松原市粮食产量预报方法初探  

A Preliminary Study on Forecasting Method of Grain Yield in Songyuan City

作　　者：李建平[1] 刘兵[1] 刘爽[1] 杜海信 王猛[1] 黄式琳 高凤岐[1] 赵秀英[1] 

机构地区：[1]吉林省松原市气象局,吉林松原138005

出　　处：《安徽农业科学》2011年第10期6227-6229,共3页Journal of Anhui Agricultural Sciences

摘　　要：通过对松原市粮食产量资料分析得出,玉米和水稻分别占总粮食产量的74.13%和12.43%,其他作物占产量的13.44%。采用等权重不同间距滑动平均得到的气象产量序列,间接反映出了因气象条件优劣造成的丰歉年景。由18个因子筛选组建的玉米产量预报方程验证结果,平均绝对误差仅0.08,平均相对误差<1%。另外,单独组建的其他粮产预报方程,复相关系数均通过0.001信度水准检验。As indicated by the analysis results on the grain yield data of Songyuan City,maize and rice yield accounted respectively for 74.13% and 12.43% of total grain yield,while other crops accounted for 13.44%.Meteorological yield series was obtained by equal weight moving average,which reflected the good and bad harvest years influenced by meteorological conditions.The mean absolute error and the relative error between the actual yield and predicted yield by prediction equation of maize yield comprising 18 meteorological factors were only 8% and less than 1% respectively.In addition,the multiple correlation coefficient of all other prediction equations constructed independently all passed the reliability test at 0.001 level.

关 键 词：粮食产量现状 粮食产量预报 逐步回归数学方法 

分 类 号：F32[经济管理—产业经济]