作 者 ：邹龙,冯浩
期 刊 ：植物营养与肥料学报 2014年 20卷 6期 页码：1413-1420
Keywords:DSSAT, spring maize, the Loess Plateau, Ansai,
摘 要 ：【目的】本研究在黄土高原丘陵沟壑区引入农业技术转化决策系统(DSSAT, Decision Support System for Agrotechnology Transfer)研究春玉米生长过程,旨在组建该区土壤、 气象、 作物品种的DSSAT模型数据库,验证和评估DSSAT模型在黄土丘陵沟壑地区的适用性,为模型在黄土丘陵沟壑地区的应用奠定基础。【方法】研究主要基于陕西安塞农田生态系统国家野外科学观测研究站2002、 2003、 2005年春玉米田间试验结果,采用DSSAT4.5内嵌的GLUE(Generalized Likelihood Uncertainty Estimation)软件,结合2002、 2003、 2005三年CK处理的大田试验数据运行GLUE软件3000次,对黄土高原丘陵沟壑区玉米品种“沈单10” 进行参数率定,然后运行DSSAT4.5中的CERES-Maize模型对其他处理条件下的春玉米生育期、 产量、 生物量进行模拟,并结合田间实测数据进行对比验证,对模型参数的适用性进行评价,从而评价CERES-Maize模型在该区春玉米种植中的适用性。【结果】 1)2002年和2005年对春玉米生育时期的模拟误差均在3天以内,产量模拟的归一化的均方根误差(n-RMSE)和平均相对误差(MRE)均小于10%,生物量模拟的误差稍大,但均在可接受范围内。2)2003年模拟的生育期时间和春玉米实测生育期时间基本吻合,唯独出苗日期相差较大,为9天;产量模拟的n-RMSE和MRE分别为6.47%和5.03%;地上生物量模拟误差稍大,模拟结果的n-RMSE和MRE分别为31.81%和30.21%。【结论】在正常降雨条件的情况下(2002年和2005年),CERES-Maize模型能够较好的模拟黄土高原丘陵沟壑区春玉米的生长过程、 地上部分生物量以及收获产量;在降雨较少的年份(2003年),模型对作物生产实际情况的再现能力相对变弱,因此CERES-Maize模型还不能很好的模拟干旱年份下作物的生长过程。
Abstract:【Objectives】The DSSAT (Crop-Environment Resource Synthesis System) model is used to simulate the growing of spring maize in Hill and Gully Area of the Loess Plateau, to establish the database of soil, weather and crop varieties of DSSAT about this area, and evaluate the applicability of this model, and also lay the foundation of the model application at this area. 【Methods】This study is based on results of spring maize field experiments conducted at Ansai Research Station of Farmland Ecosystem and National Field Scientific Observation in the years of 2002, 2003 and 2005. The GULE (Generalized Likelihood Uncertainty Estimation) which embedded in DSSAT4.5 was run 3000 times, combined with treatment CK in the years of 2002, 2003 and 2005 to calibrate parameters of “Shendan 10” which one of the maize varieties in the Loess Plateau, and then the CERES-Maize which one of the main crop models of DSSAT4.5 shell were used to simulate the growing period, yield and aboveground biomass of spring maize under others conditions. 【Results】1)the DSSAT model can be applied to simulate the growth of spring maize well and can supply an accurate simulation for the growing period and yield of spring maize in 2002 and 2005. The differences between observed data and simulated data are within 3 days for growing period, the n-RMSE and MRE of simulation yield are both less than 10%, although there’s no good work on aboveground biomass, the result could be accepted. 2)At 2003, the model can simulate the growing period of spring maize well, except the difference of 9 days for the emergence day. The yields are also simulated well, and the n-RMSE and MRE of simulation results are both less than 10%, which are 6.47% and 5.03%, respectively, while the simulation is not reasonable for the above ground biomass, the n-RMSE and MRE of simulation results are both more than 30%. 【Conclusions】The CERES-Maize model can simulate the growth period, aboveground biomass and yield of spring maize well at the Hill and Gully Area of the Loess Plateau under the condition of normal rainfall (2002, 2005). CERES can’t reappear the situation of spring maize in field accurately under the condition of less rainfall (2003), the CERES-Maize model can’t simulate the growing of spring maize well under the condition of less rainfall at this area.
全 文 ：