不同水分胁迫条件下DSSAT-CERES-Wheat模型的调参与验证  

Parameter estimation and verification of DSSAT-CERES-Wheat model for simulation of growth and development of winter wheat under water stresses at different growth stages

作　　者：姚宁[1,2] 周元刚[1,2] 宋利兵[1,2] 刘健[1,2] 李毅[1,2] 吴淑芳[1,2] 冯浩[2,3] 何建强[1,2] 

机构地区：[1]西北农林科技大学水利与建筑工程学院,杨凌712100 [2]西北农林科技大学中国旱区节水农业研究院,杨凌712100 [3]中国科学院水利部水土保持研究所,杨凌712100

出　　处：《农业工程学报》2015年第12期138-150,共13页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家高技术研究发展计划(863计划)(2013AA102904);国家自然科学基金(51209176);高等学校学科创新引智计划(No.B12007)

摘　　要：作物模型为人们认识旱区农业生境过程并对其进行调控提供了一种有效的工具。为了探讨小麦生长模拟模型DSSAT-CERES-Wheat能否准确模拟水分胁迫条件下旱区冬小麦的生长发育和产量形成过程,同时确定参数估计和模型验证的最优方案,该研究进行了连续两季（2012.10-2013.06和2013.10-2014.06）的冬小麦分段受旱田间试验。试验将冬小麦整个生育期划分为越冬、返青、拔节、抽穗和灌浆5个主要生长阶段,每相邻两个生长阶段连续受旱,形成4个不同的受旱时段水平（D1-D4）,根据小麦生育期的需水量,设置灌水定额分别为40和80 mm 2个水平（I1和I2）,共形成8个处理,每处理3次重复,在遮雨棚内采用裂区试验布置,此外在旁边设置1个各生育期全灌水的对照处理。文中设置了5套不同的参数估计和验证方案,利用DSSAT-GLUE参数估计模块得到不同的参数估计结果。通过对比分析冬小麦物候期、单粒质量、生物量、产量、以及土壤水分含量的模拟值和实测值之间的差异,以确定利用DSSAT-CERES-Wheat模型模拟旱区冬小麦生境过程的精度。结果表明,参数P1V（最适温度条件下通过春化阶段所需天数）和G3（成熟期非水分胁迫下单株茎穂标准干质量）具有较强的变异性,变异系数分别为19.07%和16.34%,受基因型-环境互作的影响较大,而其他参数的变异性则较弱,变异系数均小于10%;DSSAT-GLUE参数估计工具具有较好的收敛性,不同参数估计方案所得的参数值具有一定的一致性;不同的参数估计方案所得的模型输出结果有较大差异,其中参数估计方案1（利用两季试验中的充分灌溉处理CK数据进行参数估计,其他不同阶段受旱处理数据进行验证）的模型校正和验证精度最高,其中模型校正的绝对相对误差（absolute relative error,ARE）和相对均方根误差（relative root mean squared error,RRMSE）分别为4.89%和5.18%。在Crop growth simulation models are useful tools to help us understand and regulate the agro-ecological systems in arid areas.In this study,the CERES-Wheat,a wheat growth simulation model in the DSSAT（decision support system for agrotechnology transfer） software,was investigated for its ability to simulate the growth and yield of winter wheat（Triticum aestivum L.） in arid areas and to find the optimal plan for the estimation of genetic parameters and the model verification.Field experiments were conducted under a rainout shelter for winter wheat growing under water stresses at different growth stages in 2 growth seasons（from October 2012 to June 2013 and from October 2013 to June 2014）.The whole growth season of wheat was divided into 5 growing stages（wintering,greening,jointing,heading and grain filling）.Water stress occurred every 2 continuous stages while irrigations were applied at other stages,which resulted in 4 different levels of water stress period（D1-D4）.Two irrigation levels of 40 mm（I1） and 80 mm（I2） were applied.There were a total of 8 treatments,with 3 replicates for each,and the split-plot experiment was designed.An extra control treatment with irrigation in all 5 stages was arranged near.The experimental data were used to run the model.A total of 5 different plans for model calibration and verification were designed and the DSSAT-GLUE,a program package for parameter estimation in DSSAT,was used to estimate the relevant genetic coefficients.Then the 5 plans were compared for the discrepancies between corresponding simulated and observed values of phenological phase,single grain weight,biomass,yield and soil moisture so as to determine the accuracy of CERES-Wheat model to simulate the agro-ecological processes of winter wheat farming system in arid areas.The results showed that 2 genetic coefficients P1V（days required to complete vernalization at optimum vernalizing temperature） and G3（standard dry weight of a single tiller without stress at maturity） varied remarka

关 键 词：模型 作物 水分 冬小麦 DSSAT CERES-Wheat 

分 类 号：S512.11[农业科学—作物学]