【作者】 田慎重；

【导师】 李增嘉；

【作者基本信息】 山东农业大学 ， 农业生态学， 2014， 博士

【摘要】 本研究基于山东农业大学农学试验站的保护性耕作和秸秆还田长期定位试验地（始于2002年），于2007~2012年系统的探讨了小麦-玉米一年两熟种植条件下的土壤有机碳库演变（2002~2012）及其固碳机理、温室气体排放量估算（2007~2012）及排放机理，并利用转变耕作方式和实行轮耕进一步探讨长期保护性耕作对土壤有机碳库演变、温室气体排放变化和产量变化的影响，综合评价长期保护性耕作和秸秆还田条件下的土壤固碳潜力、温室气体减排潜力以及作物生产力水平。以期为该地区选择适宜、合理的耕作方式和轮耕组合来提高土壤碳库水平、增强土壤“碳汇”能力、选择合理的农业管理措施减缓农业源温室气体排放和提高作物生产力水平，对该地区农业经济效益、社会效益和生态效益的协同发展具有重要意义。主要研究结果如下：1.长期耕作和秸秆还田的土壤有机碳库演变及其固碳机理试验期间，经过长期的耕作和秸秆还田，4种保护性耕作措施免耕（NTP）、深松（STP）、耙耕（HTP）和旋耕（RTP）处理和对照常规耕作（CTP）处理的土壤有机碳（SOC）水平显著高于其长期无秸秆还田的处理（NTA、STA、HTA、RTA和CTA），且保护性耕作各处理较对照常规耕作处理在0~30cm土层呈现较为明显的SOC分层和表层聚集特点，常规耕作处理土壤中的SOC水平在三个土层中较为平均。各处理土壤总有机碳（TOC）中，活性有机碳（LOC）约占TOC的22.38~39.14%，稳态碳约占60.86~77.62%，RTP处理土壤TOC中所含LOC比例最高（39.14%），STP处理最低（22.38%）。经过10年连续的耕作和秸秆输入，各耕作措施的秸秆还田处理的土壤SOC库平均比无秸秆还田处理高约4.6t ha-1（2011年），原因是各耕作措施的秸秆还田处理SOC的周年累积速率（-0.27~1.34t ha-1yr-1）显著的高于无秸秆还田处理（-0.41~0.75t ha-1yr-1），且与持续的秸秆输入显著相关（R2=0.78, P<0.01）。免耕配合秸秆还田处理（NTP）能够显著的增加土壤有机碳库水平，在2002~2011年，有约13.37t ha-1的碳（C）被固定。其次为深松处理（STP），而旋耕处理在有无秸秆还田的条件下均表现为土壤的C损失，土壤中的C约以每年0.27t ha-1（RTP）和0.41t ha-1（RTA）的速率被消耗，近10年约损失掉4.08和2.74t ha-1的C，而秸秆还田在一定程度上能减缓这种损失速率。通过固碳机理可以看出，造成各处理SOC库差异的原因为各处理大团聚体数量、团聚体稳定性、团聚体关联的碳含量以及稳态碳差异，NTP和STP处理之所以能够维持较高的SOC库水平，是因为其大团聚体数量（79.19%和81.67%）、团聚体稳定性（1.89mm和1.82mm）和团聚体关联的碳库（9.6g kg-1和7.36g kg-1）以及总有机碳中稳态碳比例（71.6%和77.6%）显著高于其它处理，这导致两个处理的土壤SOC库中来自团聚体的稳定的SOC约占75.9%和65.4%，显著的高于其它处理。由回归分析得出，大团聚体关联的C库是决定土壤有机碳库水平的关键（R=0.71, P<0.05）。2.不同耕作措施条件下温室气体排放量估算及排放差异机理各耕作措施条件下的土壤表现为CH4的净吸收汇和N2O的净排放源，通过与CH4和N2O的排放相关微生物菌群的回归分析可知，CH4的吸收与土壤中CH4氧化菌（MOB,R2=0.64, P<0.01）和产CH4菌（MPB, R2=0.66, P<0.01）的数量极显著相关，且CH4氧化菌（MOB）相对于产CH4菌（MPB）为优势菌群，所以各处理表现为CH4的净吸收汇。而N2O的排放是由硝化细菌（R2=0.62, P<0.01）和反硝化细菌（R2=0.64, P<0.01）共同主导的结果。试验期间（2007.10~2012.10），各年份之间的CH4吸收通量和N2O排放通量波动性较大，不同耕作措施处理之间的CH4和N2O排放通量差异显著。试验期间CH4的总吸收通量以RT处理最高，达到10.05t ha-1，显著的高于其它处理，NT处理最低，只有6.29t ha-1，而CH4的温室效应表现为NT>HT>CT>ST>RT。造成排放差异的主要原因是因为与CH4吸收相关的CH4氧化菌（MOB）数量RT>CT>ST>NT，且排放大小主要受土壤温度、水分、pH值和NH4+-N含量的主导。各处理N2O的总排放通量为NT>CT>HT>ST>RT，温室效应表现为NT>CT>HT>ST>RT。造成各处理N2O排放差异的主要原因为硝化细菌数量的显著差异（NT>CT>RT>ST），且排放大小主要与土壤温度、水分、pH值和NO3--N有关。3.小麦-玉米一年两熟农田不同耕作措施的碳足迹分析在小麦-玉米一年两熟制中的碳足迹主要由化肥、农药、机械、灌溉等组成。在小麦季，各耕作处理之间的碳足迹差异显著，最高的是深松处理（ST），为812.8kg Ce ha-1yr-1，其次是常规耕作（CT），碳足迹为803.8kg Ce ha-1yr-1，免耕措施碳足迹最低，只有787.2kg Ce ha-1yr-1。碳足迹差异主要取决于由于农业机械进行耕地、收获和秸秆还田造成的碳足迹。在小麦季各处理碳足迹的组成中，化肥投入造成的碳足迹比例最大，占到75.2~77.6%，而由于农业机械所消耗柴油而造成的碳足迹则只占约8.1~8.3%。玉米季碳足迹中，由于玉米季不进行土壤耕作，且化肥、农药投入和灌溉量相同，所以各处理总的碳足迹都为621.2kg Ce ha-1yr-1。在其组成中，化肥投入造成的碳足迹比例最大，达到73.5%，其次是由于农业机械所消耗柴油而造成的碳足迹，占到13.8%。在小麦-玉米一年两熟体系下，不同耕作措施的粮食碳成本以免耕处理（NT）最高，为0.169kg Ce kg-1，深松处理（ST）最低，为0.144kg Ce kg-1。4.长期保护性耕作方式转变为轮耕后土壤温室气体排放、有机碳库和产量的变化长期旋耕（RT）、耙耕（HT）和免耕（NT）后进行深松（ST），使旋耕/深松（RT/ST）、耙耕/深松（HT/ST）和免耕/深松（NT/ST）这三种轮耕方式的土壤CH4吸收通量较原耕作方式增加了23.6、14.7和26.1%，但增加土壤“碳汇”能力的同时，土壤N2O的排放也分别较原耕作方式增加了10.2、12.9和78.3%。但轮耕后，NT/ST、RT/ST和HT/ST处理的CH4和N2O的总排放通量只增加了约1.08、0.06和0.17kg ha1，总温室效应增加了约0.23、0.02和0.05kg CO2ha1。通过与龙口不同耕作措施长期定位试验点的对比发现，长期单一的保护性耕作和秸秆还田后，免耕（NT）处理的30cm土层SOC库显著高于其它处理，泰安和龙口的不同耕作措施长期试验点在试验期间（10年和6年）土壤有机碳库分别增长了13.37和9.51t ha-1，而两试验点的旋耕处理（RT）都表现为C的净损失，试验期间每年每公顷约损失0.27和0.33t的C。但在泰安定位试验点实行轮耕后，HT/ST、RT/ST处理的0~30cm的土壤有机碳库累积量较原处理增加，其中RT/ST处理从原处理C的净损失变为每年每公顷约固定3.66t的C，但NT/ST处理土壤C的周年累积速率降低24.4%。泰安和龙口定位试验点经过实行连续10年和6年的单一的耕作措施后，并没有表现出明显的增产效果，作物周年生产力水平反而呈逐年下降趋势，但在泰安定位试验点实行轮耕措施后， HT/ST、RT/ST和NT/ST处理的小麦产量轮耕后分别较HT、RT和NT处理平均增长了47.9、37.7和64.9%，玉米产量平均增长9.4、15.5和19.4%，作物周年生产力的增长率逐年增加。更多还原

【Abstract】 Long-term tillage and crop residue inputs are important factors that impact soil organiccarbon (SOC) pool, soil aggregation level and greenhouse gas (GHG) emissions. Anappropriate tillage method is of significant to increase regional SOC pool level and mitigateGHG emissions and to develop the conservational and sustainable agriculture in the NorthChina Plain. This study evaluated the effects of long-term tillage and crop residuemanagement on SOC pool, GHG emissions, carbon footprint, and their changes after sometillage convertions, which aimed to assess the potentials of carbon sequestration andmitigation GHG emissions under different tillage and residue systems through the carbonsequestration mechanism that including soil aggregate stability, distribution, associated carbonand labile organic carbon (LOC) pool, and the emission mechanism including the GHGemission related bacteria population (methanogens, methane-oxidizing bacteria, nitrobacteriaand denitrifying bacteria). The research was conducted in North China between2007and2012. The treatments were five tillage systems that included conventional tillage (CT),subsoiling (ST), harrow tillage (HT), rotary tillage (RT) and no-till (NT), combination withcrop residue retention (P) or residue removal (A). The results are as follows:1. Soil organic carbon pool change by a long-term tillage and residue managementsystem and its carbon sequestration mechanismSoil organic carbon (SOC) contents of the0~30cm depth layer under no tillage (NTP),subsoiling (STP), harrow tillage (HTP), rotary tillage (RTP), conventional tillage (CTP) withcrop residue input were significant higher than that of no residue input treatments (NTA、STA、HTA、RTA and CTA). The soils under NTP, STP, HTP and RTP treatments gatheredmore SOC to the0~10cm layer compared to the CTP treatment, and their SOC pool wereincreased during the experiment. The labile organic carbon (LOC) proportion in TOC was 22.38~39.14%under all treatments, and the stabled carbon proportion was60.86~77.62%.The highest LOC proportion in TOC was the RTP treatment (39.14%), and the lowest onewas STP (22.38%). The levels of SOC pool under different tillage with residue inputtreatments were higher4.6t C ha-1than those of no residue input treatments after10-yrscontinued crop residue input and tillage, becacse the annual accumulation rate under thetillage treatments with residue input (-0.27~1.34t C ha-1yr-1) were significant higher thanthose of no residue input treatments (-0.41~0.75t C ha-1yr-1), and the SOC pool level wassignificant correlated with the input rate of crop residue biomass (R2=0.78, P<0.01). In thisstudy, SOC pool under NTP treatment was significant increased and approximately13.37t Cha-1was sequestrated from2002to2012, similarly, approximately9.59t C ha-1wassequestrated in the STP treatment. However, the soil under RTP and RTA treatments wereoccured C lost and approximately0.27t C ha-1and0.41t C ha-1were lost each year, the cropresidue returned to field could mitigate this rate of C lost.The analysis of carbon sequestration mechanism showed that the differences of SOC poolunder different tillage and residue management systems were from the changes of soilaggregate proportion, stability, associated carbon and stabled SOC pool levels. Higher levelsof SOC pool were mearsured at the NTP and STA treatments, because they had the higherproportion (79.19%and81.67%), stability (1.89mm and1.82mm), associated C content(9.6g kg-1and7.36g kg-1) of soil aggregate and stabled C pool proportion (71.6%and77.6%).Therefore, the soil macro-aggregate associated C pool contribution proportion to TOC underthe NTP and STA treatments (75.9%and65.4%) were significant higher than othertreatments. The regression analysis also showed that there was a significant positivecorrelation between the soil macro-aggregate associated C pool and soil SOC level (R=0.71,P<0.05).2. Estimation emission fluxes of GHG under different tillage systems and their emissionmechanismThe soil was observed a sink of CH4uptake and emission source of N2O under differenttillage systems. The regression analysis of the CH4uptake related bacteria showed that theCH4uptake flux was correlated with the population of methane-oxidizing bacteria (MOB,R2=0.64, P<0.01) and methanogens (MPB, R2=0.66, P<0.01). Maybe the dominant bacterium in soil was MOB that leaded to the observation a sink of CH4uptake. The population ofnitrobacteria (R2=0.62, P<0.01) and denitrifying bacteria (R2=0.64, P<0.01) were drived theN2O emission in this study.Total emission flux of CH4and N2O during the experiment (2007.10~2012.10) waslargely fluctuated in different years. The highest total flux was observed at RT treatment with10.05t ha-1, the lowest was NT treatment (6.29t ha-1). The global warming potential (GWP)of CH4was ordered NT>HT>CT>ST>RT due to the difference of MOB population(RT>CT>ST>NT), the population general related with soil temperature, moisture, pH andNH4+-N contents. Meanwhile, total emission flux of N2O was NT>CT>HT>ST>RT and theGWP of N2O was NT>CT>HT>ST>RT, the reasons may was the difference of nitrobacteriapopulation (NT>CT>RT>ST) that general related with soil temperature, moisture, pH andNH4+-N contents.3. Footprints under different tillage systems in a wheat-maize cropping systemThe footprint constitute in a wheat-maize cropping system including chemical fertilizer,agricultural chemicals, diesel, seed and irrigation. The highest proportion in footprintconstitute was chemical fertilizer (73.5~77.6%), while the diesel proportion was only8.1~13.8%. In wheat period, the highest footprint was observed at ST treatment (812.8kg Ceha-1yr-1), the lowest one was NT treatment (787.2kg Ce ha-1yr-1), while the footprints inmaize period under different tillage methods were621.2kg Ce ha-1yr-1due to the equal inputof chemical fertilizer, agricultural chemicals, diesel, seed and irrigation.The highest carbon cost in a wheat-maize cropping system was observed at NT treatment(0.169kg Ce kg-1), the lowest one was observed at ST treatment (0.144kg Ce kg-1).4. Changes of soil organic carbon pool, greenhouse gas emission and crop yield by along-term single tillage and rotational tillage systemLong-term single tillage (RT, HT, NT) convertion to rotational tillage systems (RT/ST,HT/ST, NT/ST) increased the ablity of CH4uptake sink, the CH4uptake flux was increased23.6%,14.7%and26.1%after these convertions respectively, while the N2O flux was emitedlargely than the original treatments (increased10.2%,12.9%and78.3%, respectively). Thetotal emission flux of CH4and N2O under NT/ST, RT/ST and HT/ST treatments wereincreased1.08,0.06and0.17kg ha1, respectively, and the GWP of CH4and N2O was increased0.23,0.02and0.05kg CO2ha1, respectively, compared with the treatments of NT,RT and HT.In comparison to the experiment site of Longkou, higher SOC pool level of0~30cmdepth was measured at long-term single NT treatment, this tillage method at two experimentsites (Tai’an and Longkou) were locked13.37and9.51t C ha-1in10-yrs and6-yrs,respectively. However, RT treatments of two experiment sites were lost0.27and0.33t C ha-1each year. The C accumulation after HT and RT treatments convertion to HT/ST and RT/STwere increased, especially the latter, which observation of C lost converted to C accumulation.The C accumulation rate under NT/ST treatment was decreased24.4%than that of NTtreatment.The yields of wheat and maize were not improved under continued long-term singletillage system not only at Tai’an but also at Longkou, sometimes the yields were presented thedecreased trendences. After convertion to HT/ST, RT/ST and NT/ST treatments, the wheatyield was improved47.9,37.7and64.9%, respectively, and the mazie yield was improved9.4,15.5and19.4%, respectively. The total yield in a wheat-maize cropping system was observeda gradual increase trend after these rotational tillage systems.更多还原