【作者】 王健波；

【导师】 严昌荣；

【作者基本信息】 中国农业科学院 ， 农业水资源利用， 2014， 博士

【摘要】 旱地农业在国家经济和粮食安全中发挥了重要作用，在我国农业生产中占有重要的地位。北方旱区缺水少雨且降水集中，造成土壤遭受干湿交替频繁；同时农田耕作及少量秸秆还田，造成土壤贫瘠，所以采取保护性耕作方式改善土壤质量，提高土壤生产力，进而达到作物高效用水。然而，在不同的耕作方式下土壤有机碳和土壤水分通常会有很大的变化，关于土壤有机碳和土壤水分在土壤剖面中的含量研究较多，并且对于土壤有机碳和土壤水分的研究一般都是分开的。然而土壤生态系统是极其复杂的，土壤有机质和水分中任何一方面发生改变都会对土壤的其他性质产生重大影响。尤其是在生态系统异常脆弱的北方旱区，必须综合考虑土壤水分和土壤有机碳的相互作用，进行全面系统的研究。因此，研究北方旱区土壤有机碳和土壤水分的动态变化特征及可能存在的相互依存关系对改善区域土壤质量以及高效利用土壤水分具有重要的作用。本研究以临汾20年长期定位试验为平台，采用室内培养和野外田间试验相结合的方法对土壤有机碳和土壤水分进行研究。长期定位耕作试验站位于山西省临汾市尧都区，种植作物为冬小麦。试验共设计2个处理，分别为免耕覆盖（NT）和常规耕作（CT），并于2012年6月小麦收获后在免耕覆盖耕作试验区设置不同秸秆还田量处理，包括全量秸秆还田4500kg·hm-2（NT3），2/3量秸秆还田3000kg·hm-（2NT2），1/3量秸秆还田1500kg·hm-（2NT1），秸秆不还田处理（NT0）。室内培养试验处理和处理组合包括（2秸秆处理×2耕作处理×2水分处理）：（1）加或者不加标记秸秆（δ13C=13,070‰）（2）来自免耕或者常规耕作的土壤（3）经历或者不经历干湿交替。本文研究了不同耕作方式下土壤有机碳及其组分碳的动态变化特征、有机碳矿化特征，以及土壤供水特征、作物耗水结构和水分利用状况。通过室内、田间试验相结合的方法探讨了不同数量的外源秸秆碳对土壤水分的影响，及水分交替变化对土壤有机碳矿化的影响，揭示水碳相互作用的机制，进一步完善保护性耕作技术下的保水增碳研究。本研究主要结论如下：1、免耕覆盖提高土壤表层有机碳及其组分含量，显著降低颗粒有机碳及表层微生物量碳和可溶性有机碳的季节变动性。免耕覆盖措施下土壤表层有机碳含量随小麦生育期的推进略有增大趋势，常规耕作措施下变化相对较小。土壤微生物量碳和颗粒有机碳含量在播种后增大，在冬前第一次达到最大值，随之降低，在拔节期降到最低点，在开花期达到最大值，然后降低，一直持续到小麦成熟期。可溶性有机碳含量在播种前最大，在拔节期最小，在开花期略微增大。耕作措施显著影响土壤有机碳及其组分在剖面上的分布，具体表现为：在0-5cm和5-10cm免耕覆盖处理土壤有机碳含量比常规耕作分别高82.1%和52.9%，同时免耕覆盖可以显著提高0-10cm土壤微生物量碳、颗粒有机碳和可溶性有机碳含量。而在10-50cm常规耕作措施下土壤有机碳和颗粒有机碳略微比免耕高，在20-50cm和30-50cm拔节期后其微生物量碳和可溶性有机碳分别大于免耕覆盖处理。研究发现免耕覆盖显著降低颗粒有机碳及表层微生物量碳和可溶性有机碳的季节变动性。结果表明在研究耕作措施对土壤有机碳及其组分的影响过程中应该考虑季节变动和土壤剖面分布的影响。2、不同耕作方式下土壤二氧化碳释放总量的差异主要是由秸秆覆盖引起的，土壤二氧化碳释放速率与表层土壤体积含水量和土壤温度呈正相关。土壤翻耕时二氧化碳释放速率显著大于未翻耕土壤（CT），是未翻耕土壤的2.7倍，但48h后差异不显著。在整年内，免耕覆盖处理土壤二氧化碳释放速率基本大于常规耕作和免耕秸秆不还田处理，在休闲期和小麦生育后期差异显著。土壤二氧化碳释放速率与表层土壤温度呈指数正相关，与表层土壤体积含水量呈线性正相关，但采用温、湿度双变量模型的决定系数得到提高，其中免耕覆盖处理提高幅度最大。免耕覆盖处理的土壤二氧化碳释放年总量达到7.2t C hm-2·yr-1，显著大于常规耕作和免耕秸秆不还田处理；其中常规耕作高于免耕秸秆不还田处理，但差异不显著。研究结果表明：不同耕作方式下土壤二氧化碳释放总量的差异主要是由秸秆覆盖引起的。在没有秸秆覆盖的情况下，翻耕短时间内增加土壤二氧化碳释放速率，但对全年释放总量影响较小。3、免耕覆盖提高作物水分利用效率。试验区20年统计数据显示免耕覆盖耕作下播种前0-50cm土壤平均储水量为111.1mm，比常规耕作高12.5%，两处理土壤储水量与休闲期降雨量相关性达到显著水平，并且两处理间的差异随着休闲期降雨量的增大而逐渐减小，即免耕覆盖在降雨较少的休闲期保水作用凸显。在小麦生育期内，免耕覆盖处理耗水总量比常规耕作少7%，全年棵间蒸发量比常规耕作少为47mm，抑制蒸发率可以达到24%，差异达显著水平。在小麦产量形成的关键时期开花-灌浆期常规耕作处理蒸腾速率为0.3mm·d-1，仅为免耕覆盖处理的33%。在整个小麦生育期内常规耕作的蒸发蒸散比为41%，而免耕覆盖为39%，其差异主要体现在小麦播种到分蘖期和生育后期。随着土壤有机质的不断积累和水分含量的相对提高，自2000年以后免耕覆盖耕作方式下的小麦产量平均比常规耕作提高了37%。研究表明，常规耕作下的小麦产量对降雨量的依赖性要高于免耕覆盖耕作，免耕覆盖对常规耕作的水分利用效率的增加率（NT vs CT）随年降雨量的增加逐渐降低。4、免耕条件下秸秆全量还田水分利用效率最高，但与秸秆覆盖量没有正比例关系。在免耕条件下，秸秆不还田的土壤非饱和导水率大于秸秆全量覆盖处理，但非饱和导水率与秸秆覆盖量没有比例关系。秸秆还田提高表层土壤含水量，并且随着秸秆覆盖量增加而增加，但全量还田处理在小麦收获期降低亚表层土壤含水量。免耕条件下有秸秆覆盖的各处理（不包括NT0）土壤储水量基本表现为储水量随秸秆覆盖量增加而增加，但在干旱缺水时期因秸秆不还田处理下土壤表层容易形成结壳，影响土壤储水量。秸秆全量覆盖处理下全年棵间蒸发量最小，抑制蒸发率为14%；秸秆全量覆盖处理下蒸腾强度最大，比秸秆不覆盖大24%；其蒸发蒸散比最小且值为39%。所以全量秸秆覆盖可以有效降低非常生产性耗水，其水分利用效率最大且值为13.5kg·hm-2·mm-1，但与秸秆覆盖量没有正比例关系。5、干湿交替显著提高土壤有机碳矿化速率和秸秆激发强度。干湿交替条件下土壤有机碳矿化速率大于恒湿条件，比恒湿平均大3.9mg C-CO-2kg1soilday-1。秸秆显著增加总有机碳的矿化速率（206.96%vs无秸秆），在恒湿条件下总有机碳矿化速率在2-11d显著降低，达到373.20%（vs1d）。来自秸秆部分的二氧化碳在恒湿条件下（44.45%-34.00%）显著大于干湿交替（32.84%-13.09%）。在相同水分条件下，秸秆对土壤有机碳的激发效应方向呈现出先正后负的现象。培养过程中干湿交替条件下免耕土壤的激发效应是15.32到﹣0.62mg C kg-1soil day-1，显著大于恒湿条件下的激发效应。研究还表明在0-20cm的土层中免耕和常规耕作土壤有机碳矿化没有显著差异。数据显示干湿交替对碳矿化和激发效应影响显著，秸秆和水分产生正的交互作用。研究结果首次阐明了在室内模拟田间情况下，频繁的干湿交替下秸秆对长期免耕土壤的激发效应动态变化。更多还原

【Abstract】 Dryland agriculture played an important role in the national economy and food security, occupyingan important position in Chinese agricultural production. It is heavy droughty and has concentratedrainfall in Northern arid regions, resulting in the soils subjected drying and wetting frequently. Andtillage practices and less straw returning resulted in poor soil. So the conservation tillage was taken toimprove soil quality, increase soil production, and thus achieve water use efficiency of crops. However,under different tillage practices soil moisture and soil organic carbon usually have a lot of changes.Most studies focused on soil moisture content and soil organic carbon in the soil profile, and generallyfor the studies of soil moisture and organic carbon were separated. The cycling of soil water and thecarbon in farmland ecosystem is not isolated from each other, with interdependent relationship. Theinteraction between soil organic carbon and soil moisture must be synthetically consideration, andcomprehensive and systematic study should be done. Therefore, the studies on character of soil moistureand soil organic carbon and their interdependencies in northern arid regions may plays an important rolein improving soil quality and high-efficient use of soil water.This study was based on long-term (20years) experiment in Linfen, using combined method ofincubation and field experiment on soil organic carbon and soil moisture. The long-term experiment wasconducted from1992in Lifen City, Shanxi Province, China. The long-term experimental plots wereperformed using two different treatments: conventional-tillage (CT) and no-tillage (NT) treatments. TheCT treatment included moldboard plowing without returning wheat straws to the field at the end ofAugust, but retaining10-15cm stubbles after harvest; the tillage depth was bout20cm. For the NTsystem, all of the residues from the prior winter wheat straw were flattened and mulched in the field.The winter wheat (Triticum aestivum L.) was sown at September end and harvested in middle Juneevery year. The fallow period lasted from the harvest to September end, during which, herbicides wereapplied to control weeds. In June,2012, after harvesting of wheat, straws were returned to the field inproportion on the NT plots,100%straw mulch (NT3,4500kg·hm-2),66%straw mulch (NT2),33%straw mulch (NT1) and0%straw mulch. Twelve DW cycles were implemented during the experimentalperiod (120d). Each cycle contained two periods, four days of wet conditions and six days of dryconditions at25°C according to the daily precipitation and air temperature records of the area. Thetreatments and their combinations were as follows (2straw treatments×2tillage-treated soils×2watercontents):(i) with or without straw input,(ii) soil from the CT field or NT field and (iii) with or withoutexposure to DW cycles, which was characterised by wetting at field capacity (0.186g H2O g-1soil) andair drying (0.022g H-2O g1soil) at25℃in an incubator with controlled temperature and humidity.Through the incubation and field experiment we explored the effect of fluctuation of water on soilorganic carbon mineralization and the effects of different amounts of organic carbon on soil watercapacity and crop water structures, and then revealed the mechanism of interaction between carbon andwater, and further improved the research on the carbon sequestration and water retention under conservation tillage practices.1. Soil organic C, POC, MBC and DOC were all significantly higher under a NT system whencompared with a CT system, at the surface layer. Compared to CT treatment, NT treatment markedlyreduced the fluctuation of DOC and MBC in0-30cm depths and fluctuation of POC in0-60cm depths.While the SOC increased slowly in the surface soils during crop growth with the NT treatment, itshowed less fluctuation with the CT treatment. MBC and POC increased after sowing and reached afirst peak before winter, then dropped to a minimum during the jointed stage, reaching their maximumduring the anthesis stage and then decreasing until maturity. DOC was highest before sowing, decreasedbefore the winter and jointed stages, and increased again during the anthesis stage. Long-term tillagepractices could cause differences in the SOC soil profile and in its fraction distribution. On average, theNT treatment had82.1%and52.9%higher SOC at0-5cm and5-10cm depths than did the CTtreatment, respectively. POC, MBC and DOC were all significantly higher with NT than with CT in theupper10cm. While SOC and POC were slightly higher under CT than when using the NT practice at10-50cm depths, the MBC and DOC began to increase after the jointed stage at20-50cm and30-50cmdepths. The POC, MBC and DOC were highly correlated with the SOC. In our study, we found thatlabile SOC in NT and CT soils had different season fluctuations. Compared to CT treatment, NTtreatment markedly reduced the fluctuation of DOC and MBC in0-20cm depths and fluctuation ofPOC in0-60cm depths. This study demonstrated that measuring the effect of tillage practices on SOCbased on soil organic C fractions must consider both seasonal changes and profile distribution.2. The differences of CO2emissons under different treatments were attributed to wheat straw mulching.The CO2emissions were positively correlated with soil temperatures and slightly positively with soilwater content.The CO2emission rate immediately after plowing treatment was significantly higher than CTtreatment (P＜0.001),2.7times the CT treatment, but not significantly after48hours. Tillage practiceshad a significant influence on the CO2emissions during whole years. Higher CO2emissions weredetected in NT plots than in CT and NT0treatments during fallow and late growing period. The CO2emissions were positively correlated with soil temperatures in the top20cm (R2=0.71, P＜0.001) andslightly positively with soil water content (R2=0.05, P＞0.05). While using a regression analysisprocedure, a significant multiple regression equation was developed between soil CO2emissions andwater content, temperatures (P＜0.001), and the determination coefficients (R2) of different treatmentswere raised, by the highest amplitude under the NT treatment. Annual CO2emissions were significantlyhigher under NT (7218kg CO12-C ha-1·yr-) than under CT (5870kg CO-12-C ha·yr-1) and NT0(5585kgCO-12-C ha·yr-1), but not significantly (P＞0.05) between CT and NT0. The soil CO2emission duringthe wheat growing period was significantly greater than during the fallow period (P＜0.001). Theresults indicated that although NT increased soil CO2emissions compared to CT and NT0, thesedifferences under different treatments were attributed to wheat straw mulching and there was0.3t Cha-1left in the NT soil annually which increased soil carbon sequestration.3. Water use efficiency (WUE) was increased under NT practice when compared with CT practice. From20years of statistical data the water storage under NT capacity was average111.1mm,12.5%higher than NT. The storage capacity was highly correlated with precipitation during fallowperiod, with a significant difference. The difference between them gradually decreased as the increase ofrainfall during the fallow period. So the water retention of NT during the low rainfall fallow periodhighlighted. The water consumption of NT practice was7%less than the CT practices during thewheat-growing period. Soil evaporation of NT in the whole year was47mm less than CT and the rate ofevaporation reduction can reach24%, with a significant difference. In the key period of wheat floweringand filling stage the rate of transpiration in CT was0.3mm·d-1, only33%of NT. The proportion ofevaporation to evaportransportation under CT during wheat-growing period was41%, while NT was39%, and the difference between them is mainly reflected in the wheat sowing to tillering and lategrowing period. With the increase of the accumulation of soil organic matter and moisture content, theyield of NT practices was average37%higher than CT since2000. Studies have shown that the yielddependence on rainfall under CT practice was greater than NT, and the increment rate of WUE underNT practices relative to CT decreased as the precipitation.4. Water use efficiency (WUE) was highest in100%straw mulch under no-tillage practice, but had norelationship to the different rates of straw mulch.In no-till conditions, soil unsaturated hydraulic conductivity under NT0treatment was greater thanNT3treatment, but the rate of straw mulch and unsaturated hydraulic conductivity had no proportionalrelationship. Straw returning had greater effect on soil surface moisture than the bottom and the NT3treatment improved soil surface moisture, reducing soil subsurface moisture at the harvest time. Inno-till conditions, soil water storage under straw mulch treatments basic basically performed: NT3＞NT2＞NT1and storage capacity increased with the rate of straw mulch, but during drought periods thecrusts under NT0surface soil affected soil water storage. The annual amount of soil evaporation underNT3treatment was lowest, and the rate of evaporation reduction was14%. During wheat-growingperiod the amount transpiration under NT3treatment was24%higher than NT0treatment. The ratio ofevaporation to evapotranspiration under NT3treatment was39%, and the NT3treatment could reduceunproductive water consumption with the WUE of13.5kg·hm-2·mm-1, but there is no proportionalrelationship with the rate of straw mulch.5. DW cycles have a significant effect on C mineralization rate as well as on PE.The SOC mineralisation rate in rewetted soils was greater than that in soils that were kept at aconstant water content, with an average of3.90mg C-CO12kg-soil day-1. Following its input, wheatstraw significantly (207%) increased the rate of total C mineralisation, which was sharply (373%)reduced from2d to11d under the continuously wet treatment. The proportion of CO2from strawdeclined dramatically (78%) during the first10days, and this proportion under the continuously wettreatment (44%to34%) was greater than that under the DW cycles (33%to13%) from the rewettingday. The intensity of the priming effect (PE) that was induced by wheat straw decreased over time; thepriming direction was first positive, and then negative under the same water treatment. This intensity ofthe no-tillage soil under DW cycles ranged from15.32to﹣0.62mg C kg-1soil day-1, which was higher than that of the soils that were kept at a constant water content. There was no significant difference inthe SOC mineralisation rate between the top-20-cm samples from the NT and CT soils. The dataindicate that the DW cycles had a significant effect on the SOC mineralisation rate as well as on the PE,demonstrating a positive interaction between wheat straw and moisture fluctuations. Our results suggestthat under the simulated DW cycle conditions, the CO2release rate in the rewetted native soils underthe NT system (with straw) was greater than in the soils under the CT system (without straw), and thisdifference decreased with the incubation time.. Therefore, we consider our results to be the first todemonstrate the dynamics of priming effects in a long-term no-tillage soil of wheat straw under DWcycles as in the field.更多还原