摘要
选取北方春玉米区、黄淮海春夏播玉米区、西南山地丘陵玉米区、南方丘陵玉米区、西北内陆玉米区五个区域中70个站点，利用1998—2000年玉米生长的田间试验资料和气象台站的气象观测资料，研究CERES—Maize模型在主要玉米种植区域对玉米生育期、产量的模拟能力。选用符合度指数D、RMSE等统计指标进行评价。结果表明：玉米生育天数模拟的D值为0．58—0．95，产量模拟的D值为0．66～0．88，说明CERES—Maize模型可以较好地模拟出区域内玉米生育期和产量观测值的分布趋势；生育期的模拟误差小于产量的模拟误差；区域尺度的统计分析可以更准确的表示模型的模拟效果。
The historic maize planting dataset in 70 stations was collected to analyze and validate the availability of CERES-Maize crop growth model together with the historic weather data in China from 1998 to 2000. These stations are located at North China spring maize area, Huanghua- ihai plain summer maize area, South China mountain maize area, Southwest China maize area and Northwest China maize area. The data of grow- ing period and yield were taken into account. Several statistic indexes,such as index of agreement （D） and root mean square error （RMSE）, were applied to analyze and assess the model performance. The value of D is from 0. 58 to 0.95 between the simulation and observation about growing period and the D of yield is from 0. 66 to O. 88. These indicated that the simulations by CERES-Maize had an adequately agreement with observations. The errors between simulation and observation of growing period were lower than that of yield. The results also showed that statistic analysis at regional scale could be better to express the simulation results and assessments.
出处
《中国农业气象》
CSCD
2008年第4期383-386,405,共5页
Chinese Journal of Agrometeorology
基金
国家“十一五”科技支撑计划课题(2007BAC03A02)
国家“十五”科技攻关课题(2004BA611B)
