基于MOD13产品水稻遥感估产模型研究  

Models for rice yield estimation using remote sensing data of MOD13

作　　者：程乾[1] 

机构地区：[1]浙江工商大学旅游学院资源环境与城乡规划系,杭州310035

出　　处：《农业工程学报》2006年第3期79-83,共5页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家863子课题(2002AA243011)

摘　　要：中分辨率成像光谱仪(MODIS)可以提供经过加工处理的多时相44种数据产品,其中MOD13就是经过严格算法处理后的16d内每个像元的最佳的植被指数产品。通过杭州市郊野外大田选取适当研究区,进行产量构成要素和实际理论产量的测算,建立MOD13-VI(VegetationIndex)产品与水稻理论产量以及产量构成要素之间的相关关系,结果表明:这种相关与地面光谱和水稻理论产量所建立的相关关系具有一致性。经过各种估产模型的比较,复合估产模型更能反映整个生育期的情况,准确度有所提高,MOD13-EVI(EnhancedVegetationIndex)所建立的估算模型精度显著优于MOD13-NDVI(NormalizedDifferenceVegetationIndex),研究表明利用各个生育期MOD13产品所建立的复合模型有助于提高水稻产量的估算精度。Among 44 products of MODIS, MOD13 is the best vegetation index product which was strictly calculated. Suitable study area was selected in the rice planting area in the Hangzhou city, and the number of effective spike, effective grain numbers, one-thousand-grain weight and theoretical rice yield were measured. The relationship among MOD13-NDVI, MOD13-EVI and theoretical rice yield were analyzed and the estimation models for rice yield were established according to different rice growth stages, and different models were compared. Results showed that the comprehensive estimation model of MOD13 was the best estimation model, and MOD13-EVI was superior to MOD13-NDVI in estimating rice yield.

关 键 词：MOD13水稻 估产 模型 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置] S127[自动化与计算机技术—控制科学与工程]