Simulation of corn foxtail millet and pearl millet yields using ALMANAC model in the Loess Plateau
View Fulltext  View/Add Comment  Download reader
  
DOI:10.7606/j.issn.1000-7601.2009.01.16
Key Words: crop model  loess plateau  ALMANAC  corn  foxtail millet  pearl millet
Author NameAffiliation
LIU Gang State Key Laboratory of Earth Surface Processes and Resources Ecology School of Geography Beijing Normal University,Beijing 100875,China 
XIE Yun State Key Laboratory of Earth Surface Processes and Resources Ecology School of Geography Beijing Normal University,Beijing 100875,China 
WU Ruijun Institute of Soil and Water Conservation Chinese Academy of Sciences and Ministry of Water Resources,Yangling,Shaanxi 712100,China 
GAO Xiaofei State Key Laboratory of Earth Surface Processes and Resources Ecology School of Geography Beijing Normal University,Beijing 100875,China 
ZHANG Wenbo State Key Laboratory of Earth Surface Processes and Resources Ecology School of Geography Beijing Normal University,Beijing 100875,China 
Hits: 1658
Download times: 227
Abstract:
      Crop models are helpful in the decision-making process.The purpose of this study was to evaluate the ALMANAC crop model on farm level in a small watershed, Danangou, in the Loess Plateau.The simulated crops were corn, foxtail millet and pearl millet during 2000 and 2004.The results showed that the ALMANAC crop model could simulate well the yields of three kinds of crops.Values for coefficient of determination (R2) between measured and simulated yields were more than 0.72.The relative errors of measured yields and simulated yields of corn, foxtail millet and pearl millet in five years were-2.29%, -2.32% and 8.34%.The normal mean root square deviation between measured and simulated yields of corn, foxtail millet and pearl millet were 15.7%, 11.5% and 15.8%.The simulated errors of three kinds of crops were different in terraces, northern slopes and southern slopes.The mean of simulated yields of corn was above measured yields in terraces, but below in northern or southern slopes.The mean of simulated values of foxtail millet was below measured yields, but above in southern slopes.The mean of simulated values of pearl millet was above measured yields.The standard deviation of relative errors showed that the terraces had the smallest difference between the simulated and measured yields, the second were northern slopes, and the third were southern slopes.