is as follows:  = 6  ∗  where c is the cohesion of the clay foundation soil, γFill is the unit weight of the fill and HFill is the height (thickness) of the fill. Since the rule of thumb assumes that there is no influence from groundwater in the embankment, c and γFill are effective stress parameters. The equation is helpful only very early in the design stage to make a quick preliminary estimation of whether stability may be a problem and if more detailed analyses should be conducted. It can also be used in the field during investigation. If in-situ vane shear tests are being carried out as part of the field investigation for a proposed embankment, the data from vane strength tests on the underlying soils can be used with the equation to estimate the FS in the field. This estimate can aid in directing the drilling and sampling programme either for design or construction review and help to ensure that critical strata are adequately explored and sampled. A more detailed stability analysis is needed when the FS obtained in this way is less than 2.5 or when groundwater is expected to lie within the slip circle. If the critical stability is under drained conditions, such as in sand or gravel, then effective stress analysis using a peak friction angle should be used for stability assessment. In the case of over-consolidated fine grained soils, a friction angle based on residual strength may be appropriate. This is especially true for soils that exhibit strain softening or are particularly sensitive to shear strain. If the critical stability is under undrained conditions, such as in most clays and silts, a total stress analysis using the undrained cohesion value with no friction is generally needed. Ethiopian Roads Authority Page 3-23 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Once the soil profile, soil strengths, depth of ground water table and other necessary parameters are determined by field explorations and/or laboratory testing, the stability of the embankment can be analysed using standard methods and a factor of safety estimated. The method of analysis that should be selected to determine the factor of safety depends on the soil type, strength characteristics and other parameters. General guidelines are given in Table 3-5. In general, for side slopes of any road embankment, a minimum design safety factor of 1.3 as determined by the ordinary method of slices (Chapter 4) is sufficient to maintain longterm stability. Embankments supporting or potentially impacting non-critical structures should have a minimum safety factor of 1.3. For slopes that would cause greater damage upon failure, such as slopes adjacent to bridge abutments, major retaining structures, and major roadways, the design safety factor should be increased in the range of 1.3 to 1.5. All bridge approach embankments and those supporting critical structures should have a safety factor of 1.5. Critical structures are those in which failure would result in property damage. If an embankment is provided with benches, these should be at least 1.5 m wide and not more than 6m apart vertically. If springs or seepages are found near the embankment, suitable drains should be provided to collect the flow. 3.7 Fill Slope Stabilization A variety of techniques are available to mitigate inadequate slope stability for new embankments, existing embankments or embankment widening projects. These techniques include staged construction to allow the underlying soils to gain strength, base reinforcement, ground improvement, use of lightweight fill, and construction of toe berms and shear keys. Table 3.6 presents a summary of these solutions to mitigate embankment stability problems. 3.7.1 Staged construction Where soft compressible soils are present below a new embankment and when it is not economical to remove and replace these soils with compacted fill, the embankment can be constructed in stages to increase the strength of soils under the weight of the new fill. This means that the rate of filling is governed by the increase in soil strength due to consolidation. Usually the design is carried out using the undrained strength, and vertical drains are used to increase the consolidation process. The stability and degree of consolidation can be related to the gain in strength from the tests and observations of excess pore water pressures. Page 3-24 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Table 3-5: Slope assessment guidelines for the design of embankments and cuts. From US DOT FHWA (2006B) Soil type Type of analysis Short-term Embankments on soft clays: Source of shear strength parameters • • immediate end of construction (cut), φ = 0 analysis. Cohesive Staged construction Embankments on soft clays: build embankment in stages with waiting periods to take advantage of clay strength gain due to consolidation. • • • Long-term Embankment on soft clays and clay cut slopes. • • Granular • • All types UU or field vane shear test or CU triaxial test. Use undrained strength parameters at Po (Initial effective vertical stress) Recommended methods analysis and remarks of Use Bishop method. An angle of internal friction should not be used to represent an increase of shear strength with depth. The clay profile should be and divided into convenient layers the appropriate cohesive shear strength assigned to each layer. CU triaxial test. Some samples should be consolidated to higher than existing in-situ stress to determine clay strength gain due to consolidation under staged fill heights. Use undrained strength parameters at appropriate Po for staged height. Use Bishop method at each stage of embankment height. Consider that clay shear strength will increase with consolidation under each stage. Consolidation test data needed to estimate length of waiting periods between embankment stages. Piezometers and settlement devices should be used to monitor pore water pressure dissipation and consolidation during construction. CU triaxial test with pore water pressure measurements or CD triaxial test. Use effective strength parameters. Use Bishop method with combination of cohesion and angle of internal friction (effective strength parameters from laboratory test). Obtain effective friction angle from charts of standard penetration resistance (SPT) versus friction angle or from direct shear tests. Use Bishop method with an effective stress analysis. UU = unconsolidated undrained, CU consolidated undrained, CD = consolidated drained Methods recommended represent minimum requirement. More rigorous methods such as Spencer’s and Morgenstern-Price method should be used when the design demands. Ethiopian Roads Authority Page 3-25 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Table 3-6: Design techniques useful for mitigating embankment failure. From US DOT FHWA (2006B) Relocate highway alignment. Reduce grade line. (flatten slope) Counterweight berms. Excavation of soft soil and replacement with shear key. Slow rate or staged construction A shift of the highway centre-line to an area with better soils may be the most economical solution. A reduction in grade line will decrease the weight of the embankment and will improve the stability of over-stressed soils. A counterweight berm outside of the centre of rotation provides an additional resisting moment that increases the factor of safety. Berms should be built concurrently with the embankment. The embankment should never be completed prior to berm construction since the critical time for shear failure is at the end of embankment construction. The top surface of a berm should be sloped to drain water away from the embankment. Also, care should be exercised in the selection of materials and compaction specifications to ensure that the design unit weight will be achieved for berm construction. The strength of soft soils is often insufficient to support embankments. In such cases, the soft soils are excavated and replaced with granular material that acts as a shear key. Many weak sub-soils will tend to gain strength during the loading process as consolidation occurs and pore water dissipates. For soils that consolidate relatively fast, such as some silts and silty clays, this method is practical. Proper instrumentation is desirable to monitor the state of stress in the soil during the loading period to ensure that loading does not proceed so rapidly that a shear failure occurs. Typical instrumentation consists of slope inclinometers to monitor stability, piezometers to measure excess pore water pressure and settlement devices to measure the amount and rate of settlement. This option could also be used if weak sub-soils are pre-treated with wick drains. Lightweight fills. In some countries, lightweight materials such as blast furnace slag, shredded rubber tires, or expanded shale are available. Use of such materials decreases the driving force as well as settlement. Base reinforcement Techniques such as stone columns, soil mixing, geosynthetics, soil nailing, ground anchors, and grouting can be used to increase resisting forces. Specialty contractors should be considered for these design solutions. Reinforcement of embankment soils. The embankment soils can be strengthened by incorporating reinforcements within them. The reinforcement permits steeper slopes compared to unreinforced embankments. Ground improvement Densification of the soil through a special compactor and altering the soil composition may help to minimize embankment slope failure. After the first fill placement, construction of the second and subsequent stages commences when the strength of the previous layers is sufficient to maintain stability. Computer programs are used to define the height of fill placed during each stage and the rate at which it is placed, along with the time of settlement and the percent consolidation required for stability. Different approaches are used to assess the rate of fill placement and the necessary strength gain on various types of foundation soils. These approaches differ on the basis of loading conditions such as total stress, effective stress, and rapid draw-down (as applicable). Page 3-26 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments An analysis based on total stress is often useful to simulate conditions that may exist during or shortly after construction. In this approach, the rate of embankment construction is performed using maximum fill heights and intermediate fill construction delays. During these delay periods the fill that was placed is allowed to settle until an adequate amount of consolidation of the foundation soil can occur. Once the desired amount of consolidation has occurred and pore pressure dissipated, placement of the next layer of fill can begin. The thickness of the fill and the intermediate delay periods are estimated during design. For this approach, field measurements such as the rate of settlement or the rate of pore pressure decrease should be obtained to verify that the design assumptions regarding rate of consolidation are correct. In the effective stress approach, the pore pressure increase beneath the embankment in the soft subsoil is monitored and used to control the rate of embankment construction. During construction, the pore pressure increase is not allowed to exceed a critical amount to ensure embankment stability. The critical amount is generally controlled by use of the pore pressure ratio, which is the ratio of pore pressure to total overburden stress. To accomplish the pore pressure measurement, pore pressure transducers are typically located at key locations beneath the embankment to capture the pore pressure increase caused by consolidation stress. Some judgment is needed when interpreting such data and deciding whether or not to reduce or extend the estimated delay period during fill construction, as the estimate of the parameters may vary from the actual values in the field. Also, this approach may not be feasible if the soil contains a high percentage of organic material and trapped gases, causing the pore pressure readings to be too high and not drop off as consolidation occurs. Some roadway embankments may be subjected to water ponding at the base of the slopes during flood events or nearby standing water or lakes. This situation may cause embankment and foundation soils to become saturated. Soils may not drain as quickly as the water recedes and may remain saturated for some period after the water returns to its normal lower elevation. This situation can create a critical embankment stability condition commonly referred to as a "rapid draw-down" condition (Figure 3-15). Stability analyses performed to evaluate this situation should model the embankment as being saturated up to the high water elevation, and should be performed using effective stress parameters for foundation soils and embankment materials. Figure 3-16 shows the typical result of an analysis using the total stress approach. Each time fill is added, the fill starts to consolidate, while the soft subsoil and previous fills have already had time to react to the stress increase due to the fills applied earlier. At the end of the analytical process, a weighted average of the percent consolidation that has occurred for each stage up to the point in time (or the time for full height) should be used to determine the average percent consolidation of the subsoil due to the total weight of the fill. In general, it is best to choose as small a fill height and delay period increment as practical. Typical fill height increments range from 60 cm to 120 cm, and delay period increments range from 10 to 30 days. Ethiopian Roads Authority Page 3-27 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-15: Effect of flooding and rapid-drawdown on embankment stability Figure 3-16: Concept of calculating the percent consolidation in staged construction. From Washington State DOT (2013) 3.7.2 Base reinforcement Base reinforcement may be used to increase the factor of safety against slope failure. Base reinforcement typically consists of placing a geotextile or geogrid at the base of an embankment prior to constructing the embankment. Base reinforcement is particularly effective where soft/weak soils are present below a planned embankment location. The base reinforcement can be designed for either temporary or permanent applications. Most base reinforcement applications are temporary, in that the reinforcement is needed only until the shear strength of the underlying soil has increased sufficiently as a result of consolidation under the weight of the embankment. Therefore, the base reinforcement does not need to meet the same design requirements as permanent base reinforcement regarding creep and durability. Page 3-28 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments The design of base reinforcement is similar to the design of a reinforced slope in that limit equilibrium slope stability methods are used to determine the strength required to obtain the desired safety factor. Base reinforcement materials should be placed in continuous longitudinal strips in the direction of main reinforcement. Joints between pieces of geotextile or geogrid in the strength direction (perpendicular to the slope) should be avoided. All seams in the geotextiles should be sewn and not lapped. Likewise, geogrids should be linked with pins and not simply overlapped. Where base reinforcement is used, the use of gravel borrow areas, instead of earth materials, may also be appropriate in order to increase the embankment shear strength. 3.7.3 Ground im provement Ground improvement can be used to mitigate inadequate slope stability for both new and existing embankments, as well as reduce settlement. Most foundation problems occur from high void ratios, low strength materials and unfavourable water content in the soil. Therefore, basic concepts of soil improvement include densification, cementation, reinforcement, soil modification or replacement, drainage, and other water content controls. In addition to these categories, wick drains may be used in combination with staged embankment construction to accelerate strength gain and improve stability and accelerate long term settlement. The wick drains in effect significantly reduce the drainage path length, thereby accelerating the rate of strength gain. Other ground improvement techniques such as stone columns can also accelerate strength gain in the same way as wick drains. Columns made of stone or chemically stabilized soil increase the stiffness of the foundation and can substantially increase stability and decrease settlement. 3.7.4 Lightweight fills Lightweight fills are other means of improving embankment stability. Situations where they may be required include conditions where the construction schedule does not allow the use of staged construction. 3.7.5 Removal and replacement As in the case of settlement, the very soft compressible cohesive soils are excavated and replaced with better materials (e.g. compacted sand or suitable fill) to provide a stable foundation. If the soft material is much deeper than the practical excavation depth, partial excavation and replacement is also possible. However the effect on stability and long term settlement of the remaining soft material should be considered. Sometimes partial excavation and replacement of soft material is used with ground treatment techniques to overcome the above problems. This method will be more difficult if the groundwater level lies above the base of the excavation. 3.7.6 Toe berms and shear keys Toe berms and shear keys are methods to improve the stability of an embankment by increasing the resistance along potential failure surfaces. Toe berms are typically constructed of granular materials that can be placed quickly, do not require much compaction, but have relatively high shear strength. As implied by the name, toe berms are constructed near the toe of the embankment slopes where stability is a concern (Figure 317a). The side slopes of the toe berms are often gentler than the fill embankment side slopes, but the berm itself should be checked for stability. The use of berms may increase Ethiopian Roads Authority Page 3-29 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 the magnitude of settlements as a consequence of the increased size of the loaded area. Toe berms increase the shearing resistance in the following ways: • • • By adding weight, and thus increasing the shear resistance of granular soils below the toe area of the embankment; By adding high strength materials for additional resistance along potential failure surfaces that pass through the toe berm; By creating a longer failure surface, thus adding more shear resistance, as the failure surface must pass below the toe berm rather than the embankment and the berm. Shear keys function in a manner similar to toe berms, except instead of being adjacent to the toe of the embankment, the shear key is placed under the fill (Figure 3.17b) and frequently below the toe of the embankment. Shear keys are best suited to conditions where they can be embedded into a stronger underlying formation. Shear keys typically range from 1.5 to 4.5 m in width and extend 1.2 to 3.0 m below the ground surface. They are typically backfilled with quarry waste or similar materials that are relatively easy to place below the groundwater, require minimal compaction, but still have high internal shear strength. As with toe berms, shear keys improve the stability of the embankment by forcing the potential failure surface through the strong shear key material or along a much longer path below the shear key. Figure 3-17: Use of a counterweight berm (a) and a shear key (b). From US DOT FHWA (2006B) Page 3-30 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 3.8 Road Embankments Embankments in Hilly Areas The design of embankments in hilly and mountainous regions often takes a different form from those on flat areas. This is because many embankments in these places are constructed by cut and fill operations. The procedure is that soil and rock are cut, if suitable for use as fill, and compacted to form side-long “sliver fills” (Figure 3-18). The use of these predominantly granular materials limits the amount of vertical deformation or settlement of the type observed in embankments on soft soils in flat areas. However, poor compaction techniques used on such fills located on inclined, side-long hill slopes often results embankments that are only marginally stable. Figure 3-18: Typical construction of embankments in hilly areas. From FAO (1998) The main problem of embankments in hilly areas is the overall stability of the fill and the foundation layers. The maximum permissible angle of side-slopes with which it is possible to maintain long-term stability is, therefore, a major consideration during design. An examination of embankment failures along a number of mountain roads shows that many are caused by: • • Inadequate under-drainage under conditions of pronounced seepage; Incomplete removal of vegetation and organic material and lack of benching prior to embankment construction; Ethiopian Roads Authority Page 3-31 Chapter 3 Road Embankments • • • Geotechnical Design Manual – 2013 Construction of embankments on loose spoil material derived from earlier excavations; Erosion on slopes immediately below the embankment; Presence of pre-existing shear surfaces beneath the embankment the presence of unfavourably orientated planes of weakness in the soil or rock beneath the embankment. Figure 3-19 shows the types of slope failures that are commonly observed in fills and the underlying hill slopes. Slope failures in fill slopes constructed in flat areas are often in the form of small-scale shallow translational or rotational slides, where failure is contained entirely within the embankment side slopes and maximum depth of rupture does not exceed 2 m. However, the types of failure often found in embankments in hilly areas can extend beneath the entire slope upon which the embankment is constructed, as shown in Figure 3-19. This is especially true when slope conditions are exacerbated by pore water pressure and seepage during the rainy season. Figure 3-19: Types of slope instability commonly seen in fills and the underlying hillslope. From MPWT (2008) Sometimes when valley slopes are exposed to high stresses exerted by side-cast spoil, they may start to fail. In addition, failure is sometimes associated with the interface between the natural ground and the fill. In addition, most earth embankments can be prone to erosion problems arising from rainfall runoff and road runoff, as well as land-use practices. Excessive erosion can lead to the formation of rills and gullies that ultimately affect the side slope stability. The amount of erosion is normally a function of runoff source, rainfall or runoff intensity, soil type, slope angle, length of slope, degree of compaction, and vegetation cover . Slope failures are common where rivers erode side slopes that eventually undermine the stability of fill slopes constructed on the slope above. Generally, the overall stability of a fill slope on a hillside is difficult to quantify. Unlike relatively homogenous embankments on flat areas, the mechanisms of failure of embankments in hilly terrain are very difficult to analyse using conventional limit Page 3-32 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments equilibrium methods. This is because failure surfaces often involve both the fill and underlying hill slope, each with significantly different material properties. In addition, the cross-section of embankments in hilly terrain is different from that on flat ground as they are usually constructed only on one side of the road. Hence, stability and fill side slope angles are often designed on the basis of local experience, based on the properties of the material forming the fill. Table 3-7 shows some slope stabilization techniques appropriate for fill and underlying hill slopes. In many cases, these options are site specific and are selected and designed based on the characteristics of the fill and slope materials in the project site. Generally, before constructing a fill in hilly areas, it is necessary to assess the stability condition of the slope against shallow and deep failures. Although the weakest layer is often just below the fill and the strength of soils increases with depth, the potential for failure along a deeper surface in the ground beneath should always be considered. This is especially true when the fill is placed on colluvium, or where the stratigraphy is such that weak volcanic deposits (such as tuff and ash) underlie layers of stronger material, or where marl and shale are overlain by limestone and sandstone. Table 3-7: Slope stabilization techniques for embankments on hill slopes. Modified from MPWT (2008) Instability Stabilization options Failure within fill slopes (internal) • • • • • Reduce the slope angle and remove excess material Properly compact fill material Consider the use of a retaining wall Ensure road-side drainage is controlled Bio-engineering is usually important to prevent surface erosion • • • Re-grade the slope and remove excess material Properly compact fill material Before placing fill, prepare benches on the slope to intercept potential planes of failure and to provide a key Consider the use of a retaining wall Ensure road-side drainage is controlled Bio-engineering is usually important to prevent surface erosion Failure in fill slopes and underlying hill slopes (external) Failure within underlying hill slopes • • • • • • • Reduce side slope if sufficient space exists Retaining wall Ensure road-side drainage is controlled Bio-engineering is usually important to prevent surface erosion and increase the resistance of the surface soil. Problems also frequently occur when rock strata beneath the fill are dipping parallel to the 1 ground slope , or where the groundwater table is at or very close to the surface. Adversely 1 Or are dipping out of the slope at an angle that is less than the slope angle but greater than the friction angle along joint or bedding/foliation planes Ethiopian Roads Authority Page 3-33 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 orientated rock planes can cause the slope and the fill above to slide, triggered by increased load, or increased pore pressure along the failure plane. Groundwater from different sources can soften the founding material, or cause the fill material to be undermined through seepage erosion. In these situations, fill slopes require surface and sub-surface drainage structures to keep groundwater away from the area. 3.9 Fill-slope Angles and Benches When deciding fill slope angles for design without stability analysis, it is advisable to consider the type of material (rock or soil, and the strength parameters of each) that is going to be used. For example, fills containing high amount of fines may show surface cracking when dry. The design should also consider the stability of existing fill slopes in the surroundings of project sites. Observation of the existing natural slopes should include the vegetation, in particular the types of plants that may indicate wet soil. Indirect relationships, such as subsurface drainage characteristics, may be indicated by vegetative pattern springs and seepage lines in the area should also be checked. Table 3-8 shows preliminary or provisional fill slope batters based on types of materials. In general, for embankments in hilly terrain, where fills are dominantly granular, side slopes are often designed using a slope ratio of 1.5H:1V as shown in Figure 3-20, assuming that the specifications for particle size, drainage and compaction are met. Larger-sized rock blocks (rip-rap) may be placed on the lower side of the fill to reinforce the embankment and drain surface water. Earth fill may be placed on top of rock fill, separated by a cap layer, if rock is scarce or load is an issue. Moreover, smaller sized rock fragments larger than the average size of the fill itself could be placed underneath the fill to drain water away from the embankment and foundation soil. Table 3-8: Preliminary fill slope angles Fill materials <5m Rock fill 1.5H:1V 5-10 m 10-15 m 2H:1V Well graded sand, gravels, and 1.5H:1V – 2H:1V sand or silt mixed with gravels 2H:1V Poorly graded sand 2H:1V – 3H:1V - Sandy clay soils, silty clay soils and stiff to hard clayey soils 2H:1V 3H:1V - Soft clay soils, plastic clays 3H:1V - - In more gently sloping ground, the side-slope can be relaxed to accommodate weaker material or earth fills. The common practice for side-slopes of earth fills is to design them with 2H:1V or lower angle as shown in Figure 3-21. Depending on the type of earth fill, rip-rap could be placed on the downslope side of the fill for additional slope protection. The size of the rock should be large enough that it withstands the tractive force of runoff water. A 300mm thick under-filter may also be needed to prevent potential piping. Environmentally, it is preferable to avoid high fills so that the road is less conspicuous in the landscape. Page 3-34 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-20: Typical side-slopes of a rock fill embankment Figure 3-21: Typical side-slopes of an earth fill embankment The necessity for benching the existing natural slope beneath an embankment and their width and vertical spacing is decided on the basis of the length and inclination of the slope (often > 15%), material properties, groundwater conditions and other environmental factors. Fill benches are horizontal or near-horizontal steps that are constructed by cutting into the natural slope. In high embankments, berms up to 5m width may be constructed to give access to sideslopes for maintenance or for the control of surface drainage (Figure 3-22). They are also important to prevent erosion provided the surface they collect is properly controlled. Steps or shear keys cut into the hill slope prior to water embankment construction (Figure 3-23) increases the resistance to failure along the natural ground-fill interface. Ethiopian Roads Authority Page 3-35 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-22: Fill e mbankment on a hill-side with outward dipping berms Figure 3-23: Embankment on inward inclined hill-side benches. From Keller and Sherar (2011) Page 3-36 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-24: Benched fill on a benched hill-side slope. From JKR (2010) When berms are utilised, they normally dip outwards (Figure 3-22) by up to 5%. The purpose of these benches is mainly to drain water from the embankment. Benches can also be constructed both on the hill-side and fill slopes as shown in Figure 3-24 and made to incline outward. In this case, a geotextile separator may be used to separate the fill from slope materials. In general, the direction of bench inclination should be decided by the engineer on site depending on the slope of the hill, the type of the fill (earth or rock), and material properties. 3.10 Wall-supported Embankments Sometimes, retaining walls are needed to support fill on hillside slopes. The situations where these structures are most frequently used include when fill needs to be confined either because there is not enough space for its construction or where the hillside slope is too steep to form a stable fill slope. Retaining walls may also be needed at the base of a bridge approach embankment. They may also be essential on road widening projects in urban areas. Retaining walls are generally classified as gravity, semi-gravity, and nongravity cantilevered and anchored. Gravity walls derive their capacity to resist lateral loads through the dead weight of the wall. The gravity wall type includes rigid gravity walls, mechanically stabilized earth (MSE) walls, and prefabricated modular gravity walls. Semigravity walls are similar to gravity walls, except that they rely on their structural components to mobilize the dead weight of an embankment fill to derive their capacity to resist lateral loads. Non-gravity cantilevered walls rely on structural components of the wall partially embedded in foundation materials or hill-side slope to mobilize passive resistance to resist lateral loads. Anchored walls derive their capacity to resist lateral loads by their structural components being restrained by tension elements connected to anchors and additionally by partial embedment of their structural components into existing ground. The anchors may be ground anchors (tiebacks), passive concrete anchors, passive pile anchors, or pile group anchors. Ground anchors are connected directly to the wall structural components whereas the other three types of anchors are connected to the wall structural components through tie rods. Ethiopian Roads Authority Page 3-37 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 Figure 3-25: Typical use of retaining structures in road embankments Selection of appropriate wall type for embankment support is based on an assessment of the design loading, depth to adequate foundation support, presence of deleterious soils or groundwater, physical constraints of the site, cross-sectional geometry of the site both existing and planned, settlement potential, desired aesthetics, constructability, maintenance, and cost. Rigid gravity walls may be constructed of stone masonry, mass concrete, or reinforced concrete and can be used to retain both cut and fill slopes and failed slopes. They have relatively narrow base widths, and are generally not used when deep foundations are required. Figure 3-26 presents a schematic diagram of gravity and semi-gravity retaining structures. They are most economical to support low height embankments of less than 6 m. Due to their rigidity, they should only be used where their foundations can be designed to limit total and differential settlements to acceptable values. Page 3-38 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-26: Gravity and semi-gravity retaining walls Semi-gravity cantilever, counterfort and buttress walls are constructed of reinforced concrete. Semi-gravity walls have relatively narrow base widths, and can be supported by both shallow and deep foundations. The position of the wall stem relative to the footing can be varied to accommodate right-of-way constraints. These walls can support embankments, sign structures, and cut slopes. They can accommodate drainage structures and utilities. Often, they are most economical at wall heights of up to 10 m. The most common terminology associated with the design of semi-gravity structures is given in Figure 3-27. Figure 3-27: Terminology associated with semi-gravity retaining walls Non-gravity cantilevered walls are constructed of vertical structural members, usually consisting of partially embedded soldier piles or continuous sheet piles. Soldier piles may be constructed with driven steel piles, treated timber, precast concrete or steel piles placed in drilled holes and backfilled with concrete or cast-in-place reinforced concrete. Continuous sheet piles may be constructed with driven precast pre-stressed concrete sheet piles or steel sheet piles. These types of walls are suitable to stabilize fills but are most Ethiopian Roads Authority Page 3-39 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 suitable for cut slope support. Because of the narrow base width, they are useful for situations with tight space or right-of-way constraints. 3.11 Mortared Masonry Walls The most commonly used gravity retaining walls for embankment stabilization are mortared masonry and gabion walls. These structures use locally available rock, are labour-intensive to construct and are usually very cost effective. Mortared masonry walls are useful for the support of natural, cut or fill slopes up to 6m depending on whether they are constructed as a single unit or stepped structures. A mortared masonry wall design uses its own weight and base friction to balance the effect of earth pressures. Retaining walls should be designed to withstand lateral earth and water pressures, the effects of surcharge loads, the self-weight of the wall, and in special cases, earthquake loads. Masonry walls are very rigid and cannot tolerate large settlements. They are especially suited to uneven founding levels, and where adequate founding conditions exist, usually rock. The base width to height ratio of mortared masonry walls usually lies between 0.5 - 0.75:1. If the wall foundation is stepped, movement joints should preferably be positioned to reflect the differing wall heights, so that limited differential movements can be accommodated. For high durability, it is important to ensure that the stone is of good quality and is not significantly weathered, that the cement mortar conforms to a strength criterion, and that the wall does not contain uncemented voids. The permeability of the wall is improved by providing weep-holes. These holes are generally provided at 1-2 m o o lateral and vertical intervals. They are inclined forwards at a slope of 2 – 3 , constructed preferably using a 75mm polythene pipe. The first row of weep holes should be as close to the base of the wall as possible. 3.12 Gabion Supported Embankments Gabion walls are constructed from rectangular steel wire mesh baskets that are filled on site with stone or rock to form a gravity retaining structure. They are strengthened at the corners by higher gauge wire and mesh diaphragms that divide them into compartments. The wire should be galvanized, and sometimes PVC coated for greater durability. The baskets usually have a double twisted hexagonal mesh, which allows the gabion wall to deform to an extent without the boxes breaking or significant loss their strength. Gabion is commonly used for walls of up to 6 m high. Because of their inherent flexibility, they are not favoured immediately below sealed roads due to the likelihood of movement of the backfill and subsequent pavement cracking. Where gabion walls are nevertheless used to support sealed roads, additional care should be taken to locate the base of the wall on a good foundation, in order to avoid potential movements. In general, gabion walls have the following advantages: • • • • Gabions can be easily stacked in different ways, with internal or external indentation improving stability They can accommodate some movement without significant loss of strength They allow free drainage through the wall, The cross section can be varied to suit site conditions, Page 3-40 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 • Road Embankments The boxes can take limited tensile forces to resist differential horizontal movement Their disadvantages include: Gabion walls need a large space to fit the wall base. This base width normally occupies about 40 to 60% of the height, and gabions may not be a good solution where space is limited. Their high degree of permeability can result in a loss of fines through the wall. In embankments, this can result in settlement behind the wall and on the surface of the road. Often this problem can be avoided by using a geotextile between the wall and the backfill as shown in Figure 3-28. Figure 3-28: Typical types of gabion walls Gabion walls may be stepped on either the front or back face as shown in Figure 3-28. The design of both types is based on the same principles. Often, it is not recommended to construct vertical faced gabion walls as any movement that occurs during backfilling may give the wall an appearance of bulging. Stepping is ideal where space is not an issue as it increases stability. The maximum recommended step at each course is half of the depth of o the unit. In some cases, the bases of gabion walls can be inclined by up to 6 into the slope to increase stability. In order to keep the back-slope as dry as possible, it is advisable to provide outlet drains from the lowest point of the wall, and ensure that drainage discharge can be visually inspected from these outlets into catch-pits. Ethiopian Roads Authority Page 3-41 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 3.13 Reinforced Embankments When embankments are required to be constructed with steeper than usual slopes where limited right of way and construction constraints exist, then there is a need either to reinforce the slope or construct a retaining wall. Reinforced embankments permit increased height and angle of the embankment side slopes and may also reduce any potential slope failure passing through the foundation. Often, granular soils are relatively strong under compressive stresses. When these soils are reinforced, significant tensile strength can be induced to the fill, resulting in a composite structure which can withstand both compressive and tensile forces. Reinforced slopes are a form of mechanically stabilized earth that incorporate planar reinforcing elements in the constructed embankment with face inclinations of up to 70 degrees. Metallic strips, geosynthetics, and polymer and wire grids are used as reinforcing elements. However, geosynthetics may provide a better solution because of reduced cost and ease of construction. A geosynthetic reinforced wall face can sometimes be near-vertical depending on internal stability. Reinforced embankments have been used in road construction for many decades. They are used in new construction to steepen side slopes and increase embankment heights thereby reducing fill requirements (Figure 3-29). They are also used to replace conventional retaining walls and repair failed slopes. The most prominent use of reinforcement is, however, for widening and reconstruction of existing roads. The use of reinforced steepened slopes to widen roadways improves mass stability, eliminates additional rightof-way, and often speeds construction. Reinforcement also helps to provide lateral resistance at the edges of a compacted fill. This allows compaction equipment to operate more safely near the edge of the slope. Further compaction improvements can be obtained in cohesive soils through the use of geosynthetics with in-plane drainage capabilities (e.g. non-woven geotextiles) that allow for rapid pore pressure dissipation in the compacted fill. In addition, secondary geosynthetic compaction aids placed as intermediate layers between main reinforcements in steepened slopes, as shown in Figure 3-30, may provide improved face stability and reduce the total cost of the use of geosynthetics. Page 3-42 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments Figure 3-29: Application of reinforced slopes in road construction. From New York State DOT (2007) The spacing and anchorage length into the slope of secondary geosynthetics is determined on the basis of the geometry, loading and performance requirements for the design, the degree of stability required, the types of materials in the fill, and the engineering properties of in-situ soils. Often, the spacing is about 0.5 m and the anchorage length is greater than 1.5 m. The spacing increases upward from the ground surface to the top of the embankment. In the case of primary geosynthetic reinforcement, there are a number of different calculations used to compute the spacing and anchorage lengths. In most situations, the result is that the reinforcement spacing is inversely proportional to reinforcement depth. However, this approach often adds construction difficulties. In practice it is common to divide the reinforced mass into zones of constant reinforcement spacing (increasing towards the wall top from zone to zone) (Ortigao and Sayao 2004). The anchorage length should also be greater than the deepest expected failure surface, to reinforce the soil mass properly. When the length is uniform and equal to the base width of the wall, the base width itself should satisfy all the stability conditions. The economic advantages of geosynthetic reinforced soils would normally be the resulting material and right-of-way savings. In the stabilisation of landslides it may also be possible to re-use the slide debris rather than import high quality backfills. Right-of-way savings can be a substantial benefit, especially in hilly, rural areas where acquiring alternative agricultural land could be a problem. Reinforced embankments may also provide an economical alternative to retaining walls. The use of vegetated-faced reinforced soil slopes will blend with the environment and provide an aesthetic advantage over retaining wall Ethiopian Roads Authority Page 3-43 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 structures. Generally, there is a lower risk of long-term stability problems developing in reinforced slopes. Reinforcement may also bring about strength gains in soil layers over the course of time through improved drainage. Figure 3-30: Typical construction of reinforced fills. From NY State DOT (2007) The major factors that influence the selection of a reinforced embankment alternative for any road construction project include geologic and topographic conditions, environmental conditions, size and nature of the structure, aesthetics, durability considerations, performance criteria, availability of materials, foundation condition, experience with a particular system or application and cost. There are several approaches to the design of reinforced slopes. Most use rotational limit equilibrium methods. The reason is that this geometry provides a simple means of directly increasing the resistance to failure with the inclusion of reinforcement. It is also adaptable by most slope stability computer programs and agrees well with experimental results. The overall design requirements for reinforced embankment slopes are similar to those for unreinforced slopes. That means the factor of safety must be adequate for both the shortterm and long-term conditions and for all possible modes of failure. However, there are three possible failure modes (Figure 3-31) for reinforced slopes: internal, external and compound. The internal mode of failure occurs when the slip plane passes through reinforcing elements. In the case of The the external surface passes and underneath the reinforced soil. failure ismode, said tothebefailure compound when the behind failure surface passes behind and through the reinforced soil mass (Figure 3-31). The reinforcement is represented by a concentrated force within the soil mass that intersects the potential failure surface. By adding the failure resistance provided by this force to the resistance of the soil, a factor of safety equal to the rotational stability safety factor is applied to the reinforcement. The tensile capacity of a reinforcement layer is considered as the minimum of its allowable pullout resistance behind the potential failure surface or its long-term allowable design strength. The slope stability factor of safety is Page 3-44 Ethiopian Roads Authority Chapter 3 Geotechnical Design Manual - 2013 Road Embankments taken from the critical surface requiring the maximum amount of reinforcement. Final design is performed by distributing the reinforcement over the height of the slope and evaluating the external stability of the reinforced section. The flow chart presented in Figure 3-32 shows the steps required for design of reinforced soil slopes. Figure 3-31: Failure modes for reinforced soil embankments. From US DOT FHWA (2001) The first step is to establish the geometric, loading and performance requirements for design. The geometric and loading requirements are the embankment slope height, slope angle, external surcharge loads, and traffic barriers. The performance requirements are related to external, compound and internal stabilities. Hence, external (sliding, deep seated, overall stability, local bearing failure or lateral squeeze), compound, and internal failures need a factor of safety greater or equal to 1.3. The safety factor for dynamic loading and the magnitude and time rate of post constriction settlement based on project requirements are also important. In step two, there is a need to determine the engineering properties of in situ soils. These include: the foundation and retained soil (i.e. soil beneath and behind reinforced zone), strength parameters (cu and ϕu, or c′ and ϕ′) for each soil layer, unit weights γwet and γdry, consolidation parameters (Cc, Cr, Cv and ∆p), location of the ground water table and piezometric surfaces, and for slide repair, the identification of the location of previous slip planes and cause of failure. The third step involves the determination of the properties of reinforced fill and, if different, the retained fill, namely, gradation and plasticity index, compaction characteristics (γdry, ±2% of optimum moisture content, and ωopt), compacted lift thickness, shear strength parameters (cu and ϕu, or c’ and ϕ’), and the chemical composition of the soil (pH). The fourth step is concerned with the evaluation of the design parameters for reinforcement. One of these parameters is the allowable geosynthetic strength (Tal). It is defined as the ultimate strength (T ult) divided by the reduction factor (RF) for creep, installation damage and durability. For granular backfill, RF = 7 may be conservatively used for preliminary design and routine, non-critical structures where the minimum test Ethiopian Roads Authority Page 3-45 Chapter 3 Road Embankments Geotechnical Design Manual – 2013 requirements are satisfied. The second parameter is the pullout resistance that is modelled with a factor of safety of 1.5 for granular soils, 2 for cohesive soils, and a minimum anchorage length (Le) of 1 m is also needed. Figure 3-32: Steps for design of reinforced soil slopes. From US DOT FHWA (2001) A preliminary design of reinforced embankments can also be made by the use of charts. The use of these charts can be extended for final design in low fills with a slope height of less than 6 m. Page 3-46 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 4 Roadside Slopes ROADSIDE SLOPES Unstable natural slopes and road cuts often create a considerable problem to road users in Ethiopia during the rainy season. These unstable slopes or landslides typically occur where a natural slope is over-steep or where cut slopes in weathered rock and soil encounter groundwater. Unless stabilized early, failed natural slopes and road cuts may enlarge and can affect large areas, requiring major reinstatement and/or continued maintenance. They can also be a major source of sediment to local river systems. Slope stabilization and landslide control usually utilises a number of methods to mitigate the causes of failure, together with measures to improve the stability of the slope. Techniques commonly used to control the occurrence of landslides include earthworks, bioengineering, and surface drainage. Stabilization methods that need more substantial engineering works can involve anchoring, retaining, piling and deep subsurface drainage. Other options that need to be considered during design include avoidance through alignment selection, realignment or perhaps even tunnelling, depending upon circumstances. In this chapter, roadside slopes are defined as those slopes that are either cut or fill slopes, or adjacent natural slopes, both within and outside of the Right of Way, as illustrated in Figures 4-1 and 4-2, but which can influence the stability of the road. Figure 4-1: Terms used commonly to define a road and associated slopes Ethiopian Roads Authority Page 4-1 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-2: Natural and cut slopes adjacent to a road 4.1. Natural Slopes Many roads in Ethiopia are constructed in mountainous or hilly areas. As a result, slopes that have been stable for many years may fail because of changes in stress conditions or the effects of road construction on drainage. Often, the susceptibility of natural slopes to failure is little understood in advance of construction, even with a complete geotechnical study. Table 4-1 describes different types of natural slopes and associated slope stability problems. Natural slope failures are commonly classified as actively unstable or previously unstable based on their history of failure. Inactive landslides or slope failures are those that have failed in the past but for which there is no evidence of movement currently or in recent times. These inactive slopes may be "dormant", when the causes of failure remain potentially recurrent and movement may occur again, or they may be "stabilized" when the factors causing the movement have been removed or reduced naturally or artificially. A "relict" landslide is an inactive landslide which developed under climatic or slope conditions considered to be different from those that exist at present. Figure 4-3 presents an illustration of these different terms in the case of a toppling failure. Page 4-2 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-1: Slope stability problems associated with natural slopes. Modified from Hearn and Hunt (2011) Facet Ridge top Topographic features Typical stability problems Rounded relief Deeply weathered soils likely Landslides possible Sharp relief Rock outcrop possible Costly and difficult rock excavation Rock falls and slides possible Irregular relief Asymmetric relief Ridge lines generally May be subject to intense erosion Landslides possible Slope > 40º Probably underlain by rock Landslides unlikely Slope = 25° - 35° Colluvium or landslide material Landslides likely Continuous rock slopes with constant dip Joint sets control long-term stability May be problematic for excavation Rock slides possible Valley sides Small farms or crop fields Valley floor Difficult alignment along ridge top Landslides possible Joint controlled slopes govern stability of alignment and cut slope Landslides possible Drainage problems likely May be colluvial in srcin Elongated mid-slope benches Potentially unstable Ancient river terraces or rock benches Stable or easy for road construction Landslides unlikely Vegetated areas Possibly areas of wet ground Landslides possible Slopes at margins of rivers Possibly actively unstable Very difficult for road alignment Forested slopes behind river terraces Possibly old periodically active instability Slope failures on active natural slopes may involve two or more mechanisms, occurring either at different places on the slope, at different depths or at different times due to changes in ground conditions once initial failures have occurred. On these slopes, the engineering implication of a slope failure will vary according to whether it is a first-time failure or reactivated failure. First-time failures have an immediate effect on roads in their path but may not represent a long-term maintenance problem because the failed mass may come to rest at an angle significantly lower than that from which it failed, and remain stable unless disturbed by toe erosion or road construction earthworks. Reactivated slope failures, however, are often the most problematic for road construction and maintenance. Ethiopian Roads Authority Page 4-3 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-3: Illustration of the terms used to describe stages of slope failure. From Cruden and Varnes (1996) Reactivation can be advancing or retrogressive. In an advancing landslide the rupture surface extends in the direction of movement. This is relatively uncommon. In a retrogressive landslide the rupture surface extends in the direction opposite to the movement of the displaced material, i.e. upslope. An example of this is shown in Figure 412 later. Most landslides that expand progressively in area are retrogressive. The materials most likely to exhibit retrogressive failure are often clays and shale with some cohesive strength that allow steep back scarps to form that then become subject to failure. In a widening landslide the rupture surface extends into either or both flanks. When the volume of displaced materials is decreasing downward, a landslide is said to be diminishing. Figure 4-4 illustrates these terms. Page 4-4 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-4: Graphic description of the evolution and extent of slope failure. From Cruden and Varnes (1996) In the case of inactive landslides, the knowledge that old slip surfaces exist may make it easier to understand and predict the behaviour of the slopes during road construction. The shear strength along existing slip surfaces is lower than the peak for the materials involved, and is referred to as residual strength. 4.2. Cut Slopes During road construction in mountainous areas and along valley sides, the alignment of the road and the design will usually require the excavation of slopes to accommodate or widen the roadway. Cut slopes should be designed to be stable for the anticipated groundwater regime, but they frequently become unstable when either the ground conditions exposed during excavation are different from those envisaged during design or when steeper slopes have to be constructed due to space constraints. In such cases, measures should be considered to increase stability. Most natural slopes are in a state of equilibrium under normal climatic conditions. This equilibrium is often disturbed by slope excavation. Other influences (e.g. rainfall, earthquake, material weathering, blasting, external loading and toe erosion) can also trigger slope failure. Hence cut slopes should be designed to have enough resistance against the worst credible combination of conditions that may develop after excavation. Ethiopian Roads Authority Page 4-5 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Cut slope design requires the determination of the angle at which excavated slopes will remain stable for the assumed or observed groundwater conditions that will apply. This involves the examination of soils and rock from surface exposures and studies of the performance of existing natural slopes and cut slopes in the area. The information needed for the design of cut slopes includes the nature and strength of the materials that will be excavated, the groundwater conditions, the inclinations of rock strata, the degree of weathering, and the extent of joints or any other potential weaknesses. The methods to use in the design of an individual cut slope will vary according to the conditions at the site. In many cases, slopes often contain a mix of materials and are heterogeneous, making it virtually impossible to determine the average or effective shear strength of the slope forming materials by laboratory tests. In such cases, stability analyses will help very little in determining the safe angle of cut, and the cut slope design must be accompanied by the application of judgment and experience. Often, it is difficult to adopt variable cut slope angles at a location where the soil type changes within short distances laterally or vertically. In these situations, the safe angle should be determined from an overall evaluation of the predominant soils encountered. In the event of uncertainty, it is necessary to select a conservative angle of inclination. In general, for low volume roads, the practice is that cut slopes need to be as steep as possible, and at a steeper angle than fills since they constitute relatively undisturbed in-situ soil or rock. Steep cuts, accompanied by the relevant stabilization and protection measures, are especially necessary when there are Right-of-Way or property line constraints. In areas where there is a high water table condition or erosion and sliding is apparent, it becomes more difficult to design safe cut slopes to steeper angles without the use of other measures, such as expensive retaining structures. The steepness or inclination of cut slopes is normally expressed using a slope ratio, percent or degree. In Ethiopia, the slope ratio is the ratio between the horizontal distance and the elevational change written in the form of H:V. The choice of cross-section dictates the geometric requirements for cut slope design, and options include box cut, full cut, and a balance or partial balance of cut and fill (Figure 45). The selection of cross-section in any situation is dependent on topographical and geological conditions, geometric requirements, and cost. A box cut is required where the outside of the road formation is unable to daylight on the adjacent slope, as a result of topography and vertical alignment. It is normally used where the ground must be cut through to avoid an overly steep road grade, such as on a steep hill crest. The material excavated from a box cut, as with all cuts, should as a rule be transported and used elsewhere along the road rather than side-cast as spoil. Figure 4-6 shows an example of a box cut in Ethiopia. Page 4-6 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-5: Types of cross-section design. Modified from Keller and Sherar (2011) Most roads are constructed partially in cut and partially in fill, other than those constructed in flat terrain. This type of design is normally called a ‘cut and fill’ approach whereby materials cut from a hill slope are used to build an adjacent fill for the purpose of supporting part of the road. If the amount of material from excavations roughly matches the amount of material needed to make fill slopes and embankments, then the process is called a ‘balanced cut and fill’ operation, assuming the excavated material is suitable for use as fill. Theoretically, balancing is achieved along the centreline of the road. In practice this is rarely the case as ground elevation with respect to the centreline can change abruptly and balanced cut and fill is normally used to describe the situation whereby the balance can be achieved within a short distance along the alignment. During cut and fill construction, it is important to avoid side-casting of waste material and especially to prevent it from entering streams and other watercourses. All cut material used as fill should be compacted to specification and not side-cast. For full cuts, the excavated material is either used elsewhere along the alignment as fill or is properly disposed of in designated safe and environmentally secure disposal sites. Full cut road construction is typically found in terrain with side slopes of 40 degrees and above. Ethiopian Roads Authority Page 4-7 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-6: Example of a box cut Full cut cross-sections usually result in deep cut slopes. Weak rocks and rock masses with adverse rock structure, including soft or highly fractured/closely jointed sedimentary rocks may not be suitable for full-cut designs. Cut slopes in these materials can result in significant slope instability. Similarly, silty clays and other low strength soils may also be unsuitable for forming deep cut slopes. In such situations fill slopes and retaining walls may need to be considered or the alignment modified accordingly. An example of a full cut cross-section is shown in Figure 4-7. Often cut slopes are benched, i.e. excavated in steps in order to enhance drainage control and intercept falling debris. Benched cut slopes are usually deployed in rock and weathered rock and should not be used in weak soils, such as colluvium. As a rule, excavation should be performed from the upper part of the slope to the lower part to maintain stability. Figure 4-8 shows a typical example of bench construction. The height and width of benches depends on the characteristics of the slope materials, and is usually specified in design and earthworks manuals. Water control is key in benched cut slopes and drainage from above must be transported laterally using ditches at the toe of each cut section and discharged into adjacent streams or via chutes and cascades into culverts. Page 4-8 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-7: Full cut cross-section Figure 4-8: Schematic profile of cut slope benches 4.3. Landslides A landslide is the movement of rock, debris or earth down a slope. It results from the failure of materials which make up a natural slope or cut or fill slope and is driven by the force of gravity. Landslides, both on natural slopes and in earthworks slopes, can be classified on the basis of the type of movement and material involved. The material in a landslide mass is either a rock or soil (or both). The type of movement describes the internal mechanics of the landslide mass during displacement. Types of movement categories include "fall", "topple", "slide", "spread", "flow" and "creep". Landslides are often described using two Ethiopian Roads Authority Page 4-9 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 terms that refer to material and movement type (for example: rock fall, debris flow, etc). Landslides may also form a complex failure mechanism containing both rock and soil and encompassing more than one type of movement. In terms of quantity, landslides may range from a single boulder in a rock fall or topple to tens of millions of cubic metres of material in a large, deep-seated landslide or debris flow. They can also vary in their extent, with some occurring very locally and impacting a very small area, road section or hill slope while others affect much larger slopes and drainage channels beneath them. The distance travelled by landslide material can also range significantly from a few centimetres to many kilometres depending on the volume, water content and the gradient of the slope. Moreover, the movement of land slide material can vary from abrupt collapses to slow gradual slides and at rates which range from the almost undetectable to the extremely rapid. Sudden and rapid events are the most dangerous because of a lack of warning and the speed at which material can travel as well as the force of its resulting impact. Extremely slow moving landslides might move only millimetres a year and the movement can be active over many years. Although these types of landslides are not usually a threat to people they can cause damage to roads and other infrastructure. Landslides can also be divided into shallow and deep-seated based on the depth of the slip plane. Slides with a sliding depth of less than 3-5 m are considered to be shallow. Often, these types of landslides involve the soil mantle Deep-seated landslides are those slides in which the slide plane is more likely to be within weathered rock. These are usually deeper than 5m. Further discussion on shallow and deep-seated landslides is presented in the Site Investigation Manual of the Ethiopian Road Authority (ERA). In Ethiopia, landslides often occur after periods of prolonged and heavy rainfall. Often, this is most likely to happen during the wet season between July and the end of September. In the southern part of the country, this period may extend into October and November as severe and localized rainfall can occur during this time. 4.3.1. Types of landslides Table 4-2 summarises the different types of landslides. While this classification system is useful as a general guide, classifying landslides in practice is often very difficult. For example, a landslide which is described as a flow often initiates through sliding along discrete shear surfaces. Hence, since both the type and rate of movement can change with motion, there is often a need to survey the area and understand the history of ground movement before finalizing its classification. Methods to investigate a landslide are described in the ERA Site Investigation Manual. A brief description of the different types of landslides is given below. Page 4-10 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-2: Simple classification of landslide types. Modified from Cruden and Varnes (1996) Type of material Type of movement Rock Fall Rock fall Topple Rock topple Soil Debris Earth Debris fall --- Planar Rock slide Debris slide Earth (mud) slide Rotational Rock slump Debris slump Soil slump Spread Rock spread --- Soil spread Flow Rock flow Debris flow Earth (mud) flow Creep Rock creep --- Soil creep Slide A fall is a form of movement in which the mass in motion travels most of the distance through the air after detachment. Often, it includes movement by bouncing. A fall starts with the detachment of material from a steep slope along a surface in which little or no shear displacement takes place. Toppling occurs as a result of forces that cause an over-turning moment about a point below the centre of gravity of the mass. A topple is very similar to a fall in many aspects, but does not involve a complete separation at the base of the failure A spread is characterized by the lateral extension of a more rigid mass over an underlying ground surface or through a deforming underlying material in which the controlling basal shear surface is often not well-defined. A slide is a slope movement by which the material is displaced more or less coherently along well-defined shear surfaces, either single or multiple. Slides can be rotational (the slip surface is curved), translational (the slip surface is more or less planar) or wedge (the slip surface is formed by two or more intersecting surfaces). Rotational slides are relatively infrequent and occur in cohesive soils or very weak rock masses with little or no structural control. Translational slides occur in rock masses with discontinuity planes that dip unfavourably or in granular soils. These failures are the most common. A flow is characterized by internal deformation in which the individual particles travel separately within the mass, analogous to the movements in liquids. Flows can be slowmoving, if in fine-grained soils, or rapid to very rapid in granular soils and failed rock masses. Often, they start as a slide on steep slopes. Creep is a long-term movement of non-increasing velocity without well-defined sliding surfaces. In some cases creep failures may represent a deep-seated (or viscous) flow. 4.3.2. Causes of landslides The susceptibility of a slope to the occurrence of landslides is often related to an increase in shear stress (driving force) or a reduction in shear strength (resisting force). The processes that increase the shear stress or decrease the shear strength in a slope are normally termed as landslide preparatory or triggering factors. Preparatory Ethiopian Roads Authority Page 4-11 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 (preconditioning) factors make the slope increasingly susceptible to failure without actually initiating movement. Triggering factors are events that finally initiate movement. Thus geological and topographical parameters are usually regarded as landslide preparatory factors. Triggering factors include rainfall, earthquakes, toe erosion and manmade activities. The causes of landslides on natural slopes can be divided into external and internal. External causes are responsible for an increase in shear stress through an increase in the height and steepness of the slope, for example due to road cuts or river scour and down cutting, and any structural load increases, such as those caused by fill slope construction, spoil dumping and earthquake loading. Internal causes are those which occur without any change in the external conditions of the slope. They are associated with a loss of the shear strength of materials. An increase in pore pressure or a reduction of cohesion due to weathering are considered as internal causes of landslides. Table 4-3 lists some of the more common internal and external causes of landslides. Table 4-3: Common landslide causal factors. Modified from Nettleton et al (2005) Internal causes Materials: • Soils subject to strength loss on contact with water or as a result of stress relief (strain softening). • Fine-grained soils which are subject to strength loss due to weathering. • Weak rocks with adversely orientated discontinuities characterized by low shear strength such as bedding planes and joints. Weathering: • Physical and chemical weathering of soils causing loss of strength (apparent cohesion and friction). Pore-water pressure: • High pore-water pressures causing a reduction in effective shear strength. External causes Removal of slope support: • Undercutting by water (waves and stream incision). • Washing out of soil. • Man-made excavations. Increased loading: • Natural accumulations of water, snow, talus. • Man-made pressures (e.g. fill and buildings). Transient effects: • Earthquakes and tremors. • Shocks and vibrations (blasting). Often, a number of factors contribute to the occurrence of a landslide both on natural slopes and cut slopes. However, it is usual to find that only one factor triggers the actual movement. This triggering factor may be natural or artificial. Table 4-4 shows a number of common natural and artificial preparatory and triggering factors. Page 4-12 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-4: Natural and artificial causes of landslides Natural causes Ground conditions: • Plastic and otherwise weak material • Sensitive and collapsible material • Weathered and sheared material • Jointed or fissured material orientated structural • Adversely discontinuities (including joints faults, flexural shears, lithological contacts) • Contrast in permeability and its effects on groundwater and pore-pressures • Contrast in stiffness (stiff, dense material over less dense and weaker material) • Geomorphological processes: • Tectonic and volcanic uplift • Fluvial erosion of the slope toe • Subsoil erosion (solution, piping) • Natural loading of the slope by accumulation of material from above • Undercutting of cliffs and river banks Physical processes: • Intense, short period rainfall • Prolonged high precipitation • Rapid drawdown following floods • Increase in pore water pressure • Earthquake loading • Freeze-thaw, i.e. mechanical weathering • Shrink and swell of expansive soils. Artificial causes • Removal of vegetation • Interference with, or changes to, natural drainage • Modification of slopes during the construction of roads, railways, buildings, etc • Overloading slopes • Mining and quarrying activities • Vibrations from heavy traffic, blasting, etc • Drawdown (of reservoirs) • Irrigation • Defective maintenance of drainage systems • Side casting of uncompacted spoil Table 4-5 presents another way of dividing landslide causes based on geology, topography, climate and hydrology. Geological causes may be jointed rocks exposed on rock slopes and in cliffs (Figure 4-9), or colluvium overlying rock head surfaces. Weathering may for example form a mantle of weak soils overlying harder rocks which provide an interface or potential shear surface along which landslides can initiate. Topographical causal factors may include steep hillsides which can promote gravity induced slope failures and concave slope facets which can hold moisture for long periods of time, thus leading to lower effective stress and reduced strength. Ethiopian Roads Authority Page 4-13 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-5: Geological, topographical and hydrological causal factors Geological causes Topographical causes • Rock type, weathering grade, jointing and • Steep slopes fracture patterns • Un-vegetated steep slopes, susceptible to erosion; • Presence of faults or shear zones • The direction and angle of dip and joints in • Irregular depressions or undrained underlying bedrock compared to the angle swampy areas on slopes. and orientation to the slope, particularly if Hydrological causes bedrock is exposed or at a shallow depth • Periods of prolonged and/or intense beneath the surface, rainfall that could lead to saturation of • The persistence of joints and clay filling the slope • The sequence of the underlying strata, • Anomalously high rainfall particularly if this includes weak or • The presence of a river or stream at the impermeable layers base of the slope, particularly if this presence of colluvium and • The could cause toe erosion during periods unconsolidated materials. of flood or high flow Climatic causes • The presence of a drainage course at or above the crest of the slope • Periods of prolonged and/or intense rainfall that could lead to saturation of the • Any indications of a high or slope temporarily perched water table within the slope, e.g. seepages and springs. • Anomalous high rainfall. Some landslides are triggered by a rise in groundwater level during or soon after heavy rain. The effect of groundwater on the initiation of landslides can occur in several ways. The may increase on a pore slope pressures thereby increasing total or stress or driving force.water Alternatively, it the mayweight increase within athe slope decrease the coefficient of friction on a potential sliding surface. This leads to a decrease in shear strength. Groundwater can also cause clays to hydrate and expand and make them susceptible to failure. A common observation in many areas of Ethiopia is that a road cut or a natural slope may be perfectly stable during the dry season but may fail after the rains begin. This seasonal change in stability is due mainly to the rapid increase in the level of groundwater within shallow perched aquifers. Often, a perched water table is created in unconsolidated soils that lie above an underlying impermeable rock stratum (Figure 4-10). Perched water of this nature is common in the highlands of Ethiopia where closely jointed basaltic rock masses are present in association with weak volcano-clastic rocks. Another contribution of groundwater to the occurrence of landslides is through the development of a seepage force. A seepage force is a drag force that moving water exerts on each soil particle in its path, contributing to the driving force. The concept of the seepage force may be visualized by noting how easily portions of a coarse-textured soil may be dislodged from a road cut when the soil is transmitting a relatively high volume of groundwater. Page 4-14 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-9: Adversely jointed rock mass that can fail if joints become undercut by excavation Figure 4-10: Perched water table and the formation of landslides on road cuts 4.3.3. Roadside landslides Landslides can occur either above or below the road, or through the road in the case of deep-seated failures. Landslides which occur below the road can involve fill slopes as well as the underlying natural slope material. These failures are discussed in Chapter 3. Landslides above the road can occur in the cut slope or can involve movement of the natural slope above. These types of failures are discussed below. Figure 4-11illustrates landslides commonly observed above a road. These include failures that affect the cut slope only, those which involve the cut slope and natural slope together, and those that are confined to the natural slope. Ethiopian Roads Authority Page 4-15 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-11: Landslides above a road. From Hearn and Hunt (2011) An example of a retrogressive landslide developed on a natural slope is given in Figure 412. Figure 4-12: Retrogressive landslide developed on a natural slope Most cut slope failures are shallow. Sometimes, however, even small, shallow failures can develop into major failures and result in frequent road closures. Figure 4-13 shows a landslide developed in a deep cut slope formed in colluvium. This landslide initiated in the weaker soils of the upper portion of the cut slope and travelled as a flow to block the road. Failures that involve the cut slope and the natural slope above are often shallow but can be retrogressive, ultimately affecting large areas of slope. While most develop in the cut slope and propagate upslope (retrogression) some may initiate in the natural hill slope and spread downwards (advancing). The reduction of support in the cut and the percolation of water from above through tension cracks are often responsible for failures that encompass the cut Page 4-16 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes and natural slopes. Knowledge of the presence of slip surfaces in natural slopes makes it easier to understand and predict the performance of cut slopes within previously failed material. The materials most likely to exhibit retrogressive failures that encompass both natural and cut slopes are fissured clays where positive pore pressures are allowed to develop within the discontinuities. Figure 4-14 shows a landslide which affected both the cut and the natural slope at the above. Figure 4-13: Landslide on a cut slope Figure 4-14: Landslide affecting both the cut and natural slope above Fill slope failure commonly occurs either within the fill itself or at the fill/natural ground interface. This is mainly due to inadequate compaction of the fill, leading to saturation by surface infiltration and shallow failure, or due to the development of a perched Ethiopian Roads Authority Page 4-17 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 groundwater table resulting in a preferential slip plane at the fill/natural ground interface. In the latter case, benching of the natural ground prior to filling, and the prior provision of drainage where seepages are seen to occur at the interface, should prevent such failures from occurring. Figure 4-15 depicts a fill slope failure that has extended to the road centreline. Figure 4-15: Failure of a fill slope Deep-seated landslides are common where elevational differences are large, such as in deep gorges, such as the Blue Nile, where the stratigraphical, hydrological, and topographical conditions favour their development. 4.3.4. Roadbed landslides Occasionally, landslides may occur across the width of a road as shown in Figure 4-16, with or without affecting the cut and natural slopes above. These types of failure are most frequent in areas where the entire road formation is constructed on colluvium and weathered rock, or on fill. Usually, a major triggering factor in the initiation and acceleration of these movements is toe erosion, reactivation of much larger movements of which the one affecting the road is only a part, and the presence of a water source above the road that is hydraulically connected with a spring below the road. Page 4-18 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-16: Example of a failure affecting the entire road This type of ground movement also occurs where runoff percolates through road layers and soaks the sub-grade because of a drainage problem or the absence of sufficient crossdrains. In many cases, irrigation or agricultural activity in the surrounding area can aggravate the problem through improper and uncontrolled drainage. Stream erosion below the road can also worsen the situation. Features that indicate the occurrence of landslides within and above a road are summarized in Table 4-6. Ethiopian Roads Authority Page 4-19 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-6: Indications of slope instability at or above a road Things to look for Heave in pavement. Pavement elevation rising Implication Course of action May indicate unstable slope above the road and circular failure Monitor and record locations and changes. Stabilise through Avoid cutting earthworks, retaining walls and/or drainage Surface cracking May indicate unstable slope in surrounding above the road area May indicate additional unstable area Springs Bulging or compression ridges in slope Tilted features, such as trees, poles, walls, fences Monitor and record locations Avoid disturbing until and changes. stability Properly drain evaluated May indicate unstable slope above the road May contribute water and loss of stability Avoid Monitor and record locations disturbing until and changes stability Properly drain evaluated May indicate unstable slope above the road. May indicate additional unstable area Monitor and record locations and changes. Properly drain May indicate unstable slope above the road May indicate additional unstable area Monitor and record locations and changes. Properly drain May indicate unstable slope above the road Blocked drainage May contribute water and loss of stability May cause distress Loose debris on slope. Things to avoid May indicate unstable slope above the road May indicate additional unstable area Monitor and record locations and changes Unblock drain Avoid Monitor and record locations disturbing until and changes stability evaluated 4.3.5. Landslide stabilisation Once a landslide is identified on a natural, cut or fill slope, the remedial measures must provide a reduction in the driving forces, an increase in the resisting forces, or both. The study and identification of landslides is normally carried out using the methods discussed in the ERA Manual. or Table 4-7 the contains a list of the most common methods that Site can Investigation be used to stabilize remedy effects of landslides. While one remedial measure may be sufficient to minimize the effect of a landslide, most remedial works usually involve a combination of two or more methods. The selection of appropriate remedial measures depends on engineering feasibility, economic viability, and environmental acceptability. Detailed discussions on some of the methods listed in Table 4-7 are given in Sections 4.5 and 4.6. Page 4-20 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-7: Common landslide remedial measures. From Sassa and Canuti (2008) Modification of slope geometry: Retaining structures: • Removing material from the area adjacent • Gravity retaining walls to the crest of the slope(with possible • Crib walls replacement by lightweight fill) • Gabion walls • Adding material to the area at the base of • Passive piles, piers and caissons the slope(counterweight berm or fill) • Cast-in situ reinforced concrete walls • Reducing the overall slope angle. • Reinforced earth retaining structures • Retention nets for rock slopes Drainage: • Surface drains to divert water from • Shear key. flowing into the slide area Internal slope reinforcement: • Shallow or deep trench drains filled with • Rock bolts free-draining materials (coarse granular • Micropiles fills and geosynthetics) • Soil nailing • Buttress counterforts of coarse-grained • Anchors materials • Grouting • Vertical (large diameter) wells with • Stone columns gravity draining • Freezing • Sub-horizontal drains • Bioengineering. • Drainage tunnels, galleries or adits • Dewatering by pumping • Drainage by siphoning. 4.4. Soil slope cuts Soil slopes contain a variety of materials that may range from loose colluvial deposits to cohesive residual clays. The stability of these slopes in a road cut is usually a function of groundwater and drainage regime, the size and composition of the constituting materials, the processes of consolidation that these materials have undergone throughout their depositional history in the case of sediments, the inherent properties and structure of the parent rocks in the case of residual soils, and any secondary precipitates or cementing materials. Normally, soil slopes that can stand for some time after excavation are those which contain appreciable amounts of clay with associated negative pore pressures. Depending upon the rate of reduction in factor of safety, landslides may occur suddenly as the slope is being excavated, or after the slope has been standing for some time. In cohesive soils, wetting and drying (swelling and shrinkage) can result in the development of shrinkage cracks that allow water ingress, thus exacerbating the situation. In granular soils, saturated slopes may be susceptible to debris slides and flow slides. Shallow failures and ravelling of surface materials are a common feature of Ethiopian roads. Generally, the stability of cut slopes consisting of granular soils depends on the slope angle, the angle of internal friction, the unit weight of soil, and pore pressures. When failure is confined to large landslides, the safe angles of excavation can be determined using simple slope analyses. The shear strength parameters for these stability analyses are determined from laboratory tests and can be corroborated from empirical correlations. In areas where long lengths of road are affected by shallow instability, the determination of cut slope angles based on experience is a more practical approach. Ethiopian Roads Authority Page 4-21 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 In cohesive soil slopes, stability is a function of consolidation history, soil density, slope angle, and pore pressures. Figure 4-17 shows the general variations of shear strength, pore pressure, load and factor of safety (stability) with time for a clay cut slope. The initial shear strength is often equal to the undrained shear strength on the assumption that no drainage occurs during excavation. In contrast to embankment slopes, the pore pressure within a clay cut slope increases over time. This increase is accompanied by swelling, which results in reduced shear strength. Thus, the factor of safety decreases over time until an unstable condition is reached. This, for the most part, explains why clayey cut slopes sometimes fail a long time after initial excavation. In situ shear strength is a direct function of the maximum past overburden pressure for cuts in over-consolidated clays. The higher this pressure is, the greater the shear strength. However, over the long term, its strength no longer depends on the prior loading and will decrease with time. The loss in strength is attributed to reduction of negative pore pressure after excavation, and its magnitude depends on the rate of dissipation of this pore pressure. Long-term cut slope stability is also dependent on seepage forces and, therefore, on the ultimate groundwater level in the slope. The presence of groundwater within or just below a proposed cut will affect the slope angle required to achieve and maintain stability. After excavation, the groundwater surface will usually drop slowly to a stable zone at a variable depth below the new cut surface. This drawdown usually occurs rapidly in cut slopes made up of granular materials but is usually much slower in clay. Generally, groundwater adds to seepage forces, raises pore pressures, and causes a corresponding reduction in effective stress and shear strength. It also adds weight to the soil mass, increasing driving forces. Hence it is important to identify seepage within proposed cut slopes so that adequate drainage designs are prepared in advance. Figure 4-17: Stability condition of a clay cut slope over time From Bishop and Bjerrum (1960) reproduced in Abramson et al (2002) Page 4-22 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes 4.4.1. Design considerations Cuts slopes in soils up to a height of about 3 to5 m are generally designed based on past experience with similar materials in the surrounding area. Observations and experience in the performance of cut slopes in the region provides knowledge of stable cut slope angles and common stability issues affecting large cuttings in soil. Cut slopes greater than 5 m in height may require geotechnical analysis, depending upon the risk they pose. Any cut slope where failure would result in large rehabilitation costs or threaten public safety should normally be designed using rigorous techniques. In addition, deep cuts andwhere cuts with geometry and varying stratigraphy (especially layers), cuts high irregular groundwater or seepage forces are likely, cuts involvingwith soilsweak with questionable strength, and cuts in old landslides or in areas susceptible to landslides should also be analysed using standard geotechnical approaches. The major parameters in relation to design of cut slopes are the slope angle and height of the cut and its constituent materials and water condition. For dry granular soils in which the anticipated mode of potential failure is planar, stability is largely independent of height, and therefore slope angle becomes the parameter of most interest. For purely cohesive soils, the height of the cut becomes a critical design parameter. For soils with both cohesion and friction, and saturated soils, slope stability is dependent on both slope angle and the height of the cut. Also critical to the proper design of cut slopes is the groundwater level (pore pressure) and its fluctuation. In principle, the stability of cut slopes in clay soils during design should be evaluated for both the end-of-construction and the long-term conditions. For major roads and highways, the long-term condition is usually the more critical. As shown in Figure 4-17, the stability of a cut slope in clay soil decreases with time after excavation as pore pressures increase and the soils within the slope swell and become weaker. As a result, the critical condition for stability of cut slopes is normally the long-term condition, when increase in pore water pressure and swelling and weakening of soils is complete. If the materials in which the cut is made are so highly permeable that these changes occur completely as excavation and construction proceeds, as will mainly be the case for Ethiopian conditions, the end-ofconstruction and the long-term conditions are the same. The short-term condition is computed using total stress methods (undrained condition). If the cut is designed to be permanent, the long-term condition should be analysed using effective stress methods assuming that excess pore pressures dissipate completely during that time. The strength parameters φ = 0 and cu (undrained strength) are used for total stress analysis assuming that the soil behaviour is exclusively cohesive. For effective stress analyses, drained strength (c' and φ'), along with pore pressure, u, will be required. Cuts along mountain roads in Ethiopia are usually in rock, granular soil or in situ weathered residual soil. In many cases it would be acceptable to assume c' = 0. In principle, it is advisable to analyse stability using effective stress methods since the strengths of soils are governed by effective stresses under both undrained and drained conditions. In practice, however, it is virtually impossible to determine accurately what excess pore pressures will result from excavation. Because of this, performing accurate analyses of stability using effective stress procedures is often difficult. Ethiopian Roads Authority Page 4-23 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 4.4.2. Cut slope angles, profiles and benches Table 4-8 shows cut slope angles (H:V) recommended for preliminary design purposes. A schematic illustration of these slope ratios is shown in Figure 4-18. These values are fairly close to the long-term equilibrium conditions, and can be especially important in places where the heterogeneity of the soil and rock layers makes any kind of stability analysis very difficult. They can normally be used for relatively dry slopes. If near surface groundwater or seepage water exists or is expected to occur during the design life of the slope, then the slope angles should be flatter than those given in this table, or appropriate drainage measures should be taken. When two different soil layers are present, the cut slope angle should be varied, where possible, to take advantage of stronger materials. Figure 4-18: Typical cut slope ratios in most soil types Generally, the use of the cut slope angles shown in Table 4-8 or other customized values should target a design factor of safety of 1.3. This factor of safety is considered as a minimum value for static roadside slope stability. The reason for using a factor of safety greater than one is to take into account the variation of soil properties in slopes and the uncertainty over longer term groundwater conditions. If temporary road closures and debris clearance can be tolerated, as would be the case for low trafficked roads, then steeper slopes with a factor of safety of 1.1 may be used, even for long-term cases. The advantages of steep slopes is that they result in less Right-of-Way width, less excavated material, less spoil disposal, and a shorter length of slope exposed to erosion. Their disadvantages are that they are difficult to vegetate and are more prone to failure. Page 4-24 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-8 Soil cut slope ratios (H:V) for preliminary design purposes Type of Material Cutslope total height (m) Remark 3–6 6 - 10 10 - 15 Residual clay soils 1:1 1:1 2:1 Consider benching when the slope height is above 6 m. Vegetation cover is highly recommended. Heavy, plastic clay soils 1.5:1 2:1 ----- Keeping the slope dry is extremely important. 1.5:1 2:1 ----- Keep a constant slope. Appropriate drainage and vegetation is necessary. 0.75:1 1.5:1 2:1 Reduce the upper portion to 1:1 to limit gully formation or widening. 1:1 1.5:1 2:1 Cover the slope with grass and other suitable plants and keep the slope dry. Granular soils with some clays Dense transported soils (subangular cobbles, gravels and sands in a fine matrix) Loose to medium dense transported soils (boulders, sub-angular cobbles and gravels in a fine matrix), or talus Note: Slope angles are indicative and require site-specific assessment. Roadside cut slopes are generally not designed for seismic conditions unless slope failure could impact major structures. This issue and whether higher factors of safety are needed, should be decided on a case by case basis depending on the consequences of failure, past experience with similar soils, and uncertainties in the analyses. Cut slope profiles can be single-sloped, multi-sloped or benched (Figure 4-19). Singlesloped profiles are usually cut in dense soils with enough resistance against failure. Their height is often limited to 6 m. Multi-sloped profiles are cut where an excavation encounters soil overlying rock or where the stratigraphy consists of two or more soil or rock layers with different strength characteristics. Benched slopes are designed when there is a need to slow down and intercept surface runoff, or to contain falling debris from one bench to another, as would be the case for an excavation in jointed rock. There is no fixed rule regarding the appropriate dimension of a bench. Good practice might be to provide a bench width equal to one third of the height of the cut immediately above it. Ethiopian Roads Authority Page 4-25 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-19: Types of cut-slope profiles For benches to perform properly, in terms of drainage and collecting falling debris, they should be designed with an inward slope of 3-5% and have a longitudinal gradient. Care should be taken to avoid undercutting planes of weakness or over-steepening weak soil or rock layers. The benched slope should be designed accordingly, or a uniform slope selected instead. Benches should be well maintained and cleaned regularly to ensure that they will not create a problem by blocking drainage. 4.4.3. Colluvium and talus slopes Colluvium is a term used to describe a fine-grained slope deposit. It has become widely misused to include heterogeneous transported soils of mixed srcin and mixed sizes to the extent that any transported soil on a slope is now acceptably referred to as colluvium. Talus (scree) on the other hand is a term commonly used to describe broken rock fragments derived from rock cliffs and deposited at their angle of repose in a cone or an apron below the exposed face. Both colluvium and talus deposits are characterized by high permeability and compressibility, and potential for renewed movement. The composition of colluvial deposits varies according to the nature of the bedrock sources and the climate under which weathering and transport occurred. Generally, they consist of a heterogeneous mixture of soils and rock fragments ranging in size from clay particles to rock boulders. Colluvium and talus deposits are often only marginally stable, and can be rendered unstable by heavy rainfall, temporarily perched water tables, or other external factors. Among the possible causes of movement is the reduction in shear strength at the interface between the colluvium and the residual soil/bedrock interface due to the increase in pore pressures. Reactivation is sudden and can occur anytime during heavy rain or if support is removed in excavations. Most failures in colluvium are planar, although slumps and flows can also be encountered. Retrogression is also common in these soils. Page 4-26 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-20: Road cut on a colluvial slope The shear strength of colluvium is generally governed by the matrix of granular and finegrained fractions. On most mountain hillsides, colluvium is treated as a cohesionless material. Reliable representation of pore pressure distribution is critical in the stability analysis of colluvial slopes. The pore pressure distribution can be represented by the pore pressure ratio (ru), which is the ratio of the water pressure to the weight per unit area of the overburden at any point within the soil mass. More frequently, the pore pressure distribution may best be represented by a groundwater or perched water table. Often, good indication of the strength of colluvium is obtained by pit excavation and Standard Penetration Tests, although the results can be misleading if large boulders are present. Colluvial deposits frequently occur in the highlands of Ethiopia. They are especially common on the hill sides of deep gorges such as those of the Blue Nile basin. Figure 4-20 shows an example of a road cut on a colluvial slope along the Gundewein to Mekhane Selam road in the Blue Nile Gorge. Road cuts in colluvium and talus deposits are expected to be stable if they have a slope of 33o (1.5:1) for a cut height of less than 10m as shown in Table 4-8. If a relatively thin colluvial cover overlies bedrock or a residual soil layer on inclined ground, it is advisable to remove it during cutting if it is practical to do so. 4.4.4. Residual soils Residual formed by rocks,onand usually always) contain asoils highare proportion ofin-situ clays. weathering They oftenofoccur flat ridge (though tops andnotstructural benches where in situ weathering has predominated over erosion and transportation. Unlike colluvium, these soils have not been disturbed and can exhibit an additional cohesion associated with the residual rock structure, which increases their strength. Figure 4-21 illustrates the formation of residual soils and the associated effect of erosion and consolidation. Ethiopian Roads Authority Page 4-27 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-21: Simplified illustration of the formation of residual soils Depending on rock type and climate, residual soils (Weathering Grade IV and V) may present special problems with respect to slope stability and erosion. Such soils may contain relict structures from the srcinal bedrock that act as planes of weakness and their strength properties may vary significantly over short distances due to variations in weathering grade. Weathering Grade V materials are often highly prone to erosion. Because of this, it may be difficult to determine design shear strength parameters from laboratory tests. Groundwater also tends to be complex within in situ weathered soils, again due to relict structural variations in the soil mass and variations in weathering grade and rock head level. In this case, shear strength parameters should be determined by back-analyzing slope failures and by using empirical design procedures based on local experience. Generally, slopes formed in in situ weathered soils should be designed using both total and effective stress approaches to assess short-term strength and also long-term stability. Because of the cohesive nature of the fine grained component of Weathering Grade V-VI materials and the maintenance of soil suctions when slopes are kept largely dry, these soils can often be cut steeply for appreciable heights. Figure 4-22 shows a 1:1 road cut slope through residual red clays. Page 4-28 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-22: Relatively steep 1:1 cut slope in residual clay In Ethiopia, residual soils are found in many parts of the country usually associated with basaltic rocks that can weather easily. Pore pressures in residual soil slopes often react quickly to heavy rainfall and any infiltration can eliminate soil tensions and increase positive pore pressure by raising either perched water or groundwater tables. On slopes, failure can occur either during or sometime after the rainfall ends. Along roads, any percolation from the top of cuttings can lead to softening and slope failure, as illustrated in Figure 4-23. Figure 4-23: Failure in black cotton soil due to infiltration of water 4.4.5. Collapsible soils Collapsible soils undergo a reduction in volume upon wetting. They have usually a low dry density and low moisture content. Such soils have considerable strength and stiffness in their dry natural state and can withstand a large applied vertical stress with a small compression. When they are wet, however, they experience much larger settlements, even with no increase in vertical pressure. Collapsible soils are present in the central and southern part of the Ethiopian Rift Valley in the form of silty loess-type deposits. These soils contain a large amount of void space, and particles are held together by the clay component. They can stand at steep slopes (0.5:1) Ethiopian Roads Authority Page 4-29 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 indefinitely provided the moisture content remains low. However, upon wetting, the soil loses its strength and because of the open internal structure, may undergo large rapid deformations that can result in slope failures. Slope failures in collapsible soils can occur as either shallow planar slides or minor slumps. Surface erosion and subsurface piping are most common problems on slope cuts containing collapsible, in situ weathered and other sensitive soils. All cut slopes should be designed with adequate drainage to limit erosion and piping as much as possible. The amount of erosion that occurs on a slope is a function of soil type, rainfall intensity, slope angle, length of slope, and vegetative cover. For cuts in collapsible soils, consideration should be given to preventing earthwork activity during the wet season. 4.4.6. Latosols Latosols (often referred to incorrectly as laterites) form a group comprising a wide variety of red, brown, and yellow, fine grained in situ weathered tropical soils (Weathering Grade IV and VI) as well as nodular gravels (concretions) and cemented soils (ferricretes and bauxites). They vary significantly in density and texture, depending upon weathering history, from medium dense soils to rock-like material, known as laterite. They are characterized by the presence of iron and aluminium oxides or hydroxides responsible for the red colour of the soil. When some of these soils (plinthites) are exposed to the air, irreversible hardening occurs, producing a laterite material suitable for use as a building or road stone. Often, strength under field conditions tends to be much higher than is indicated by the laboratory tests. The stiff nature of some in situ weathered tropical soils allows steep cuts to be formed to a height of 3 to 6 m. These slopes can remain stable for long periods of time, as long as they remain dry and protected from surface erosion and piping through drainage control. Figure 4-24: Cut through weathered rock and residual soils In Ethiopia, lateritic soils are distributed in the north western, western, and southern part of the country, including especially areas around Assossa and many places in western Gojam. Page 4-30 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes 4.5. Rock Cut Slopes Roads in mountainous areas often require rock excavation. Unlike soil slopes, rock cut slope stability is influenced by a number of factors such as the type of rock, the strength of intact rock blocks, the characteristics of discontinuities, the degree and extent of the rock mass weathering, and the groundwater conditions. Intact rock strength is the strength of a block of rock or specimen that is not affected by discontinuities. It depends on the physical properties of the constituent minerals and the bond and cementation that exist. It is often difficult to reproduce field conditions in the laboratory. Hence, in situ tests such as point load or Schmidt Hammer rebound tests may be needed to estimate intact rock strengths and correlate them with unconfined compression values. Intact rock made up of minerals susceptible to weathering is usually weak. Others composed of minerals such as quartz are relatively strong. In most rock masses (except where discontinuities are widely spaced) intact rock strength usually exceeds the strength of discontinuities, and discontinuity orientation and friction angle usually control rock slope stability. 4.5.1. The role of discontinuities The term discontinuity is used for all structural breaks in geological materials with negligible or zero tensile strength. They include bedding planes, joints, faults, foliations and shear zones. Table 4-10 summarizes the characteristics of these discontinuities. Discontinuities present in a rock mass are the most significant factors affecting rock slope stability and the primary controlling parameters for rock cut slope design. The influence and shear strength of discontinuities in a rock mass are controlled by their orientation, persistence, spacing, aperture, infillings, and roughness. Orientation is measured in the field using dip and dip direction (or strike) as shown in Figure 4-25. The orientation of discontinuities with respect to applied loads can be critical to deformation or stability. If the orientations of the discontinuity surfaces are inclined out of the slope (daylight), they can have a significantly negative influence on the stability of slopes. Seepage pressures are also controlled by discontinuity orientation and connectivity. Persistence refers to the lateral extent of a discontinuity as is shown in Figure 4-25. It is an important parameter because it defines the potential volume of the failure mass. Recording trace lengths to describe joint persistence is useful in determining the stability and ultimate cutting angles of deep excavations. Ethiopian Roads Authority Page 4-31 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-9: Types of discontinuities and their characteristics. From GWP Consultants (2008) Discontinuity Bedding planes Descriptions Arising from the deposition of sediments in layers, they are distinct physical discontinuities. They may occur at the interface between different rock types at various spacings within a single rock unit. They may be persistent and generally extend over greater areas than any other type of discontinuity. In some rock types, movements along bedding planes may have developed weakened shear zones. The nature and inclination of bedding is always of Joints prime importance when considering slope stability in sedimentary rocks. Joints are developed to some degree in almost all rocks. They are planar fractures formed to relieve stresses, across which there has been little or no movement. Jointing plays some part in the majority of slope failures in rock masses since intact rock is generally stronger than the discontinuities. Faults Faults occur less frequently than joints and may have undergone substantial displacements. Faulting often produces continuous or persistent planes of weakness. Fault zones may develop in which the fault occurs as a series of displacement surfaces in an area of distorted, crushed and often weathered material (termed ‘gouge’). Faulting can occur in any rock. Faults can provide the shearing or release surfaces for several modes of failure. Foliation Foliation is a structural property exhibited only in metamorphic rock types. Slate, crystalline metamorphic rock and tightly folded sedimentary rocks show closely spaced laminations which are not directly related to bedding features. Discontinuities associated with cleavage are likely to be smooth and continuous. Within the rocks affected, they are likely to be a major factor controlling slope stability. Shear zones A structural break where differential movement has occurred along a surface or zone of geological failure; characterized by polished surfaces, striations, slicken sides, gouge, breccia, mylonite, or any combination of these. Shear zones involve volumes of rock deformed by shear stress under brittle-ductile or ductile conditions. They often occur at the edges of tectonic blocks, forming joints that mark distinct terrains. Spacing is the distance between two discontinuities of the same set measured normal to the discontinuity surface. Spacing affects block size and geometry in the rock mass. The spacing of adjacent joints largely controls the size of individual blocks of intact rock. It also controls the mode of failure. Spacing is a required input to several rock mass classification systems. When a set can be distinguished from parallel or sub-parallel joints, the true or apparent spacing can be measured as illustrated in Figure 4-25. Page 4-32 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-25: Description of different discontinuity parameters Aperture is the openness or separation within a given discontinuity and is measured normal to the discontinuity surfaces. Aperture affects the strength, deformability, and seepage characteristics. Materials that fill the aperture of discontinuities are termed as infillings. The presence of infillings affects the permeability and shear strength of a discontinuity. Secondary minerals such as quartz and calcite may provide significant cohesion to the rock mass if present as infillings. Descriptions of infillings are site specific but must address the thickness of fillings, their type or composition, degree of alteration, and hardness. The shape and roughness of a discontinuity constitute its surface characteristics. Roughness is a measure of the surface unevenness and waviness of the discontinuity relative to its mean plane. It can be a small scale surface irregularity or unevenness, or a large scale undulation or waviness. It can also be described as stepped, undulating, and planar in metre scale, or rough, smooth, and slickensided in centimetre scale. Roughness affects the shear strength of the discontinuity directly and is an important parameter for stability analyses. Discontinuity strength is rarely the same in all directions, since even small steps or undulations, known as asperities, strongly influence the potential for sliding along that surface. This effect is most pronounced when the asperities are oriented perpendicularly to the direction of sliding. The shear strength and any associated slope failure are also controlled by the relative positioning of these asperities. Hence, while movement through two fully interlocked joints is very unlikely, failure can occur where waves and undulations are in point contact. The degree of fracturing of a rock mass is described by fracture density, fracture frequency and rock quality designation (RQD). Fracture density is based on the spacing between all natural fractures in an exposure or in cores from boreholes. Fracture frequency on the other Ethiopian Roads Authority Page 4-33 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 hand is defined as the number of discontinuities per metre length. Common descriptions of fracture density are summarized in Table 4.11, but it should be noted that core recovery is also dependent upon the strength of the rock mass and the method of drilling. Table 4-10: Fracture density. Modified from US DOI Bureau of Reclamation (1998) Fractured density Description Unfractured No observed fractures, or core recovered mostly in lengths greater than 1 m. Slightly fractured Moderately fractured Core recovered mostly in lengths from 300 mm to 1 m with few scattered lengths less than 300 mm or greater than 1 m. Core recovered mostly in lengths from 100 to 300 mm with most lengths about 200 mm. Highly fractured Lengths average from 30 to 100 mm with fragmented intervals. Core recovered mostly in lengths less than 100 mm. Extremely fractured Core recovered mostly as chips and fragments with a few scattered short core lengths. RQD is a measure of the degree rock fracturing. It is defined as the percentage of rock cores that have a length equal or greater than 100 mm over the total drill length.  = ∑ ℎ ≥ 100  × 100%  If cores are broken by handling or the drilling process, the broken pieces are fitted together and counted as one piece (provided that they form the requisite length of 100 mm); highly weathered, soft, fractured, sheared, and jointed rocks yield lower RQD values. The length of cores is measured along the centreline of the core. Figure 4-26 illustrates the system of determining RQD from rock cores and outcrops. Page 4-34 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-26: Graphical illustration of the rock quality designation (RQD) 4.5.2. The role of weathering Rock weathering takes place by mechanical and chemical processes and involves the disintegration and decomposition of a rock mass. The degree of weathering is governed by the interaction between the rock mass and different internal and external factors such as geology and climate. Weathering in rock masses often starts along discontinuity surfaces. Hence, as it advances deep into rock masses through joints and bedding planes, it affects the physical and mechanical properties of rock masses. Each weathering grade boundary marks sharp or gradational changes in geotechnical properties. Weathering effects generally decrease with depth, although zones of differential weathering can occur and may modify the weathering profile. Examples include differential weathering within a single rock unit, apparently due to relatively higher permeability along discontinuities, differential weathering due to compositional or textural differences, differential of contact zones associated with thermal effects such as interflow zones within weathering volcanic rocks, directional weathering along permeable joints, faults, and bedding planes, along which weathering penetrate more deeply into the rock mass, and differential weathering because of topographic effects. Weathering is important in rock slope design because it may be the primary criterion for determining cut slope angle, depth, method and ease of excavation, and use of excavated materials. Porosity, absorption, compressibility, shear and compressive strengths, density, and resistance to erosion are the major engineering parameters influenced by weathering. Weathering generally is indicated by changes in the colour and texture of the body of the Ethiopian Roads Authority Page 4-35 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 rock, colour, and condition of fracture fillings and surfaces, grain boundary conditions, and physical properties such as hardness. The ERA Site Investigation Manual describes weathering classification in the field. 4.5.3. The role of groundwater Groundwater occupying the discontinuities within a rock mass can significantly reduce the stability of a rock slope. Water pressure within a discontinuity reduces the effective normal stress acting on the plane, thus reducing the shear strength. Water pressure within discontinuities that run roughly parallel to a slope also increases the driving forces. Further, thethus presence water in bothorintact rockinfilling and discontinuities can promote weathering reducingofshear strengths, can erode materials. Where a rock mass has many discontinuity sets, which are closely spaced, the groundwater may behave much as it would in a soil; there is a high degree of connection of voids and groundwater levels vary only gradually over large areas. However, in rock masses with only a few sets of discontinuities and where the discontinuity spacing is large, water pressures can vary appreciably from one fracture to the next. Seemingly erratic groundwater levels can also develop when geological dykes and sills, high angle faults or steeply dipping strata act as barriers. Water levels may be significantly different on either side of these barriers. Figure 4-27 illustrates the effect of type, orientation and persistence of discontinuities on groundwater levels. Surface recharge can have a marked effect on groundwater levels in rock slopes. A low porosity closed jointed rock mass may experience a rapid rise in groundwater level of several metres as a result of only a few millimetres of rain since infiltrating water is concentrated in a small number of fractures with low aperture. Where groundwater intercepts the slope face, the flow can erode material below the seepage level, which can contribute to slope failure. Page 4-36 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-27: Effect of discontinuity characteristics on groundwater level Determination of groundwater levels and pressures includes measurements of the elevation of the groundwater surface and variations of this elevation based on seasonal fluctuation. Also important is the location of perched water tables, the location of aquifers, and the presence of artesian pressures. Water pressure in rock slopes reduces the stability of the slope by reducing the available shear strength of potential failure surfaces. Changes in moisture content of the rock, particularly those rock types with low slake durability, causes materials to lose strength over time. Thermal expansion or freezing of groundwater causes ice wedging and may effectively block drainage of discontinuities in the rock mass, increasing pore pressures and resulting in an increased rock fall potential and the potential for more large scale failures. Determination of the permeability of the rock strata is important because the discharge of water from slopes along a highway can necessitate the requirement for increased maintenance as the result of pavement deterioration and the need for higher capacity drainage seepage. systems. Figure 4.28 shows an example of a rock slide initiated by groundwater Ethiopian Roads Authority Page 4-37 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-28: Rock slide initiated by groundwater seepage out of the slope 4.5.4. Mode of failures The mode of rock slope failure is primarily controlled by the orientation and spacing of discontinuities within the rock mass as well as the orientation of the excavation and the angle of inclination of the slope. The modes of failure common in rock cuts can be divided into planar (translational), wedge, toppling, and circular. Planar failure is governed by a single discontinuity surface dipping out of the slope face (Figure A single block potentialis for sliding along a single represents the simplest4-29). of planar sliding. Thewith mechanism kinematically possible in plane cases where at least one joint set strikes approximately parallel to the slope strike and dips toward the excavation slope. Failure will occur if the joint plane intersects the slope and dips at an angle greater than the angle of internal friction (ᵠ) of the joint surface. Page 4-38 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-29: Modes of failure in rock slopes. Modified from GWP Consultants (2008) Planar failure along stepped planes is possible in cases where a series of closely spaced parallel joints strike approximately parallel to the excavation slope and dip toward the excavation The parallel not in be failure. continuous. However, least one joint plane slope. must intersect the joints slope may planeortomay result In the case of at continuous parallel joints, a second set of joints is necessary to act as release joints. These release joints must also strike more or less parallel to the slope and the magnitude and direction of the joint dip angle must be such that the joint plane does not intersect the slope plane. Wedge failure involves a failure mass defined by two discontinuities with a line of intersection that is inclined out of the slope (Figure 4-29). In addition, this mode of failure requires that the dip angle of at least one joint-intersect is greater than the friction angle of the joint surfaces and that the line of joint intersection intersects the plane of the slope. Ethiopian Roads Authority Page 4-39 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Multiple wedges can be formed by the intersection of four or more sets of discontinuities. Although conceptually possible, the sliding failure of a multiple wedge system rarely occurs because of the potential for kinematic constraint. Toppling failure involves slabs or columns of rock defined by discontinuities that dip steeply into the slope face. It involves overturning or rotation of rock layers. Closely spaced, steeply dipping discontinuity sets that dip away from the slope surface are necessary prerequisites for toppling. In the absence of cross jointing, each layer tends to bend down-slope under its own weight thus generating flexural cracks. If frequent cross joints are present, the layers can topple as rigid columns. In either case, toppling is usually initiated by layer separation with movement in the direction of the excavation. Layer separation may be rapid or gradual. Rapid separation is associated with block weight and/or stress relief while gradual separation is linked with environmental processes such as thermal expansion and freeze/thaw cycles. An example of toppling failure in a layered and vertically jointed limestone is given in Figure 4.30. Circular failures are commonly associated with soil slopes, but they may also occur in highly weathered and decomposed rock masses, highly fractured rock masses, or in weak rock such as marls and shale. 4.5.5. Design considerations Successful rock slope design demands a sound and complete understanding of the combined influence of intact rock strengths, characteristics of discontinuities and groundwater conditions. In practice, the design process is a balance between stability and other considerations, such as constructability, economics, potential environmental impacts, and the accepted level of risk. Steeper cut slope angles result in lower construction costs, require less Right-of-Way, require a smaller volume of rock excavation, and because the cut height is generally less and the excavation footprint smaller, may be perceived as a more environmentally sensitive design. This needs to be contrasted, however, with the increased risk of slope failure. Page 4-40 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-30: Example of toppling failure Usually, rock cuts made steeper than their discontinuities may require expensive measures to maintain stability. Alternatively, provision would need to be made for rock fall containment, although in steep terrain there is a limit to the area that can be made available for this purpose and a compromise may be required. Superimposing a slightly steeper cut onto an already steep natural slope may present problems associated with difficult access to begin the excavation. Furthermore, a ‘sliver cut’ may be unsafe to excavate because the narrow work area provides inadequate room for construction equipment. As in the case of soil slopes, a rock slope is assessed to be stable or unstable on the basis of the factor of safety. The factor of safety is primarily dependent on the geometry of the potential slip plane (mode of failure) with respect to the cut slope orientation and profile and the shear strength along this plane. In addition, other factors such as the potential for rock fall and aspects of drainage, land take (Right of Way), ease of construction and maintenance should also be considered. Table 4-12 gives a summary of the factors that need to be considered during design. Rock falls from cut slopes are generally more dominant in the early years after excavation when the rock slope has been freshly cut. Weathering, relaxation and rainfall can lead to rock slope failures exposing fresh rock and inducing further rock falls. Over a period of time the rock fall quantity should ordinarily reduce. The natural establishment of vegetation also tends to slow any rock fall after a period of time. Ethiopian Roads Authority Page 4-41 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-11: Aspects to be considered during rock cut slope design. Stability The overall cut slope profile (slope angle and possible use of benches) should be determined to ensure stability, taking into account the stratigraphy and characteristics of discontinuities in the rock mass, groundwater condition, and weathering. Major discontinuities such as adversely dipping persistent joints, bedding planes and sheet joints, weak weathered seams, and fault zones should be given special attention in the design of rock cuts. The practice of relying on generalised assumptions about strength of discontinuities and groundwater may not adequately cater for local zones in the rock mass. In areas where failure can lead to major damage to properties and structures or significant road blockage, the rock face should be mapped in detail by experienced engineering geologists. Benches should be provided between batters of significantly different gradients. Benches are generally not necessary in massive hard rock from a stability point of view. Potential rock fall Where benches are provided, possible bouncing of rock fall should be assessed. Benches if provided should be as wide as possible (preferably at least 2 min fresh rock) in order to contain small rock falls and debris and allow access for maintenance. For large rock slopes, the potential risk of rock fall can be difficult to assess. If practicable and economically feasible, a catch ditch or fence should be provided along the toe of the cut to contain falling rocks. Unless it is considered that there is no potential for surface erosion, drainage ditches should be provided along benches. Bench drainage Drainage ditches would reduce the of runoff on the constructed slope, with consequent reduction of velocity erosion and andvolume infiltration. Benches without drainage provision may encourage infiltration of surface water, leading to reduced stability Land take (Right of Way) A rock slope without benches can minimize land take. If stability is assured, this should be considered. The option of providing a catch ditch or a rock fall barrier at road level may not necessarily take up significantly more space than the provision of benches on the slope. This should be considered as an alternative. Construction The construction process (e.g. blasting) often weakens the rock mass, particularly at bench locations if they are considered in the design. Controlled techniques (e.g. hand scaling, pre-splitting, trimming) should be employed to minimize damage. Vegetation The slope gradient and the topsoil should support vegetation growth. Aspects of maintenance inspections and works, including provision of safe access should be considered and agreed during design. Maintenance Benches can facilitate access is forsuch maintenance. The maintenance of benches that they should be regularly cleaned, especially during rainy seasons. Generally, the design of rock cut slopes is a progressive process. It starts with adequate exploration of the project site within which each rock slope along the road is divided into design units. The classification of the slope and the formation of design units are generally based on slope material properties, discontinuity conditions, the stratigraphy of the slope, the degree of weathering, and the influence of groundwater. Page 4-42 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes A design unit is defined as a portion of a roadside slope that can be cut at a consistent angle with or without benches. It may contain a single rock type or different rock types of similar intact rock strength. The difference in discontinuity characteristics (orientation, persistence, spacing, infillings, etc) in the entire unit should be allowed for in the design. The degree of weathering should also be the same throughout, and groundwater seepage requiring the same drainage measures. Once design units are formed, the safest and most economical design which best accommodates the site geology and project-specific constraints is prepared. Like soil slopes, any design option in rocks should target a minimum factor of safety of 1.3. 4.5.6. Safe angles of cut and benches In a procedure that does not involve detailed rock slope stability analyses, rock quality designation (RQD) or fracture frequency can be used in the field to assess the overall stability of a design unit and the safe angle of excavation. Alternatively, a suitable cut slope for a design unit can also be decided by inspecting existing cuts in similar materials along the proposed route or adjacent alignments. Other factors being equal, new cuts can be formed at the same slope as stable existing cuts. However, material properties, discontinuity characteristics, groundwater conditions and the degree of weathering should also be the same. Table 4-13 provides cut slope angles developed from the condition of rock cuts in Ethiopia and considering the type of rocks in the country. In general, when the geological conditions affecting slope stability are favourable and constructability is not a controlling factor, a steep cut (0.25H:1V) may be possible. It is important to note that changes to the orientations of cut slopes, even slight, may decrease slope stability significantly by increasing the feasibility of movement along joint sets. This outcome might require rock fall mitigation and protection measures to compensate. In such cases, the economic advantages afforded by a steeper slope can be lost and a flatter, more stable slope (0.5H:1V or less) may be needed. Ethiopian Roads Authority Page 4-43 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-12: Indicative maximum cut slope angles (H:V) for rock slopes (without discontinuity control) Type of Material Cut-slope height Remark <10m 10-15m 15-20m Fractured and slightly to moderately weathered basalt 0.25:1 0.75:1 1.00:1 Fractured and weathered basalt in blocks, cobbles and gravels of all sizes in a matrix of silt or clay 1.00:1 1.25:1 1.50:1 Weathered tuff, volcanic breccias or agglomerate 1.00:1 1.25:1 1.50:1 Fractured, slightly weathered, thickly bedded limestone with intercalations 0.25:1 0.50:1 1.00:1 Apply controlled excavation; benches and drainage are needed at different heights. Remove hanging blocks before cutting. This minimizes the formation of overhangs Slightly weathered and fractured mudstone and 1.00:1 marl 1.25:1 1.50:1 Remove weathered and unconsolidated portion at low angles. Drainage is important to keep the slope dry. Reinforcement works and erosion controlling measures may be needed. Slightly weathered, poorly cemented sandstone, with poorly defined stratification 0.25:1 0.50:1 1.00:1 Controlled or trim is recommended so asblasting not to damage the rock behind the cut face. Weathered, highly friable shale and jointed 1.00:1 mudstone 1.50:1 2.00:1 Shale has a characteristic of being deformed when wet. Hence, drainage systems are needed to keep the slope dry. Benches may also be required. For discontinuity-controlled slopes, each will need to be assessed on a case by case basis. When the rock is fresh and no significant weathering is observed, similar cut slope angles may be used for one or more design units. If differential weathering is observed or anticipated, vertically or laterally, the use of variable cut slope angles is advisable. Critical factors to consider when fixing bench dimension are summarized in Table 4-13. Page 4-44 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-13: Factors controlling bench dimension on rock slopes. Modified from GWP Consultants (2008) Bench width The optimum width of a bench depends on geotechnical parameters and the purpose of the bench itself. In relatively unweathered rock, a minimum bench width of 2 m is recommended. In all cases, benches must have suitable drainage measures included in the design. Benches usually require access for maintenance (ditch cleaning, debris removal) and this should be accommodated in the design. The height of benches is dictated generally by stability issues (including Bench height rock fall and working arrangements scaling the face). For ease of working, it is safe advisable to limit the heightfor of benches to 6m. Bench inclination The bench face angle, or slope, depends on stability and rock properties (including permeability and weathering). Generally, the following points should be considered during the design of benched cut slopes. • • • • • • • • Benches on rock slopes should be at least 2 m wide and sloped in wards to retain falling blocks and to facilitate drainage For incompetent materials thicker than 3 m, benches should be made wider as necessary based on specific conditions. Engineering judgment must be used to determine site-specific minimum thickness of a weathered rock unit or a weak layer bed that requires benching. For inter-layered rocks, engineering judgment should be used to determine the sitespecific bench size required. Where permeable formations overlie impermeable ones (including areas of fractured flow), the configuration of benches must consider drainage issues. Benches must be adjusted during construction to follow changes in the character of discontinuities and bedding surfaces. Installing a bench drain along the contact between competent and incompetent rock units where groundwater is anticipated is advisable. Access for maintenance equipment should be provided to the lowest (foot) bench, and if feasible to all higher benches. Benches should always be maintained and regularly cleaned. 4.5.7. Methods of rock excavation Rock excavation for road cuts is normally carried out by techniques of digging, ripping, breaking or blasting. Figure 4.31 shows the situations when these activities are applied during rock excavation. Digging is used in high weathered and relatively weak rocks. Ripping is a process of breaking up rock and soil with a large tooth or teeth attached to the back of a bulldozer. Ripping is generally preferred over breaking and blasting because it is considerably less expensive. Ripping is also less dangerous and requires fewer precautions. Ripping can be done in close proximity to populated areas where noise and vibration from breaking and blasting are restricted. Ethiopian Roads Authority Page 4-45 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-31: General guide to select a method of rock excavation. From Pettifer and Fookes (1994) However, ripping is limited to weak and fractured rock. In hard rock with few discontinuities, light blasting is sometimes performed before ripping or breaking. Once rock is loosened by ripping, an excavator can be used to construct the designed slope. Breaking is carried out with a hydraulic hammer (also known as a breaker or hoe ram) fitted to an excavator. It is used to break up rock in areas where blasting is prohibited due to environmental or other constraints. Like a ripper, a hydraulic hammer can be used in most rock types, although when shaping of a slope face is needed, it works best in weak and moderately to highly fractured rock. Existing discontinuities in the rock act as presplit lines, minimizing hammer induced scars and fractures while creating a slope face that appears to be naturally weathered. In areas where the amount of rock needed to be excavated is high, rock blasting may be used. Blasting is accomplished by discharging an explosive placed in a borehole (confined) or in mud capping boulders (unconfined). A confined charge uses high gas energy, whereas an unconfined charge works by shock energy output. 4.5.7.1 Blast design The degree of rock blasting depends on blast design, which in itself is controlled by factors such as side burden, sub-drilling, stemming (collar distance), spacing, hole depth, delays, and blast-hole diameter and pattern. The meaning of some of these terms is graphically described in Figure 4-32. Page 4-46 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes The side burden is defined as the distance from a blast hole to the nearest free face of the rock mass at the instant of detonation. Too large a side burden will produce inadequate fragmentation, toe problems, and excessive ground vibrations. An insufficient side burden will cause excessive air-blast and fly-rock. Sub-drilling is the distance drilled below the floor level to ensure that the full face of rock is removed. In road construction, sub-drilling is generally limited to 10% or less of the bench height (H). Stemming is the distance from the top of the explosive charge to the collar of the blast hole. This zone is usually filled with an inert material to give some confinement to the explosive gases and to reduce air-blast. Well-graded, crushed gravel works best as stemming, but it is common practice to use drill cuttings because of availability and economics. A stemming that is too short results in excessive violence in the form of airblast and fly-rock and may cause back-break (breaking beyond the desired limiting wall). Stemming that is too long creates large blocks in the upper part of the rock mass. Spacing is defined as the distance between adjacent blast holes, measured perpendicular to the side burden. Spacing is calculated as a function of the side burden and also depends on the timing between holes. Spacing that is too close causes crushing and ponding between holes. Spacing that is too wide causes inadequate fracturing between holes, toe problems, and is accompanied by humps on the rock face In any blast design, the side burden and the blast-hole depth (or bench height) must be reasonably compatible. The rule of thumb for bench blasting is that the hole depth-to-side burden ratio should be between 1.5 and 4.0. A hole depth less than 1.5 times the side burden causes excessive air blast and fly rock and, because of the short, thick shape of the side burden, gives coarse and uneven fragmentation. Hole depths greater than four times the side burden are also undesirable, as poorly controlled blasting will result. Figure 4-32: Blast design parameters. Modified from US DOI Bureau of Reclamation (1998) Ethiopian Roads Authority Page 4-47 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 4.5.7.2 The role of geology Rock mass properties are critical variables affecting the design and results of a blast. A detailed drill log that contains information on drilling resistance and the depth of various geological boundaries and discontinuities is required to effectively design a blast. Slow penetration, excessive drill noise, and vibration indicate a hard rock that will be difficult to break. Fast penetration and a quiet drill indicate a weaker, more broken zone of rock. Total lack of resistance to penetration, accompanied by a lack of cuttings or return of water or air, indicates that the drill has encountered a void. Lack of cuttings or return water may also indicate the presence of an open bedding plane, joint or fissure. The density of rock is a major factor to determine how much explosive is needed to displace a given volume (powder factor). The burden-to-charge (hole) diameter ratio varies with rock density, changing the powder factor. Aside burden-to-charge diameter ratio of 25 to 30 is used for average density rocks. Denser rocks such as basalt require smaller ratios (higher powder factors). Lighter materials such as some sandstone can be blasted with higher ratios (lower powder factors). Figure 4-33 shows the effect of blasting in limestone, the density and tensile strength of which is in between those of basalt and sandstone. Blasting in limestone usually results in fine fragmentation and dust. Jointing is probably the most significant parameter in blasting design . Close jointing usually results in good fragmentation. Widely spaced joints, especially where the jointing is persistent, often results in a very blocky blasted mass because the joint planes tend to isolate large blocks in place. Where possible, the perimeter holes of a blast should be aligned with principal joint sets. This produces a more stable excavation. Rows of holes perpendicular to a primary joint set produce an unstable perimeter. Bedding has a pronounced effect on both the fragmentation and the stability of the excavation perimeter. Open bedding planes, open joints, or beds of weaker materials should be treated as zones of weakness. In a bed of hard material greater than 1 m thick, it is often beneficial to load an explosive of higher density than is used in the remainder of the hole. In bench blasting, the presence of a pronounced horizontal foliation, bedding plane, or joint can be conveniently used for the bench floor. Page 4-48 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Figure 4-33: Blasting in limestone Steeply dipping joint sets that daylight in the cut face can cause stability problems and difficulty in breaking the toe of the excavation. By contrast, when joints dip into the excavation wall, the stability of the slope is enhanced. However, the inclination can still create toe problems because the toe rock tends to break along bedding or foliation planes. In this case, advancing the opening perpendicular to dipping beds may be a compromise. Blast hole cut-offs (part of a column of explosives not fired) caused by differential movement along joints may also occur. However, undetected voids and zones of weakness such as solution cavities, “mud” seams, and shears pose serious problems to blasting because explosive energy always seeks the path of least resistance, resulting in poor fragmentation. When charging the blast-hole, inert stemming materials rather than explosives should be loaded through these weak zones and voids. If the presence and exact depth of voids is in doubt, the top of the powder column should be checked frequently as loading proceeds. A void exists if the column fails to rise as expected. The presence of weak rocks can be known by observing adjacent exposures or information from boreholes. Alternate zones of hard and weak rock exposed for the same explosive energy usually result in unacceptably blocky fragmentation and unnecessary displacement of the hard rock. Displacement can also occur if there is a void between the bedding in hard rock. Figure 4-34 shows an example of a thickly bedded rock mass displaced outward by about 1 m because of excessive blasting. In this case blasting has also widened vertical joints and solution cavities, and created some overhangs. In this situation, the best approach is to use smaller blast holes with smaller blast pattern dimensions. This allows a better powder distribution in the rock mass. The explosive charges should always be concentrated in the hard strata and stemming should be added through weak zones and voids. Alternatively, it is possible to use techniques of controlled blasting. Ethiopian Roads Authority Page 4-49 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Figure 4-34: Rock displacement as a result of uncontrolled blasting 4.5.7.3 Controlled blasting Depending on the purpose of rock excavation and the degree of stability required, two types of blasting are commonly used for road construction. These are bulk (massive or heavy) blasting and controlled blasting. Bulk blasting uses large explosive charges that are designed to fragment a large amount of side burden. Hence, it typically creates radial fractures around the blast hole and back-breaks (fractures that extend into the final slope face), which reduce the strength of the remaining rock mass and increase its susceptibility to slope ravelling and rock fall. Bulk blasting should only be used for road aggregate production where stability is not a concern. Controlled blasting is used for removing material along the final slope face without disturbing the general stability of the rock mass. It creates less back-break than bulk blasting because it removes less side burden and uses more tightly spaced drill holes with lighter charges. In some cases, it is also used before bulk blasting to create an artificial fracture along the final cut slope, which will prevent the radial cracks caused by the latter from penetrating back into the finished face. There are different types of controlled blasting techniques. The difference between these techniques is most importantly in the amount of side burden they remove and the type of powder they use. The most common techniques are pre-splitting blasting, smooth blasting, and cushion blasting. A summary of the advantages and limitations of these methods is given in Table 4-15. Page 4-50 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-14: Controlled blasting techniques. From US DOT FHWA CFLHD (2011) Procedure Description Pre-splitting Protects the final cut by Presplit holes are blasted. producing a fracture plane Procedure uses small along the final slope face. diameter holes at close Can produce steeper cuts spacing and lightly loaded with less maintenance with distributed charges. issues. Performs well in hard competent rock Smooth blasting Smooth blast holes are blasted after main blasts. Procedure uses small diameter holes at close spacing and lightly loaded with distributed charges. Produces smooth, stable slope. Can be done on slopes years after initial construction. Drill hole traces are less apparent than pre-splitting. Performs best in hard, competent rock. Cushion blasting is done after main blasts. Larger drill holes are used with small diameter, lightly loaded distributed loads. Space around the explosive is filled with crushed rock to cushion Reduces the amount of radial fracturing around the borehole and also reduces borehole traces. The large diameter holes allow blasting depths up to 30 m. Produces a ragged final slope face. Performs well in the explosive force. all rock types Cushion blasting Advantages Limitations The small diameter borings limit the blasting depth to shallow depths (15 m). Borehole traces are present for entire length of boring. Does not perform well in highly fractured, weak rock. The small diameter boring limits blasting to shallow depths (15 m). Borehole traces are present for much of the boring length. Does not protect the slope from damage caused by main blasting. Does not perform well in highly fractured, weak rock. Radial fractures are more abundant than presplit and smooth blasting. Slope face is more prone to ravelling. Borehole traces still apparent in hard, competent rock. Pre-splitting is used before bulk blasting to protect the final rock face from damage caused by the latter. Pre-splitting creates a fracture plane along the final slope face, which prevents the radial cracks created by bulk blasting from penetrating into the finished face. Without pre-splitting, bulk blasting damage can extend up to a considerable depth into the final slope face. Pre-splitting also allows for steeper and more stable cuts than any other blasting procedure. In massively bedded, competent rock, such as the limestone shown in Figure 434, a properly charged presplit blast will contain drill hole half cast for almost the entire length of the blast line and will have no back-break because the energy from the blast will travel uniformly, thus creating a continuous fracture between holes. Pre-splitting requires relatively small drill holes, from 50 to 100mm in diameter because the goal is to create discrete fractures, not massive breaking. However, because the small hole diameter allows the drill bit to deviate from the anticipated line more readily than a larger drill diameter, the maximum depth of pre-splitting is usually about 15 m. For this reason pre-splitting is used only for relatively small blasting operations. Because of these limitations, pre-splitting is most often used on slopes steeper than 1H:1V, which helps the drillers to maintain adequate hole alignment at depth. Pre-splitting performs best in competent, hard to extremely hard rock. It is most difficult in highly fractured, weathered, and/or weak rocks, where there is a requirement for the use of closely spaced drill holes and/or uncharged guide holes. Table 4-16 gives a summary of the parameters used to carry out pre-splitting. Theoretically, the side burden for presplit blasting is unlimited. But in reality, variations in geology that are not visible on the outer Ethiopian Roads Authority Page 4-51 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 face of the slope can limit that burden. Thus, the condition of the rock in the interior of the slope needs to be ascertained before determining the blasting design. In any case, a minimum of 10 m of side burden is recommended for any presplit blasting procedure. Table 4-15: Parameters used for drilling in presplit, smooth and cushion blasting. From US DOT FHWA CFLHD (2011) Blasting method Pre-splitting Smooth blasting Cushion blasting Hole (mm) diameter Spacing (m) Side Burden (m) Explosive charge (kg/m) 38-44 0.3-0.46 --- 0.03-0.1 50-64 75-90 0.46-0.6 0.6-1.0 ----- 0.03-0.1 0.05-0.23 100 0.6-1.2 --- 0.23-0.34 38-44 0.6 1 0.05-0.55 50 0.75 1.06 0.05-0.55 50-64 1.0 1.2 0.03-0.1 75-90 1.2 1.5 0.05-0.2 100-115 1.5 1.8 0.1-0.3 127-140 1.8 2.1 0.3-0.45 152-165 2.1 2.7 0.45-0.7 Smooth blasting, also called contour blasting or perimeter blasting, can be used before bulk blasting as an alternative to pre-splitting. It is also used after bulk blasting, either as an entirely different event or as the last delay of the bulk blast. Smooth blasting uses drill holes with roughly the same diameter and depth as those used in pre-splitting, spaced slightly further apart and loaded with a slightly larger charge density. If the side burden is adequately reduced, smooth blasting produces a more ragged slope face with minimal back-break. Smooth-blasted slopes may require more maintenance than presplit slopes due to increased radial fractures from the controlled blasting and overall fracturing from bulk blasting. Smooth blasting is best preformed in hard, competent rock, although it can be used in weak or highly fractured rock by increasing the spacing of the drill holes and/or adding uncharged guide holes to the pattern. The advantages of reducing over-break usually outweigh the cost of the additional perimeter or guide holes. Smooth slope blasting can be used on a variety of cut slope angles and is effective in developing contoured slopes with benches or other slope variations. The spacings and side burdens commonly used to drill for smooth blasting operation are given in Table 4-16. Cushion blasting, sometimes referred to as trim blasting, uses a row of lightly loaded “buffer” holes filled with crushed stone over the entire depth of the hole, which reduce the impact on blasting holes and protect the surrounding rock mass from the shock caused by the blast, thus minimizing the stress and fractures in the finished slope face. The maximum diameter for cushion holes used in road projects is typically 75 mm. The drill steel used to advance these smaller holes tends to drift at depth, meaning the maximum depth is usually held to 12 m. Cushion blasting creates some back-break, which can make a slope more prone to ravelling and rock fall. Page 4-52 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Cushion blasting is more demanding than presplit or smooth blasting because the hole spacing, burden, and charge density must be carefully chosen and continually reassessed in order to minimize back-break. It can also be more time consuming because more drilling is required and charges take more time to load. Cushion blasting produces better results than smooth blasting or pre-splitting in poorly lithified, moderately to highly fractured and weathered rocks. Suggested blast parameters for conducting cushion blasting are summarized in Table 4-16. 4.6. Soil Slope Stability Analyses Soil slope stability analysis is an iterative process through which traditionally a factor of safety (FS) is computed and potential sliding surfaces are determined. In road construction, the primary purpose of slope stability analysis is the contribution to the safe design of cut and embankment slopes. It is concerned with identifying critical soil properties that govern the stability of these slopes. Some of the objectives of slope stability analyses are summarized as follows: • • • • • • To understand the development of natural slopes and the processes responsible for modifying these slopes, and the influence of environmental factors. To evaluate the possible occurrence of landslides. To assess the stability of existing landslides and natural slopes. To assess the stability of cut and embankment slopes under short-term (often during construction) and long-term conditions. To study the effect of groundwater conditions, seismic loadings and other factors on natural slopes, cut slopes and embankments. To enable the redesign of failed slopes and the planning and design of preventive and remedial measures (Chowdury et al 2009). A slope stability analysis in soil commonly requires the following parameters to be determined: The depth and configuration of the failure surface in the slope section. • • • The strength and density of soil materials and the configuration of soil layers. The groundwater table and/or soil moisture condition. An accurate cross-section for analysis. For failed slopes, back analysis is undertaken to determine the condition of the slope at the time of failure. A factor of safety of 1.0 is assumed and, if the pre-failure topography can be reasonably deduced, then the unknown parameters are reduced to the failure surface location, the strength parameters and the groundwater condition. Both in initial stability (first-time failure) and back analyses procedures, sensitivity analysis is sometimes performed by varying the water table according to drainage measures and the strength parameters to assess their effects on factor of safety. There are different methods that can be used to analyze the stability of soil slopes. These include conventional limit equilibrium methods, numerical (finite element and finite difference) analyses, probabilistic approaches, and stability charts. Ethiopian Roads Authority Page 4-53 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 The conventional limit equilibrium methods of soil slope stability analysis in geotechnical practice investigate the equilibrium of a soil mass tending to move down slope under the influence of gravity. A comparison is made between forces, moments, or stresses tending to cause instability of the mass, and those that resist instability. Often two-dimensional sections are analyzed and plane strain conditions are assumed. These methods assume that the shear strengths of the materials along the potential failure surface are governed by linear (Mohr-Coulomb) or nonlinear relationships between shear strength and the normal stress on the failure surface. The finite element method can be used to compute stresses and displacements in earth structures. The method is particularly useful for soil-structure interaction problems, in which structural members interact with a soil mass. The stability of a slope cannot be determined directly from finite element analyses, but the computed stresses in a slope can be used to compute a factor of safety. Use of the finite element method for stability problems needs software developed for this purpose. With the advance in technology, the use of relatively complex finite difference computer programs for stability analyses is also becoming common to refine results obtained from traditional methods. Probabilistic approaches consider the magnitudes of uncertainties related to shear strengths and other parameters involved in computing the factors of safety. In the traditional (deterministic) approaches, the shear strength, slope geometry, external loads, and pore water pressures are assigned specific values. In probabilistic methods, the possibility that these parameters may vary is considered, providing a means of evaluating the degree of uncertainty associated with the computed factor of safety. Although probabilistic techniques may not be required for routine design purposes, their use allows addressing issues beyond those tackled by deterministic methods. The stability of slopes can be analyzed quickly using stability charts. Although the charts assume simple slopes and uniform soil conditions, they can be used to obtain reasonably accurate answers for the analysis of a variety of short-term and long-term conditions if irregular slopes are approximated by simple slopes, and carefully determined average values of unit weight, cohesion, and friction angle are used (Duncan et al 1987). This manual focuses on limit equilibrium methods. Discussions on numerical and probabilistic approaches are available in many geotechnical engineering reference books. Users are strongly encouraged to refer to these books. 4.6.1. Limit equilibrium methods Most limit equilibrium methods are based on the principles of statics, i.e. summation of moments, vertical forces, and horizontal forces. They assume the validity of Coulomb's failure criterion along an assumed failure surface and utilize the Mohr-Coulomb expression to determine the shear strength along the sliding surface. Most of the methods of stability analysis currently in use for slope design fall in this category. The procedure in limit equilibrium methods is that a free body of the slope is considered to be acted upon by known or assumed forces. Shear stresses induced on the assumed failure surface by the body and external forces are compared with the available shear strength of the material. The shear stress at which a soil fails in shear is defined as the shear strength of the soil. The requirements for static equilibrium of the soil mass are used to compute a Page 4-54 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes factor of safety with respect to shear strength. Limit equilibrium methods do not account for the load deformation characteristics of the materials. The factor of safety (FS) is defined as the ratio of the available shear resistance to that required for equilibrium. Limit equilibrium analyses assume the FS is the same along the entire slip surface. An FS of greater than 1.0 indicates the available strength exceeds the required resistance and that the slope is stable with respect to sliding along the assumed slip surface. The slope will be unstable if the FS is less than 1.0.The slope is said to have reached limit equilibrium when the FS is equal to 1.0. Figure 4-35 shows some other means of defining the FS using force and moment equilibrium. In most limit equilibrium methods, the soil mass above the assumed slip surface is divided into vertical slices for purposes of convenience in the analysis. Each of these methods may result in different values of FS because they employ different assumptions to make the problem statically determinate, and some of the methods do not satisfy all conditions of equilibrium mentioned in Figure 4.35. Figure 4-35: Equilibrium conditions used to define the factor of safety Table 4-17 presents different types of limit equilibrium method useful for soil slope stability analyses. Some methods are simple, and are suitable for preliminary design while others are rigorous and should be favoured for evaluation of final designs. The infinite slope method assumes that the slope is of infinite longitudinal extent and that sliding occurs along a plane surface parallel to the face of the slope. For slopes composed of uniform cohesionless soils (c' = 0), the critical slip surface will be parallel to the slope. The Infinite slope method is a special case of the force equilibrium procedure, with one Ethiopian Roads Authority Page 4-55 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 slice. With only one slice, two equations are available (horizontal and vertical force equilibrium), and two unknowns must be evaluated (the factor of safety and the normal force on the base of the slice). The ordinary method of slices relies on solving moment equilibrium conditions and is applicable to circular failure surfaces only. This method does not solve either vertical or horizontal force equilibrium conditions. This method is relatively simple and can be solved by hand calculators. Hence, it is rarely used for final design. The simplified Bishop method was developed by Bishop in 1955. It solves two of the equilibrium moment andvertical vertical. This assumes horizontal forces are not onlyequations, perpendicular to the sides of method the slice, but arethat equal and opposite. Therefore, the horizontal forces are assumed to cancel each other out during mobilization. Since horizontal equilibrium is not satisfied, the simplified Bishop method is rarely used in designs that include provision for seismic forces. Like the ordinary method of slices, the simplified Bishop methods can only be used on circular failure surfaces. Simplified Janbu is a force equilibrium method in which the moment equilibrium is either ignored or assumed to be zero. Like any other force equilibrium methods, the simplified Janbu solves both the horizontal and vertical forces. The main assumption required in using this method is the inclination of the horizontal forces on the given slice. The inclination of the horizontal forces acting on a slice may be either the slope angle or the average slope angle if multiple slopes are involved. This method uses non-circular failure surfaces and, therefore, may be solved graphically. Spencer’s method solves all three conditions of equilibrium and is therefore termed a complete limit equilibrium method. The method was srcinally developed to determine the stability of circular failure surfaces, but can now be used for non-circular failure surfaces as well. The method assumes that the inter-slice forces are parallel and act on a certain angle from the horizontal. This angle is one of the unknowns in this method. Therefore, the first approximation of the angle should be the slope angle. The other unknown is the factor of safety that is solved through an iterative process by using computer programs. The Morgenstern and Price method is very similar to the Spencer approach. The main difference between the two methods is that Spencer solves for the inter-slice angle, while the Morgenstern and Price method solves for the scaling parameter that is used as a function that describes the slice boundary conditions. The Morgenstern and Price method provides added flexibility using the inter-slice angle assumptions. Page 4-56 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes Table 4-16: Commonly used limit equilibrium methods Method Infinite slope Ordinary methods of slices Bishop Janbu Failure surface Assumptions Advantages Limitations Recommendation Suitable for plane failures over long slopes, especially for those with a thin layer of weathered soil over rock. Planar (Straight line) Assumes a slope of infinite longitudinalRelatively simple extent and that sliding occurs along a for manual plane surface parallel to the slope. calculation. Failure surface assumptions are always an approximation. Circular Forces on the sides of the slice are neglected. The normal force on the base Useful when of the slice is calculated by summing calculation must be forces in a direction perpendicular to done by hand using the bottom of the slice. Moments are a calculator. summed about the centre of the circle to compute FS. Does not consider the forces on the sides of the slice. The method does not satisfy equilibrium of forces in either the vertical or horizontal directions. The method also may be used to overcome problems that may develop near the toe of steep shear surfaces. Circular Considers force and moment equilibrium for each slice. The interslice forces are horizontal. Simplified methods assume the resultant of the vertical forces is zero on each slice. Rigorous method assumes values for the vertical forces on the sides of each slice until all equations are satisfied. Circular failure Simplified method surface may not be compares well with representative of all other advanced failures; horizontal methods. Very equilibrium of forces efficient and is not satisfied (not relatively common. suitable to analyze Computer programs the effect of readily available. earthquakes for example). Useful where circular failure surface is assumed. Final design should be checked using Spencer or Morgenstern-Price methods. Non-circular Simplified procedure satisfies force equilibrium in both the horizontal and vertical directions, but not moment equilibrium. Generalized procedure considers force and moment equilibrium on each slice. Realistic shear surfaces can be used. Suitable for shallow landslides. Routine analysis can be easily handled by a The method is recommended for use in hand calculations where noncircular slip surfaces are being analyzed Ethiopian Roads Authority Computed values are for homogeneous materials (can give large errors in slopes composed of more than one material). Factor of Safety is Page 4-57 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 programmable calculator or by hand. Spencer Non-circular (Rigorous) Morgenstern- Non-circular Price (Rigorous) Handles any geometry and The side forces are parallel (all side loads. Gives Requires computer forces are inclined at the same angle). statically complete software to perform Satisfies fully the requirements for bothsolution. the calculations. force and moment equilibrium. Implemented in many computer programs. Handles any geometry and loads. It can simulate internal Considers forces and moments on each shearing. Often slice like Janbu generalized procedure. considered as a benchmark. Computer programs available. Page 4-58 underestimated in simplified procedure. Computer help is necessary. It should also be used as a check on final designs where the slope stability computations were performed by simpler methods. Most useful for back analyzing failed slopes and to refine analyses performed by simple methods. Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 Roadside Slopes 4.6.2. The use of computer programs Many of the limit equilibrium methods discussed above are at present implemented in most slope stability computer programs. The use of reliable and verified slope stability analysis software is essential in cases where conditions are complex. Computer programs provide a means for efficient and rapid detailed analysis of a wide variety of slope geometry and load conditions. Issues that should be remembered when using any computer program are: • • • • • • A thorough knowledge of the capabilities of the software and knowledge of the theory of limit equilibrium slope stability analysis methods is important to determine if the software is appropriate for any given situation . The software analyzes a failure geometry that reasonably reflects the actual condition. An understanding of the possible modes of failure is crucial to the successful application of the result of the analysis. This is particularly important in profiles where the mode of failure is governed by geological factors. Failures of colluvium over bedrock or failures in weathered rock most frequently occur along the surfaces dictated by structure. In such cases, circular failures do not generally occur and shallow non-circular analysis would be appropriate. The analytical program being used must be compatible with the critical elements of the slope problem to be investigated, for example drainage condition, loading condition, or layering of materials within the soil-rock mass. Appropriate shear strength and pore water parameters must be used for the analyses. In cases where the accuracy of parameters is in doubt, it is appropriate to undertake a sensitivity analysis to determine the effects on factor of safety of variations in these parameters. Back analysis of similar existing failures may also be an issue to consider. It is often recommended to check results from computer programs, if possible by hand or spreadsheet methods. If this is not possible then a sample parallel check using another program is recommended. The program output should be checked to ensure that results are reasonable and consistent. Important items to check include the weights of slices, shear strength properties, and pore water pressures at the bottom of slices. The user should be able to determine if the critical slip surface is passing through the relevant material. Any search scheme employed in computer programs is restricted to investigating a finite number of slip surfaces. In addition, most of these schemes are designed to locate one slip surface with a minimum factor of safety. The schemes may not be able to locate more than one local minimum. The results of automatic searches are dependent on the starting location for the search and any constraints that are imposed on how the slip surface is moved. Automatic searches are controlled largely by the data that the user inputs into the software. Regardless of the software used, a number of separate searches should be conducted to confirm that the lowest factor of safety has been calculated. 4.6.3. Determination of shear strength parameters The shear strength parameters (c, φ, c' and φ') are normally determined using laboratory tests. Depending on drainage allowed, three types of test are possible; unconsolidatedundrained (UU), consolidated-undrained (CU), and consolidated-drained (CD). The choice Ethiopian Roads Authority Page 4-59 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 depends on the state of stability at the time of testing (end-of-construction (short term), steady-state seepage (long term), or rapid drawdown). Cohesive soil shear strength parameters should be obtained from undisturbed soil samples using consolidated-undrained (CU) triaxial tests with pore pressure measurement if portions of the proposed slope are saturated or might become saturated in the future. Effective strength parameters from these tests should be used to analyze cohesive soil cut slopes and to evaluate long term effects. Consolidated-drained (CD) loading procedures are used to determine the effective stress shear parameters of freely draining soils. These will drain with closest relatively short strength testing times and the consolidated-drained loadingsoils procedure comes to representing the loading for long-term, drained conditions in the field. Consolidateddrained tests procedures are also used to measure the residual shear strength of clays using a direct shear test. Unconsolidated undrained (UU) triaxial tests can be used to obtain undrained shear strength parameters for short term stability analysis, or when it is determined that total stress (strength) parameters are sufficient. It should be noted that for unsaturated soils, particularly cohesive soils, the natural moisture content of the soil at the time of testing must be determined since this will affect the shear strength parameters. Consideration should be given to future changes in moisture content, particularly if field testing was performed during the dry season. If laboratory test data are not available, or if the interest is a preliminary design only, strength parameters may be estimated using parameters in situ testsand suchother as the Penetration Test (SPT). Correlations between strength soilStandard properties such as grain size and plasticity may also assist in selecting approximate values. 4.6.4. Stability analysis procedures • • • • • Determine soil strength parameters. This usually involves laboratory testing. Distinguish clearly between first-time landslides and possible renewed movements. Rely only on the residual strength along parts of assumed slip surfaces that correspond to existing, or prior shear zones (Chowdury et al 2009). Establish the geometry of the slope, including the ground surface and subsurface boundaries between various materials. Establish seepage and groundwater conditions. Consider the possibility of artesian pressures and perched water tables by examining geological details. The assumption of a single groundwater level, above which all slope materials are dry, is an oversimplification in most situations. A reasonable representation of groundwater conditions can be obtained by combining observed water tables with knowledge of subsurface variations in geology and soil permeability. Select loading conditions for analysis. Make decisions concerning the use of effective or total stress types of analysis. In particular, consider the type of materials, whether analysis is for short-term or long-term conditions, whether reliable estimates of pore pressures can be made in advance, and whether pore pressures are to be monitored in the field (Chowdury et al 2009). Select trial slip surfaces and compute factors of safety. Attempt to visualize the probable shape of the slip surface or surfaces. Special attention must be given to the Page 4-60 Ethiopian Roads Authority Chapter 4 Geotechnical Design Manual - 2013 • • • • Roadside Slopes existence of major discontinuities, existing slip surfaces, stratification, nonhomogeneity, tension cracks, and open joints. In homogeneous soil slopes without discontinuities, assume a slip surface of circular shape unless local experience dictates otherwise. In embankments, give consideration to method of construction, zones of different materials, and nature of foundation in order to visualize the probable shape of the slip surfaces (Chowdury et al 2009). Repeat selecting slip surfaces until the “critical” slip surface has been located. The critical slip surface is the one that has the lowest factor of safety and which, therefore, represents the most likely failure mechanism. Compare the computed factor of safety with experienced-based criteria The design assumptions should be verified during construction. This may require repeating the above steps and modifying the design if conditions are found that do not match previous assumptions Following construction, the performance of the constructed slope should be monitored. Actual groundwater levels based on pore water pressure measurements should be compared with those assumed during design. Table 3-6 in Chapter 3 summarizes the types of analysis, sources of shear strength data and the type of limit equilibrium methods suitable for cohesive and granular soils. 4.7. Rock slope stability analyses The stability of hard rock slopes is normally controlled by discontinuities (joint and joint sets) within the rock. Failures tend to occur as discrete blocks. Therefore, kinematic analysis of the discontinuities is performed first to determine the most likely mode of failure, and second by slope stability analyses to determine the factor of safety. 4.7.1. Kinematic analysis A kinematic analysis is the first step in evaluating slope stability. This analysis establishes the possible failure modes of the blocks that comprise the slope. The analysis determines if the orientations (dip and dip direction) of the various discontinuities will interact with the cut slope orientation and inclination to form discrete blocks with the potential to fail without regard to any forces that may be involved. Failure modes typically fall within one of three categories: plane failure, wedge failure, or toppling. Where a rock mass is highly fractured by randomly orientated discontinuities or composed of very weak rock, the mode of failure may be circular as in a soil slope. The analysis involves a comparison of the orientations of the dominant discontinuity sets with the orientation of the cut slope. Where discrete blocks are formed and where the failure surfaces that bound these blocks dip out of the slope at an angle steeper than the shear strength along the discontinuity, failure is kinematically possible. A stereonet is used to display the discontinuity and slope data in this analysis. The advantages and limitations of conventional rock slope analyses are summarized in Table 4-18. Ethiopian Roads Authority Page 4-61 Chapter 4 Roadside Slopes Geotechnical Design Manual – 2013 Table 4-17: Advantages and limitations of conventional rock slope analyses. From Eberhardt (2003), modified from Coggan et al (1998) Analysis method Kinematic (using stereographic interpretation) Limit Equilibrium Critical parameters Advantages Critical slope and discontinuity geometry; representative shear strength characteristics. Representative geometry and material characteristics; soil or rock mass shear strength parameters (cohesion and friction); discontinuity shear strength characteristics; groundwater conditions; support and reinforcement characteristics. Relatively simple to use; gives initial indication of failure potential; may allow identification and analysis of critical keyblocks block theory; using links are possible with limit equilibrium methods; can be combined with statistical techniques to indicate probability of failure. Wide variety of commercially available software for different failure modes (planar, wedge, toppling, etc.); can analyse factor of safety sensitivity to changes in slope geometry and material properties; more advanced codes allow for multiple materials, reinforcement and/or groundwater profiles. Limitations Only really suitable for preliminary design or design of non-critical slopes; critical discontinuities must be identified; needdiscontinuity to be used with representative or joint shear strength data; primarily evaluates critical orientations, neglecting other important joint properties. Mostly deterministic producing single factor of safety (but increased use of probabilistic analysis); factor of safety gives no indication of instability mechanisms; many methods available all with varying assumptions; strains and intact failure not considered; probabilistic analysis requires well-defined input data to allow meaningful evaluation. 4.7.2. Limit equilibrium analyses Limit equilibrium analyses are routinely used in the analysis of rock slope stabilities where movements occur on distinct failure surfaces. Analyses are undertaken to provide either a factor of safety or, through back-analysis, a range of shear strength parameters at failure. In general, these methods are the most commonly adopted in rock engineering, though many failures involve complex deformation and fracturing. All limit equilibrium techniques share a common approach based on a comparison of resisting forces/moments mobilized and the disturbing forces/moments. Methods may only vary with respect to the failuretomechanism (e.g. translational rotational sliding), assumptions adopted in order achieve a determinate solutionor(Eberhardt 2003). and the Page 4-62 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 5 Geotechnical Report and Checklist GEOTECHNICAL REPORT AND CHECKLIST A geotechnical report is a document used to communicate the site conditions and design and construction recommendations to the design and construction personnel. Site investigations for transportation projects provide specific information on subsurface soil, rock, and water conditions. This information is normally compiled in a site investigation report. Interpretation of the site investigation information by a geotechnical engineer, results in design and construction recommendations that should be presented in the project’s geotechnical report. The geotechnical report summarizes all of the pertinent information that will be used in road design. It provides findings from the field explorations, laboratory testing, preliminary and final geotechnical evaluations as they relate to the project, and construction planning reviews. It also contains design (and construction) recommendations for the alignment and Right of-Way issues, and for structures such as bridges. The primary audience for the report are highway designers and construction supervisors and contractors. Depending on the project stage and the extent of information needed, geotechnical reports can be divided into preliminary and final level reports as discussed below. 5.1 Preliminary Level Geotechnical Report A preliminary level geotechnical report is typically used to provide geotechnical input in the early stage of project development and reconnaissance studies (pre-feasibility, feasibility, and possibly even preliminary design stages). It may also be prepared for rapid assessment or emergency repair needs (during the occurrence of landslides, rock-fall, bridge foundation scour, etc.).Early submittal of geotechnical information and recommendations or engineering evaluation of preliminary data may be necessary to establish basic design concepts or design criteria. This is commonly the case on roads passing through difficult terrain where alignment and/or grade changes may be necessary because of stability issues or excavation problems. A preliminary level geotechnical report is prepared based on a minimum of a desk study review of existing geotechnical data for the site, and generally consist of feasibility assessment and identification of geologic hazards. Geotechnical design for a preliminary level report is typically based on engineering judgement and experience at the site or similar sites. For preliminary level design, a geotechnical reconnaissance of the project site and a limited subsurface exploration is often sufficient, as well as some detailed analyses to characterize key elements of the design, to assess potential alternatives, and to estimate preliminary costs. For preliminary design of more complex road projects with potentially unusual subsurface conditions, a geotechnical reconnaissance of the site should be conducted in addition to the desk study review to assess the site conditions. A preliminary level geotechnical report should contain the following contents: • • • • A general description of the project, project elements, and project background; A brief summary of the regional and site geology. The amount of detail included will depend on the nature of the project; A summary of the field exploration and laboratory testing conducted; A description of the project soil and rock conditions. For preliminary design reports in which new borings have been obtained, soil profiles for key project features (e.g. bridges, retaining walls, etc.) may need to be developed; Ethiopian Roads Authority Page 5-1 Chapter 5 Geotechnical Report and Checklist • • • 5.2 Geotechnical Design Manual – 2013 A summary of geological hazards identified that may affect the project design (e.g. landslides, rock-fall, debris flows, liquefaction, soft ground, expansive soils, seismic hazards, etc.); A summary of the conceptual or preliminary geotechnical recommendations; Appendices that include any boring logs and laboratory test data obtained, soil profiles developed, any field data obtained, and any photographs. Final Level Geotechnical Report A final level geotechnical report is developed for final design and construction review purposes. It is prepared based on a desk study review of existing geotechnical data, a detailed geological assessment of the site, complete subsurface investigation and laboratory programmes, and detailed analyses and interpretations. The development of a final geotechnical report will not normally be completed until the design has progressed to the point where specific recommendations can be made for all of the geotechnical aspects of the work. In road construction, final alignment, grade, and geometry will usually have been selected prior to issuance of the final geotechnical report. A final geotechnical report should contain the following contents: • • • • • • • • A general description of the project, project elements, and project background; Project site surface conditions and topographic assessments; A summary of geotechnical conditions that briefly describes the subsurface and groundwater conditions for key areas of the project where foundations, cuts, fills, etc., are to be constructed. This document should also describe the impact of these subsurface conditions on construction. Regional and local geology. This section should describe the site stress history and depositional/erosional history, bedrock and soil geological units, etc. Regional and site seismicity for major bridges, potential source zones, potential magnitude of shaking, frequency, historical activity, and location of nearby faults; A summary of the site data available from project or site records (e.g. final construction records for previous construction activity at the site, as-built bridge or other structure layouts, existing test hole logs, geological maps, previous or current geologic reconnaissance results, etc.); A summary of the field exploration conducted, if applicable, with a description of the methods and standards used as well as a summary of the number and types of explorations that were conducted. A description of any field instrumentation (e.g. piezometers) installed and its purpose should be included. A summary of the laboratory testing conducted with the description of the methods and standards used, as well as a summary of the number and types of tests that were conducted; A description of the soil/rock units encountered at the project site, groundwater conditions including the identification of any confined aquifers, artesian pressures, perched water tables, potential seasonal variations, any influences on the groundwater levels observed, and direction and gradient of groundwater, if known. If multiple groundwater level readings were obtained over time, the dates and depths measured, or as a minimum the range of depths measured and the dates the highest and lowest water level readings were obtained. Also a brief description of the method used to obtain groundwater levels (open standpipe, vibratory piezometer, pneumatic piezometer, etc.). Page 5-2 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 • • • • • • Geotechnical Report and Checklist The descriptions of soil and rock conditions illustrated with subsurface profiles (i.e. parallel to road centreline) and cross-sections (i.e. perpendicular to roadway centreline) of the key project features, as appropriate. A subsurface profile or crosssection is defined as an illustration that shows the spatial distribution of the soil and rock units encountered in the borings and probes. As such, the profile or crosssection will contain the existing and proposed ground line, the boring logs (including SPT values, soil/rock units, etc.), and the location of any water level(s). Interpretive information contained in these illustrations should be kept to a minimum. What appears to be the same soil or rock unit in adjacent borings should not be connected together with stratification lines unless that stratification is reasonably certain. The potential for variability in the stratification must be discussed in the report. A subsurface profile for bridges, viaducts, and other significant structures. For retaining walls, subsurface profiles should always be provided for soil nail walls, anchored walls, and non-gravity cantilever walls, and all other walls in which there is more than one boring along the length of the wall. For other wall situations, judgement may be applied to decide whether or not a subsurface profile is needed. For cuts, fills, and landslides, soil profiles should be provided for features of significant length, where multiple borings along the length of the feature are present. Subsurface cross-sections must always be provided for landslides, and for cuts, fills, structures, and walls that are large enough to warrant multiple borings to define the underlying geology. A summary of geological hazards identified and their impact on the project design (e.g. landslides, rock-fall, debris flows, liquefaction, soft ground or expansive soils, etc.), if any. The location and extent of the geologic hazards should be described. For the ofdata unstable slopes (including existing settlement areas), cuts, and fills, the analysis following is needed: o The analytical approach and assumptions used, o Values of the design parameters, o A description of any back-analyses conducted, the results of those analyses, and comparison of those results to any laboratory test data, o Any definition of acceptable factors of safety or discussion of acceptable risk of failure. Proposed cuts and excavations should be considered in terms of temporary (shortterm) and long term and stability analyses performed for those that have a potential for failure. Global and local stability conditions should be analysed as appropriate. The level of analysis should be consistent with the consequences of slope failure. Special attention is required for very high cuts and fills, steep cuts, and cuts with adverse geological structure. The stability analyses used must be appropriate for the slope conditions. For example, a circular failure model should not be used to analyze a cut in rock where discontinuities will control the stability. The method of analysis should be stated, along with the input data and any assumptions made. If stereographic analyses are used, the stereo-nets should be appended and the results of the analyses summarised. Geotechnical recommendations for earthwork (fill design, cut design, usability of on-site materials as fill). The design of embankment features such as fill slope angles, the foundation, and subsurface drainage should include analysis of settlement, slope stability, groundwater conditions, subsidence, compaction characteristics and potential problems with the materials to be used in the embankments. Embankment design recommendations should include the slope Ethiopian Roads Authority Page 5-3 Chapter 5 Geotechnical Report and Checklist Geotechnical Design Manual – 2013 required for stability, any measures that need to be taken to provide a stable embankment (geosynthetic reinforcement, wick drains, controlled rate of embankment construction, light-weight materials, etc.), embankment settlement magnitude and rate, and the need for and extent of removal of any unsuitable materials below. Cut design recommendations should contain the slope angle required for stability, seepage and piping control and erosion control measures needed, as appropriate, and any other special measures required to produce a stable slope. In addition, cut slope and other on-site materials should be identified as to their feasibility for use as fill, with a discussion on the type of fill material for which they could be utilised, the need, if any, for aeration to reduce the moisture • • • content, and the effect of environmental factors on their usability. Geotechnical recommendations for rock slopes and rock excavation. Such recommendations should include any special measures to produce a stable rock slope such as rock bolting/dowelling as well as any recommendations to prevent erosion and undermining of intact blocks of rock, internal and external slope drainage requirements, feasible methods of rock removal and rock excavation, and the need for controlled blasting or any other special techniques that may be necessary. Geotechnical recommendations for stabilization of unstable slopes (e.g., landslides, rock-fall areas, debris flows, etc.). This section should provide a discussion on mitigation options, and detailed recommendations regarding the most feasible methods for mitigating the unstable slopes, including a discussion of the advantages, disadvantages, and risks associated with each option. Geotechnical recommendations for retaining walls and reinforced slopes with a discussion on considered wall/reinforced slope options, the recommended wall/reinforced slope options, foundation type and and design (for strength limit state: ultimate bearing resistance, lateral upliftrequirements resistance if deep foundations have been selected; for service limit state: settlement limited bearing, and any special design requirements), seismic design parameters and recommendations (e.g., design acceleration coefficient, extreme event limit state bearing, uplift and lateral resistance if deep foundations have been selected), design considerations for scour when applicable, and lateral earth pressure parameters. For reinforced slopes requiring internal stability design (e.g., geosynthetic walls, and soil nail walls), recommendations on minimum width for external and overall stability, embedment depth, bearing resistance and settlement, soil reinforcement spacing, strength, and length requirements, and dimensions to meet external stability requirements are needed. For other retaining walls, minimum width for overall stability, embedment depth, bearing resistance, settlement, and design parameters for determining earth pressures should be provided. For anchored walls, achievable anchor capacity, no-load zone dimensions, and design earth pressure distribution. • • Recommendations on aggregate and borrow materials, including sketches of local sources and regional location maps, the quality of materials and their suitability for the different road structures, and estimated quantity. The limits of the material source relative to the proposed alignment should be defined, the approximate quantity of material available described, the amount of overburden to be stripped, and material excavation characteristics. Geotechnical recommendations for bridges and hydraulic structures, foundation options considered, foundation design requirements (for strength limit state: the ultimate bearing resistance and depth, and lateral and uplift resistance; for service Page 5-4 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 • • • 5.3 Geotechnical Report and Checklist limit state: settlement limited bearing, and any special design requirements), seismic design parameters and recommendations (e.g., design acceleration coefficient, soil profile type for response spectra development, liquefaction mitigation requirements, extreme event limit state bearing, uplift, and lateral resistance, and soil spring values), design considerations for scour if applicable, earth pressures on abutments and walls in buried structures, and recommendations regarding bridge approach slabs. Construction considerations. Address issues of construction staging, shoring needs and potential installation difficulties, temporary slopes, potential foundation installation problems, earthwork constructability issues, dewatering, etc. Long-term or construction monitoring needs which should include recommendations on the types of instrumentation required to evaluate long-term performance or to control construction, and the zone of influence for each instrument. Appendices. Typical appendices should include layouts showing boring locations relative to the project features and stationing, subsurface profiles and typical crosssections that illustrate subsurface stratigraphy at key locations, all boring logs used for the project design (includes older borings as well as new borings), including a boring log legend for each type of log, laboratory test data obtained, instrumentation measurement results, and special provisions needed, design charts for foundation bearing and uplift, design detail figures. General Geotechnical Report Outline The objective of using any standard outline to prepare a report is to provide a degree of uniformity in the presentation of geotechnical data, while retaining enough latitude in the contents to reflect the individuality of each project. A sample outline with contents that need to be covered in a geotechnical report for road construction is given below. 1. Project scope: • Proposed construction, reconstruction, resurfacing or rehabilitation • Project limits of the various construction operations • Proposed alignment changes • Specific geometric design features (number of lanes, widths, turn lanes, bypass lanes) • Proposed structures (bridges, culverts, interchanges, etc.) type, number, locations, description of any proposed rehabilitation 2. Project background information: • Site topography Type and amount of relief Land use • Seismicity • Water courses, ponding areas and wetland locations 3. Surface geology and features of interest • General soil types • Extent and uniformity • Bedrock features • • Ethiopian Roads Authority Page 5-5 Chapter 5 Geotechnical Report and Checklist • • • • • • • • Geotechnical Design Manual – 2013 Rock type/formation/extent Weathering conditions General subsurface conditions (depth to bedrock, refusals, etc.) Drilling information (field logs and observations) Soil and bedrock profile, topsoil horizons In-situ (SPT, DCP, CPT, etc.) test results Geophysical information Instrumentation 4. Groundwater conditions • Levels (including types of piezometers) • Springs • Nearby wells 5. Laboratory test data • Optimum moisture contents, gradations, Proctor compaction results, maximum densities, Atterberg Limits, textural classifications, percent passing No. 200 Sieve, CBR, Modulus, etc. 6. Subgrade conditions • Sub-cuts (length, depth, drainage, backfill material, typical section) • Road, widened sections • Turn lanes and curvatures Embankments and road cuts Fill-slope analyses Cut-slope analyses • The use of excavated materials for fill 7. Rock excavation • Recommended techniques of excavation (blasting, ripping, etc.) • Description of excavated material and their use 8. Aggregate and borrow sources • Possible sources of aggregate • Materials available (quality, quantity, use) 9. Landslides and unstable areas • Description of landslides • Causes of landslides • • • Stability analyses • Correction measures 10. Retaining walls and reinforced slopes Design requirements Earth pressures • Foundation analysis 11. Shallow and deep foundations (walls and bridges),construction considerations • • Page 5-6 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 Geotechnical Report and Checklist Appendices • • • Subsurface profiles Laboratory results Photos, etc. • References • • 5.4 Project location map Borehole logs and descriptions Checklist The checklist given in Table 5-1 covers important information that should be presented in geotechnical reports. The advantage of using this checklist is to ensure that pertinent data are not forgotten or overlooked, and to provide an overview of aspects that need to be investigated further. Upon completing the checklist, the geotechnical engineer should summarize the responses and use the result to determine if additional follow-up actions are appropriate. Ethiopian Roads Authority Page 5-7 Chapter 5 Geotechnical Report and Checklist Geotechnical Design Manual – 2013 Table 5-1: Checklist of important information in geotechnical reports Tasks to be completed for geotechnical report Y N Laboratory test data: Were lab soil classification tests such as natural moisture content, gradation, Atterberg limits, performed on selected representative samples to verify field visual soil identification? Are laboratory test results such as shear strength, consolidation, etc., included and/or summarized? Subgrade: Has the subsurface investigation adequately characterized the soil or rock? Are there weak or problem soils at the subgrade level and have soft-spots been properly identified throughout the road alignment? If these materials are to be removed and replaced, have the station limits, depth, and lateral limits for the planned removal been provided? Do subsurface investigations and existing moisture contents for the proposed subgrade soils indicate the need for subgrade stabilization? If stabilization is needed, is the detail of this treatment shown on the report, including depth, station limits, lateral extent, and specifications? Is a design CBR value (or an equivalent of it) provided? If drainage or groundwater is an issue with the proposed subgrade, has an appropriate drainage system (pipe, under drains, etc.) been recommended? Embankments: If soiladdressed? conditions and project requirements warrant, have settlement and stability issues been Have the consolidation properties of the foundation soils been determined? Has the total expected embankment settlement and the time of consolidation been estimated? If differing foundation soil and/or loading conditions occur throughout the embankment area, have sufficient analyses been completed to evaluate consolidation at locations representative of the most critical conditions? Have the total settlement and the time of consolidation analyses indicated acceptable values at all locations for the scope of the embankment work? If total settlement or time of consolidation is unacceptable, have the stations and lateral extent of the problem areas been defined? Has a method (change alignment, lower grade, using stabilizing counter-berms, excavate and replace weak subsoil, lightweight fill, geotextile fabric reinforcement, etc.) been chosen as a solution to the settlement issues? Based on accepted design practices, were there any attempts to evaluate the effectiveness of the chosen solution(s)? Has an economic analysis been performed to evaluate the cost benefits of the recommended solution compared to other methods? For bridge approach embankments, are recommendations made to accelerate the settlement before the bridge abutment is constructed (waiting period, surcharge, or wick drains)? Has the effect of any foundation soil consolidation (including differential settlement) been evaluated with regard to adjacent structures (e.g., bridges, buildings, culverts, utilities) which may also undergo settlement due to the consolidation of the surrounding Page 5-8 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 Geotechnical Report and Checklist soil? Has the total (short term) and effective (long term) shear strength of the foundation soils been determined? Have calculations been performed to determine the F.S. for stability? Are the following Factor of Safety (F.S.) criteria achieved or exceeded, as determined by the calculations, for the given stability conditions: 1.30 for short and long term conditions, 1.10 for rapid drawdown (flood condition), and 1.50 for embankment supporting bridge abutments? When differing soil or loading conditions occur throughout the embankment area, are station to station recommendations included for fill slope design? If the F.S. was not met or exceeded, have the stations and lateral extent of the problem areas been defined? Has a method been chosen as a solution to the stability issues? Has an economic analysis been performed to evaluate the cost benefits of the recommended solution compared to others? Has the effect of the solution being used been evaluated with regard to structures (e.g., bridges, buildings, culverts, utilities) which may be subject to unusual stresses or require special construction considerations? If geotechnical instrumentation is proposed to monitor fill stability and settlement, are detailed recommendations provided on the number, type, and specific locations of the proposed instruments? If piezometers will be used, has the critical groundwater table been determined and the appropriate information included in the reports Are water-bearing zones properly identified and their impact addressed? Cuts and excavations: Does drilling provide continuous stratigraphic sections for the range of elevations that represent proposed cut slope areas? Do the cut slopes have a minimum stability F.S. of 1.30 If there is a potentially unstable soil or rock layer within the cut slopes, was this layer considered as a possible failure zone? Have erosion protection measures been addressed for back slopes, side slopes, and ditches (including riprap recommendations or special slope treatments)? Is the usage of excavated soils and rocks properly addressed? Are station to station recommendations included for excavation limits of unsuitable materials, erosion protection measures for back slopes, side slopes, and ditches, and specific drainage structures? Are there evidences of springs and excessively wet areas? Did the design consider additional drainage in the cut slope (springs / seeps) and roadway base? Are recommended slope designs and blasting specifications provided for rock slopes? Are rock slope designs based on orientation of major rock joints and weathering conditions rather than “template” design procedures (such as designing all rock slopes at 0.25:1)? Is the effect of blast-induced vibrations on adjacent structures and slopes evaluated? Was controlled blasting considered in the design? Landslides: Has a site investigation been conducted to define the limits and depth of the landslide? Is a site plan and scaled cross-section provided showing ground surface conditions both Ethiopian Roads Authority Page 5-9 Chapter 5 Geotechnical Report and Checklist Geotechnical Design Manual – 2013 before and after failure? Was the history of the landslide area studied, including movement history, maintenance work, and past corrective measures? Has a site specific geotechnical investigation been performed to investigate the landslide area? Has the vertical and lateral extent of defined landslide conditions been included on cross sections and profile sheets? Are detailed slide features, including location of ground surface cracks, head scarp, and toe bulge, shown on these cross sections and profiles? In the case of rock-falls, were the cause and effect of any existing rock-fall conditions determined? Were bedding and jointing of the bedrock formations identified as a significant factor affecting the slope stability? Was there any groundwater monitoring programme to identify the phreatic surface through the landslide area? Is the landslide failure plane and mode of failure determined from field observations or instrumentation? If groundwater (static or flowing) significantly influences the stability of the landslide, has the source of recharge been identified? Is a stability analysis carried out to determine the F.S. for stability? Is a Factor of Safety (F.S.) of 1.3 achieved or exceeded, as determined by the calculations, for the given stability conditions? Is a landslide correction or stabilization method determined? Has a cost comparison been performed to evaluate the effectiveness of stabilization alternatives? Is long-term monitoring necessary? Retaining walls: Is/are the most suitable and cost-effective wall type(s) selected for the project site conditions? Are reasons given for the choice and/or exclusion of certain wall types (gravity, reinforced soil, tieback, cantilever, gabion, etc.)? Are the soil strength parameters used to compute the design factor of safety for overturning, sliding, and external slope stability? Are the groundwater level and proper loading conditions known? If applicable, has the influence of groundwater been taken into account with regards to soil unit weights and active pressures? Are standard methods used to determine the lateral earth pressures? Was an economic analysis performed to evaluate the cost benefits of the chosen wall type compared to others? Does the design lateral earth pressure include the effects of soil backfill strength, slope geometry, and surcharge loads? Are all the required F.S. computed? Do the F.S. values meet or exceed the minimum: Bearing Capacity (F.S. = 3.0), External Stability (F.S. = 1.3) Overturning (F.S. = 2.0), Sliding (minimum F.S. = 1.50)? If poor foundation soils are present, has a solution been determined with respect to excessive settlement, inadequate bearing capacity, sliding, and global stability? If special drainage details are needed behind and/or beneath the wall, are recommended details provided in the geotechnical report? Page 5-10 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 Geotechnical Report and Checklist Are excavating requirements covered (safe slopes for open excavations, need for sheeting or shoring)? Has the effect of the wall design and construction procedure been determined and accounted for on the construction schedule of the road? Are all the necessary notes, specifications, special provisions, and details for the construction of the wall system included in the geotechnical report? Ethiopian Roads Authority Page 5-11 Chapter 5 Geotechnical Design Manual - 2013 6 Geotechnical Report and Checklist REFERENCES AND BIBLIOGRAPHY References AASHTO (1993). Guide for Design of Pavement Structures . American Association of State Highway and Transportation Officials, Washington D.C., USA. ASTM 4546. Standard Test Methods for One-Dimensional Swell or Collapse of Cohesive Soils. ASTM International, West Conshohocken, PA, 2003, DOI: 10.1520/C0033-03, www.astm.org Abramson, LW, Lee, TS, Sharma, S and Boyce, GM. (2001). Slope Stability and Stabilisation Methods, J Wiley. Austroads (1998). Guide to Stabilisation in Roadworks. Austroads, Sydney, Australia. BRAB (1978). The Design and Construction of Residential Slabs-on-Ground. State of the Art. Building Research Advisory Board, Washington, DC, USA. Bishop, AW and Bjerrum, L. 1960. The relevance of the triaxial test to the solution of stability problems. Proc ASCE Conf on Shear Strength of Cohesive Soils. Boulder, Colorado, 437-501. Briaud, JL, James, RW and Hoffman, SB. (1997). Settlement of Bridge Approaches. NCHRP Synthesis 234. Transportation Research Board/NRC, Washington, DC. CED Engineering. (1998). Geosynthetic Engineering: Geosynthetic Separators. Course No G04-005 Chowdury, R, Flentje, P and Bhattacharya. (2009). Geotechnical Slope Analysis, CRC Press. Coggan, J.S., Stead, D., and Eyre, J.M. (1998). Evaluation of Techniques for Quarry Slope Stability Assessment. Institute of Mining and Metallurgy, London, UK. Cruden, D and Varnes, DJ. (1996) Landslide types and processes. In: Landslides, Investigation and Mitigation. Spec report 247, TRB/NRC Turner, AK and Schiuster, RL eds. Eberhardt, E. (2003). Rock Slope Stability Analysis – Utilization of Advanced Numerical Techniques. Earth and Ocean Sciences, University of British Columbia, Vancouver, Canada. FAO. (1998). Watershed management field manual. Ch 6 in: Road Construction Techniques 13/5. GWP Consultants. 2008. Slope Design. Quarry Design Handbook, pre-publication draft, Appendix 4-4 Guyer, P. (2011). Soil Stabilisation for Pavements. An Engineering SoundBite. Hearn, G.J. and Hunt, T. (2011). C1. Route corridor and alignment selection . In: Hearn, G.J. (ed). Slope Engineering for Mountain Roads . Engineering Geology Special Publication No 24, Geological Society of London, 135-144. Holtz, R.D. and Kovacs, W.D. (1981). An Introduction to Geotechnical Engineering. Prentice-Hall Inc., Eaglewoods Cliffs, N.J., USA. Ethiopian Roads Authority Page 6-1 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 JKR. (2012). Guidelines for Slope Design. JKR 21500-0011-10. Slope Engineering Branch, Malaysia. Keller, G and Sherar, J. (2011). Slope stabilisation and stability of cuts and fills. Ch 11 in: Low Volume Roads Engineering. Best Management Practices. Field Guide, US Dept Agriculture Forest Service. MPWT (2008). Slope Maintenance Transport, Vientiane, Laos. Manual. Ministry of Public Works and Montana DOT. (2008). Laboratory testing. Ch 9 in: MDT Geotechnical Manual. Montana Department of Transportation. National Cooperative Highway Research Program. (1997). Harmonised Test Methods for Laboratory Determination of Resilient Modulus for Flexible Pavement Design. 1-28A, Transportation Research Board, Washington DC, USA. National Cooperative Highway Research Program (2001). Guide for Mechanistic and Empirical - Design for New and Rehabilitated Pavement Structures. FHWA/TRB/NRC. National Cooperative Highway Research Program (2004). Development of Project 1-37A Design Guide Using Mechanistic Principles to Improve Pavement Design. Transportation Research Board, Washington D.C., USA. National Lime Association (1972). Lime Stabilization Manual. National Lime Association, Arlington, Virginia, USA. Natural Resource Managament. 2008. Dispersive Soils - High Risk of Tunnel Erosion, Southern Tasmania. Nettleton, IM, Martin, S, Hencher, S and Moore, R. (2005). Debris flow types and mechanisms. In: Debris Flow Risk Assessment and Mitigation on the Scottish Trunk Road Network. Scottish Executive, Crown Copyright. Ortigao, JAR and Sayao, A. (2004). Handbook of Slope Stabilisation, Blackwell/Springer, Berlin. Paige-Green, P. 2008. Dealing with road subgrade problems in Southern Africa. 12th International Conference of the International Association for Computer Methods and Advances in Geomechanics, 4345-4353. Pettifer, GS and Fookes, PG. (1994). A revision of the graphical method for assessing the excavatibility of rock. Quart Jnl Engng Geol, 27, 145-164. Rollings, M.P. and Rollings, R.S. (1996). Geotechnical Materials in Construction. New York, McGraw-Hill, USA. Sassa, K and Canuti, P. (2008). Landslides - Disaster Risk Reduction. Springer. US ACE. (2002). Use of Geogrids in Pavement Construction. ETL - 1 - 188. Department of the Army, US Army Corps of Engineers. US ACE. (2003). Use of Geogrids in Pavement Construction. ETL - 1 - 189. Department of the Army, US Army Corps of Engineers. US DOI Bureau of Reclamation. (1991). Characteristics and Problems of Dispersive Clays. P-91-09. Materials Engineering Branch, US Department of the Interior, Denver, Colorado US DOI Bureau of Reclamation. (1998). Engineering Geology Field Manual. Vol 1 and 2., Page 6-2 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 Geotechnical Report and Checklist US Department of the Interior, Denver, Colorado.. US DOT FHWA (2006A). Geotechnical Aspects of Pavements. NHI 05-037. Dept of Transportation, Federal Highways Administration, Washington DC, USA. US DOT FHWA (2006B). Soil and Foundations. Reference Manual. CED Engineering, FHWA NHI-06-088. Dept of Transportation, Federal Highways Administration, Washington DC, USA. US DOT FHWA CFLHD. (2011). Rock excavation methods. Ch 3 in: Context Sensitive Rock Slope Design Solutions. Publ No Federal Highways Authority, FHWA-CFL/TD-11002, Central Federal Lands Highway Division, Colorado. University of Iowa. 2013. Foundations on weak and/or compressible soils. Foundation Engineering, Foundations of Structures, University of Iowa, 53:139. Washington State DOT. (2013). Soil cut design. Ch 10 in: Geotechnical Design Manual M46-03.08, WS Department of Transportation. Bibliography APAI (undated). Asphalt Paving Design Guide. Asphalt Paving Association of Iowa. Abramson, LW, Lee, TS, Sharma, S and Boyce, GM. (2001). Slope Stability and Stabilisation Methods, J Wiley. BRAB (1978). The Design and Construction of Residential Slabs-on-Ground. State of the Art. Building Research Advisory Board, Washington, DC, USA. CED Engineering. (1998). Geosynthetic Engineering: Geosynthetic Separators. Course No G04-005. Caltrans. (2004). Retaining Walls. Ch 3 in: Bridge Design Specification. California Department of Transportation. Caltrans (2013). Geotechnical Manual. California Department of Transportation. Eberhardt, E. (2003). Rock Slope Stability Analysis – Utilization of Advanced Numerical Techniques. Earth and Ocean Sciences, University of British Columbia, Vancouver, Canada. GWP Consultants. (2008). Slope Design. Quarry Design Handbook, pre-publication draft, Appendix 4-4. Illinois DOT. (2005). Subgrade Stability Manual. Bureau of Bridges and Structures, Illinois Department of Transportation. Indiana DOT. (2008). Design Procedures for Soil Modification or Stabilisation. Indiana Dept of Transportation. Iowa State. (2013). Embankment Construction. Ch 6D in: Geotechnical Design Manual. Revised Edition, Statewide Urban Design and Specification (SUDAS), Iowa State. Lexington-Fayette Urban County Government. (2005). Engineering Analysis and Evaluations. Ch 6 in: Geotechnical Manual. Lexington, Kentucky. Montana DOT (2009). Measurement and Evaluation of Subgrade Parameters: Phase 1 Synthesis of Literature. State of Montana Dept of Transportation and US DOT FHWA, MDT Research Programs. Ethiopian Roads Authority Page 6-3 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 National Cooperative Highway Research Program. (2004). Guideline and Recommended Standard for Geofoam Applications in Highway Embankments. NCHRP Report 529. Transportation Research Board, Washington, DC, USA. New York State, DOT. (2013). Reporting and Documentation. Ch 26 in: Geotechnical Design Manual. New York State Department of Transportation. Pavement Interactive (2007). Triaxial Test. www.pavementinteractive.org/article/triaxialtest Geotechnical South Carolina Design Manual.Dept of Transportation. (2010). Embankments. Ch 17 in: US ACE. (1984). Stabilisation with lime. Ch 4 in: Soil Stabilisation for Pavements, Mobilisation, Construction. EM 1110-3-137. US Army Corps of Engineers, Department of the Army. US ACE.(1994). Cut slope stability. Ch 8 in: Slope Stability. EM 1110-2-1902, Department of the Army, US Army Corps of Engineers. US ACE. (2002). Use of Geogrids in Pavement Construction. ETL - 1 - 188. Department of the Army, US Army Corps of Engineers. US ACE. (2003). Use of Geogrids in Pavement Construction. ETL - 1 - 189. Department of the Army, US Army Corps of Engineers. US ACE. (2010). Constructively Speaking, Collapsible Soils, Issue No 3, Afghanistan Engineer District. US Army Corps of Engineers. Military Soils US ACE. (2012). Soil stabilisation3-34.64. for roads and airfields. Ch 9 in: Engineering. MCRP 3-17.7G/TM FM 5-410. US Army Corps of Engineers. First published 1992. US DOI Bureau of Reclamation. (1991). Characteristics and Problems of Dispersive Clays. P-91-09. Materials Engineering Branch, US Department of the Interior, Denver, Colorado. US DOI Bureau of Reclamation. (1998). Engineering Geology Field Manual. Vol 1 and 2., US Department of the Interior, Denver, Colorado. US DOT FHWA. (1975). A Review of Engineering Experiences with Expansive Soils in Highway Subgrades. Report No FHWA-RD-75-48. US Army Engineer Waterways Experiment Station. US DOT FHWA (2001). Mechanically Stabilized Earth Walls and Reinforced Soil Slopes Design and Construction Guidelines. Department of Transportation, Federal Highway Authority Publication No. NHI-00-043. Washington D.C., USA. US DOT, FHWA. (2003). Checklist and guidelines for review of geotechnical reports and preliminary plans and specifications. Department of Transportation, Federal Highways Authority Publ No ED-88-053, Washington DC, USA. US DOT FHWA (2006A). Geotechnical Aspects of Pavements. Dept of Transportation, Federal Highways Authority, NHI 05-037, Washington, DC, USA. US DOT FHWA (2006B). Soil and Foundations. Reference Manual. CED Engineering, FHWA NHI-06-088. Dept of Transportation, Federal Highways Authority, Washington DC, USA. Page 6-4 Ethiopian Roads Authority Chapter 5 Geotechnical Design Manual - 2013 Geotechnical Report and Checklist US DOT FHWA CFLHD. (2011). Rock excavation methods. Ch 3 in: Context Sensitive Rock Slope Design Solutions. Publ No Federal Highways Authority, FHWA-CFL/TD-11002, Central Federal Lands Highway Division, Colorado. University of Iowa. 2013. Foundations on weak and/or compressible soils. Foundation Engineering, Foundations of Structures, University of Iowa, 53:139. Washington State DOT. (2013). Soil cut design. Ch 10 in: Geotechnical Design Manual M46-03.08, WS Department of Transportation. Ethiopian Roads Authority Page 6-5 Appendix A Geotechnical Design Manual - 2013 Soil Stabilization APPENDIX A - SOIL STABILIZATION Soil stabilization is a general term that involves the use of mechanical or chemical modifiers to enhance the strength of soils and reduce the change in moisture. The process is often called soil modification when the purpose is to change the physical properties and thereby improve the quality of the subgrade soil. Soil stabilization is usually performed for the following reasons: • • As a construction platform to dry very wet soils and facilitate compaction of the upper layer. For this case, the stabilized soil is usually not considered as a structural layer in the pavement design process. To strengthen a weak soil and restrict the volume change potential of a highly plastic (expansive) or compressible soil. For this case, the modified soil is usually given some structural value in the pavement design process. The thickness, depth or zone of the subgrade that may be selected for soil improvement depends upon a number of factors. Among these are the anticipated traffic loads, the importance of the transportation network, constructability, the drainage characteristics, the geometric design, and the purpose of stabilization. When only a thin zone or short roadway length is subject to improvement, removal and replacement can usually be the preferred alternative, unless a suitable replacement soil is not economically available. The thickness of the subgrade to be treated is based primarily on the project economics and the objective of stabilization. Table A-1 presents a summary of the stabilization methods used in pavement design and constructions the types of soils for which they are most appropriate, and their intended effects on soil properties. Mechanical stabilization using thick gravel layers, in conjunction with geotextiles or geogrids is an effective technique for improving roadway support over soft, wet subgrades. The use of gravel layers has been discussed in Section 2.4.2. Blending gravel and, more recently, recycled pavement material with poorer quality soils can also provide a working platform. The gravel acts as filler, creating a dryer condition and decreasing the influence of plasticity. However, if saturation conditions return, the gravel blend can take on the same poorer support characteristics of the natural subgrade. Stabilization with admixtures, such as lime, cement, and asphalt, has been used outside Ethiopia to control the swelling and frost heave of soils and improve the strength characteristics of unsuitable soils. Literally hundreds of chemicals or additives have been tried in the past few decades. To stabilize expansive subgrades and reduce volume change for example, ion exchange (addition of divalent and trivalent salts), cation fixation in expanding clays (with potassium), deactivation of sulphates (with calcium chloride), waterproofing (silicones, asphalts), cementation (silicates, carbonates, phosphoric acid), and alteration in permeability and wetting properties (surface active agents, wetting agents) have all been attempted. However, lime continues to be the most widely used and most effective additive for stabilization of subgrades and rapid gain of strength. A flow chart for the determination of chemical treatment options for soil stabilization based on the percent of materials that pass the No. 200 sieve (75 µm) and the PI of the soil is shown in Figure A-1. This chart should only be used as a broad guideline. Ethiopian Roads Authority Page A- 1 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 Table A-1 Stabilization methods for pavements. From Rollings & Rollings (1996) reproduced in US DOT FHWA (2006A) Stabilization method and materials used More gravel la ic n Blending a h ec M Geosynthetics Portland cement Soil type Silts and clays Moderately plastic Bitumen Remarks None Reduce dynamic stress level None Too difficult to mix Improved gradation, Others reduced plasticity, reduced breakage Silts and clays Strength gain through minimum disturbance and consolidation Fast, plus provides longterm separation Plastic ------ Less pronounced hydration of cement Coarse ------ Hydration of cement Plastic Rapid drying, rapid strength gain, reduce plasticity rapidly, coarsen texture rapidly, slow long-term pozzolanic cementing ------ Coarse with fines Same as plastic soils Non-plastic None of plastic fines No reactive material Same as lime Same as lime Covers broader range Same as lime Same as lime Covers broader range Coarse Strengthen/bind waterproof Asphalt cement or liquid asphalt Lime es rx tu i m Lime-fly ash d A Lime-cementfly ash Improvement Dependent on quantity Some fines Same as coarse Liquid asphalt Fines None Cannot mix Pozzolans and slags Silts and coarse grained material Acts as a filler, cementing of grains Dense and strong, slower than cement Chemicals Plastic Strength increase and volume stability Difficult to mix Plastic and Reduce change in Long-term moisture collapsible Plastic and collapsible moisture Reduce change in moisture migration problem Long-term moisture migration problem sr Asphalt tera o feo r W p Geomembranes Page A- 2 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization Figure A-1 Guide for selection of admixture stabilization methods. From Austroads (1998), reproduced in US DOT FHWA (2006A) A.1 Lime modification Lime treatment or modification consists of the application of 1 – 3% hydrated lime to aid drying of the soil and permit compaction. Lime modification may also be considered to condition a soil for follow-on stabilization with cement or asphalt. Lime treatment of subgrade soils is intended to speed up construction, and no reduction in the required pavement thickness should be made. Lime may also be used to treat expansive soils. If it has been determined that a soil has potential for excessive swell, lime treatment may be appropriate. Lime will reduce swell in an expansive soil to greater or lesser degrees, depending on the activity of the clay minerals present. The amount of lime to be added is the minimum amount that will reduce swell to acceptable limits. In the case of expansive subgrades, efforts have been directed elsewhere towards modifying deeper layers using electrical, drill-hole, pressure injection, and deep-plough techniques. This is in addition to the conventional mix in-place or batch mixing approaches. Generally, the use of an electric potential to increase the rate of lime migration has showed little success and is rarely used. The use of drill-holes consists basically of drilling holes into the subgrade and backfilling with a lime slurry or lime slurry-sand mixture. Once placed in the holes, the lime migrates or diffuses into the soil initiating the soil-lime reactions. However, this diffusion process can be quite slow and time-consuming before a substantial quantity of the soil is affected. This technique has been mainly used for rehabilitation projects and new construction. Generally, 300 mm diameter holes with depths ranging from 1.8 m to 6 m, depending upon the extent of treatment desired, on 1.5 m by 1.8m grids, are needed for better results. Some Ethiopian Roads Authority Page A- 3 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 experiences showed that lime migration was quite limited to the periphery of the hole. However, a definite improvement in serviceability index was also noted for treated sections compared to companion untreated areas. Such techniques are better suited for treating existing failed pavements on highly trafficked roads due to cost. For new roads other more economic measures should be adopted such as excavation and replacement and drainage. In an attempt to obtain greater distribution of lime in swelling subgrades, the technique of lime slurry pressure injection (LSPI) was developed. The technique consists of pumping lime slurry under pressures of up to 1400KPa, depending upon soil conditions, through hollow injection rods into the subgrade. The injection rods penetrate the soil in approximately 300 mm intervals, and the slurry, 1 kg to 1.5 kg of lime per gallon of water, is injected to refusal. Refusal is reached when soil will not take additional slurry, the slurry is running freely either around the pipe or out of previous injection holes, or the slurry has fractured the surface and is flowing out. A wetting agent is often added to the slurry to assist in migration. The lime slurry left on the surface immediately following injection is mixed into the top 100-150mm of soil and re-compacted. A.2 Lime stabilization Lime stabilization is a technique for improving soft and weak subgrades beneath flexible pavements. A primary benefit of lime stabilization is a greatly increased stiffness as a function of the lime content. Consequently, the effect of lime content on subgrade properties and the thickness of the stabilized subgrade are the primary variables for stabilization design. Lime is applicable in clay soils (CH, CL, MH and ML and in granular soils containing clay binder (GC and SC, with a PI greater than 10 and with at least 10% passing the No. 40 sieve. Lime reduces the PI and renders a clay soil less sensitive to moisture changes. The most common varieties of lime for soil stabilization are hydrated lime (Ca(OH) 2), quicklime (CaO), and the dolomitic variations of these high-calcium limes (Ca(OH) 2⋅MgO and CaO⋅MgO). Often, hydrated lime is used in powder form, or as slurry (a mix of water and lime). While hydrated lime remains the most commonly used lime stabilization admixture in many countries, use of the more caustic quicklime has also grown steadily over the past two decades. Lime is usually produced by calcining limestone or dolomite, although some lime, typically of more variable and poorer quality, is also produced as a by-product of other chemical processes. For lime stabilization of clay (or highly plastic) soils, the lime content should be from 3 – 8% of the dry weight of the soil, and the cured mass should have an unconfined compressive strength of at least 350 KPa within 28 days. The optimum lime content should be determined with the use of unconfined compressive strength and Atterberg limits measured for different lime-soil mixtures moulded at varying percentages of lime. The pH values can be used to determine the initial, near optimum lime content value. The pozzolanic strength gain in clay soils depends on the specific chemistry of the soil (if it has sufficient silica and alumina minerals to support the pozzolanic reactions). Plasticity is a rough indicator of reactivity. A PI of about 10 is commonly taken as the lower limit for suitability of inorganic clays for lime stabilization. The lime-stabilized subgrade should be compacted to a minimum density of 95%, as defined by AASHTO T99. Page A- 4 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization During stabilization, hydration of the lime absorbs water from the soil and causes an immediate drying effect. The addition of lime also introduces calcium (Ca+2) and magnesium (Mg+2) cations that exchange with the more active sodium (Na+) and potassium (K+) cations in the natural soil water chemistry. This cation exchange reduces the plasticity of the soil, which, in most cases, corresponds to a reduced swell and shrinkage potential, diminished susceptibility to strength loss with moisture, and improved workability. The changes in the soil-water chemistry also lead to agglomeration of particles and a coarsening of the soil gradation; plastic clay soils become more like silt or sand in texture after the addition of lime. These drying, plasticity reduction, and texture effects all occur very rapidly (usually with one hour after addition of lime), provided there is a thorough mixing of the lime and the soil. When soils are treated with lime, it has been observed that the lime-soil mixture may be subject to durability problems, the cyclic freezing and thawing of the soil. The durability of lime stabilization on swell potential and strength may be adversely affected by environmental influences such as water, leaching processes, carbonation, and sulphate attacks. Although most lime stabilized soils retain 70% to 85% of their long-term strength gains when exposed to water, there have been instances of poor strength retention when exposed to soaking. Therefore, testing of stabilized soils in the soaked condition is prudent. Freeze and thaw cycles can lead to strength deterioration, but this is not a problem in many most parts of Ethiopia. If the problem is expected, then the most common design approach is to specify a sufficiently high initial strength gain to retain sufficient residual strength after freeze and thaw damage. Leaching of calcium can decrease the cation exchange in lime stabilized soils, which, in turn, can reverse the beneficial reduction in plasticity and swell potential. The potential for these effects is greater when low lime contents are used. Carbonation is another problem that is common after lime stabilization. If atmospheric carbon dioxide combines with lime to form calcium carbonate, the calcium silicate and calcium aluminate hydrate cements may become unstable and revert back to their srcinal silica and alumina forms, reversing the long-term strength increase resulting from the pozzolanic reactions. Carbonation may be minimized by the use of ample lime content, careful selection, placement, and compaction of the stabilized material to minimize carbon dioxide penetration, as well as prompt placement after lime mixing, and good curing. Sulphates present in the soil or groundwater can combine with the calcium from the lime and the alumina from the clay minerals to form ettringite (a hexacalcium aluminate trisulphate hydrate, common in hydrated Portland cement), which has a volume that is more than 200% larger than that of its constituents. Massive irreversible swelling can, therefore, occur in the subgrade, and the damage it causes can be quite severe. It is difficult to predict the exact combinations of sulphate, lime, clay mineralogy, and environmental conditions that will trigger sulphate attack. Consequently, if there is a suspicion of possible sulphate attack, the lime stabilized soil should be tested in the laboratory to see whether it will swell when mixed and exposed to moisture. Conventional soil-lime construction techniques are normally limited to maximum depths of 200 – 300 mm. However, if greater depths of stabilization are required in one lift, these conventional techniques are inadequate. The deep plough lime stabilization method was created in light of this problem. The technique consists of: Ethiopian Roads Authority Page A- 5 Appendix A Soil Stabilization • • • • • • • Geotechnical Design Manual – 2013 ploughing the subgrade to a depth of 300 mm prior to spreading the lime; spreading the lime required for stabilization of the layer; mixing the lime and soil with three passes of the plough to a depth of 600 mm; spraying water over the subgrade after initial dry mixing; final mixing using a deep ripper; compacting the 600 mm depth of stabilized material in one lift using either sheepsfoot or vibratory sheepsfoot rollers, and finally; compacting and test rolling using a 50-ton roller making six passes. A special three-toothed ripper attachment with a trapezoidal shaped shoe-plough bolted to the teeth is normally used for ripping operations. Tests taken at various depths (0 – 200 mm, 200 - 400 mm, and 400 - 600 mm) should reveal that adequate densities (greater than 95% at standard compaction), are obtained within the subgrade. In order to choose the preferred lime content in a subgrade soil, the preferred method is to prepare several mixtures at different lime treatment levels and determine the pH of each mixture after one hour. The pH is a good indicator of the desirable lime content of a soillime mixture. In this technique, the lowest lime content producing the highest pH of the soil-lime mixture can be taken as the initial design lime content. Most lime, when placed in a water solution, has a pH of about 12.4. The reaction that takes place when lime is introduced to a soil generally causes a significant change in the plasticity (plastic and liquid limits) of the soil. The following is the procedure to determine the optimum lime content using a pH test. • • • • • • • • • • A sufficient amount of lime should be added to soils to produce a pH of 12.4 or equal to the pH of lime itself. The optimum lime content should be determined corresponding to the maximum pH of the lime-soil mixture. Representative air-dried samples, equal to 20 g of oven-dried soil, are weighed to the nearest 0.1 g and poured into 150-ml (or larger) plastic bottles with screw on tops. It is advisable to set up five bottles with lime content of 3, 4, 5, 6, and 7 %. This will insure that the percentage of lime required can be determined in one hour. Weigh the lime to the nearest 0.01 g and add it to the soil. Shake the bottle to mix the soil and dry lime. Add 100 ml of CO2 free distilled water to the bottles. Shake the soil-lime mixture and water until there is no evidence of dry material on the bottom. Shake for a minimum of 30 seconds. Shake the bottles for 30 seconds every 10 minutes. After one hour, transfer part of the slurry to a plastic beaker and measure the pH. The pH meter must be equipped with an electrode and standardized with a buffer solution having a pH of 12.00. Record the pH for each of the lime-soil mixtures. If the pH readings go to 12.40, then the mix with the lowest percent of lime will be chosen to stabilize the soil. If the pH does not go beyond 12.30 and two or more mixes gave the same readings, the mix with the lowest percent of lime is the amount required to stabilize the soil. If the highest pH is 12.30 and only one mix gave this value, then additional test bottles should be prepared with larger percentages of lime. Besides, additional tests are needed if the pH is anything below 12.30 for various reasons. Page A- 6 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization An alternate method of determining initial design lime content is by the use of Figures A.2. Specific values required to use this figure are the PI and the % of material passing the No. 40 sieve. These properties are determined from the Atterberg limit and gradation tests on the untreated soil. The following procedures are needed in order to estimate the initial lime content from Figures A-2. • • • • • • • Determine the PI (wet method) and the percent of soil binder (% passing No. 40 sieve). Mark the % of soil binder on the Y-axis of Figure A.2 and the PI on the upper Xaxis. Enter the plot from the upper X-axis where the PI value is marked. Follow the curved line starting from the PI value down to the intersection point with the horizontal line (drawn from the percent of soil binder mark on the Y-axis). At the intersection with the percent of soil binder mark, move vertically upward to the 100% soil binder line. Read the percent of lime represented at the intersection with the 100% soil binder line. For example, for a soil having a P1 of 39% and 55% soil binder, the lime required is about 4.25% as shown in Fig A.2. Figure A-2 Chart to determine the initial lime content. From National Lime Association USA (1972) Ethiopian Roads Authority Page A- 7 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 A.3 Cement stabilization Portland cement is used widely for stabilizing low-plasticity clays, and sandy and granular soils, and improves strength and stiffness. Increasing the cement content increases the quality of the mixture. At low cement contents, the product is termed a cement-modified soil, which has improved properties of reduced plasticity or expansive characteristics. At higher cement contents, the end product is termed soil-cement. High cement contents will unavoidably induce high incidences of shrinkage cracking caused by moisture and temperature changes. For withof cement, proper mixing the soilshighly have aplastic PI of less soils than to 20 be andstabilized a minimum 45% passing the No. requires 40 sieve.that However, clays that have been pre-treated with lime are sometimes suitable for subsequent treatment. For cement stabilization of granular and/or non-plastic soils, the cement content should be 3 to 10% of the dry weight of the soil, and the cured material should have an unconfined compressive strength of at least 1050 KPa within seven days. The Portland cement should meet the minimum requirements of AASHTO M 85. The cement-stabilized subgrade should be compacted to a minimum density of 95%, as defined by AASHTO T 134. Only fine-grained soils can be treated effectively with lime for marginal strength improvement. Small amounts of Portland cements may reduce swell potential of subgrade soils. However, Portland cement generally is not as effective as lime, and may be considered too expensive for this sort of application. The determination of cement content to reduce the swell potential of fine-grained plastic soils can be accomplished by moulding several samples at various cement contents and soaking the specimens along with untreated specimens for four days. The lowest cement content that eliminates or reduces the swell potential to the minimum is the design cement content. Procedures for measuring swell characteristics of soils are found in ASTM D 4546. The cement content that accomplishes soil modification should be checked to see whether it provides an unconfined compressive strength great enough to qualify for a reduced pavement thickness design (Guyer 2011). A.4 The application of bitumen Often, bitumen-stabilized soils have been used elsewhere in the world for sub-base construction for engineering and economic reasons. Use of bitumen as a stabilizing agent produces varied effects, depending on the soil, and may be divided into three major groups: • • • Sand-bitumen, which increases strength in cohesionless soils, such as clean sands, or acts as a binder or cementing agent. Soil-bitumen, which stabilizes the moisture content of cohesive fine-grained soils. Sand-gravel bitumen, which provides cohesive strength and waterproofing for pitrun gravelly soils with inherent frictional strength. The durability of bitumen-stabilized mixtures can be assessed by measurement of their water absorption characteristics. Bitumen increases the cohesion and load-bearing capacity of the soil and renders it resistant to the action of water. Usually, bituminous stabilization is performed in place with the bitumen being applied directly on the soil or soil-aggregate system with the mixing and compaction operations being conducted immediately thereafter. Page A- 8 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization A.5 Geosynthetics Geosynthetics are a class of geo-materials that are used to improve soil conditions for a number of applications. They consist of polymeric materials manufactured in different forms. The most common applications to roads are for reinforcing embankments and foundation soils, creating barriers to water flow in liners and cut-offs, and improving drainage. A.5.1 Types of geosynthetics The generic term “geosynthetic” often used to cover a and widegeomembranes. range of different artificially manufactured materials, includingisgeotextiles, geogrids, A geotextile is a permeable geosynthetic comprised solely of textiles. These materials are also known as engineering fabrics, which are usually created from polymers, most commonly polypropylene, but also potentially including polyester, polyethylene, or nylon. Geotextiles are usually classified as either woven or non-woven. Both use a polymer fibre as raw material. Depending on the application, the fibres may be used singly or spun into yarns by wrapping several fibres together, or created by a slit film process. Woven geosynthetics are manufactured by weaving fibres or yarns together in the same way as any form of textile, although generally only fairly simply weaving patterns are used. Nonwoven geosynthetics are made by placing fibres in a bed, either in full-length or in short sections. The fibres are then bonded together, either by raising the temperature, applying an adhesive chemical, or by mechanical means. Geogrids, as their name suggests, consist of a regular grid of plastic with large openings (called apertures)soil between the tensile elements. Thethe function of the is to allow the surrounding materials to interlock across plane of the apertures geogrid. Hence, the selection of the size of the aperture is partially dependent on the gradation of the material into which it will be placed. The geogrid is manufactured using high-density polymers of higher stiffness than are common for geotextiles. These polymers are then punched in a regular pattern and drawn in one or two directions. Alternatively, a weaving process may be used in which the crossing fibres are left wide apart and the junctions between them are reinforced. Existing commercial geogrid products include extruded geogrids, woven geogrids, welded geogrids, and geogrid composites. Geomembranes are used to retard or prevent fluid from penetrating the soil and as such consist of continuous sheets of low permeability materials. These materials are made by forming the polymer into a flat sheet, which may have a roughened surface created to aid in the performance of the membrane by increasing friction with the adjacent soil layer. Geocells are designed to protect slopes against erosion, stabilize steep slopes, provide protective linings for channels, support heavy construction traffic on weak subgrade soils, and provide multi-layered earth-retaining structures. Geocells are typically constructed of high-density polyethylene (HDPE). The cells in three dimensional panels are opened and filled with granular material, which adds weight to make the multi-layer system act as a gravity retaining wall. Geocomposite materials are often created by combining two or more of the products described above to take advantage of multiple benefits. Furthermore, geocomposites may be formed by combining geosynthetics with more traditional geo-materials, the most Ethiopian Roads Authority Page A- 9 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 common example being the geosynthetic clay liner. A geosynthetic clay liner consists of a layer of bentonite sandwiched together with geomembrane or geotextile materials to create a very low permeability barrier. A.5.2 The use of geosynthetics in pavements The three primary uses of a geosynthetics in a pavement system are to serve as a construction aid over soft subgrades, improve or extend the estimated service life of the pavement, and reduce the thickness of the structural cross section for a given design period. Some or all of these objectives are normally achieved through at least one of the four functions in Figure A-3. (separation, reinforcement, filtration (drainage), and containment) as shown Figure A-3 Main functions of geosynthetics in pavement systems Geotextile and geogrid materials are the most commonly used geosynthetics in pavement design. This is especially true when only the pavement itself is considered without fills and cut slopes, abutments, or drainage facilities. Stabilization using these materials is achieved through a combination of separation, filtration, and reinforcement. The separation function of a geotextile prevents the subgrade and the sub-base from intermixing (Figure A-4), which might occur during construction and later in-service due to pumping of the subgrade by traffic loads. The filtration function is required because soils requiring stabilization are usually wet. By acting as a filter, the geotextile retains the subgrade without clogging, while allowing waterand from the subgrade to pass the sub-base, permitting the pore pressure to dissipate, promoting strength gain up dueinto to consolidation. Page A- 10 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization Figure A-4 Use of geosynthetic to separate a sub-base from the subgrade Geosynthetics also provide some level of reinforcement by laterally restraining the base or sub-base and improving the bearing capacity, thus decreasing shear stresses on the subgrade. Soft subgrade soils provide very little lateral restraint (containment). Hence, when the granular material moves laterally, ruts develop on the surface and also in the subgrade. A geogrid with good interlocking capabilities or a geotextile with high frictional capacities can provide tensile resistance to lateral aggregate movement. Geosynthetics also increase the system bearing capacity by forcing the potential bearing surface under the wheel load to develop along alternate, longer mobilization paths and, thus, higher shear strength surfaces. Geotextiles serve best as separators, filters and, in the case of non-woven geotextiles, drainage layers, while geogrids are better at reinforcing. Geogrids, as with geotextiles, prevent the sub-base from penetrating the subgrade. When geogrids are used, either the sub-base has to be designed as a separator or a geotextile must be used in conjunction with the geogrid, either separately or as a geocomposite. Table A-2. Function of geosynthetic with respect to subgrade properties. From CED Engineering (1998) Undrained shear strength (KPa) of subgrade CBR% of subgrade Geosynethetic function 60-90 2-3 Filtration, some separation 30-60 1-2 Filtration, separation, some reinforcement <30 Below 1 Filtration, separation, reinforcement In general, as indicated in Table A-2, the specific function to be provided by a geosynthetic in pavements depends on soil properties, and the range of functions potentially served increases as the subgrade strength decreases. Table A-3 lists subgrade conditions that are considered to be the most appropriate for the use of geosynthetics. These are conditions where the subgrade is unlikely to support conventional pavement construction without substantial rutting. Ethiopian Roads Authority Page A- 11 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 Table A-3 Appropriate subgrade conditions for geosynthetic use. Modified from CED Engineering (1998) Condition Related Properties Poor soils USCS: SC, CL, CH, ML, MH, OH, or PT soils; or AASHTO: A-5, A-6, A-7, or A7-6 soils Low strength Undrained shear strength < 13 KPa, CBR < 3, or MR< 30 MPa High water table Within zone of influence of surface soils High sensitivity High undisturbed strength compared to remoulded strength When used as reinforcement, it is very important to place a geosynthetic in a taut or stretched condition so that it can develop full tensile resistance. If placed with wrinkles or in a loose condition on a weak subgrade, the wheel load will create excessive rutting in the underlying soil before the geosynthetic develops enough tension to provide the required restraint. Generally, high elastic modulus geosynthetics undergo less deformation, and hence less soil rutting, to develop the tensile resistance required for subgrade restraint. Thus, the elastic modulus could be used as an important factor for selecting geosynthetic products of similar cost. A.5.3 Design considerations There are at least three design approaches which are in use for incorporating geosynthetics in pavements. These are: design by cost, design by specification, and design by function. Often, the use of geosynthetics for stabilization is completed using the design-by-function approach. A key feature of this approach is the assumption that the structural pavement design is not modified in the procedure. The pavement is designed according to standard procedures, as if the geosynthetic was not present. Reinforced flexible pavement subgrade design Different combinations of geosynthetics are recommended for use in flexible pavements based upon the subgrade soil conditions. Geosynthetics used to construct roads over very soft subgrade conditions typically serve to mechanically stabilize the subgrade, and any design is determined by the subgrade strength. As the subgrade strength increases, the application of the geosynthetics transforms from mechanical subgrade stabilization to base reinforcement. The first step in designing an effective reinforced pavement system is to determine the properties of the subgrade including the grain-size distribution, Atterberg limits, and in-situ shear strength or bearing capacity. In addition, the location of the groundwater table, AASHTO and/or USCS classifications and sensitivity are also needed for comparison with the information given in Table A-3. The in-situ shear strength can be measured directly using vane shear devices or indirectly using bearing capacity correlations from CBR. The design subgrade shear strength is defined as the 25th percentile shear strength, which corresponds to the value at which 75% of the recorded soil strength readings of the top 450 mm of the subgrade are higher. Once the design subgrade conditions have been determined, an assessment of the applicability of geosynthetics should be conducted using the following guidance and Table A-4. Page A- 12 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 Soil Stabilization Table A-4 Recommendations for geosynthetic use in flexible pavements. From US ACE (2003) Design subgrade soil strength and geosynthetic applicability CBR < 0.5 0.5 < CBR < 4.0 4.0 < CBR < 8.0 CBR > 8.0 Use a geotextile and a geogrid at the subgrade/base interface. No aggregate thickness A geotextile is recommended for finegrained subgrades with a design CBR of less than 4. Geogrid A geotextile is generally NOT recommended unless prior experience has indicated separation problems. Design an The design subgrade strength exceeds the existing practice. Do not use this design reduction is recommended. Use the aggregatesurfaced procedure to design a working platform with geosynthetics. Then, use local procedures for flexible pavement thickness design. reinforcement should evaluated. Design an be unreinforced flexible pavement. Then, use this design procedure to determine the reinforced aggregate thickness reduction. Perform a life cycle cost analysis. unreinforced flexible pavement. Then use this design procedure to determine the appropriate aggregate thickness reduction. Perform a life cycle cost analysis. procedure. Use geosynthetic reinforcement to solve site-specific construction problems. Perform a life cycle cost analysis. • • • • For design subgrade CBR strengths of 0.5 or less, the primary application is mechanical subgrade stabilization. At these soil strengths, it is recommended that a construction platform be designed to facilitate the construction of the flexible pavement. The construction platform will serve as the sub base for the flexible pavement system. For design subgrade CBR strengths of 4.0 or less, both the mechanical subgrade stabilization and base reinforcement applications are appropriate. A non-woven geotextile is recommended for separation of fine-grained subgrades, and the use of a geogrid for reinforcement should be considered. Thus, for this subgrade strength level both a geotextile and a geogrid may be necessary and the aggregate thickness can be reduced using empirical reinforced pavement thickness equivalency charts. For subgrade CBR strengths greater than 4.0, a geotextile separator is not recommended unless there is an experience of separation problems with the construction materials. For design subgrade CBR strengths between 4.0 and 8.0, the primary geogrid application is base reinforcement. Research has indicated substantial extensions in pavement service life and significant potential for base thickness reductions. A life cycle cost analysis should be made to determine the cost effectiveness of geogrid reinforcement. When the subgrade CBR is higher than 8.0, a geotextile separator is not recommended unless prior separation problems have been noted for specific construction materials. The primary application of a geogrid at high subgrade soil strengths is base reinforcement. If the use of a geotextile and/or geogrid is warranted based upon the applicability assessment, the following procedure can be used to design the reinforced flexible pavement: 1. The subgrade soil CBR can be determined using Fig. A-5 based upon shear strength (c). The shear strength (c) can be directly measured using vane shear devices. Each aggregate layer, base and sub-base must meet strength and gradation requirements. Ethiopian Roads Authority Page A- 13 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 2. The next step is to determine the design traffic. The design traffic should be determined according to local practices which results in a design index (DI). The design index combines the effect of average vehicle axle loadings and expected traffic volume as expressed by the road classification systems used for design purposes. 3. At the end, there is a need to design an unreinforced flexible pavement for a given subgrade conditions. This can be done using a graph prepared earlier using local environmental conditions. The graph should contain various curves represent design indexes (DI), and correlate the subgrade CBR strength with required pavement thickness. Using the CBR and the appropriate DI, it is possible to empirically determine the required unreinforced pavement thickness above the subgrade from the graph. 4. The reinforced aggregate thickness is determined by using again a locally prepared reinforced pavement thickness equivalency chart. Entering the chart with the unreinforced flexible pavement thickness, a line is drawn to the intersection of the equivalency curve to get the thickness of the equivalent reinforced thickness. Figure A-5 Relationship between CBR and shear strength (c). Modified from US ACE (2003) Aggregate-surfaced reinforced pavement design Geosynthetics in aggregate-surfaced roads can be used to support two pavement applications: mechanical subgrade stabilization and aggregate base reinforcement. The application is pre-determined by the subgrade soil strength. The type of geosynthetics recommended for use in aggregate-surfaced roads is based upon the subgrade soil conditions. Geosynthetics used to construct pavements over very soft subgrade conditions typically serve to mechanically stabilize the subgrade. As the design subgrade strength increases, the primary application of the geosynthetic changes from mechanical subgrade stabilization to base reinforcement. Once the index and other relevant properties of the subgrade are determined as mentioned above, the decision to use geosynthetics is made on the basis of the following procedure: • For design subgrade CBR strengths of 0.5 or less, the primary application is mechanical subgrade stabilization. At these soil strengths, the use of a non-woven geotextile is recommended for separation, and a biaxial geogrid is recommended Page A- 14 Ethiopian Roads Authority Appendix A Geotechnical Design Manual - 2013 • • • Soil Stabilization for aggregate reinforcement. At these low material strengths, the full depth of the aggregate fill should be used and no reduction in aggregate thickness is recommended. Thus, the unreinforced aggregate thickness design should be used for subgrade strengths of 0.5 CBR or less. The non-woven geotextile is placed directly on the subgrade followed by the geogrid and then the aggregate fill. The construction platform serves as a bridge over very soft material, a compaction aid for obtaining target densities, and a construction expedient. For design subgrade CBR strengths of 2.0 or less, both the mechanical subgrade stabilization and base reinforcement applications can be carried out. A non-woven geotextile is recommended for separation at subgrade strengths with a CBR of 2 or less. The use of a biaxial geogrid for reinforcement is also generally cost-effective in terms of aggregate savings. Thus, for this subgrade strength level, both a geotextile and geogrid are generally recommended, and the aggregate thickness can be reduced using the appropriate reinforced bearing capacity factor as described in the following paragraphs. The use of a non-woven geotextile for separation is generally recommended for fine-grained subgrades with design CBR values of less than or equal to 4. A nonwoven geotextile should also be used for separation when the designer has experienced separation problems with the construction materials during previous construction projects. For design subgrade CBR strengths between 2.0 and 4.0, the primary geogrid application is base reinforcement. However, the cost effectiveness of using a geogrid at these subgrade strengths should be determined by performing a life-cycle cost analysis. The primary geosynthetic application for subgrade CBR strength of greater than 4 is base reinforcement. Geogrid reinforcement is generally considered costprohibitive for these types of site-specific subgrades. Instead, geogrids used as construction expedient to solve problems, such ascan sitebe mobility anda localized soft soil deposits. Once it is known that geosynthetics are needed based upon the above applicability assessment, the following procedure can be used to design the reinforced aggregatesurfaced pavement: 1. If the use of a geotextile and/or geogrid is necessary, the subgrade soil strength must be converted from CBR to shear strength (c) using Figure A-5. The shear strength (c) can also be directly measured using vane shear devices. 2. The next step is to determine the design traffic. The design traffic gear should be based upon the gear configuration of the heaviest vehicle expected in the traffic mix, defined as either a single wheel load, a dual-wheel load, or tandem-wheel gear load. The combined weight on the selected gear is used as the design vehicle weight. For example, use one-half of the single- or dual-wheel axle load for singleaxle vehicles. For multiple-axle heaviest two neighbouring axles. vehicles, use one-half of the total load on the 3. Next it is necessary to determine the reinforced bearing capacity factor (Nc). Both the unreinforced and reinforced bearing capacity factors should normally be determined using empirical data obtained from local test sections. The unreinforced bearing capacity factor (Nc) is usually 2.8. The reinforced bearing capacity factor for a geotextile alone is 5.0. The bearing capacity factor for a geotextile separator and geogrid reinforcement is 6.7. Recommended bearing capacity factors are summarized in Table 2-10. Selecting Nc based on allowable subgrade ruts is also Ethiopian Roads Authority Page A- 15 Appendix A Soil Stabilization Geotechnical Design Manual – 2013 common: Nc = 5 for a low rutting (< 50 mm), 5.5 for moderate rutting (50 – 100 mm), and 6 for large rutting (> 100 mm). These values reduce to Nc = 2.8, 3.0, or 3.3, respectively without a geotextile. 4. Determine the subgrade bearing capacity by multiplying the reinforced or unreinforced bearing capacity factor (Nc) by the shear strength of the subgrade (c). 5. In order to find out the required aggregate thickness, the procedure requires the use of curves drawn by using the subgrade bearing capacity (cNc) and the expected single, double and tandem wheel loads. Choosing the curve for specific design wheel loads and using an appropriate cNc enables determining the required aggregate thickness. Table A-5 Recommendations for geosynthetic use in aggregate-surfaced pavements. From US ACE (2002 and 2003) Design subgrade soil strength and geosynthetic applicability CBR < 0.5 0.5 < CBR < 2.0 2.0 < CBR < 4.0 CBR > 4.0 Use a geotextile and a geogrid at the subgrade/base interface. No aggregate thickness reduction recommended. Both a geogrid and a geotextile are recommended. Use this design procedure for aggregate thickness reduction. A geotextile is required for fine-grained subgrades. A geogrid may also be costeffective. Perform a life cycle cost analysis. Perform a cost analysis. Nc1 Nc Geotextile Geogrid Both2 Geotextile 5.0 6.7 6.7 5.0 Geogrid 6.7 Both 6.7 1The 2 unreinforced bearing capacity factor, Nc, is often 2.8. Both a geotextile and a geogrid are recommended, with the former serving primarily as a separation fabric. Page A- 16 Ethiopian Roads Authority