基于同化植被净初级生产力的区域玉米产量估测

Regional Maize Yield Estimation Based on Assimilating Net Primary Production

为了评价植被净初级生产力NPP作为同化变量提高区域玉米产量估测精度的有效性及其应用于区域玉米产量估测的潜力,选择黑龙江省双鸭山市友谊农场玉米种植区为研究对象,以WOFOST为作物生长动态模型,分别以叶面积指数LAI和植被净初级生产力NPP为同化变量,选用MODIS LAI和NPP产品为遥感观测数据,开展基于遥感观测数据与作物生长模型同化的区域玉米产量估测研究。重点比较了分别以叶面积指数LAI和植被净初级生产力NPP为同化变量的区域玉米产量估测结果精度。结果表明,相较以叶面积指数LAI为同化变量的区域玉米产量估测统计结果(均值为7755 kg/hm^(2),标准差为1303 kg/hm^(2)),以植被净初级生产力NPP为同化变量的区域玉米产量估测统计结果(均值为9214 kg/hm^(2),标准差为190 kg/hm^(2))与研究区域统计结果(均值为8970 kg/hm^(2))更为接近,但在表现玉米产量空间异质性方面稍显不足。以植被净初级生产力NPP为同化变量开展区域作物产量估测是一种可行的数据同化策略,具有较大的应用潜力。

In order to evaluate the effectiveness of NPP(Net Primary Production)as an assimilation variable to improve the accuracy of regional maize yield estimation and its potential application to regional maize yield estimation,the study selected the maize growing area of Youyi Farm,located in Shuangyashan,Heilongjiang,as the research area,and used WOFOST as a dynamic model of crop growth,and the Leaf Area Index(LAI)and the Net Primary Production(NPP)as the assimilation variables,respectively.MODIS LAI and NPP products were used as remote sensing observation data.This study focused on comparing the accuracy of regional maize yield estimation results using leaf area index(LAI)and net primary production(NPP)as assimilation variables.The results showed that compared with the statistical results of regional maize yield estimation using leaf area index LAI as the assimilation variable(mean value=7755 kg/hm^(2),standard deviation=1303 kg/hm^(2)),the statistical results of regional maize yield estimation using NPP as the assimilation variable(mean value=9214 kg/hm^(2),standard deviation=190 kg/hm^(2))were more accurate.They were closer to observations of the study area(mean value=8970 kg/hm^(2)),but were slightly less effective in showing the spatial heterogeneity of maize yield.This study concludes that using NPP as an assimilation variable for regional crop yield estimation is a feasible data assimilation strategy with greater potential for application.