Labile carbon fraction Recalcitrant carbon fraction Organic amendment

Changes in soil organic carbon pools after10years of continuous manuring combined with chemical fertilizer in a Mollisol in ChinaXueli Ding,Xiaozeng Han*,Yao Liang,Yunfa Qiao,Lujun Li,Na LiKey Laboratory of Mollisols Agroecology,Northeast Institute of Geography and Agroecology,Chinese Academy of Sciences,Harbin150081,PR China1.IntroductionMaintaining soil organic carbon(SOC)is particularly importantfor sustaining the productivity of agroecosystems,because SOCplays a central role in soil quality and functioning throughinfluencing soil physical,chemical,and biological properties(Lalet al.,1999;Carter,2002).SOC levels are governed by the balancebetween carbon(C)input and output,and strongly influenced bysoil management practices(Paustian et al.,2000).Studies haveshown that maintenance of SOC at optimum level can be managedthrough crop rotation,reduced tillage,and fertilization practices(Six and Jastrow,2002;Verma and Sharma,2007;Gong et al.,2009).Most research indicated that the application of organicmanure in combination with inorganic fertilizer linearly increasedSoil&Tillage Research122(2012)36–41A R T I C L E I N F OArticle history:Received29December2011Received in revised form14February2012Accepted16February2012Keywords:Labile carbon fractionRecalcitrant carbon fractionOrganic amendmentLong-term fertilization experimentSoil carbon sequestrationA B S T R A C TIt is still unclear that whether substantial amounts of organic manure amendment could increase soilorganic carbon(SOC)sequestration in a Chinese Mollisol with relatively high organic C content.Further,changes accompanied by different organic carbon(C)fractions are not well understood based on long-term experiment.The knowledge of this kind is important for assessing the potential for C sequestrationof a high SOC soil and selecting effective management practices for increasing soil C sequestration andproductivity in agroecosystem of northeastern China.This study was aimed to assess the effects of10years’organic manuring at various rates combined with chemical fertilizer on organic C sequestration intopsoil(0–20cm)and to identify changes in different SOC(total,labile,and recalcitrant)pools.Fivefertilization treatments were included:(1)CK,unfertilized control;(2)OM0,only chemical fertilizer,nomanure added;(3)OM1,organic manure added at7.5Mg haÀ1yearÀ1plus chemical fertilizer;(4)OM2,organic manure added at15Mg haÀ1yearÀ1plus chemical fertilizer;and(5)OM3,organic manure addedat22.5Mg haÀ1yearÀ1plus chemical fertilizer.Chemical fertilizer was supplied with the same rate ineach fertilized treatment.We found that the application of graded rates of manure from OM1to OM3significantly enhanced total SOC,labile C pools,and recalcitrant C pool as compared with OM0and CK.The C storage(in top20cm)in the OM0,OM1,OM2,and OM3was increased by3.19%,12.5%,14.5%,and18.2%,respectively,over the CK treatment,suggesting that the chemical fertilizer addition had less effectson C-sequestration in topsoil compared to manure plus mineral fertilizer amendment.Moreover,topsoilC-sequestration increased with organic manure addition rates with an order of OM3(10.5Mg haÀ1)>OM2(8.4Mg haÀ1)>OM1(7.2Mg haÀ1).A positive relationship between C sequestra-tion and organic manure input indicates that the soil has not reached its maximum capacity of Csequestration.Application of organic manure with chemical fertilizer was found to produce greater sizeof both labile and recalcitrant pools than application of mineral fertilizers alone.In most cases,theincreases in these C fractions were greater when organic manure was supplied at higher rates.Moreover,increase in recalcitrant C(10.5–29.5%)was significantly higher than labile C(5.6–10.2%)in manuretreated plots as compared with no amendment plot.This indicated that a majority of organic Csequestered due to C inputs was accumulated and stabilized in recalcitrant C pool.In general,organicmanure combined with inorganic fertilizer exerted greater influence on topsoil C storage and crop yieldthan chemical fertilizer alone.Based on crop yield and soil C storage,applying organic manure at the rateof22.5Mg haÀ1yearÀ1was the most effective way to improve soil productivity and C sequestration in theagroecosystem of northeastern China.However,taking use efficiency of organic inputs andenvironmental factors into consideration,application of manure at moderate rate(about15Mg haÀ1yearÀ1)may be more feasible in this region.ß2012Elsevier B.V.All rights reserved.*Corresponding author.Tel.:+86045186602940.E-mail address:xzhan@(X.Han).Contents lists available at SciVerse ScienceDirectSoil&Tillage Researchj ou r n a l h o m e p a ge:w w w.e l s e v i e r.co m/l o c a t e/s t i l l0167-1987/$–see front matterß2012Elsevier B.V.All rights reserved.doi:10.1016/j.still.2012.02.002SOC levels(Paul et al.,1997;Bo¨hme et al.,2005;Rudrappa et al., 2006;Li et al.,2010).Nevertheless,some experiments showed little or no increase in SOC levels for a high C soil even with two or three fold increases in extraneous C inputs(Campbell et al.,1991; Paustian et al.,1992).Six and Jastrow(2002)ascribed this to C saturation phenomenon in these soils,i.e.,once the capacity for a soil to protect C is reached,excessive C inputs will not be sequestered as SOC.Therefore,to gauge the potential for C sequestration of a soil,it is of great importance for assessing the C status in relation to C saturation(Kong et al.,2005).Generally,changes in SOC induced by management practices occur slowly;these changes are relatively small as compared to the vast SOC pool size and also vary both spatially and temporally (Paustian et al.,1992;Salinas-Garcia et al.,1997).SOC fractions with different stabilities and turnover rates are important variables to detect the influence of agricultural management on soil quality(Silveira et al.,2008).Therefore,SOC dynamics are usually described by compartmentalizing SOC into different pools with various function and bioavailability(Campbell et al.,1997; Silveira et al.,2008).Acid hydrolysis has been demonstrated as an effective technique to separate and estimate different C fractions or pools with various degradabilities(Paul et al.,1997;McLauchlan and Hobbie,2004;Belay-Tedla et al.,2009).The acid-hydrolyzable C fraction,defined as labile C pool,is largely comprised of polysaccharides,proteins,nucleic acid,and some carboxyl C (Rovira and Vallejo,2002).Changes in this fraction could provide an earlier indicator of soil management effects on soil organic matter(SOM)than changes in total C alone(Franzluebbers and Stuedemann,2002;Banger et al.,2009).The non-hydrolyzable fraction,defined as recalcitrant C pool,is typically comprised of lignin and related compounds as well as fats,waxes,and resins, which could provide clues regarding the potential of SOC sequestration over the long term(Paul et al.,1997;McLauchlan and Hobbie,2004;Lima et al.,2009).The alteration in the size of labile and recalcitrant pools would influence the timeframe and longevity of soil C sequestration(Paul et al.,2003).Therefore, fractionating and quantifying the labile and recalcitrant C pools could provide valuable information for better understanding SOC changes and the underlying mechanisms(Belay-Tedla et al.,2009).Mollisol was once very fertile,noted for its high SOC content and crop productivity(Liang et al.,2009).It is often called‘The king of soils’in China and serves as a key base of grain production. Mollisol is mainly distributed in northeastern China and has a total area of 5.9Â106ha.However,about 4.4Â106ha of it was cultivated(National Soil Survey Office,1998).A significant depletion of C stocks in the soil occurred during the last decades because of active agricultural utilization,low organic return,and intensive tillage.It was estimated that about50%of the initial C content has been lost in this region due to intensive cultivation(Liu et al.,2003).The serious losses of SOC have deteriorated the soil quality and sustainable crop productivity,which spurred scientist to explore the ways to restore SOC status(Liu et al.,2010;Ding et al.,2011).Long-term fertilizer experiments in southern China showed that applying chemical fertilizer in combination with farmyard manure significantly increased SOC content more than using chemical fertilizer alone in subtropical cropping system(Li et al.,2010).However,organic inputs(crop residues or farmyard manure)at normal level(2–4.5Mg haÀ1)in northeastern China did not have significant effects on restoring SOC level in temperate Mollisol(Li et al.,2002).It is still unclear whether higher level of organic inputs could lead to a corresponding increase in soil C sequestration in a high SOC soil.Further,the responses of various SOC fractions to increased organic manure are not well docu-mented in the Mollisol based on long-term experiments.The present study aimed to:(1)evaluate the changes in topsoil C storage and sequestration in a Mollisol in China under chemical fertilization and organic manuring at various rates;and(2) quantify the responses of different organic C pools(total,labile, and recalcitrant)to different fertilization treatments.We hypoth-esized that C storage and the accumulation of different SOC pools might be affected by fertilization treatments for the Mollisol in a temperate agroecosystem.Here,we paid special attention to the changes in SOC and its fractions.The information would help us understand the processes involved in soil C sequestration and the accumulation patterns of labile-and recalcitrant-C fractions in relation to different manuring rates in the Mollisol.2.Materials and methods2.1.Study siteThe study site was set up in2001at Hailun National Field Station,Chinese Academy of Sciences,located in Hailun,Heilong-jiang province(478260N,1268380E).The region has a typical temperate continental monsoon climate,characterized by hot summers and cold winters.The mean annual rainfall is550mm, with about65%occurring from June to August.Mean annual temperature(MAT)is1.58C,and mean monthly air temperature varies fromÀ238C in January to218C in July.The soil,derived from loamy loess,is classified as Udolls according to the USDA Soil Taxonomy System(Soil Survey Staff, 1998).The soil texture is25.7%sand,32.4%silt,and40.8%clay.The soil at the start of the experiment had a pH of 6.2(soil:-water=1:2.5)and contained26.9g C kgÀ1soil and2.0g N kgÀ1 soil.The soil bulk density was1.03g cmÀ3,Bulk determined by hand with a known volume steel cylinder(5cm inner diameter and 20cm height)driven into0–20cm soil layer.Soil cores were oven-dried at1058C for48h.The bulk density was calculated by dividing the weight of the dried soil by the volume of the soil.More detailed information about the soil properties and management can be obtained from Liu et al.(2003).2.2.Experimental designThe treatments selected for the present study are:(1)CK, unfertilized control;(2)OM0,only chemical fertilizer,no manure added;(3)OM1,organic manure added at7.5Mg haÀ1yearÀ1plus chemical fertilizer;(4)OM2,organic manure added at 15Mg haÀ1yearÀ1plus chemical fertilizer;and(5)OM3,organic manure added at22.5Mg haÀ1yearÀ1plus chemical fertilizer.The treatments were arranged in a randomized block design with three replicates.Each plot was12mÂ5.6m.There was a buffer zone of 0.7m between each plot.Chemical fertilizer was supplied annually with the same application rate within the treatments(Table1). Urea was the nitrogen(N)source and was applied at rates of 30kg haÀ1yearÀ1for soybean and150kg haÀ1yearÀ1for maize. Ammonium hydrogen phosphate was the phosphorus(P)and potassium(K)sources and was applied at rates of36and 32kg haÀ1yearÀ1for soybean and maize,respectively.The pig manure was collected from the same source every year and analyzed for C,N,P,and K contents before application.The manure had an average composition of265Æ15.1g C kgÀ1, 21Æ1.8g N kgÀ1, 2.6Æ0.2g P kgÀ1,and 2.4Æ0.3g K kgÀ1on a dry-weight basis(Liang et al.,2011).The annual application rates of manure and chemical fertilizer in the different treatments are listed in Table1.Cropping systems had been soybean[Glycine max (Merrill.)L.]and maize(Zea mays L.)rotation since2001.Intensive tillage is a common soil management practice used in the farming ly,all plots were ridged by rotary tillage in October after harvest,and all above-ground crop residues were removed from field plots.In the next spring,crop was planted on the top of ridgesX.Ding et al./Soil&Tillage Research122(2012)36–4137with a conventional planter in early May.The second ridge was carried out at about half a month after planting.2.3.SamplingSoil samples were collected from surface layers (0–20cm soil depth)of each plot in April posite soil samples (more than five random subsamples)in the same plot were pooled together,placed in polyethylene bags,and transported to the laboratory immediately.In the laboratory,visible roots and plant fragments were removed.The composite samples were passed through a 2-mm sieve.One part of processed sample was air-dried for analysis of various C fractions.The remaining soil was oven-dried at 508C and ground to pass a 0.25-mm sieve for total C analysis.At crop maturity,10individuals in each plot were selected randomly for measuring grain anic matter input to soil was through crop root biomass and organic manure.Annual biomass of crop roots was estimated in proportion by the average annual soybean and maize yields from 2001to 2011.Generally,proportion of grain yield to root biomass in the top 20cm is 1:0.13for maize and 1:0.19for soybean,respectively,in northeastern China (Zhang and Zhu,1990).2.4.Total soil organic C analysisTotal soil C of bulk soils was analyzed with a VarioEL CHN elemental analyzer (Heraeus Elementar Vario EL,Hanau,Germany).Because the soils were free of carbonates,the total C content was equivalent to SOC content.C storage (Mg C ha À1)of soil at the 0–20cm depth was calculated from horizon thickness and soil bulk density value.The amount of sequestered organic C due to manuring (Mg C ha À1)in 0–20cm soil depth was determined after deducting the C storage in control from other treatments.bile and recalcitrant C poolsIn the present study,a two-step acid hydrolysis method was used to determine labile and recalcitrant C pools (Rovira and Vallejo,2002).Briefly,500mg air-dried soil was hydrolyzed with 20mL of 2.5M H 2SO 4at 1058C in sealed Pyrex tubes for 30min.The residue after hydrolysis was separated from supernatant by centrifugation (15min at 6037Âg )and then decantation,washed with 20mL deionized water.The combined supernatant and washings were pooled together;which was taken as labile pool I carbon (LPI-C).The remaining residue was washed three times with deionized water and dried at 508C.Then,the dried residue was hydrolyzed with 2mL of 13M H 2SO 4at room temperature under continuous shaking overnight.After this,the acid concen-tration was diluted to 1M with deionized water and then thesample was hydrolyzed for 3h at 1058C with shaking per 30min.The hydrolysate was also recovered by centrifugation and was taken as labile pool II carbon (LPII-C).The residue fraction was washed again with deionized water,dried at 508C,being taken as recalcitrant pool carbon (RP-C).Organic C in both hydrolysates (LPI and LPII)was analyzed using Elementar liquid TOC II analyzer (Elementar Analysensysteme Hanau,Germany)and total C in the residue fraction was measured using VarioEL CHN elemental analyzer (Heraeus Elementar Vario EL,Hanau,Germany).2.6.Statistical analysisOne-way analysis of variance (ANOVA)procedures,with Tukey’s honestly significant difference (HSD)as a post hoc,were used to test for significant differences in variables among treatments.Correlations between variables were calculated with the Pearson correlation coefficients.Statistical analysis was performed using the software package SPSS 13.0for Windows (SPSS Inc.,Chicago,USA).Figures were generated using Sigmaplot 10.0(Systat Software Inc.).3.Results3.1.SOC storage and sequestrationDifferent fertilization treatments had a significant influence on both SOC storage and sequestration.Topsoil (0–20cm)C storage was significantly higher in the manure applied treatments (OM 1,OM 2,OM 3)compared with the CK and OM 0treatments (P <0.05,Fig.1).The plots under OM 3treatment stored 14.5%higher (P <0.05)organic C than the plots under OM 0treatments in topsoil.No differences were found in organic C storage between OM 2and either OM 1or OM 3treatments (Fig.1).SOC concentra-tions varied in the same way as organic C storage,with a range of 25.6–30.3g kg À1(Table 2).The amount of sequestered organic C was highest under OM 3treatment (10.5Mg ha À1),followed by OM 2(8.4Mg ha À1)and OM 1(7.2Mg ha À1)treatments,and lowest under OM 0treatment (1.8Mg ha À1)at 0–20cm soil depth (Fig.1).bile and recalcitrant C poolsDifferent fertilization treatments resulted in significant differ-ences in the stock of both labile and recalcitrant C (Table 2).On average,organic fertilized treatments increased LPI-C,LPII-C,and RP-C by 7.8,12.9,and 19.4%,respectively,as compared to CK.OM 3soils contained the highest labile (considering both pools)and recalcitrant C pools.OM 1and OM 2treatments resulted in different LPI-C and LPII-C pools (P <0.05),but similar RP-C pool (P >0.05).No differences were found in those three pools between OM 2and OM 3treatments.RP-C pool accounted for a larger proportion in total SOC (54.0–59.3%)than both LPI-C (18.0–19.2%)and LPII-CTable 1Experiment design and addition rates of organic manure and mineral fertilizer in different treatments.Treatment aSoybeanMaizeManure (kg ha À1)Mineral fertilizerManure (kg ha À1)Mineral fertilizerUrea (kg N ha À1)Ammonium hydrogen phosphate (kg P ha À1)Urea (kg N ha À1)Ammonium hydrogen phosphate (kg P ha À1)CK 000000OM 003036015032OM 175003036750015032OM 215,000303615,00015032OM 322,500303622,50015032aCK,unfertilized control;OM 0,only chemical fertilizer,no manure added;OM 1,organic manure added at 7.5Mg ha À1year À1plus chemical fertilizer;OM 2,organic manure added at 15Mg ha À1year À1plus chemical fertilizer;OM 3,organic manure added at 22.5Mg ha À1year À1plus chemical fertilizer.X.Ding et al./Soil &Tillage Research 122(2012)36–4138(24.4–25.2%)pools.Moreover,increase in recalcitrant C (9.3–29.5%)was significantly higher than labile C (2.8–10.2%)due to fertilization,as compared with CK.Significant correlations were found among SOC,LPI-C,LPII-C,and RP-C (P <0.05,Table 3).4.DiscussionOur results clearly support the hypothesis that organic fertilization has greatly positive impact on soil C storage and sequestration.In the present study,the above-ground biomass was removed from field plots and there was no crop residue incorporation into the soil.Thus the input of organic matter was mainly through root biomass and organic manure.The increase in C storage and sequestration in organically fertilized treatments can be attributed to greater C inputs through organic manure and root biomass due to better crop growth (Table 4).Similar results have been reported from other long-term experiments with low SOC content (Potter et al.,1998;Rudrappa et al.,2006;Gong et al.,2009;Li et al.,2010).For example,Sanchez et al.(2004)observed a significant increase in SOC when low rate of dairy manure (4Mg ha À1)applied annually in arable soils (SOC <7g kg À1)of SW Michigan.Nevertheless,Campbell et al.(1991)suggested that varying C inputs showed no effect on topsoil organic C levels at Melfort,Saskatchewan.They speculated that the lack of SOC response was attributed to its high organic matter content in soil (about 64Mg C ha À1in the top 15cm).Significant increase in C storage was still observed in treatments of higher organic manure addition in the current study (Fig.1),suggesting that soil C sequestration could improve after substantial organic matter inputs even in high C soils (62.1Mg C ha À1in top 20cm).The discrepancies between trials may be caused by differences inS o i l o r g a n i c C (M g h a -1)CK OM 0OM 1OM 2OM 3Fig.1.Soil carbon storage and sequestration in soils after 10years of different fertilization treatments.CK,unfertilized control;OM 0,only chemical fertilizer,no manureadded;OM 1,organic manure added at 7.5Mg ha À1year À1plus chemical fertilizer;OM 2,organic manure added at 15Mg ha À1year À1plus chemical fertilizer;OM 3,organic manure added at 22.5Mg ha À1year À1plus chemical fertilizer.Values reported as means Æstandard deviations,n =3.Bars with different letters show a significant difference at P <0.05(HSD).Table 2Soil organic carbon (SOC),labile pool I carbon (LPI-C),labile pool II carbon (LPII-C),recalcitrant pool carbon (RP-C)and their proportion in SOC (%)in soils after 10years’different fertilization treatments.Treatment a SOC (g kg À1)LPI-C (g kg À1)LPII-C (g kg À1)RP-C (g kg À1)LPI-C (%of SOC)LPII-C (%of SOC)RP-C (%of SOC)CK 25.63Æ0.22c b 4.93Æ0.11d 6.35Æ0.08d 13.84Æ0.46c 19.21Æ0.4524.76Æ0.3554.00Æ2.26OM 026.45Æ0.24c 5.07Æ0.03cd 6.56Æ0.05d 15.13Æ0.39c 19.15Æ0.2724.78Æ0.1557.19Æ1.24OM 128.83Æ1.73b 5.11Æ0.09c 6.91Æ0.13c 15.40Æ0.27b 18.61Æ0.3924.71Æ0.5657.41Æ1.21OM 229.36Æ1.40ab 5.30Æ0.05ab 7.21Æ0.04ab 16.34Æ1.03ab 18.48Æ0.6525.18Æ1.1556.92Æ1.73OM 330.25Æ1.88a5.43Æ0.09a7.39Æ0.26a17.94Æ0.87a17.95Æ0.9924.42Æ0.6859.26Æ1.76aCK,unfertilized control;OM 0,only chemical fertilizer,no manure added;OM 1,organic manure added at 7.5Mg ha À1year À1plus chemical fertilizer;OM 2,organic manure added at 15Mg ha À1year À1plus chemical fertilizer;OM 3,organic manure added at 22.5Mg ha À1year À1plus chemical fertilizer.bValues reported as means Æstandard deviations,n =3.Different letters within each column show significant differences at P <0.05.Table 3Linear correlation coefficient (r )between soil organic carbon (SOC),labile pool I carbon (LPI-C),labile pool II carbon (LPII-C),and recalcitrant pool carbon (RP-C).SOCLPI-CLPII-CRP-CSOC 1LPI-C 0.993**1LPII-C 0.985**0.991**1RP-C0.966**0.965**0.943*1*Correlation is significant at the 0.05level (2-tailed).**Correlation is significant at the 0.01level (2-tailed).Table 4Mean crop yield and estimated input of root biomass over 10years’different fertilization treatments.Treatment aMaize yield (kg ha À1year À1)Soybean yield (kg ha À1year À1)Estimated inputs of root biomass (kg ha À1year À1)CK 5351Æ495.6d b 1793Æ45.0d 562.6OM 07085Æ547.9c 1968Æ86.3c 704.4OM 17593Æ151.9b 2254Æ101.9b 769.3OM 27824Æ142.9ab 2343Æ42.2ab 784.7OM 38132Æ372.2a2417Æ140.6a824.3aCK,unfertilized control;OM 0,only chemical fertilizer,no manure added;OM 1,organic manure added at 7.5Mg ha À1year À1plus chemical fertilizer;OM 2,organic manure added at 15Mg ha À1year À1plus chemical fertilizer;OM 3,organic manure added at 22.5Mg ha À1year À1plus chemical fertilizer.bValues reported as means Æstandard deviation,n =3.Different letters within each column show significant differences at P <0.05(HSD).X.Ding et al./Soil &Tillage Research 122(2012)36–4139climatic factors,topographic position,soil types,textures,and spatial variability in soil(Salinas-Garcia et al.,1997;Hassink,1997; Datta et al.,2010;Li et al.,2010).It is worthy to note that in our case increasing annual addition rates of organic manure from OM1to OM2did not produce significant increase in the amounts of sequestrated organic C(Fig.1).This could be attributed,in part,to the relatively small SOC changes induced by lower extraneous C addition as compared to the vast background of total SOC.Another reason for this observation may be due to the opposing effect of continuous tillage,i.e.,larger manure applications would be offset by the effects of annual tillage,which led to acceleration of SOC mineralization(Blair et al.,2006).The increases in SOC concentration from OM1to OM3were significant,addressing the significance of the considerable C supply for building up SOC pool (Table2).Interestingly,compared with the OM2treatment,extra manure in OM3treatment did not result in significant increase in SOC. Meanwhile,nonlinear regression indicated a significant positive relationship between soil C sequestration and manure input (R2=0.955,P<0.05).The positive relationship implies that the soil has not reached an upper threshold of C sequestration(i.e.,not saturated).Six and Jastrow(2002)suggested that every soil has its own maximum protective capacity of SOC,which is defined by soil inherent physicochemical properties.High clay content in the Mollisol(>40%)could be an important reason contributing to this high potential of C sequestration(Liu et al.,2003).This aspect requires further research into composition and distribution of soil aggregate and particle-size fractions and their association with organic C in this region.The higher organic C sequestration in manure treated plots than mineral treated plot has greater implications in regional level, particularly in temperate northeastern China.Generally,combination of organic and chemical fertilizers could improve crop yields and biomass production(Carter,2002; Banger et al.,2009;Gong et al.,2009).Similar result was observed in our study as shown in Table4.The organically fertilized plots produced significantly higher crop yield as compared with the CK and OM0plots,reflecting the positive interaction of integrating organic and inorganic fertilizers on crop yield and residue input. This could be interpreted as a physically/biologically favorable soil environment created by organic manure application(Pura-kayastha et al.,2008).In contrast,Edmeades(2003)stated that relative to the same amounts of nutrients application as chemical fertilizers,applying organic manures(farmyard manure,green manure,slurry)had no obvious advantages to soil productivity, measured as crop yields,despite the positive effects of these organic manures on soil physical and biological properties.The discrepancies between trials may be caused collectively by climatic factors,experiment duration,soil inherent fertility(Li et al.,2010),and different quality and quantity of added organic materials.Further,significant increases in crop yield and soil C sequestration were observed with increasing manure addition rates from7.5Mg haÀ1to22.5Mg haÀ1,addressing the importance of manure inputs at higher rate for improving soil productivity and C storage even in high organic matter soils.Thus,based on crop yield and SOC,applying organic manure at the rate of22.5Mg haÀ1 to soil seems to be effective in the studied agroecosystem. However,we should keep in mind that application of high rates of manure could lead to potentially greater greenhouse gas emissions and runoff or leaching of nutrients and increase environmental risks(Brouwer and Powell,1995;Hao and Chang,2003;Blair et al., 2006).Rochette and Gregorich(1998)reported that CO2emissions from surface soil were more than doubled in high-level manure addition treatment(100Mg haÀ1based on wet weight)compared with fertilizer treatment.In the current study,the gains of SOC with increasing manure addition rate from15Mg haÀ1to 22.5Mg haÀ1were not significantly(Table2),suggesting less effectiveness of higher rate of manure addition in further improving SOC level.Therefore,taking use efficiency of organic inputs and concerns for environmental effects into consideration, manure addition at moderate rate about15Mg haÀ1yearÀ1may be more feasible in this region.This is yet to be evaluated in future study.We are currently conducting a study to further investigate greenhouse gas emissions.SOC fractions with different stabilities and turnover rates are important variables to detect the influence of agricultural manage-ment on organic matter quality(Silveira et al.,2008).Manure application had beneficial effects on SOC sequestration in the labile pools(LPI-C and LPII-C)and to a greater extent in the recalcitrant pool(Table2).This result suggests a markable change in SOM quality between the different fertilization treatments after an experimental period of10years.The observed increase in labile C pool size in higher organic input systems could be attributable to differences in inputs of external organic matter.Ourfinding was in conformity with the observations made by Cambardella and Elliott(1992)and Janzen et al.(1992),who found that labile C levels were high in high substrate input systems and low in those with low substrate input. In addition,chemical fertilization combined with organic manuring at intermediate rate(OM2)did not change total SOC levels,but produced significant higher levels of labile C as compared with OM1. This validates that the labile C can be used as a sensitive indicator for SOC dynamics as affected by management practices(Rovira and Vallejo,2002).Meanwhile,a majority of the SOC storage due to extraneous C inputs was more sequestered in recalcitrant C pool (10.5–29.5%)than in both labile C pools(5.6–10.2%).Thesefindings indicate that large amounts of C inputs from manure became gradually stabilized in the soil,i.e.,a great portion of recalcitrant C may be formed within thefirst several years of experiment establishment.However,on average,unhydrolyzable C in many ecosystems is older than bulk soil C(Paul et al.,1997).Further investigation into the changes of this recalcitrant C pool in longer period is necessary in order to clarify its stable characteristics.The increase in recalcitrant C in manuring treatments could be related to biochemical resistance of organic-C compounds contained either in organic manure or plant materials(McLauchlan et al.,2006; Yan et al.,2007).Studies had shown that farmyard manure application resulted in an increase in lignin and lignin-like products, which are major components of the resistant C pool(Paul et al., 1997;Rovira and Vallejo,2002;Belay-Tedla et al.,2009).Therefore, the significant increases in recalcitrant C and its proportion in SOC for manure added soils(Table2)suggest that adding manure is of great importance for improving the SOM stabilization in the long term.Besides the higher organics inputs,the greater increase in amounts of recalcitrant C than that of labile C could partially be attributed to preferential degradation of labile compounds and accumulation of recalcitrant materials over time(Lopez-Capel et al., 2008).The soils under study were subjected to annual plowing, which could cause prevalent mineralization of labile fractions of SOC.In this regard,continuous cultivation could lead to decrease in the labile C pools,while it has little influence on the recalcitrant pool due to its relatively high stability(Shrestha et al.,2008).The significantly positive relationship between labile and recalcitrant C pools with total SOC(Table3)indicates that changes in SOM caused by agricultural management can be better understood by alterations in the different C fractions separated by acid hydrolysis;and also a transfer of compounds between different C fractions was expectable.5.ConclusionUse of organic manure with chemical fertilizer had positive effects on sequestering organic C in0–20cm layer of a Mollisol, cultivated intensively in a temperate agroecosystem in China. Moreover,higher addition rates of organic manure benefited more prominent increase in the total SOC and its various C fractions.X.Ding et al./Soil&Tillage Research122(2012)36–41 40。