Transcript
Master Plan Jakarta, Indonesia: The Giant Seawall and the need forstructural treatment of municipal waste water Simon A. van der Wulp a, ⁎ , Larissa Dsikowitzky b , Karl Jürgen Hesse a , Jan Schwarzbauer b a Forschungs und Technologiezentrum Westküste, Christian-Albrechts-Universität zu Kiel, Hafentörn 1, D-25761 Büsum, Germany b Institute of Geology and Geochemistry of Petroleum and Coal, RWTH Aachen University, Lochnerstraße 420, D-52056 Aachen, Germany a b s t r a c ta r t i c l e i n f o Article history: Received 30 September 2015Received in revised form 3 May 2016Accepted 19 May 2016Available online 25 May 2016 In order to take actions against the annual ooding in Jakarta, the construction of a Giant Seawall has been pro-posed in the Master Plan for National Capital Integrated Coastal Development. The seawall provides a combina-tionoftechnicalsolutionsagainst ooding,butthese willheavily modifythe masstransportsinthenear-coastalarea of Jakarta Bay. This study presents numerical simulations of river ux of total nitrogen and N,N-diethyl- m -toluamide, a molecular tracer for municipal waste water for similar scenarios as described in the Master Plan.Model results demonstrate a strong accumulation of municipal wastes and nutrients in the planned reservoirstoextremely high levelswhich will result indrastic adverseeutrophication effects if thetreatment of municipalwaste water is not dealt with in the same priority as the construction of the Giant Seawall.© 2016 Elsevier Ltd. All rights reserved. Keywords: Jakarta BayRiver dischargesFlow modellingNitrogenDEETGiant Seawall 1. Introduction ThemunicipalitiesoftheJakartaMetropolitanAreaarefacingagreatchallenge to deal with the issues of overexploitation of groundwater,land subsidence and annually reoccurring oods, lacking expansion of infrastructureincluding,amongothers,sanitationandtreatmentofmu-nicipalwastewater(Steinberg,2007).Theuncontrolled owofuntreat-ed municipal waste water has led to a strong deterioration of waterquality in the rivers of Jakarta and along the shores of Jakarta Bay(Damar, 2003; Ari n, 2004; Thoha et al., 2007).Van der Wulp et al. (2016-in this issue) have identi ed persistentlyhighconcentrationsoftotalnitrogenandtotalphosphorusalongtheJa-karta shorelinedue to high initial river concentrationsand sub-optimal ow characteristics in those areas where eutrophication issues are fre-quently reported.The Master Plan “ National Capital Integrated Coastal Development ” focusesonthedesignandfunctioningof oodrisksolutionsincorporat-ed in an integral design for socio-economic urban development(MinistryforEconomic Affairs,2015a).Thisoffshoreapproachincorpo-ratesthephasedconstructionofaGiantSeawallandlargestoragebasinstoprotectJakartaCityagainst oodsfromseaandrivers.Variousarticlesin the Jakarta Post show public concerns expressed about theenvironmentalandsocio-economicconsequencesoftheMasterPlanin-cluding the deterioration of the water quality in the intended storagebasins.Inthepresentstudy,anexistingcombinationofnumericalhydrolog-ical and ow modelling techniques as previously described in Van derWulp et al. (2016-in this issue) was used to simulate river discharges, ow, and dispersion of nutrient ux, in the form of total nitrogen (TN)and total phosphorus (TP), into Jakarta Bay. The ow and mass tracermodel was adapted to simulate scenarios similar to the three phasesoftheMasterPlantoillustratethefateofriverboundnutrientsandmu-nicipal waste water. In addition, the simulation of N,N-diethyl- m -toluamide (DEET) ux was introduced as a tracer substance. DEET isan organic compound, which is commonly used as insect repellent.Due to its intensive usage in the study area and its properties, DEET isuseful as molecular marker for municipal wastewater discharges(Dsikowitzky et al., 2014). 2. Materials and methods A number of scenarios were simulated by the adaptation of theexisting Jakarta Bay ow model as described by Van der Wulp et al.(2016-in this issue). The development phases A, B, and C, similar tothe phases described in the Master Plan National Capital IntegratedCoastal Development (Ministry for Economic Affairs, 2015a) were im-plemented in the ow model (Fig. 1). Marine Pollution Bulletin 110 (2016) 686 – 693 ⁎ Corresponding author.http://dx.doi.org/10.1016/j.marpolbul.2016.05.0480025-326X/© 2016 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul 2.1. Flow model The Delft3D modelling suite was used to simulate ow characteris-tics for Jakarta Bay. The computational grid, with a spatial resolutionof 300 m, covers Jakarta Bay and stretches out 60 km toward thesouth tip of the Seribu Islands archipelago. Along the vertical, the do-main was divided into 15 xed layers with a higher resolution closertothesurfacetoresolvemultiplelayersfortherelativelyshallowJakartaBay. Bathymetrical data were obtained from nautical charts of the JavaSeaandJakartaBay.Themodelincorporatedforcingbytides.Tidalforc-ing was added based on thirteen astronomical tidal constituents ex-tracted from the Global Tidal Model (Egbert and Erofeeva, 2002).Additionally, Sea Surface Height (SSH) and daily, three-dimensionalvalues of salinity and temperature were provided as boundary condi-tions by the HAMSOM Indonesian model (Mayer et al., 2010, 2015;Mayer and Damm, 2012). Spatiotemporal varying sea surface windand pressure elds were derived from the GME model (Majewski andRitter,2002)andimposedonthemodelgridtoaccountforwindeffects.Simulations were run for the year 2012 with a warm up time coveringthe last quarter of 2011. A detailed description of the ow model setupand validation is described in Van der Wulp et al. (2016-in this issue). 2.2. River discharges and nutrient loads Riverdischargesde nedforthe owmodelweretakenfromthehy-drologicalmodelasdescribedbyVanderWulpetal.(2016-inthisissue).Atotalaveragedischargeof205±97m 3 s − 1 owtoJakartaBaydistrib-uted over 13 rivers and streams (Cisadane, Cengkareng, Banjir KanalBarat, Muara Angke, Muara Baru, Ciliwung, Sunter, Cakung, Blencong,Banjir Kanal Timur, Cikarang-Bekasi-Laut (CBL), Keramat, Citarum)(Fig. 2a). Most of the river discharges enter Jakarta Bay through theCitarum River (137 ± 64.2 m 3 s − 1 ) and the Cisadane River (36 ±17 m 3 s − 1 ). River TN ux waschosen asthepotential driver of possibleeutrophication effects. TN loads for individual rivers were quanti edbased on river discharges, river water quality measurements taken inOctober 2012 and calibration of simulated nutrient gradients in JakartaBay (Van der Wulp et al., 2016-in this issue). TN loads ranged between39 and 174 tons d − 1 with an average of 91 ± 45 tons d − 1 . The highestTN loads entered Jakarta Bay through the Citarum River (46%) followedby the Ciliwung River (18%) and CBL (12%) (Fig. 2b).Similarly, TP loads ranged between 14 and 60 tons d − 1 with an av-erage of 31.9 ± 15.7 tons d − 1 . The highest TP loads entered JakartaBay through the Citarum river (51%), followed by the CBL (17%) andthe Ciliwung River (13%) (Fig. 2c). 2.3. DEET as tracer for municipal wastewater discharges The organic contaminant DEET was used as a tracer for municipalwastes DEET was found in exceptionally high concentrations in JakartaBay and its adjacent rivers with concentrations in the range of 30 ng L − 1 in the Citarum River and up to 24,000 ng L − 1 in the rivers owing through the central part of Jakarta City (Dsikowitzky et al.,2014). DEET loads were quanti ed based on river discharges andDEET concentrations in river water sampled in October 2012 as de-scribed in Van der Wulp et al. (2016-in this issue). DEET loads were inthe order of 44 ± 23 kg d − 1 , of which the highest loads were foundfor the Ciliwung River (16 ± 8 kg d − 1 ) and Sunter River (11 ±7 kg d − 1 ) (Fig. 2c). 2.4. Scenarios Thereferencescenarioconsistedofasimulationcoveringtheperiod2012, without land reclamations.PhaseAinvolveslandreclamationsintheformof16islandssituatedin the near-shore area of the Jakarta coastline (Fig. 1). In between theislands, a uniform depth of 3 m was speci ed.PhaseBcontainedtheconstructionoftheGiantSeawallandPortEx-tension Areas (Fig. 1). The closing of the western reservoir provides a Fig. 1. An overview of Jakarta Bay and the development phases of the construction of a Giant Sea wall similar to the Master Plan of National Capital Integrated Coastal Development.Differentiation can be made between land reclamations of phase A (cross-hatched), phase B (gray), and the closing of the eastern reservoir in phase C (dotted lines). Pumping stationsare placed to control the water level of the western and eastern reservoirs by pumping water from the reservoirs to the sea (black dots). Jakarta city limits topography ©OpenStreetMap-contributors.687 S.A. van der Wulp et al. / Marine Pollution Bulletin 110 (2016) 686 – 693 retentionlake,whichcovers70km 2 .Waterdepthaverages9.6mwithamaximum depth of 16.7 m in the main reservoir. Within the model,three pumps were de ned (Fig. 1) with a total ow rate equal to theinput by the adjacent rivers to arti cially keep the water level withinthe reservoir constant and discharge the excess of water on the otherside of the seawall. The annual average river discharges entering thewestern reservoir were equal to 11.4 ± 5.9 m 3 s − 1 with a maximum,monthly averaged, discharge of 22.2 m 3 s − 1 . It should be denoted thatriver discharges are monthly averaged quantities and do not includepeak discharges of shorter timescales. On average 22.6 ±11.3 tons TN d − 1 , 5.4 ± 2.7 tons TP d − 1 and 19.6 ± 7.8 kg DEET d − 1 were discharged into the western reservoir from the Cengkareng,Kanal Banjir Barat, Muara Angke, Muara Baru, and Ciliwung catchmentareas.For Phase C, the eastern reservoir was closed creating a retentionlake of 47 km 2 with an average depth of 7.5 m. An additional pumpingstation was added to keep the water level in the eastern reservoir con-stant (Fig. 1). The adjacent rivers, Cakung, Blencong, Banjir KanalTimur, CBL, equal a discharge of 17.1 ± 10.1 m 3 s − 1 and a maximum,monthly averaged, discharge of 35 m 3 s − 1 . Corresponding TN, TP, andDEET loads were 11.8 ± 7.1 tons d − 1 , 5.7 ± 3.4 tons d − 1 , and 9.0 ±5.5 kg d − 1 , respectively. 3. Results 3.1. Total nitrogen Model results of the reference scenario show a distribution of TNconcentrations affecting the entire bay (Fig. 3a). Elevated TN levelsdominate the shoreline of Jakarta city with high concentrations foundin the vicinity of the Cengkareng and Ciliwung river mouths. Nutrientconcentrations decrease with increasing distance from the shore withthe 10 μ M isoline, as reference, situated at the outer edge of JakartaBay.AwestwardplumebulgesintotheJavaSeafromwhereitisrapidlydilutedtowardbackgroundconcentrations.AmoredetaileddescriptionofthissimulationcanbefoundinVanderWulpetal.(2016-inthisissue).Model results of the second scenario, phase A of the Master Plan,showthatinbetweenthenewlyconstructedislands,TNconcentrationsdoubledlocallyto120 μ M(Fig.4a).Riverdischargesarechannelledpastthe islands causing a northward shift of the TN gradient by approxi-mately 5 km when compared to the reference scenario. The 10 μ M iso-lineliesattheapproximatelysamelocationindicatingthatTNgradientsat the outer edge of Jakarta Bay is not shifted with exception of a smallplume around the second island from the west (Fig. 3b).For the realisation of the seawall and the closingof thewestern res-ervoir in phase B, modelresults show a strongaccumulation of TN (Fig.3c). Within the western reservoir, TN concentrations increase by a fac-tor 56 – 780 μ M and approach a water quality similar to those of the ad- jacent rivers. The pumps discharge the excess of water withcorresponding concentration into the open waters of Jakarta Bay fromwhere it is dispersed rapidly towards background levels (Fig. 4b). The10 μ M reference isoline does not show any seaward expansion. In thecontrary,TNconcentrationsdrop in frontof theseawall with exceptionofthepumpingstationsdischargepointswhereasmallplumeisvisible(Fig.3c).TheconstructionoftheportexpansionareaintheeastofJakar-ta Bay creates con ned areas which experience less ushing and in-creased TN levels up to 150 μ M in front of the Sunter river mouth and45 μ M in front of the CBL (Fig. 4b and c).Modelresults of PhaseC showsimilarities with PhaseB withexcep-tionoftheeasternreservoirwhichhasbeenclosed(Fig.3d).TNconcen-tration within the eastern reservoir increased to 540 μ M between themainland and the island and 500 μ M in the eastern reservoir, equal toafactor15whencomparedtophaseB,priortotheclosingoftheeasternreservoir (Fig. 4). 3.2. Total phosphorus Model simulations with respect to TP show distributions similar tothose of TN (Fig. 5). Elevated TP levels were commonly found alongtheshorelineofJakartacitywithhighconcentrationsfoundinthevicin-ity of the Cengkareng and Ciliwung river mouths. Nutrient concentra-tions decrease with increasing distance from the shore with the 2 μ Misoline, as reference, situated at the outer edge of Jakarta Bay. River in-puts are rapidly diluted toward background concentrations in theouter parts of Jakarta Bay. Model results of phase A show that in be-tween the newly constructed islands, TP concentrations increase to Fig. 2. Relative distributions of river discharges (a), TN loads (b), TP loads (c), and DEETloads (d).688 S.A. van der Wulp et al. / Marine Pollution Bulletin 110 (2016) 686 – 693 12 μ M (Fig. 6a). The constructed islands cause a northward shift of theTP gradient by approximately 5 km when compared to the referencescenario. The 2 μ M isoline indicates that the TP gradient at the outeredge of Jakarta Bay, however, remains stable (Fig. 5b).The realisation of the seawall and the closing of the western reser-voir (phase B) show a strong accumulation of TP (Fig. 5c). Within thewestern reservoir, TP concentrations increase by a factor of 50 – 82 μ Mandapproachawaterqualitysimilartothoseoftheadjacentrivers.Out-side the Giant Seawall, TP is rapidly dispersed towards backgroundlevels (Fig. 6b). The 2 μ M reference isoline does not show any seawardexpansion, whereas TP concentrations drop in front of the seawallwith exception of the pumping stations discharge points (Fig. 6c). TheconstructionoftheportexpansionareaintheeastofJakartaBaycreatescon nedareaswhichexperienceless ushingandincreasedTPlevelsof up to 12 μ M in front of the Sunter river mouth and 8 μ M in front of theCBL (Fig. 6b and c). Model results of Phase C show similarities withPhase B with exception of the eastern reservoir, which has been closed(Fig. 5d). TP concentration within the eastern reservoir increases to100 μ M, equal to a factor of 16 when compared to phase B, prior tothe closing of the eastern reservoir (Fig. 6). 3.3. DEET as a tracer form municipal wastewater discharges RiverdischargesfromthecentralpartofJakartaCityaccountfor90%of the total DEET ux. With the construction of the Giant Seawall, 44%and 20% of the total DEET load accumulate in the western and easternreservoirs, respectively. 26% of the DEET is discharged from the Sunterriver in betweenboth reservoirs,from where it will be ushed towardstheopenwatersofJakartaBay.Theremaining10%enterJakartaBayviathe Cisadane and Citarum rivers. DEET concentrations simulated in thereference scenario are highest in the near-shore area with concentra-tions in the order of 700 ng L − 1 (Fig. 7a). The reference isoline, at100 ng L − 1 , shows that the DEET footprint is situated within JakartaBay. Concentrations rapidly drop beyond this border due to increasingwater depth and water movement.The inclusion of islands within the model (Phase A) results in anorthward shift of the DEET gradient over the rst 5 – 7 km, similar tothepatternoftheTNsimulations.Inbetweentheislands,DEETconcen-trationsdoubletoapproximately1200ngL − 1 .The100ngL − 1 referenceisolineindicates noincreasedin uence in direction of the Java Sea(Fig.7b).ForPhaseB (Fig. 7c),accumulationof DEEToccurs where44%of thetotal DEET loadenters thewestern reservoir, resultinginDEET levels of up to 9500 ng L − 1 , 60 times higher as the concentration found for thereference scenario. Land reclamations for port expansion in the east of Jakarta Bay limit the water exchange causing elevated DEET levels inthe order of 2500 ng L − 1 in front of the Sunter river mouth and1100 ng L − 1 in front of the CBL river mouth. Simulations of scenarioPhaseCallow21%ofthetotalDEETloadstoentertheeasternreservoir.Model simulation results in DEET concentrations of 9000 and6100 ng L − 1 for the area in between the near-shore island and themain eastern reservoir, respectively. 4. Discussion In the present study, numerical simulations regarding ow andtracerswereappliedtoillustratetheeffectsofriverdischargeswithcor-responding loads of nutrients (total nitrogen (TN), total phosphorus Fig. 3. Annual averaged total nitrogen (TN) concentrations for the reference scenario (a), initiated land reclamation of phase A (b), construction of the Great Garuda and closing of theWestern reservoir during phase B (c), and completion of the initial design with closure of the eastern reservoir during phase C (d).689 S.A. van der Wulp et al. / Marine Pollution Bulletin 110 (2016) 686 – 693
