Preview only show first 10 pages with watermark. For full document please download

Biochar Impact On Development And Productivity Of Pepper And Tomato Grown In Fertigated Soilless Media

The impact of additions (1–5% by weight) of a nutrient-poor, wood-derived biochar on pepper (Capsicum annuum L.) and tomato (Lycopersicum esculentum Mill.) plant development and productivity in a coconut fiber:tuff growing mix under optimal

   EMBED


Share

Transcript

  REGULAR ARTICLE Biochar impact on development and productivity of pepperand tomato grown in fertigated soilless media Ellen R. Graber  &  Yael Meller Harel  &  Max Kolton  &  Eddie Cytryn  &  Avner Silber  & Dalia Rav David  &  Ludmilla Tsechansky  &  Menahem Borenshtein  &  Yigal Elad Received: 15 June 2010 /Accepted: 17 August 2010 /Published online: 2 September 2010 # Springer Science+Business Media B.V. 2010 Abstract  The impact of additions (1  –  5% by weight)of a nutrient-poor, wood-derived biochar on pepper ( Capsicum annuum  L.) and tomato (  Lycopersicumesculentum  Mill.) plant development and productivityin a coconut fiber:tuff growing mix under optimalfertigation conditions was examined. Pepper plant development in the biochar-treated pots was signifi-cantly enhanced as compared with the unamendedcontrols. This was reflected by a system-wideincrease in most measured plant parameters: leaf area,canopy dry weight, number of nodes, and yields of  buds, flowers and fruit. In addition to the observedincreases in plant growth and productivity, therhizosphere of biochar-amended pepper plants hadsignificantly greater abundances of culturablemicrobes belonging to prominent soil-associatedgroups. Phylogenetic characterization of unique bac-terial isolates based on 16S rRNA gene analysisdemonstrated that of the 20 unique identified isolatesfrom roots and bulk soil from the char-amendedgrowing mix, 16 were affiliated with previouslydescribed plant growth promoting and/or biocontrolagents. In tomato, biochar treatments positivelyenhanced plant height and leaf size, but had no effect on flower and fruit yield. The positive impacts of  biochar on plant response were not due to direct or indirect effects on plant nutrition, as there were nodifferences between control and treatments in leaf nutrient content. Nor did biochar affect the fieldcapacity of the soilless mixture. A number of organiccompounds belonging to various chemical classes,including  n -alkanoic acids, hydroxy and acetoxyacids, benzoic acids, diols, triols, and phenols wereidentified in organic solvent extracts of the biochar.We conjecture two related alternatives to explain theimproved plant performance under biochar treatment:(i) the biochar stimulated shifts in microbial popula-tions towards beneficial plant growth promotingrhizobacteria or fungi, due to either chemical or  physical attributes of the biochar; or (ii) low dosesof biochar chemicals, many of which are phytotoxicor biocidal at high concentrations, stimulated plant growth at low doses (hormesis). Plant Soil (2010) 337:481  –  496DOI 10.1007/s11104-010-0544-6Responsible Editor: Jorge Vivanco.E. R. Graber ( * ) :  M. Kolton :  E. Cytryn : A. Silber  : L. TsechanskyInstitute of Soil, Water and Environmental Sciences,The Volcani Center, Agricultural Research Organization,POB 6, Bet Dagan 50250, Israele-mail: [email protected]. Meller Harel : M. Kolton : D. Rav David : M. Borenshtein : Y. EladInstitute of Plant Protection, The Volcani Center,Agricultural Research Organization,POB 6, Bet Dagan 50250, IsraelM. KoltonInstitute of Plant Sciences and Genetics in Agriculture,The Robert H. Smith Faculty of Agriculture,Food and Environment,P.O. Box 12, Rehovot 76100, Israel  Keywords  Soilless media.Biochar .Pepper .Tomato.Plantgrowthpromotion.PGPR .PGPF.Tar .Intensiveagriculture.Fertigation.Plantproductivity.Hormesis.Biocontrol.Antibiotics.Bacterialisolates Introduction There is a world-wide drive to develop affordableenergy from renewable resources in order to reducenet greenhouse gas emissions and diversify energysources. Biomass surpasses many other renewableenergy sources due to its abundance, high energyvalue, and versatility (Digman et al. 2009). However,much first generation biofuel production (i.e., ethanolfrom sugar and starch crops, and biodiesel from oilcrops) in temperate climates is almost entirely carbon positive due to heavy inputs of fossil fuels for agricultural production (Bruun and Luxhoi 2008; Lal 2005; Searchinger et al. 2008). As a result, there is continued interest in developing second and thirdgeneration biofuels which will be at least carbonneutral, if not carbon negative (Mathews 2008). Pyrolysis of waste biomass is one such method; it involves the thermal decomposition (exothermic) of  biomass in the absence of oxygen to solid (charcoal,or biochar), liquid (bio-oil) and gas biofuels. Insteadof burning the produced biochar for energy, biochar isapplied to the soil as a conditioner, where it remainsin an essentially permanent form and leads to net removal of carbon from the atmosphere (Laird 2008; Lehmann 2007). As a soil additive along with organicand inorganic fertilizers, biochar has been reported tosignificantly improve soil tilth, nutrient availability to plants, and plant productivity (Glaser et al. 2002). Pre-Columbian Amazonian Indians used charcoal toenhance soil productivity, and it is still found in largeconcentrations in fertile Amazon soils abandonedthousands of years ago. Currently, however, verylittle biochar is utilized in modern agriculture, and itsagronomic value in terms of crop response and soilhealth benefits has yet to be quantified.For the most part, the available literature docu-ments a general improvement in plant response under  biochar amendment (Chan et al. 2007, 2008; Glaser et  al. 2002; Iswaran et al. 1980; Steiner et al. 2007; Wardle et al. 1998), although there are notable exceptions (Kishimoto and Sugiura 1985; Van Zwieten et al. 2010). The wide range of reported plant responses (e.g.,  − 29 to 324% increase in drymatter) can be attributed to the large variability instudied systems (Glaser et al. 2002). In particular, it is worth noting that the effect of biochar amendments isoften examined in poor, depleted soils (Lehmann et al. 2003; Rondon et al. 2007; Steiner et al. 2008; Van Zwieten et al. 2010) or in systems with sub-optimalfertilization regimes (Chan et al. 2008; Hossain et al. 2010; Van Zwieten et al. 2010). Thespecificmechanismsunderlyingthecontributionof biochar to plant response are poorly understood.Regional conditions including climate, soil chemistryand soil condition affect biochar agronomic benefits. Inaddition, different biomass feedstocks and pyrolysisconditions create biochars with different physical andchemical properties (Keiluweit et al. 2010), affecting  plant response (Chan et al. 2007, 2008). Biochar can improve plant productivity directly as a result of itsnutrient content and release characteristics, as well asindirectly, via: (i) improved retention of nutrients(Lehmann et al. 2003; Wardle et al. 1998); (ii) improvements in soil pH (Rondon et al. 2007); (iii) increased soil cation exchange capacity (Liang et al.2006); (iv) improved soil physical properties (Chan et  al. 2008), including an increase in soil water retention(Laird et al. 2010); and (v) alteration of soil microbial  populations and functions (Pietikainen et al. 2000). These effects may also act in concert to result inimproved crop performance.Given the complexity of the soil-plant-water-envi-ronmentsystem,itisdifficulttoisolatethosefactorsthat actuallyplayaninstrumentalroleinthe ‘ charcoaleffect  ’ (Wardle et al. 1998). Our goal in the present study was to test whether biochar can impact plant growth whennutritional and soil physical aspects of biochar amend-ment are neutralized. This was achieved by examiningthe impact of a nutrient-poor, wood-derived biochar ontomato (  Lycopersicum esculentum  Mill.) and pepper ( Capsicum annuum  L.) development in a commercialcoconut fiber:tuff soilless mixture under an optimalfertigation (fertilization plus irrigation) regime in agreenhouse. If improved plant growth would beobserved under such optimal conditions, it woulddemonstrate that biochar-induced plant growth stimu-lation goes beyond obvious contributions to plant nutrition and improved soil physical and chemical properties. This would then provide an opportunity toevaluate specific factors potentially responsible for the ‘ charcoal effect  ’ . 482 Plant Soil (2010) 337:481  –  496  Materials and methods Biochar and plant growth mediumBiochar prepared from citrus wood in a traditionalcharcoal pit (lump charcoal) was obtained, ground intoa powder of less than 0.5 mm particles, and stored in asealed metal box until use. Plants were grown in acommercial soilless mixture composed of a mixture of coconut fiber and tuff at a 7:3 v:v ratio (the tuff isunsorted to a size of 8 mm). Biochar at the desired rates(1, 3 or 5% by weight) was mixed into the commercialsoilless mixture. The influence of biochar at 4 amend-ment levels (0, 1, 3, 5% w:w) on the water holdingcapacity of the soilless mixture at field capacity wastestedgravimetricallyasfollows.Pre-weighed0.5Lpotswere filled with a known weight of media to a certain bulk density, the filled pots were weighed, and the porousmediaineachwasthenthoroughlysaturatedwitha large excess of water. The pots wereleft to drain freely,and their weight was measured every 30 min until it ceased to change (2 h). Field capacity was calculated bycomparing the weight of the media at 2 h with the dryweight of the media. There were five replicates (n isnumber of replicates) per amendment level.Biochar characterizationThe biochar was characterized chemically and phys-ically in order to define the material as fully as possible. Aqueous extracts of biochar (1:10 w:w)were prepared in double distilled water by stirring at 25°C for 30 min, and by autoclaving (121°C, 1 atm.,35 min.). The extracts were filtered through a0.22  μ  m filter (Milex-HV, Millipore) and stored at 4°C until analyzed. The extracts were characterizedfor pH, electrical conductivity (EC), nutrient content (B, Ca, Cu, Fe, K, Mg, Mn, P, S, Si, Zn, Na) byinductively coupled plasma (ICP-AES Arcos, EOPmodel, Spectro Ltd.), Cl by digital chloridometer (Labconco), N-NO 3  and N-NH 4  by injector autoan-alyzer (Lachat Instruments), and total organic carbon(TOC) content by total organic carbon analyzer (Skalar Analytical B.V.). A concentrated acid extract (0.1 g biochar in 10 mL 70% HNO 3 , digested at 120°C for 24 h) was also analyzed by ICP for B, Ca,Cu, Fe, K, Mg, Mn, P, S, Zn, and Na.Ash content (in duplicate) was determined byweight loss after heating to 800°C in air for 4 h.Total C, H, N, O, and S were determined in triplicate by element analyzer (Thermo Flash EA-1112 Ele-mental Analyzer). X-ray diffraction analysis wascarried out on a Philips XRD diffractometer, and biochar samples (both uncoated and gold coated)were observed by scanning electron microscope(SEM; model JSM-5410, JEOL Ltd). Element iden-tification on uncoated samples was by Link ISIS X-Ray Analysis detector (Oxford Instruments). FTIR absorbance spectra of KBr pellets prepared with 0.6%wt biochar were recorded between 400 and4000 cm − 1 with one hundred scans averaged with aresolution of 4 cm − 1 (Bruker Tensor 27 FTIR Spectrometer). The specific surface area (SSA) of the biochar was determined by N 2 -BET adsorption at The Israel Ceramic & Silicate Institute, TechnionCity, Haifa, Israel.Biochar tars were extracted by automated Soxhlet (VELP) in a dichloromethane:methanol (DCM:MeOH) 95:5 v:v mixture. Biochar (0.5 g) was boiledfor 1 hr in 40 mL of solvent and rinsed for 1 h in thecondensed solvent vapors. The solvent extract wasthen evaporated to dryness and derivatized using0.5 mL of   N,O -Bis(trimethylsilyl)trifluoroacetamide(BSTFA) and 1 mL acetonitrile for 15 min at 60°C.Analysis was carried out by gas chromatography/massspectrometry (GC/MS; Agilent, model number 6890N/5973N) in scan mode using temperature programming (oven 170°C, initial hold 5 min, rampto 300°C at 5°C/min, final hold 10 min) and a 30 mlong capillary column with a (5%-phenyl)-methylpo-lysiloxane phase, 0.25 mm inner diameter, and0.25  μ  m film thickness. Compound identificationwas via the NIST98 library.Plants and fertigation regimePlants of tomato cv. 1402 (Hazera Genetics, Ltd.,Brurim, M.P. Shikmim, Israel) and sweet pepper cv.Maccabi (Hazera Genetics, Israel) were obtained froma commercial nursery (Hishtil, Ashkelon, Israel) at 40  –  50 days after seeding, and transplanted into 1-L pots containing the potting medium without or with biochar at 1, 3 or 5% by weight. Plants werefertigated with drippers 2  –  3 times per day with5:3:8 NPK fertilizer (irrigation water prepared to have N (total), P and K concentrations of 120, 30, and150 mgL − 1 , respectively; EC 2.2 dS m − 1 ), allowingfor 25  –  50% drainage. Plants, trained on bamboo Plant Soil (2010) 337:481  –  496 483  sticks or hooked to polyethylene ropes as trainingsystems, were maintained at 20  –  30°C in a pest- anddisease-free greenhouse for up to 3 months.Experimental design and statistical analysisA total of six sets of greenhouse experiments werecarried out. Two sets (one for each crop pepper (P)and tomato (T)) compared 0, 1, and 3% biochar byweight (experiments 013P [ n =9] and 013T [ n =7]),two sets compared 0, 3, and 5% biochar by weight (experiments 035P [ n =6] and 035T [ n =6]), and twosets compared 0 and 5% biochar by weight (experi-ments 05P [ n =10] and 05T [ n =10]). Replicate plantsin each experimental set were arranged randomly inthe greenhouse. Data was averaged per leaf level(node), plant, and replicate. In order to compensatefor initial variability in plant parameters, individualdata at each evaluation date is presented as the percent change compared with the value at the initialevaluation of the same plant organ. Calculation of data at each sampling date was carried out accordingto the formula C=100×B/A where A=value at initialevaluation date, B=value at a given evaluation date,and C=calculated percentage of change. Data in percentages was arcsin-transformed before further analysis. Data was analyzed using ANOVA andFisher  ’ s protected LSD test. Standard errors (SE) of the means were calculated and growth levels werestatistically separated following a one-way analysis of variance. The standard errors (SE) are marked in thefigures with error bars and stated in the tables andtext. Results marked with different lower case lettersare statistically different at a significance level of   P  ≤ 0.05. Statistical analysis was done using R version2.10.1 software (http://www.r-project.org).Plant growth parametersPlant growth parameters were evaluated every 1  –  3 weeks or when plants reached a stage for evaluation(e.g., presence of mature leaves, blossoms, set fruits,and mature fruits). Plant height was measured fromstem base to top. Leaves at a certain fixed node height were marked and evaluated. Area of leaf blade(pepper) and terminal leaflet (tomato) were calculatedfrom width and length measurements according tothe formula A=W×L×R where W and L refer towidth and length and R is a constant factor equalto 0.78 and 0.85 for tomato and pepper leaves,respectively. These factors were derived by measur-ing width, length and area of 25 leaves of each plant type. The length of the leaf axis was measured intomato leaves.For flower and fruit counting, buds were regardedas flowers as long as fruit set was not visible. Fruitswere regarded as set fruits until they reached the sizeof 12 mm diam. Fruits were harvested when theyreached the color of mature grade according tocommon practice. Canopy fresh weight of pepper  plants was evaluated after detaching the fruits. Plant material was dried in an air circulation 60°C oven for 5  –  6 days until dry weight was unchanged.Leaf nutrient sampling and measurementsLeaves (1 each from the top and bottom of each plant)were sampled 35, 52, and 65 days following plantingin experiment 013P, and 52 and 65 days following planting in experiment 013T. Leaves were washedwith distilled water, dried for 1 week in a ventilatedoven at 60°C, and stored pending chemical analysis.The dry tissue was ground to pass a 20-mesh sieve,and 100-mg samples were wet ashed with H 2 SO 4 -H 2 O 2  and analyzed for Na, K, organic-N, and P.Ashing in HClO 4 -HNO 3  was used to analyze for Ca,Mg, and micronutrients. Element concentrations weredetermined as follows: NO 3 -N and P by autoanalyzer;K by flame photometer (M410, Sherwood Sci. Ltd);Ca, Mg, Fe, Zn, and Mn by Analyst 800 atomicabsorption spectrophotometer (Perkin Elmer).Culturable microbe abundances in pepper experimentsFrom experimental set 013P, potting medium clingingto root surfaces (rhizosphere) and located around theroots (bulk medium) was sampled non-destructivelyfrom 3 replicate plants 76 days after planting, andfrom 3 other replicate plants 84 days after planting.Replicates were not pooled. A 10 g (wet weight)aliquot of the sampled bulk medium was suspended in4.5 ml of sterile distilled water by gentle agitation(25 rpm) for 30 min at 20°C. In parallel, sampledroots were washed under tap water to remove the bulk of the attached soil, excess water was thoroughly blotted off, and 10 g (wet weight) were gently crushedwith a pestle in a clean plastic bag. The root tissue 484 Plant Soil (2010) 337:481  –  496  was then transferred to a sterile Erlenmeyer vialcontaining 4.5 ml of sterile distilled water andsubjected to gentle agitation as described above.Ten-fold serial dilutions were prepared from eachsample. Three replicate 100  μ  L aliquots from eachdilution were plated for fungi and bacteria on agar media as detailed below.General bacterial were quantified on nutrient agar supplemented with 1 ppm benomyl (Dhingra andSinclair  1986),  Pseudomonas  spp. on Kings Bmedium (Elad and Baker  1985), filamentous fungi, yeast and  Trichoderma  spp. on PDA medium supple-mented with 50 ppm Rose Bengal (4,5,6,7-tetra-chloro-2 ′ ,4 ′ ,5 ′ ,7 ′ -tetraiodofluorescein) stain (Elad et al. 1981), and actinomycetes species on alkaline water agar (Ho and Ko 1980).  Bacillus  spp. were quantifiedon nutrient agar following boiling for 20 min to killvegetative cells but not endospores (Dhingra andSinclair  1986). Microbe abundances are presented ascolony forming units (CFU) per g sample. Resultsfrom the two sampling events were similar, therefore,those from the first event are presented.Taxonomic characterization of isolated bacteriaBacterial isolates from biochar-amended and control potting mixtures showing distinct colony morphologieswere characterized based on 16S rRNA gene sequence phylogeny. Colonies were resuspended in 100  μ  L of sterile distilled water, boiled for 10 min and centrifugedat 15000 g for 5 min at room temperature. PCR amplification was performed using 2.5  –  5  μ  l of thesupernatant as a template with the general bacterial primer pair 11F (5 ′ GTTTGATCCTGGCTCAG (Kaneet al. 1993)) and 1392R (5 ′ ACGGGCGGTGTGTRC(Lane 1991)). PCR reactions (final volume of 50  μ  L)contained the following components: 1.5 U Taq DNA polymerase (DreamTaq; Fermentas), Taq buffer con-taining a final magnesium concentration of 2.5 mM,dNTPs (20 nmol each), 12.5  μ  g bovine serum albuminand 25 pmol of each primer. The PCR program wascarried out as previously described (Cytryn et al. 2006) with slight modifications: an initial denaturation step of 95°C for 180 s followed by 30 cycles of denaturationat 95°C for 30 s, annealing at 55°C for 30 s andelongation at 72°C for 80 s. Cycling was completedwith a final elongation step at 72°C for 5 min. The presence and size of PCR amplicons were determined by agarose gel electrophoresis (1%). The PCR productsof colonies that gave a single distinct band weresequenced using the general bacterial primer 907R (5 ′ CCGTCAATTCMTTTGAGTTT (Muyzer et al. 1998). Partial 16S rRNA gene sequences (630 bp) werecharacterized using the Ribosomal Database Project (RDP) on-line sequence analyses package. After elimination of chimeric sequences using Chimera-CHECK (Cole et al. 2009), partial 16S rRNA genesequences were compared to sequences from theGenBank and RDP databases using Blast and theRDP Seqmatch online classifier  (Wang et al. 2007) (www.ncbi.nlm.nih.gov/BLAST/  and www.rdp.cme. msu.edu respectively). Sequences were deposited intothe Genbank nucleotide sequence database under accession numbers HM481450-HM481472. Results Biochar characterizationThe biochar exhibited very weak absorbance in the IR spectrum, with the major observable bands of the FTIR spectrum occurring at about 1600 and 1430 cm − 1 denoting aromatic ring C=C stretching, and a weaker  band at about 1700 cm − 1 , indicative of aromaticcarbonyl/carboxyl C=O stretching (Guo and Bustin1998) (Fig. 1a). Ash content of the biochar was 10.9%. The only definable mineral phases identified by X-raydiffraction of the biochar were quartz and calcite, bothat low, non-quantifiable levels. These results corre-sponded well with SEM-EDX results, which did not reveal any separate mineral phases. The pore structureof the precursor wood can clearly be seen in the SEMimages (Fig. 1b). The surface area of the biochar, asdetermined by N 2 -BET, was 46.2 m 2 g − 1 . Despite its porous nature, the biochar had no impact on the water-holding capacity of the soilless media at any amend-ment level (measured field capacity 31.9±4.6, 31.9±3.1, 30.4±2.2, and 32.4±4.5%, for 0, 1, 3 and 5% biochar, respectively). Accounting for ash content, theelemental composition of the biochar was found to be70.6% C, 0.6% N, 2.3% H, and 15.5% O, giving an O/ C atomic ratio of 0.16, H/C ratio of 0.40, C/N ratio of 130.69, and an H/O ratio of 2.41. These results aresimilar to those for wood chars reported in theliterature (Krull et al. 2009).Thechemicalcharacterizationofthebiocharaqueousextracts is presented in Table 1. pH of the extracts was Plant Soil (2010) 337:481  –  496 485