基于不同空间分辨率土壤数据库的中国江苏省水稻土有机碳的动态模拟

A number of process-based models have been developed for quantifying carbon(C)sequestration in agro-ecosystems.The DeNitrification-DeComposition(DNDC)model was used to simulate and quantify long-term(1980-2008)soil organic carbon(SOC)dynamics in the important rice-producing province,Jiangsu,China.Changes in SOC storages were estimated from two soil databases differing in spatial resolution:a county database consisting of 68 polygons and a soil patch-based database of 701 polygons for all 3.7 Mha of rice fields in Jiangsu.The simulated SOC storage with the coarse resolution county database ranged between 131.0-320.6 Tg C in 1980 and 170.3-305.1 Tg C in 2008,respectively,while that estimated with the fine resolution database was 201.6 and 216.2 Tg C in 1980 and 2008,respectively.The results modeled with the soil databases differing in spatial resolution indicated that using the soil input data with higher resolution substantially increased the accuracy of the modeled results;and when lacking detailed soil datasets,the DNDC model,parameterized with the most sensitive factor(MSF) method to cope with attribute uncertainty,could still produce acceptable results although with deviations of up to 60% for the case study reported in this paper.     

不同秸秆还田率情境下的亚热带水田土壤的“碳汇”贡献模拟研究

明确不同秸秆还田量对土壤“碳汇”的贡献大小是合理制定农业碳中和措施的基础。以我国典型亚热带地区——福建省水田土壤为研究对象,基于2016年15833个土壤样点实测数据和目前该地区最详细的1︰5万大比例尺土壤数据库,运用农业生态系统中广泛使用的DNDC(DeNitrification and DeComposition)模型模拟了不同秸秆还田率下全省未来的土壤有机碳动态变化。结果表明,2017—2053年传统管理(15%)以及秸秆还田30%、50%和90%下水田土壤的年均固碳速率分别为173、302、478和838 kg·hm^-2,固碳总量分别为11.56、20.15、31.90和55.95 Tg。从土壤亚类来看,咸酸和盐渍水稻土的年均固碳速率最大,不同秸秆还田率下介于220～920kg·hm^-2·a^-1之间;而渗育和潴育水稻土的固碳量最大,不同秸秆还田率下合计介于9.45～45.52 Tg之间,约占研究区总固碳量的81%。从行政区来看,龙岩、泉州两个地级市的固碳速率和总量均最大,不同秸秆还田率下均分别在202～937 k...     

美国路易斯安那州土壤碳估算的尺度效应——美国STATSGO和SSURGO两大最新土壤数据库的比较研究

Estimation of soil organic carbon (SOC) pools and fluxes bears large uncertainties because SOC stocks vary greatly over geographical space and through time.Although development of the U.S.Soil Survey Geographic Database (SSURGO),currently the most detailed level with a map scale ranging from 1:12 000 to 1:63 360,has involved substantial government funds and coordinated network efforts,very few studies have utilized it for soil carbon assessment at the large landscape scale.The objectives of this study were to 1) compare estimates in soil organic matter among SSURGO,the State Soil Geographic Database (STATSGO),and referenced field measurements at the soil map unit;2) examine the influence of missing data on SOC estimation by SSURGO and STATSGO;3) quantify spatial differences in SOC estimation between SSURGO and STATSGO,specifically for the state of Louisiana;and 4) assess scale effects on soil organic carbon density (SOCD) estimates from a soil map unit to a watershed and a river basin scale.SOC was estimated using soil attributes of SSURGO and STATSGO including soil organic matter (SOM) content,soil layer depth,and bulk density.Paired t-test,correlation,and regression analyses were performed to investigate various relations of SOC and SOM among the datasets.There were positive relations of SOC estimates between SSURGO and STATSGO at the soil map unit (R2=0.56,n=86,t=1.65,P=0.102;depth:30 cm).However,the SOC estimated by STATSGO were 9%,33% and 36% lower for the upper 30-cm,the upper 1-m,and the maximal depth (up to 2.75 m) soils,respectively,than those from SSURGO.The difference tended to increase as the spatial scale changes from the soil map unit to the watershed and river basin scales.Compared with the referenced field measurements,the estimates in SOM by SSURGO showed a closer match than those of STATSGO,indicating that the former was more accurate than the latter in SOC estimation,both in spatial and temporal resolutions.Further applications of SSURGO in SOC estimation for the entire United States could improve the accuracy of soil carbon accounting in regional and national carbon balances.     

秸秆源性碳在黑土和褐土中的转运及其对氮肥的响应

采用室内培养结合同位素13C标记技术,探讨了添加小麦秸秆和氮肥后黑土和褐土的碳素矿化特征,分析了土壤原有碳（Cs）和秸秆源性碳（Cstr）在土壤不同有机碳组分（轻组、闭蓄态组分、重组）中的转运特征。结果表明：土壤培养过程是碳矿化损失的过程,施氮肥对黑土和褐土中碳矿化均有一定的抑制效应,褐土中该抑制效应在施秸秆时较不施秸秆时更显著。土壤有机碳组分中Cs损失率表现为由高到低依次为重组、闭蓄态组分、轻组,施秸秆使黑土重组Cs损失率由2.83%增加至5.53%（P<0.05）,使褐土闭蓄态组分中Cs损失率由1.86%减少至0.82%（P<0.01）。随培养时间的进行,土壤轻组中Cstr经降解逐渐向闭蓄态组分和重组中转移,180 d后转移缓慢;至培养结束（300 d）时,Cstr总残留率在黑土与褐土之间无显著差异,轻组中Cstr残留率为褐土（4.98%～8.52%）显著高于黑土（1.71%～2.47%）。与不施氮相比,施氮处理使褐土轻组的Cstr残留率增加了近一倍。综上,施氮肥对黑土和褐土中碳矿化均有一定的抑制效应,重组是土壤碳损失的主要来源,添加秸秆对黑土重组的碳矿化有激发效应,而对褐土闭蓄态组分的碳矿化有抑制作用。褐土轻组中外源秸秆的降解程度低于黑土,并且施氮肥抑制其降解。     

基于Century模型和1:500 000土壤数据库的华东旱地土壤有机碳动态模拟

Changes in soil organic carbon (SOC) in agricultural soils influence soil quality and greenhouse gas concentrations in the atmosphere. Dry farmland covers more than 70% of the whole cropland area in China and plays an important role in mitigating carbon dioxide (CO2) emissions. In this study, 4109 dry farmland soil polygons were extracted using spatial overlay analysis of the soil layer (1:500000) and the land use layer (1:500000) to support Century model simulations of SOC dynamics for dry farmland in Anhui Province, East China from 1980 to 2008. Considering two field-validation sites, the Century model performed relatively well in modeling SOC dynamics for dry farmland in the province. The simulated results showed that the area-weighted mean soil organic carbon density (SOCD) of dry farmland increased from 18.77 Mg C ha1 in 1980 to 23.99 Mg C ha1 in 2008 with an average sequestration rate of 0.18 Mg C ha1 year?1. Approximately 94.9% of the total dry farmland area sequestered carbon while 5.1% had carbon lost. Over the past 29 years, the net SOC gain in dry farmland soils of the province was 19.37 Tg, with an average sequestration rate of 0.67 Tg C year1. Augmentation of SOC was primarily due to increased consumption of nitrogen fertilizer and farmyard manure. Moreover, SOC dynamics were highly differentiated among dry farmland soil groups. The integration of the Century model with a fine-scale soil database approach could be conveniently utilized as a tool for the accurate simulation of SOC dynamics at the regional scale.     

基于DNDC模型的福建省不同利用类型耕地土壤固碳潜力未来预估

目的明确不同利用类型耕地土壤“碳源/汇”的贡献大小,为合理制定农业碳中和措施提供基础。研究不同利用类型耕地土壤有机碳(Soil Organic Carbon,SOC)的时空演变特征,探讨影响SOC变化的主控因子及其与SOC变化速率（d_SOC）的关系,从而定量表征福建省不同利用类型耕地土壤固碳潜力。方法基于农业农村部福建省测土配方施肥项目的230,445个实测样本数据所建立的1∶5万大比例尺土壤数据库,结合DNDC (Denitrification and Decomposition) 模型模拟了2021 ~ 2033年福建省不同利用类型耕地的固碳速率和固碳潜力。结果全省耕地年均d_SOC为1.82 kg C hm⁻² a⁻¹,固碳总量（T_SOC）为39.40 Gg C。其中水田、旱地和水浇地的年均d_SOC分别为16.58、−78.08和−126.22 kg C hm⁻² a⁻¹,T_SOC分别为306.93、−195.21和−72.33 Gg C。从空间分布来看,福建省耕地d_SOC总体上呈现西北内陆低、东南沿海高的分布格局,且水田高于旱地和水浇地。从时间序列来看,水田d_SOC、T_SOC和土壤有机碳密度（Soil Organic Carbon Density,SOCD）的变化特征与全省耕地相似,且显著高于旱地和水浇地。初始SOC含量是影响各利用类型耕地年均d_SOC的主要驱动因素,降雨通过影响初始SOC含量及土壤pH值间接影响d_SOC。结论福建省耕地土壤表现为弱的碳汇,其中以水田碳汇为主,旱地和水浇地表现为碳源。因此,应在保持经济效益和作物生长的前提下增加水田有机肥投入量,针对旱地和水浇地注重雨季排水、施用生物炭肥。     

基于地形与遥感辅助信息的小流域尺度高分辨率有机碳空间分布预测研究

土壤有机碳(soil organic carbon,SOC)既是衡量土壤质量的重要指标,也是影响全球碳氮循环的关键因素之一。作为数字土壤制图(digital soil mapping,DSM)研究中起主要作用的环境变量,地形元素在SOC预测制图中也是无可替代的。应用机器学习模型,通过引入不同超参数设置下获得的高分辨率(5 m)Geomorphons(GM)地形分类图作为丘陵地形特征信息的补充,结合数字高程模型(digital elevation model,DEM)衍生变量和光学、合成孔径雷达(synthetic aperture radar,SAR)遥感数据对句容市黄梅镇北部小流域尺度(1︰25 000)丘陵地貌区地表层SOC含量进行预测制图,并评估不同GM变量在SOC含量预测中的表现。基于74个土壤样本和不同环境变量组合,分别采用袋装决策回归树(bagged classification and regression tree,bagged CART)、随机森林(random forest,RF)和立体派(cubist)三种方法构建SOC含量预测模型,并通过四个精度验证指标,采用十...     

土壤碳氮对多年生能源草、覆盖作物和氮施肥的响应

Cover crop and nitrogen(N) fertilization may maintain soil organic matter under bioenergy perennial grass where removal of aboveground biomass for feedstock to produce cellulosic ethanol can reduce soil quality. We evaluated the effects of cover crops and N fertilization rates on soil organic carbon(C)(SOC), total N(STN), ammonium N(NH_4-N), and nitrate N(NO_3-N) contents at the0–5, 5–15, and 15–30 cm depths under perennial bioenergy grass from 2010 to 2014 in the southeastern USA. Treatments included unbalanced combinations of perennial bioenergy grass, energy cane(Saccharum spontaneum L.) or elephant grass(Pennisetum purpureum Schumach.), cover crop, crimson clover(Trifolium incarnatum L.), and N fertilization rates(0, 100, and 200 kg N ha~(-1)). Cover crop biomass and C and N contents were greater in the treatment of energy cane with cover crop and 100 kg N ha~(-1) than in the treatment of energy cane and elephant grass. The SOC and STN contents at 0–5 and 5–15 cm were 9%–20% greater in the treatments of elephant grass with cover crop and with or without 100 kg N ha~(-1)than in most of the other treatments. The soil NO_3-N content at 0–5 cm was 31%–45% greater in the treatment of energy cane with cover crop and 100 kg N ha~(-1)than in most of the other treatments.The SOC sequestration increased from 0.1 to 1.0 Mg C ha~(-1)year~(-1)and the STN sequestration from 0.03 to 0.11 Mg N ha~(-1)year~(-1)from 2010 to 2014 for various treatments and depths. In contrast, the soil NH_4-N and NO_3-N contents varied among treatments,depths, and years. Soil C and N storages can be enriched and residual NO_3-N content can be reduced by using elephant grass with cover crop and with or without N fertilization at a moderate rate.     

基于进化遗传优化的自适应神经模糊推理新系统与地理信息系统的非洲山地土壤有机碳储量估算和绘图

Soil organic carbon (SOC) pool has the potential to mitigate or enhance climate change by either acting as a sink,or a source of atmospheric carbon dioxide (CO2) and also plays a fundamental role in the health and proper functioning of soils to sustain life on Earth.As such,the objective of this study was to investigate the applicability of a novel evolutionary genetic optimization-based adaptive neuro-fuzzy inference system (ANFIS-EG) in predicting and mapping the spatial patterns of SOC stocks in the Eastern Mau Forest Reserve,Kenya.Field measurements and auxiliary data reflecting the soil-forming factors were used to design an ANFIS-EG model,which was then implemented to predict and map the areal differentiation of SOC stocks in the Eastern Mau Forest Reserve.This was achieved with a reasonable level of uncertainty (i.e.,root mean square error of 15.07 Mg C ha-1),hence demonstrating the applicability of the ANFIS-EG in SOC mapping studies.There is potential for improving the model performance,as indicated by the current ratio of performance to deviation (1.6).The mnapping also revealed marginally higher SOC stocks in the forested ecosystems (i.e.,an average of 109.78 Mg C ha-1) than in the agro-ecosystems (i.e.,an average of 95.9 Mg C ha-1).     

基于Landsat遥感影像和1∶50 000土壤数据库的福州市耕地有机碳动态变化研究

耕地土壤碳库是全球碳库中最为活跃的部分,其变化对全球气候变化产生重要影响。目前对耕地土壤有机碳估算多采用中、小系列比例尺的土壤数据库,较少结合遥感影像与大比例尺土壤数据库进行估算。基于此,本研究采用Landsat遥感影像和1∶50 000高精度土壤数据库,以福建省福州市为例,基于遥感与碳循环过程模型对1987年和2016年耕地土壤有机碳动态变化进行研究。结果表明,利用Landsat影像反演得到的耕地土壤基础呼吸与土壤有机碳相关性强,建立的1987年和2016年模型R²分别为0.637和0.752。研究期间,全市耕地土壤有机碳密度从东部沿海向西部内陆地区递增,整体发挥着“碳汇”作用,有机碳密度和储量分别增加0.20 kg·m^-2和2.946×10⁵ t。从不同土壤类型比较得出,黄壤、红壤和水稻土是“碳汇”,有机碳密度分别增加0.70 kg·m^-2、0.40 kg·m^-2和0.19 kg·m^-2;其他土类为“碳源”,其中,水稻土碳储量最大,两期在全市总碳储量中占比均超过90%。从不同行政区比较得出,仓山区、长乐区、马尾区和连江县为“碳源区”,其他地区为“碳汇区”,其中,仓山区碳储量一直为全市最低,两期占比均不足0.5%,而福清市则一直居于全市首位,占比均高于20%。总体而言,福州市耕地土壤有机碳30年间空间动态变化显著,在不同土类和行政区间存在差异,今后应根据不同耕地土壤类型和行政区的有机碳情况有针对性进行耕地管理。     

制图尺度对CO2浓度升高情景下旱地土壤有机碳模拟的影响

基于土壤数据库的动态模型预测未来二氧化碳(CO_2)浓度升高下农田有机碳变化是实施农业固碳的基础,但目前基于不同制图尺度土壤数据库对旱地有机碳模拟结果的影响尚不清晰,一定程度上增加了农业管理措施制定的风险性。基于此,选择江苏北部(简称\"苏北地区\")3.90×10~6 hm~2旱地为例,运用生物地球化学过程模型(Denitrification and Decomposition,DNDC)模拟未来CO_2浓度升高下该地区1:5万、1:25万、1:50万、1:100万、1:400万、和1:1000万制图尺度的土壤有机碳变化。结果表明:2010—2039年间CO_2浓度在目前正常增加速率(1.9ppm a~(-1))的基础上提高0.5倍、1倍和2倍,苏北旱地数据最详细的1:5万尺度年均固碳速率分别为357 kg hm~(-2)、360 kg hm~(-2)和365 kg hm~(-2)。但进一步从其他制图尺度来看,由于使用的土壤数据库不同导致有机碳模拟结果差异很大。以1:5万尺度年均固碳速率为基准,3种CO_2浓度情景处理下1:25万~1:1000万尺度的模拟误差分别在0.89%~60.55%、0.81%~60.71%和0.15%~61.02%之间,这说明未来CO_2浓度升高的大背景下我国旱地土壤有机碳模拟中选择适宜的制图尺度非常重要。     

不同耕作措施下江苏省稻田土壤固碳潜力的模拟研究

以江苏省稻田为对象,整合DNDC和1:100万土壤数据库,以土壤图斑为基本模拟单元,定量估算少耕、免耕和综合措施（少耕 + 30% 秸秆还田）下江苏省稻田土壤的固碳潜力（0 ~ 30 cm）。模拟结果表明：相对于传统耕作,采用少耕、免耕和少耕 + 30% 秸秆还田均可明显地增加稻田SOC的积累,其在2009—2050年间的固碳潜力分别为24.5、47.7和43.8 Tg。免耕和少耕 + 30% 秸秆还田条件下稻田固碳速率大约是少耕的2倍。结合实际情况,少耕 + 30% 秸秆还田将是最可行的固碳措施之一。     

县域尺度红壤丘陵区水稻土有机碳模拟

区域尺度土壤有机碳储量的时空变化及其管理是全球气候变化和农业可持续发展研究的重要内容。本文以中亚热带红壤丘陵区的江西省余江县为例,基于12a的长期试验和1998年、2001年的野外定位采样对比研究,利用反硝化分解模型?DNDC(Denitrification-Decomposition)在田块和县域尺度研究了县域尺度表层(0~20 cm)水稻土有机碳储量的时空变化规律。结果表明,以长期试验数据验证,DNDC模型可以较好地模拟水稻土表层有机碳的长期动态变化。2001年农田水稻土(面积为3.6×108m2)表层(0~20 cm)有机碳总储量为2.9×109kg,平均土壤有机碳密度为6.0 kg m-2。1998年至2001年余江县水稻土表层土壤有机碳库逐年增加,年际平均变化量为3.0×107kg。通过对余江县水稻田模拟不同碳投入的情景,分析预测1998年至2017年土壤有机碳储量,种植绿肥提高秸秆还田比率同时减少化肥的投入,可有效地增加红壤区域有机碳蓄积。     

未来30年川东平行岭谷区县域农田SOC动态模拟

以位于川东平行岭谷的典型县——垫江县为研究对象,探讨在特定气候模式下未来30 a研究区农田土壤有机碳(SOC)及其动态变化,为研究区未来耕地可持续利用与管理提供数据支持和科学依据。利用生物地球化学模型DNDC,选取IPCC AR4报告中的BCCR_BCM 2.0的B1模式,在基于研究区土壤性质和农业管理制度等建立的GIS区域数据库的支持下,模拟研究区2011—2041年SOC动态变化。结果表明:1)DNDC模型能够较好地模拟特定气候条件下SOC及其动态变化,模拟值和观测值的相关系数r为0.981,达到0.01水平下的极显著相关关系;模拟值和观测值的RMSE值为16%,模拟结果较好。2)未来30 a研究区农田0~20 cm土层SOC密度和储量均呈显著增加态势,单位面积碳增量2 637.07~8 091.55 kg(C)·hm~(-2),增幅为10%~34%,新增固碳量2.7×10~5~8.3×10~5 t,年均增速87.9~269.7 kg(C)·hm~(-2)·a~(-1)。3)未来30 a川东平行区县域农田土壤总体呈持续碳增汇状态,研究区固碳、丢碳以及相对平衡三者间的差异将逐渐凸显。     

不同比例尺农田土壤碳库模拟的最佳栅格单元分辨率

基于太湖地区1∶50 000、1∶1 000 000、1∶14 000 000三种比例尺土壤矢量图斑单元以及通过上述矢量图斑单元转换生成的不同分辨率的栅格单元,在1982年水稻土土壤有机碳库的基础上,利用DNDC(Denitrification-Decomposition)模型模拟了2000年水稻土表层(0~20 cm)土壤有机碳库,对比分析了不同分辨率栅格单元与3种比例尺矢量图斑单元的水稻土类型数量、面积、表层土壤有机碳储量以及有机碳密度的变化特征,并以矢量图斑单元获得的这4个指标结果为基准,用相对变异百分数(VIV)来判别基于三种比例尺DNDC模型模拟的最佳栅格单元分辨率。结果表明,在4个指标的│VIV│1%前提下,基于1∶50 000、1∶1 000 000、1∶14 000 000三种比例尺的最佳栅格模拟单元分辨率分别为0.2 km×0.2 km、2 km×2 km、17 km×17 km,既能保证模型模拟过程中的精度要求,又可以避免数据冗余,提高模拟效率。建立的土壤碳库模拟研究的比例尺与其最佳栅格单元分辨率对应转换关系,对区域土壤碳模拟研究具有重要参考价值。     

Regional simulation of long-term organic carbon stock changes in cropland soils using the DNDC model: 2. Scenario analysis of management options

The development of successful agricultural policies in response to the Kyoto Protocol is aided by the identification of regions where the effects of soil organic carbon (SOC) sequestration measures can be maximized. We describe a modelling approach which incorporates a spatial analysis of the results from regional simulation under different management alternatives. SOC stock changes in Flemish cropland soils, simulated with the DNDC model, were previously fitted to a large data set of SOC measurements for the period 1990–2000. Using the results of this study, simulations with DNDC of SOC stock changes during the period 2006–2012 including the Kyoto commitment period were carried out at the community level for a business‐as‐usual (BAU) scenario and seven alternative agricultural management options for SOC sequestration. The baseline SOC stock decreased during that period by 0.15 t OC ha⁻¹ year⁻¹ compared with 0.48 t OC ha⁻¹ year⁻¹ in the 1990s. All alternative scenarios resulted in net SOC storage compared with the BAU scenario, but none of the individual scenarios were able to increase the average absolute SOC stock. Overall, spatial variability in SOC storage for the selected management options was strongly dependent on the current distribution of crops and associated management. The modelling approach used in this study provides a case study in regional scale modelling of SOC sequestration and is applicable to other regions in Europe with comparable intensive agriculture.     

黑土有机碳变化的DNDC模拟预测

为探讨黑土有机碳的长期变化规律及DNDC模型在土壤有机碳预测方面的适用性,本文利用吉林省公主岭地区黑土不同施肥措施下的长期定位试验数据,选取不施肥(CK)、单施化肥(NPK)、配施有机肥(NPKM)和增施有机肥(M2+NPK)4个处理进行土壤有机碳分析,并将数据用作DNDC模型验证。验证结果表明:各处理DNDC验证中RMSE值均小于10%(分别为5.09%、6.11%、9.38%、8.36%),说明模拟值与观测值一致性良好,模型可用于该地区土壤有机碳模拟。选取了化肥施用、有机肥施用、秸秆还田比率、温度及降水5个因子进行模型的敏感性分析,结果表明:有机肥的施用对土壤有机碳含量的影响最显著,且这种影响具有持久性。最后模拟了4种施肥情境下未来(至2100年)的土壤有机碳变化情况。结果表明:对照不施肥处理(CK)土壤有机碳含量略有下降,至2100年土壤有机碳含量为11.55 g·kg-1,较试验前土壤初始有机碳(13.2 g·kg-1)下降约12.5%。单施化肥处理(NPK)土壤有机碳含量较为稳定,并未出现土壤有机碳含量下降。配施有机肥(NPKM)和增施有机肥(M2+NPK)处理土壤有机碳含量增加明显,至2100年土壤有机碳含量为24.4 g·kg-1和27.6 g·kg-1,分别较初始有机碳含量上升84.8%和109.1%。