11 pages

Quantification of the contribution of biological nitrogen fixation to tropical green manure crops and the residual benefit to a subsequent maize crop using 15N-isotope techniques

Please download to get full document.

View again

of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Quantification of the contribution of biological nitrogen fixation to tropical green manure crops and the residual benefit to a subsequent maize crop using 15N-isotope techniques
  Journal of Biotechnology 91 (2001) 105–115 Quantification of the contribution of biological nitrogenfixation to tropical green manure crops and the residualbenefit to a subsequent maize crop using  15 N-isotopetechniques Margarita G. Ramos  a , Maria Antonieta A. Villatoro  b , Segundo Urquiaga  c ,Bruno J.R. Alves  c , Robert M. Boddey  c, * a Instituto Nacional de Ciencias Agrı´colas ,  Carretera de Tapaste  –  San Jose´  ,  Ga  eta Postal No .  1 ,  C  . P .  32700  , San Jose´ de Las Lajas ,  Ha  ana ,  Cuba b Faculdad de Medicina Veterinaria e Zootecnia ,  Edifı´cio M  - 6  ,  Ciudad Uni   ersitaria ,  Zona  12  ,  Guatemala c Embrapa Agrobiologia ,  Caixa Postal   74  . 505  ,  Serope´dica ,  Rio de Janeiro ,  Brazil  Received 19 September 2000; received in revised form 6 February 2001; accepted 13 February 2001 Abstract In this study the contribution of biological N 2  fixation (BNF) to leguminous green manures was quantified in thefield at different sites with different  15 N methodologies. In the first experiment, conducted on a Terra Roxa soil inCuba, the BNF contribution to three legumes ( Crotalaria juncea ,  Mucuna aterrima  and  Cana  alia ensiformis ) wasquantified by applying  15 N-labelled ammonium sulphate to the soil. The second experiment was planted in a very lowfertility sandy soil near Rio de Janeiro, and the  15 N natural abundance technique was applied to quantify BNF in  C  .  juncea ,  M  .  ni   eum  and soybean. In both studies the advantages of using several non-N 2 -fixing reference plants wasapparent and despite the much greater accumulation of the  C  .  juncea  in the experiment performed on the fertile soilof Cuba, the above ground contributions of BNF at both sites were similar (40–80 kg N ha − 1 ) and greater than forthe other legumes. In a further experiment the possible contribution of root-derived N to the soil / plant system of twoof the legumes was quantified using a  15 N-leaf-labelling technique performed in pots. The results of this studysuggested that total below-ground N could constitute as much as 39 to 49% of the total N accumulated by the legumecrops. © 2001 Elsevier Science B.V. All rights reserved. Keywords :   Biological N 2  fixation; Green manures; Quantification; Root residues; Tropical / locate /  jbiotec 1. Introduction Many small-holders, especially food crop pro-ducers, in Third World countries have limitedaccess to chemical fertilisers. In some countries * Corresponding author. Fax:  + 55-21-2682-1230. E  - mail address : (R.M. Boddey).0168-1656 / 01 / $ - see front matter © 2001 Elsevier Science B.V. All rights reserved.PII: S0168-1656(01)00335-2  M  . G  .  Ramos et al  .  /   Journal of Biotechnology  91 (2001) 105  –  115  106 prices have risen sharply as government subsidieshave been removed in response to economic aus-terity plans and this has been re fl ected in a reduc-tion in fertiliser use and, in consequence, cropyields (Raussen, 1998). For the correction of soilacidity and P and K de fi ciency there are fewalternatives to chemical fertiliser, as such small-holders rarely have access to suf  fi cient organicmatter (Quin ˜ ones et al., 1998). However, in manyfarming systems the use of leguminous green ma-nures is traditional, and the inputs from biologicalnitrogen  fi xation often promote signi fi cant in-creases in subsequent grain or other crops. Theextent to which a legume crop can bene fi t asubsequent crop depends on the quantity of bio-logically  fi xed N which is incorporated into thesystem by the legume, the proportion of residualN left over for the subsequent crop, and its ef  fi -ciency of utilisation (Boddey et al., 1997; Giller etal., 1998).Only  15 N-based techniques have the potential toseparately evaluate the inputs of BNF- and soil-derived N to the legume based on a single harvestof the legume crop at maturity. Two variants of the  15 N dilution technique are available; a) thatwhere the soil is amended with  15 N-enriched N( 15 N enrichment technique), and b) where thetechnique is applied with no addition of isotopi-cally-enriched N and the BNF estimate is derivedfrom the difference in the natural abundance of  15 N of the legume crop and neighbouring non-N 2 - fi xing reference plants ( 15 N natural abundancetechnique). In both cases it is not possible todetermine directly if any particular non-N 2 - fi xingreference crop accumulates N with the same  15 Nenrichment / abundance as the legume crop. Forthis reason several non-N 2 - fi xing reference plantswere utilised in both studies to produce severalindependent estimates of the BNF contribution(Boddey et al., 1995, 2000). The range of theseestimates is then considered to be an index of their accuracy.A further concern is that considerable quanti-ties of biologically  fi xed nitrogen are to be foundin the roots and nodules of the legumes. Even insandy soils, recovery of all roots, especially in the fi eld, is almost impossible. Roots continuouslysenesce during plant growth so some of theirnitrogen, whether acquired from the soil or BNF,is released into the soil as these roots decompose,and this N is not accounted for when roots aremanually separated from the soil. Russell andFillery (1996) and McNeill et al. (1997) developeda  15 N leaf-labelling technique to assess the totalcontribution of root derived N to the plant / soilsystem. This technique is based on the idea that if the aerial tissue of the plants is labelled with  15 N,then a proportion of the  15 N label is translocatedto the roots. If these roots senesce and the N isreleased into the soil by mineralisation, then thesoil N will also become labelled with  15 N. It isassumed that the  15 N enrichment of the N re-leased into the soil is the same as that of therecoverable roots and hence from the quantity of excess  15 N found in the soil, quantity of total Nreleased by the roots can be quanti fi ed (Russelland Fillery, 1996).In this paper we report two studies which wereperformed to quantify the BNF inputs to thelegume using  15 N techniques in the  fi eld and thebene fi t to a subsequent maize crop, and also a potexperiment to estimate the proportion of N in thebelow ground N of two green manure legumesusing a  15 N leaf-labelling technique. 2. Materials and methods 2  . 1 .  Experiment  1 The experiment was performed in 1996 at thecentral experimental station of the Instituto Na-cional de Ciencias Agr ı´ colas, San Jose ´  de LasLajas, Province of Havana, Cuba, (23 °  00   N, 32 ° 12   W, altitude 138 m above sea level) on a darkred latosol (Haplustalf) with the following chemi-cal characteristics (0  –  30 cm): organic matter,2.4%; pH, 6.5; available P (Truog), 384 mg kg − 1 ;and exchangeable cations (cmol kg − 1 ): K, 0.63;Ca, 11.6; Mg, 2.6; Al, 0.0. The treatments con-sisted of   fi ve crop species, three legumes(Sunnhemp  —  Crotalaria juncea , Mucuna  —  Mu - cuna aterrima , and  Cana  alia ensiformis ), andmaize and sorghum as non-N 2 - fi xing referencecrops. The plots (3.6 × 6.0 m) were arranged in arandomised complete block design with fourreplicates.  M  . G  .  Ramos et al  .  /   Journal of Biotechnology  91 (2001) 105  –  115   107 The legumes and the reference crops wereplanted on May 1st and the legumes were inocu-lated with the rhizobium as a peat-based inocu-lant (10 11 cfu g − 1 ) mixed with solution of 10%sugar using the following strains from the Em-brapa Agrobiologia culture collection:  C  .  juncea  —  BR2001,  M  .  aterrima  —  BR2811,  C  .  ensi  -  formis  —  BR2003. Seed density was 0.9, 5.0 and9.75 g per linear metre for the  C  .  juncea ,  M  . aterrima  and  C  .  ensiformis , respectively. At 10days after germination (DAG) a central area(2.0 × 0.90 m) of each plot was amended with 15 N-labelled (10 atom%) ammonium sulphate at arate of 2 g N m − 2 (20 kg N ha − 1 ). 2  . 1 . 1 .  Har  ests and analyses At 60 DAG, samples were taken from a centralarea of 0.45 m 2 of the  15 N-labelled sub-plots,dried (  72 h at 65  ° C) and weighed. All crops(legumes and reference crops) were then incorpo-rated into the soil with a disc harrow, then theseed bed was prepared for planting of maize(using a mouldboard plough followed by a chiselplough) in the whole area of the experiment. Themaize (variety Francisco Mejorado) was plantedat 25 cm spacing in rows 0.90 m apart, 15 daysafter the incorporation of the green manures. Themaize was harvested on 1st October and the sam-ples divided into stems, leaves and cobs, dried(  72 h at 65  ° C) and weighed.All plant material was ground to   0.85 mmusing a Wiley mill and analysed for total nitrogenusing the semi-micro Kjeldahl technique followedby steam distillation and quanti fi cation of ammo-nium in the distillates using Nessler ’ s reagent. Theammonium sulphate resulting from the steam dis-tillation of the digested samples of the greenmanure crops taken from the  15 N-labelled sub-plots was dried and aliquots were prepared foranalysis of their  15 N enrichment using a NOI-6emission spectrometer according to the proceduredescribed by Fielder and Proksch (1975). Theproportional contribution (%Ndfa) of BNF to thethree legume crops was calculated using Eq. (1),using either maize or sorghum as the non-N 2 - fi xing reference crop, hence producing two indi-vidual estimates of %Ndfa for each legume(Chalk, 1985).%Ndfa = 100  [1 − (Atom%  15 N excess of legume / Atom%  15 N excess of reference plant)] (1) 2  . 2  .  Experiment  2 2  . 2  . 1 .  Experimental design This experiment was planted at the  fi eld stationof Embrapa Agrobiologia, 60 km west of Rio deJaneiro, (22 °  47   S, 43 °  40   W, altitude 33 m abovesea level) and the soil (Series Ecologia-Hapludult)consisted of 80% sand. Before the application of chemical fertilisers this soil had the followingchemical characteristics (0  –  20 cm): organic C(Walkley-Black), 0.33%; total N, 0.030%; pH, 4.9;available P (Mehlich I), 5 mg kg − 1 ; and ex-changeable cations (cmol kg − 1 ): K, 0.02; Na,0.02; Ca, 0.3; Mg, 0.3; Al, 0.3. The soil wasploughed incorporating mainly Guinea grass( Panicum maximum ) and 1 t ha − 1 of dolomiticlime 2 months before planting. In April (the cooland usually dry season), oats ( A  ena strigosa )were planted over the whole area by surface appli-cation of 37.5 kg ha − 1 of seeds along with 500 kgha − 1 of 4-14-8 fertiliser and 40 kg ha − 1 of frittedtrace elements (Type BR 12) in an attempt toincrease soil organic matter. After 70 days of growth (before grain was produced) the area wastreated with the herbicide Glyphosate(RoundUp ® , Monsanto, 3 l ha − 1 ) and the oatslaid down with a knife roller.The experiment consisted of six treatments ar-ranged in a randomised complete block designwith four replicates:1. Soybean followed by direct-drilled maize2. Mucuna cinza ( Mucuna ni   eum ) followed bydirect-drilled maize3. Sunnhemp ( Crotalaria juncea ) followed by di-rect-drilled maize4. Sunnhemp followed by conventional tilledmaize5. Unplanted  —  oat residues left on the soil sur-face followed by direct-drilled maize6. Unplanted  —  oat residues incorporated intothe soil followed by conventional tilled maizeThe legume seeds were inoculated with peat-based inoculants of rhizobium strains BR29 +  M  . G  .  Ramos et al  .  /   Journal of Biotechnology  91 (2001) 105  –  115  108 BR96, BR2811 and BR2001 + 2003 for the soy-bean, mucuna and sunnhemp, respectively fromthe Embrapa Agrobiologia collection. The plots(8 × 10 m) consisted of 16 rows of each legumespaced 0.5 m between rows. The sunnhemp wasplanted at 40 kg seeds ha − 1 , the mucuna at 65 kgha − 1 and the soybean 30 kg ha − 1 to obtain,respectively 15  –  20, 5 and 10 plants per linearmetre.Harvests of 1 m 2 of all aerial tissue were takenfrom each plot at 15, 21, 28, 35, 42, 49, 60 and 75days after planting (DAP). In the same area rootswere recovered to a depth of 15 cm. At eachharvest, samples were taken from neighbouringplots of aerial tissue of maize and sorghum andthree different non-leguminous weed speciesgrowing with the plots (nutgrass  —  Cyperus rotun - dus , Bermuda grass  —  Cynodon dactylon  and  Sida glazio  ii   ) to serve as non-N 2 - fi xing referenceplants to apply the  15 N natural abundancetechnique.All plant samples were dried (  72 h, 65  ° C),weighed and ground (  0.85 mm) using a Wileymill and analysed for total nitrogen using thesemi-micro Kjeldahl technique followed by steamdistillation using a Kjeltec Model 3010 automaticdistillation / titration unit (Tecator, Ho ¨ go ¨ nas, Swe-den) as described by Urquiaga et al. (1992). Sub-samples of the ground plant material were furtherground to a  fi ne powder using a roller mill similarto that described by Smith and Myung (1990) andaliquots containing between 35 and 70   g of Nwere weighed into tin capsules for analysis of   15 Nisotope abundance using a continuous- fl ow iso-tope-ratio mass spectrometer consisting of a CarloErba Model EA 1108 automatic C and N analysercoupled to a Finnigan Delta Plus mass spectrome-ter (Finnigan MAT, Bremen, Germany).In this case the calculation of the BNF contri-bution was based on the equation:%Ndfa = 100.(  15 N ref  −  15 N fi xing plant ) / (  15 N ref  − B  )(Shearer and Kohl, 1986) (2)where  B   is the  15 N abundance of N derivedfrom N 2  fi xation which may differ slightly fromthat of atmospheric N 2 . 2  . 3  .  Experiment  3  The same Ecologia series soil used in Experi-ment 2 was used in this pot experiment with nofertiliser addition. Ten pots were  fi lled with 12.5kg soil and  fi ve were planted with  Cana  alia  and fi ve with mucuna. Each pot received  fi ve seedswhich 10 days after germination were thinned totwo vigorous plants per pot. Thirty days afterplanting one fully expanded leaf from the middleof the canopy of each plant was labelled with  15 Nas follows (McNeill et al., 1997): the leaf wastrimmed so as to leave a section approximately5 × 10 mm attached to the petiole and this wasinserted into a small vial containing 1 ml of anaqueous solution (5 mg ml − 1 ) of   15 N-enrichedurea (71 atom%  15 N) for 24 h. The vial hangingfrom the leaf  / petiole was wrapped in aluminiumfoil to reduce evaporation losses. This procedurewas repeated two more times at 3 day intervals.Care was taken to not allow the  15 N-labelledsolution to drip onto the soil and any senescentleaves which fell onto the soil surface during plantgrowth were immediately removed, again to avoidcontaminating the soil with  15 N enriched N, inthis case derived from these leaves. The aerialtissue, roots, nodules and soil of all plants wereharvested at 70 DAP, separating the plants intogreen and senescent leaves, petioles, stems, (recov-erable) roots and nodules. The soil was entirelyremoved from the pots, air dried and sieved toremove all roots then weighed and sub-sampled.All plant material was dried (  72 h, 65  ° C),weighed and ground with a Wiley mill as in theprevious experiments. Aliquots of the soil andplant samples were  fi nely ground using the rollermill for subsequent analysis of   15 N enrichmentusing the automated continuous- fl ow mass spec-trometer as described above (Section 2.2). 3. Results 3  . 1 .  Experiment  1 The total dry matter production of the differentgreen manure crops after 60 days of growthranged from the equivalent of 2.1 t ha − 1 (mu-  M  . G  .  Ramos et al  .  /   Journal of Biotechnology  91 (2001) 105  –  115   109Table 1Accumulation of dry matter and N, and  15 N enrichment (atom%  15 N excess) of   fi ve green manure crops (three legume + sorghumand maize) grown in soil amended with 20 kg of   15 N-labelled ammonium sulphate (10 atom%  15 N)Total N kg ha − 1 Atom%  15 N excessDry matter mg ha − 1 Crop195.1 a 0.650 cSunnhemp 11.1 a64.0 c2.1 d 0.235 dMucuna4.4 cCanavalia 57.9 c 0.638 c99.9 b 0.820 bSorghum 7.6 b76.8 bc 0.980 a6.7 bMaizeCoef  fi cient of variation (%) 9.5 10.0 13.5Experiment 1, Cuba. Harvest at 60 days after planting. Values are means of four replicates. cuna) to 11.1 t ha − 1 (sunnhemp) (Table 1). Asexpected the N content (% N) of the referencecrops, maize and sorghum, was considerablylower than that of the leguminous green manures(data not shown), but these cereal crops accumu-lated more N than either the mucuna or the Cana  alia  even though the latter are capable of obtaining N from BNF.The  15 N enrichment of all three legume cropswas signi fi cantly lower than that of either themaize or sorghum indicating that all three ob-tained signi fi cant contributions of unlabelled Nfrom BNF (Table 1). Using these data to calculatethe proportion of N derived from BNF, the esti-mates for the sunnhemp were 39 and 27% of plantN when maize and sorghum were used as thereference crop, respectively (Fig. 1). While themucuna and the  Cana  alia  accumulated far lessdry matter and N than the sunnhemp, the muchgreater proportion of N derived from BNF by themucuna resulted in the total BNF contribution tothe sunnhemp and the mucuna being very similarand almost double that of the  Cana  alia . Thesedata illustrate the advantage of directly quantify-ing the BNF contribution using  15 N techniques inthat these results show that a large proportion(between 71 and 74%) of the (approximately) 200kg N ha − 1 accumulated by the sunnhemp wasderived from the soil.These data explain why, despite the fact thatthe sunnhemp accumulated between three and  fi vetimes as much N as that accumulated by themucuna, the maize yield achieved in the plotsfollowing sunnhemp yielded less than 1 t ha − 1 more grain, probably because the sunnhemp re-moved more N from the soil prior to the plantingof maize (Fig. 2). 3  . 2  .  Experiment  2  The oats pre-crop yielded a mean of 160 g drymatter m − 2 (1.6 t ha − 1 ) containing 2.5 g N m − 2 (25 kg N ha − 1 ). Despite the fact that the legumeseeds showed a high rate of germination (  80%)when tested in the laboratory, germination in the fi eld was considerably lower owing to the lowwater holding capacity of this very sandy soil. Inconsequence, dry matter production during the fi rst 35 days was low, and dry matter yields at the fi nal harvest (75 DAP) reached only 3.7, 2.8 and1.0 t ha − 1 for the sunnhemp, mucuna and soy- Fig. 1. Estimates of the proportion of N derived from BNFand from soil using the isotope dilution technique on soilamended with  15 N (Experiment 1, Cuba). Estimates werederived from the  15 N enrichment of the legume plant andeither (M) maize or (S) sorghum as references. Values aremeans of four replicates.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!