Review of greenhouse gas__ emissions from crop production systems andfertilizer management__ effects

ReviewReview of greenhouse gas emissions from crop production systems and fertilizer management effectsC.S.Snyder a ,*,T.W.Bruulsema b ,T.L.Jensen c ,P.E.Fixen daInternational Plant Nutrition Institute,P.O.Drawer 2440,Conway,AR 72033,USAbInternational Plant Nutrition Institute,18Maplewood Drive,Guelph,Ontario,Canada N1G 1L8cInternational Plant Nutrition Institute,102-411Downey Road,Saskatoon,Saskatchewan,Canada S7N 4L8dInternational Plant Nutrition Institute,2301Research Parkway,Suite 126,Brookings,SD 57006,USAContents 1.Introduction.....................................................................................................2482.Background .....................................................................................................2482.1.Greenhouse gases ...........................................................................................2482.2.Agricultural share ...........................................................................................2482.3.Agricultural sources .........................................................................................2492.3.1.Emissions of nitrous oxide (N 2O)from cropland ...........................................................2492.3.2.Emissions of CO 2from lime ................. (251)Agriculture,Ecosystems and Environment 133(2009)247–266A R T I C L E I N F O Article history:Received 30May 2008Received in revised form 28August 2008Accepted 27April 2009Available online 3June 2009Keywords:Greenhouse gas Nitrous oxide FertilizerBest management practice NitrogenCropping system Climate change Global warming Carbon dioxide MethaneA B S T R A C TFertilizer nitrogen (N)use is expanding globally to satisfy food,fiber,and fuel demands of a growing world population.Fertilizer consumers are being asked to improve N use efficiency through better management in their fields,to protect water resources and to minimize greenhouse gas (GHG)emissions,while sustaining soil resources and providing a healthy economy.A review of the available science on the effects of N source,rate,timing,and placement,in combination with other cropping and tillage practices,on GHG emissions was conducted.Implementation of intensive crop management practices,using principles of ecological intensification to enhance efficient and effective nutrient uptake while achieving high yields,was identified as a principal way to achieve reductions in GHG emissions while meeting production demands.Many studies identified through the review involved measure-ments of GHG emissions over several weeks to a few months,which greatly limit the ability to accurately determine system-level management effects on net global warming potential.The current science indicates:(1)appropriate fertilizer N use helps increase biomass production necessary to help restore and maintain soil organic carbon (SOC)levels;(2)best management practices (BMPs)for fertilizer N play a large role in minimizing residual soil nitrate,which helps lower the risk of increased nitrous oxide (N 2O)emissions;(3)tillage practices that reduce soil disturbance and maintain crop residue on the soil surface can increase SOC levels,but usually only if crop productivity is maintained or increased;(4)differences among fertilizer N sources in N 2O emissions depend on site-and weather-specific conditions;and (5)intensive crop management systems do not necessarily increase GHG emissions per unit of crop or food production;they can help spare natural areas from conversion to cropland and allow conversion of selected lands to forests for GHG mitigation,while supplying the world’s need for food,fiber,and biofuel.Transfer of the information to fertilizer dealers,crop advisers,farmers,and agricultural and environmental authorities should lead to increased implementation of fertilizer BMPs,and help to reduce confusion over the role of fertilizer N on cropping system emissions of GHGs.Gaps in scientific understanding were identified and will require the collaborative attention of agronomists,soil scientists,ecologists,and environmental authorities in serving the immediate and long-term interests of the human population.ß2009Elsevier B.V.All rights reserved.*Corresponding author at:International Plant Nutrition Institute,P.O.Drawer 2440,Conway,AR 72034,USA.Tel.:+15013368110;fax:+15013292318.E-mail address:csnyder@ (C.S.Snyder).Contents lists available at ScienceDirectAgriculture,Ecosystems and Environmentj o u r n al h o m e p a g e :w w w.e l se v i e r.co m /l oc a t e /a g e e0167-8809/$–see front matter ß2009Elsevier B.V.All rights reserved.doi:10.1016/j.agee.2009.04.0212.3.3.Emission of CO2from urea (251)2.3.4.Emissions of CH4from rice culture (251)2.4.Agricultural sinks (251)2.5.Soil carbon stabilization (252)2.5.1.Role of N in stabilizing soil C (252)2.6.Factors affecting GHG emissions from soil (253)2.6.1.Soil physical properties and conditions (253)2.6.2.Soil compaction (253)3.Cropping and fertilizer practices and the associated BMPs (253)3.1.Tillage systems (253)3.2.Tile drainage (254)3.3.Cropping systems (254)3.4.Fertilizer use and crop yield enhancement (255)3.5.Fertilizer production and transport (256)3.6.Nitrogen source impact on N2O emissions from soil (257)mercial fertilizer sources (257)3.6.2.Biologicallyfixed legume-derived N (258)3.6.3.Livestock manures (258)3.6.4.Inhibitors and controlled-release fertilizer sources or enhanced-efficiency fertilizers (258)3.7.Application management—rate,timing,and placement (259)3.7.1.N rate impact on N2O emissions from soil (259)3.7.2.N placement (260)3.7.3.N timing (261)3.8.Balanced fertilization including other required nutrients along with N (261)4.Fertilizer BMPs (262)4.1.General practices (262)4.1.1.Equipment,proper application,and application technology (262)4.1.2.Crop management,system planning,and evaluation (262)4.1.3.Inhibitors and enhanced-efficiency sources (262)4.1.4.Research and development needs (262)5.Conclusions (263)Acknowledgements (263)References (263)1.IntroductionThe objective of this literature review is to assess best management practices(BMPs)in relation to their potential to mitigate the greenhouse gas(GHG)emissions associated with fertilizer use in cropping systems.This document is organized as follows:Background information that describes the global warming potential(GWP)of GHG emissions from agriculture and fertilizer use,and how they compare to other sources.A description of how crop and fertilizer management practices affect GHG emissions.A conclusion recommending how the BMPs described might be implemented to mitigate GHG emissions.2.Background2.1.Greenhouse gasesGreenhouse gases are those that absorb infrared radiation in the atmosphere,trapping heat and warming the surface of the Earth. The three greenhouse gases(GHGs)associated with agriculture are carbon dioxide(CO2),methane(CH4),and nitrous oxide(N2O). Other important GHGs include water vapor and many halocarbon compounds,but their emissions are not considered to be influenced by agriculture.Fossil fuel combustion is considered responsible for more than 75%of human-caused nd use change(primarily deforestation)is responsible for the remainder.Human activities are thought to have more than doubled the rate of emission of CH4 over the last25years(Denman et al.,2007).Atmospheric concentrations of N2O are reported to have risen from about 270parts per billion(ppb)during the pre-industrial era to319ppb in2005.According to Hirsch et al.(2006),and the United Nations Educational,Scientific and Cultural Organization(UNESCO)and the Scientific Committee on Problems of the Environment of the International Council on Science(SCOPE)(2007),emissions of N2O from the Earth’s surface have increased by about40–50%over pre-industrial levels as a result of human activity.2.2.Agricultural shareThe U.S.Environmental Protection Agency(EPA,2007)has estimated that agriculture generates less than10%of the total emissions of GHGs in the U.S.,and its share is not increasing (Fig.1).In Canada,the share attributed to agriculture accounts for less than8%of the emissions inventory(Environment Canada, 2007).While the major GHG issue for the total economy is CO2,for agriculture the most important is N2O,mainly from soils and N inputs to crop and soil systems(Fig.2).Emissions of CH4,mainly from livestock,are also substantial in their contribution to GWP.So even though N2O is a small part of the overall GHG issue,it becomes the major focus of this review because agriculture is considered to be its major source,and it is linked to soil management and fertilizer use.Although animal manures, legumes,and other sources are also important,the majority of this report will address fertilizer use.The emissions shown in Figs.1and2are estimated by the U.S. EPA(2007)to be only partially mitigated by sinks including forests and agricultural soils.The U.S.EPA estimate of these sinks in2005 amounted to a total of829million t(Tg)of CO2equivalents, dominated by forests,with agricultural soils accounting for a sink of only32Tg of CO2equivalents.C.S.Snyder et al./Agriculture,Ecosystems and Environment133(2009)247–266 2482.3.Agricultural sourcesThe three GHGs associated with agriculture –CO 2,CH 4,and N 2O –differ in their effectiveness in trapping heat and in their turnover rates in the atmosphere.Calculating their GWP therefore depends on the timeframe considered.For a 100-year timeframe,unit masses of CH 4and N 2O are considered to have 23and 296times the GWP,respectively,as a unit of CO 2(IPCC,2001).Although older sources (IPCC,1996)may use the GWP CO 2equivalent values of 21for CH 4and 310for N 2O,for the purposes of this review we will use the more recent values of 23and 296,respectively (IPCC,2001).Carbon dioxide,in comparison to CH 4and N 2O,is cycled in the largest amounts through agricultural cropping systems.Plants consume large amounts of CO 2through photosynthesis to make food,feed,fiber,and fuel—but all these plant products eventually convert back to CO 2when consumed or when they decompose.The net emission of CO 2is small in comparison to its total cycling in agriculture,and is mostly due to energy use on-farm and in the manufacture and transport of agricultural products.Methane is emitted from rice cultivation and from ruminant livestock.It can be captured from manure using anaerobic digesters,collection from covered anaerobic manure lagoons,and emissions can be reduced by livestock diet modification.Non-rice agricultural soils are a sink for atmospheric CH 4,but to a smaller extent than soils in their native condition.Nitrous oxide emissions result from two microbial soil processes:nitrification and denitrification.Both processes gen-erally release small amounts relative to the soil N supply.The N 2Ois produced through the activity of soil microorganisms (Bange,2000).The interaction of controlling factors for nitrification and denitrification are complex because the amount of N 2O produced depends on the range of oxygen (O 2)concentrations in the soil.Temperature determines the rate at which the soil microorganisms nitrify or denitrify.At cooler temperatures the rate of N species conversion is slow.It increases to a maximum as temperatures rise.The O 2concentration in the soil is influenced by its moisture content.Other important factors affecting N 2O emissions are:soil texture,the amount of ammonium (NH 4+)available for nitrifica-tion,and the amount of nitrate (NO 3À)available for denitrification (Granli and Bøckman,1994;Firestone,1982).Nitrification occurs when NH 4+oxidizing bacteria such as Nitrosomonas sp.catabolize NH 4+and transform it to nitrite (NO 2À)which is further changed into NO 3Àby Nitrobacter sp.and Nitrospira sp.bacteria (Norton,2008).Nitrous oxide and nitric oxide (NO)are minor by-products of the transformation from nitrite (NO 2À)under oxygen-limited conditions,when nitrifiers use NO 2Àas a terminal electron acceptor (IFA/FAO,2001).About three decades ago,Bremner and Blackmer (1978)reported emissions of N 2O arising from nitrification under fully aerobic conditions,at rates ranging from 0.04to 0.45%of N added.Nitrification determines the form of N present and therefore how N is absorbed,utilized,or dispersed into the environment;which has large implications for plant productiv-ity and environmental quality.During nitrification,the relatively immobile NH 4+is converted to the highly mobile NO 3À.Conversion of NH 4+to NO 3Àstrongly influences N utilization by plants,because the NO 3Àformed is highly susceptible to loss from the root zone by leaching and or denitrification (Subbarao et al.,2006),but is also highly available to plants and is often the major uptake form of N.Denitrification –as described by Firestone (1982),Firestone and Davidson (1989),and Robertson and Groffman (2007)–occurs when NO 3Àis transformed to dinitrogen (N 2)gas as described in the following pathway,NO 3À!NO 2À!NO !N 2O !N 2.The conversion of NO 3Àto N 2can be complete,but a small and variable portion of the N is often emitted as N 2O gas.Emissions are sporadic,occurring before,during,and after the growing season.Flushes of N 2O can occur when previously well-aerated soils become moistened or saturated from precipitation or irrigation,or when frozen soils thaw (e.g.during snowmelt).Bedard-Haughn et al.(2006)reported that the composition of the microbial population exerts a dominant control on emissions and remains relatively constant over time,whereas interactions among spatially and temporally variable environmental drivers [NO 3Àconcentrations,temperature,water-filled pore space (WFPS),available carbon (C),etc.]control the magnitude of N 2O.2.3.1.Emissions of nitrous oxide (N 2O)from croplandThe International Fertilizer Industry Association and the Food and Agriculture Organization of the United Nations (IFA/FAO,2001)provided estimates of emissions of N 2O from cropland in Canada,the U.S.,and the world.These estimates are shown in Table 1.The IFA/FAO estimates showed that although North American field crop agriculture accounted for roughly 16%of the world field cropland area and 17%of the world N consumption in 1995,its N 2O emissions were smaller,at about 12%.This would tend to indicate that N 2O emissions may be higher in other regions of the world.If one assumes that 1%of fertilizer N is emitted as N 2O-N (as in IPCC,2006)then fertilizer-induced emissions account for about 33%of the estimated total in North America.The fraction of applied N actually emitted as N 2O varies widely on a site-specific basis.Thornton and Valente (1996)reported that coefficients of variation for N 2O emissions measurementstypicallyFig.1.Greenhouse gas emissions from the U.S.economy by sector,in billion (109)t of CO 2equivalents.Calculated from Table 2-16in U.S.EPA (2007).Fig.2.Distribution of greenhouse gas emissions from U.S.agriculture and total.Calculated from Table 2-16in U.S.EPA (2007).C.S.Snyder et al./Agriculture,Ecosystems and Environment 133(2009)247–266249range between 100and 300%.The current Tier 1method emission factor is 0.01with an uncertainty range of 0.003–0.03(IPCC,2006).In terms of GWP,this is equivalent to 4.65kg CO 2per kg of N applied,with an uncertainty range of 1.4–14.0.The Environment Canada (2007)GHG inventory reported that recent research indicated low N 2O emissions in arid regions.The research found that in the Prairie regions,which account for about 80%of the land area and N applications (Grant and Wu,2008),an average of 0.16and 0.8%of N fertilizer applied was emitted as N 2O in the Brown-Dark Brown and Grey-Black soil zones,respectively,compared to 1.19%in Eastern Canada.The emission coefficient used in the calculation of Canada’s GHG inventory is thus based on regression to an aridity index linked to these observations.McSwiney and Robertson (2005)reported that a set emission factor is appropriate only when crops are fertilized at N rates less than or equal to those required for maximum yields,because the percentage of fertilizer N that is emitted as N 2O becomes more variable at higher N rates.Halvorson et al.(2008b),Del Grosso et al.(2008),and Burton et al.(2008)observed that set emission factors,including the IPCC (2006)factor,over-estimated seasonal N 2O emissions.Emission factors may increase when the N rate exceeds the crop and soil uptake capacity (Grant et al.,2006;Halvorson et al.,2008b ).Influence of fertilizer N rate is further discussed in Section 3.7.The most important factors affecting N 2O emissions from fertilized fields were listed by IFA/FAO (2001)as (1)climate,soil organic C (SOC)content,soil texture,soil drainage,abundance of NO 3-N and soil pH and (2)management related factors including:N application rate per fertilizer type,and type of crop,with major differences between grasses,legumes and other annual crops.The length of measurement period and the frequency (and intensity)of measurements were also cited as key factors in any local or large-scale estimations of N 2O emissions,in agreement with Parkin (2008).Eichner (1990)listed the following factors affecting fertilizer-derived N 2O emissions:(1)management factors—ferti-lizer type,application rate,application technique,application timing,tillage system,use of other chemicals,crop type,irrigation,and residual N and C from crops and fertilizer and (2)environ-mental factors—temperature,precipitation,soil moisture content,SOC content,soil O 2status,soil porosity,soil pH,freezing and thawing cycles,and microorganism abundance and activity.Del Grosso et al.(2006)pointed out that in addition to increasing direct soil N 2O emissions,agricultural practices can also increase ammonia (NH 3)volatilization and NO 3Àleaching.Volatilized N can affect N 2O emissions because a portion of this N will be deposited on agricultural and non-agricultural soils and in water and be subjected to transformations that may result in N 2O emissions.A portion of the NO 3Àthat is leached or discharged in drainage can also be denitrified and result in N 2O emissions.Crutzen et al.(2008),considering the global rate of increase in atmospheric N 2O in comparison to the total reactive N produced by human activity,concluded that the sum of direct and indirect emissions could amount to 3–5%of N applied—a total emission three to five times larger than the direct emission coefficient of IPCC (2006).They stated,‘‘The large difference between the low yield of N 2O in agricultural fields,compared to the much larger average value derived from the global N 2O budget,implies considerable ‘‘background’’N 2O production occurring beyond agricultural fields but,nevertheless,related to fertilizer use,from sources such as rivers,estuaries and coastal zones,animal husbandry and the atmospheric deposition of ammonia and NO x .’’In contrast,Nevison et al.(2007)reported that a 2%emission factor for global anthropic N input was sufficient to explain the atmospheric N 2O increases over the last 150years.Galloway et al.(2004),using a similar analysis,stated that ‘‘of the $156Tg N year À1created by human action in the early 1990s,$2.5%can be accounted for by tropospheric accumulation of N 2O.’’Assumptions that natural background N 2O emissions have remained relatively constant in natural estuarine and marine systems since the pre-industrial era may be questioned,since Fulweiler et al.(2007)reported decreases in estuarine primary production attributed to climate change.They measured decreased organic matter deposition to coastal marine sediments,and a reversal in net sediment N flux in Narragansett Bay,near Rhode Island in the U.S.Fulweiler et al.(2007)suggested that some estuaries may no longer remove N from the water column,but are actually net N contributors through N fixation.Crutzen et al.(2008)reported the pre-industrial,natural N 2O sinks and sources were ‘‘equal to 10.2Tg N 2O-N year À1,’’which included marine emissions.They noted that the decrease in global N 2O emissions arising from deforestation since 1860was ‘‘an uncertain 0to 0.9Tg N 2O-N year À1.’’Though it is not the intent of these authors to minimize the importance of N 2O emissions associated with agricultural soil management,it is interesting to note this uncertainty exceeds the 0.47Tg N 2O-N year À1worldwide fertilizer-induced emissions,calculated from the estimate pro-vided in IFA/FAO (2001).A number of comments can be made in regard to the global N budget approaches (i.e.,Crutzen et al.,2008;Galloway et al.,2004;Nevison et al.,2007).The analyses are subject to uncertainties in the estimates of natural biological N fixation and natural rates of N 2O emissions.Crutzen et al.(2008)also assume that the fraction of ‘‘new’’anthropogenic N released as N 2O is the same for all four sources:fertilizer N,industrial N,biological N fixation in agricultural crops,and the N emission resulting from fossil fuel combustion.Their approach attributes all the increase in N 2O production to these new sources,and none to the effect of human activities on the rate of N 2O release from the huge pool of N in soils.Nonetheless,the possibility of this approach’s accuracy must be acknowledged.The indirect emissions arising from N use merit greater consideration.These N 2O emission estimation uncertainties underscore the importance of appropriate methodology and measurements of N 2O and N 2emissions associated with denitrification.‘‘Denitrification is difficult to measure because of the analytical difficulty in detecting small increases in N 2concentrations against the large background in the environment.A number of recent advances in approaches to directly quantify denitrification in aquatic andTable 1Estimates of N 2O emissions from cropland in the U.S.,Canada,and the world in 1995(adapted from IFA/FAO,2001).RegionArea (million ha)Fertilizer Napplied (million t)Animal manure N applied (million t)N 2O-N emitted Total (million t)Fertilizer-induced a Million t%of total Canada 46 1.580.210.0670.01624U.S.19011.15 1.580.3160.11235World143673.4820.66 3.1500.73523aEstimated using IPCC emission factor of 1%.C.S.Snyder et al./Agriculture,Ecosystems and Environment 133(2009)247–266250terrestrial environments have been made,but few people are trained to use these newer approaches.’’(m.with Dr.Eric Davidson,Woods Hole Research Institute,10December2007).Chapuis-Lardy et al.(2007)noted that while much research has concentrated on net emission,there are numerous reports of net negativefluxes of N2O,fluxes from the atmosphere to the soil.They concluded that the factors regulating N2O consumption are not yet well understood and merit further study.Such study could potentially provide valuable information toward management of soils and soil biology to enhance conditions favorable to the consumption of N2O.2.3.2.Emissions of CO2from limeWhen agricultural lime is added to soil to increase pH,some fraction of its carbonate may be released as CO2.The current IPCC Tier2methodology for the U.S.greenhouse gas inventory(USEPA, 2007)uses emission factors of0.059kg C kgÀ1of lime and 0.064kg C kgÀ1of dolomite(West and McBride,2005).Thus,the GWP of lime use would average0.22kg CO2kgÀ1of limestone. Hamilton et al.(2007),citing evidence from cropping systems in Michigan,concluded that use of lime in agriculture could provide a sink for CO2,through enhanced export of bicarbonate in drainage water.They pointed out that acidity generated from nitrification would reduce this sink.However,their assumption regarding the amount of acidity generated by urea and anhydrous ammonia fertilizers exceeds that commonly used in soil fertility literature by more than threefold(Havlin et al.,2005).The most common N fertilizers(urea,anhydrous NH3, ammonium nitrate)generate at most3.6kg of lime need per kg of N to balance the acidity they ing IPCC Tier2 methodology,this amount of lime works out to an additional GWP of3.6Â0.22=0.84kg CO2per kg of N applied—small compared to the estimate of N2O emission resulting from the application.This factor would be of relevance only to N use on soils that have low pH and low pH buffering capacity.The median pH of North American soils sampled in2005was6.3,with69%of the soils testing above pH 6.0(PPI/PPIC/FAR,2005).The majority of sampled soils in North America do not have a large lime requirement.In many states and provinces in North America, soils are calcareous and might benefit from some acidification that would result in the soils becoming more neutral in pH,enhancing availability of some plant nutrients.The need for lime is not only due to the use of ammoniacal fertilizer.Soil acidification is also associated with plant root proton excretion(i.e.exudation of H+ions)(Marschner,1991)and removal of basic cations in crop harvests(Jackson and Reisenauer, 1984)and the acidity of rainfall.One possible way to avoid the emission associated with lime use is to apply oxide(e.g.quicklime or slaked lime)rather than carbonate materials,if they can be produced with CO2recovery.If relatively pure oxides of calcium or magnesium could be used,lime use would likely result in a small GHG sink,since CO2could be precipitated from the air in response to the addition of the material.These materials may be prohibitively expensive,how-ever.These oxide and hydroxide forms of lime may also present unacceptable inhalation,dermal exposure,and eye injury concerns in broad commercial use because of their caustic nature.2.3.3.Emission of CO2from ureaApplying urea to soils leads to a loss of CO2that wasfixed in the industrial production process.Urea(CO(NH2)2)is converted into NH4+,hydroxyl ion(OHÀ),and bicarbonate(HCO3À),in the presence of water and urease enzymes.Similar to the soil reaction following addition of lime,bicarbonate that is formed evolves into CO2and water.This source category is included because the CO2 removal from the atmosphere during urea manufacturing is estimated in the Industrial Processes and Product Use Sector(IPCC, 2006).Similar to lime,all C in urea is considered to be emitted as CO2, but a default uncertainty factor ofÀ50%of the chemical maximum may be applied.Since urea contains12g C for every28g N,this works out to a GWP of1.6kg CO2kgÀ1of urea-N applied,or0.8if the uncertainty factor is applied.This GWP cost could be similar to that of the lime requirement generated by the ammoniacal fertilizers,but is still small compared to the estimated effect on N2O emission.2.3.4.Emissions of CH4from rice cultureFlooded rice(Oryza sativa L.)culture is one of the major anthropogenic sources of CH4emissions.Urea and ammonium sulfate are the principal N sources used on the roughly3million ha of rice grown in the U.S.annually,and urea is the most common N source(Snyder and Slaton,2001).Compared to corn(Zea mays L.)at roughly38million ha in2007,and over101million ha of allfield and forage crops,rice may be considered a relatively insignificant contributor to the total GHG emissions from U.S.agriculture.On a global scale,rice culture is a more important GHG contributor than in North America.Bufogle et al.(1998)cited work reporting that CH4emissions were less when ammonium sulfate rather than urea was the fertilizer N source for rice.Asflooded soils become more reduced,sulfate-reducing bacteria effectively compete with methanogenic bacteria.Research by Jugsujinda et al.(1995)in Louisiana showed that CH4emissions were reduced when NO3-N was applied to rice,because added NO3-N increased the redox potential.Other work with rice in China,where manure and urea were the N sources,showed that CH4emissions occurred at lower redox(more reduced)potentials(<À100mV)than did N2O emissions(>+200mV),and there was a significant inverse relationship between emissions of these two GHGs(Hou et al., 2000).These results indicate the risk of promoting N2O emissions whenflooded rice soils are drained.Hou et al. (2000)stated that maintaining the soil redox potential between À100to+200mV would prevent CH4production and would also be low enough to encourage N2O reduction to N2.Lindau et al. (1990)reported greater CH4emissions in urea-treated rice plots as opposed to unplanted urea-treated plots.They also stated that N2O emissions due to fertilizer N addition were low,and N2O emissions after urea application were not above untreated control levels.The presence of rice plants and the type of fertilizer N applied were reported to affect the emissions of N2, N2O,and CH4.Immediate wetting and continuousflooding(‘‘permanent flood’’)of soil after application of urea or ammonium-based fertilizer N is an effective measure to reduce nitrification and N2O emissions(Hutsch et al.,1999).This is a common best manage-ment practice for drill-seeded,flood-irrigated rice culture in the U.S.(Wilson et al.,2006).2.4.Agricultural sinksSince C is cycled in large amounts in cropping systems,small increases in C capture combined with small decreases in respiration (C release)can result in large changes in the balance—the net emission or sequestration.Essentially,crops(including pastures and rangelands)can capture and store CO2by converting it to organic forms of C that are stored in the soil.When agronomic practices increase soil organic matter(SOM),CO2is removed from the atmosphere in the long-term(Lal et al.,1998,2003;Follett et al., 2001).Robertson(2004)listed four strategies by which net CO2 emissions from agriculture could be reduced.They include:C.S.Snyder et al./Agriculture,Ecosystems and Environment133(2009)247–266251。