基于积温-辐射与LAI积分面积模型的玉米成熟期预测

Forecasting of Regional Maize Maturity Using Accumulated Temperature-Solar Radiation Model and Leaf Area Index Integral Area Model

为预测区域尺度的玉米成熟期,以4 d的MODIS叶面积指数产品(LAI)为数据源,选择黑龙江省、吉林省和辽宁省3省玉米为研究对象,结合农业气象资料和全球多模式集合预报资料(THOPREX interactive grand gloabal ensemble,TIGGE),采用积温-辐射和LAI曲线积分面积两种模型,提前10 d对东北地区玉米成熟期进行逐日动态预测。结果表明,LAI曲线积分面积模型的预测结果在时效和精度上均为最优,该模型决定系数R 2达到0.87,均方根误差(RMSE)为2.5 d,并且有效地克服了当前成熟期预测方法空间分辨率低和预测时效性差等局限性。LAI曲线积分面积模型适用于大面积农作物成熟期预测。

Accurate prediction of maize maturity is of great significance for efficient harvesting and agricultural machinery dispatch management.In order to predict the maturity of maize at the regional scale in advance,the 4-day MODIS leaf area index(LAI)product was used as the data source,selecting the corn in Heilongjiang,Jilin and Liaoning Provinces as the research object,combined with agricultural meteorological data and global multi-model ensemble forecast data.The dynamic prediction of maize maturity in Northeast China was carried out ten days in advance by using the accumulated temperature solar radiation model and the LAI curve integral area model.The research showed that the prediction results of the integral area model of LAI curve were optimal in terms of time efficiency and precision.The coefficient of determination(R 2)of the LAI curve integral area model was 0.87,the root mean square error(RMSE)was 2.5 d,and it was effectively better than that of the accumulated temperature solar radiation model.The current maturity prediction method had limitations such as low spatial resolution and poor prediction timeliness.It showed that the LAI curve integral area model method had applicability in the prediction of large-area crop maturity.