Optimizing Parameters of CSM-CERES-Maize Model to Improve Simulation Performance of Maize Growth and Nitrogen Uptake in Northeast China  

作　　者：LIU Hai-long YANG Jing-yi HE Ping BAI You-lu JIN Ji-yun Craig F Drury ZHU Ye-ping YANGXue-ming LI Wen-juan XIE Jia-gui YANG Jing-min Gerrit Hoogenboom 

机构地区：[1]Agricultural Information'Institute, Chinese Academy of Agricultural Sciences~Key Laboratory of Agri-lnformation Service Technology, Ministry of Agriculture, Beijing 100081, P.R. China [2]Greenhouse and Processing Crops Research Centre, Agriculture & Agri-Food Canada, Ontario NOR 1GO, Canada [3]Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences~Key Laboratory of Crop Nutrition and Fertilization, Ministry of Agriculture, Beijing 100081, P.R.China [4]Agricultural Environment and Resources Research Center, Jilin Academy of Agricultural Sciences, Changchun 130124, P.R.China [5]College of Resource & Environment Sciences, Jilin Agricultural University, Changehun 130118, P.R.China [6]AgWeatherNet, Washington State University, WA 99350-8694, USA

出　　处：《Journal of Integrative Agriculture》2012年第11期1898-1913,共16页农业科学学报（英文版）

基　　金：funded by the National Basic Research Program of China (2007CB109306 and 2013CB127405);The authors acknowledge Ministry of Education,China,for providing the scholarship (2008325008)

摘　　要：Crop models can be useful tools ibr optimizing fertilizer management for a targeted crop yield while minimizing nutrient losses. In this paper, the parameters of the decision support system for agrotechnology transfer （DSSAT）-CERES-Maize were optimized using a new method to provide a better simulation of maize （Zea mays L.） growth and N upfake in response to different nitrogen application rates. Field data were collected from a 5 yr field experiment （2006-2010） on a Black soil （Typic hapludoll） in Gongzhuling, Jilin Province, Northeast China. After cultivar calibration, the CERES-Maize model was able to simulate aboveground biomass and crop yield of in the evaluation data set （n-RMSE=5.0-14.6%）, but the model still over-estimated aboveground N uptake （i.e., with E values from -4.4 to -21.3 kg N ha-~）. By analyzing DSSAT equation, N stress coefficient for changes in concentration with growth stage （CTCNP2） is related to N uptake. Further sensitivity analysis of the CTCNP2 showed that the DSSAT model simulated maize nitrogen uptake more precisely after the CTCNP2 coefficient was adjusted to the field site condition. The results indicated that in addition to calibrating 6 coefficients of maize cultivars, radiation use efficiency （RUE）, growing degree days for emergence （GDDE）, N stress coefficient, CTCNP2, and soil fertility factor （SLPF） also need to be calibrated in order to simulate aboveground biomass, yield and N uptake correctly. Independent validation was conducted using 2008-2010 experiments and the good agreement between the simulated and the measured results indicates that the DSSAT CERES-Maize model could be a useful tool for predicting maize production in Northeast China.Crop models can be useful tools ibr optimizing fertilizer management for a targeted crop yield while minimizing nutrient losses. In this paper, the parameters of the decision support system for agrotechnology transfer （DSSAT）-CERES-Maize were optimized using a new method to provide a better simulation of maize （Zea mays L.） growth and N upfake in response to different nitrogen application rates. Field data were collected from a 5 yr field experiment （2006-2010） on a Black soil （Typic hapludoll） in Gongzhuling, Jilin Province, Northeast China. After cultivar calibration, the CERES-Maize model was able to simulate aboveground biomass and crop yield of in the evaluation data set （n-RMSE=5.0-14.6%）, but the model still over-estimated aboveground N uptake （i.e., with E values from -4.4 to -21.3 kg N ha-~）. By analyzing DSSAT equation, N stress coefficient for changes in concentration with growth stage （CTCNP2） is related to N uptake. Further sensitivity analysis of the CTCNP2 showed that the DSSAT model simulated maize nitrogen uptake more precisely after the CTCNP2 coefficient was adjusted to the field site condition. The results indicated that in addition to calibrating 6 coefficients of maize cultivars, radiation use efficiency （RUE）, growing degree days for emergence （GDDE）, N stress coefficient, CTCNP2, and soil fertility factor （SLPF） also need to be calibrated in order to simulate aboveground biomass, yield and N uptake correctly. Independent validation was conducted using 2008-2010 experiments and the good agreement between the simulated and the measured results indicates that the DSSAT CERES-Maize model could be a useful tool for predicting maize production in Northeast China.

关 键 词：DSSAT CERES-Maize model maize growth simulation model evaluation fertilizer N experiment 

分 类 号：S513.01[农业科学—作物学] S511