基于高光谱特征参数的棉花长势参数监测  

Monitoring of Cotton Growing Based on Hyperspectral Parameters

作　　者：金秀良[1,2] 李少昆[1,2] 王克如[1,2] 肖春华[1,2] 王方永[1,2] 陈兵[1,2,3] 陈江鲁[1,2] 吕银亮[1,2] 刁万英[1,2] 王琼[1,2] 王楷[1,2] 

机构地区：[1]石河子大学新疆兵团绿洲生态农业重点开放实验室,新疆石河子832003 [2]中国农业科学院作物科学研究所,农业部作物生理生态与栽培重点开放实验室,北京100081 [3]新疆农垦科学院棉花研究所,新疆石河子832000

出　　处：《西北农业学报》2011年第9期73-77,共5页Acta Agriculturae Boreali-occidentalis Sinica

基　　金：国家自然基金(30860139);国家科技支撑项目(2007BAH12B02);国家“863”项目(2006AA10Z207,2006AA10A302)

摘　　要：通过小区棉花密度和水分对比试验,分析不同密度和水分处理的棉花整个生育期生物量和LAI(叶面积指数)与高光谱特征参数的相关性,建立棉花生物量和LAI光谱估算模型。选取的所有植被指数与棉花LAI和生物量相关性均达到极显著水平,其中光谱参量NDVI(890,670)与LAI的相关性和光谱参量Height(920)与生物量相关性最好,用这2个参量与生物量建立棉花长势估测模型,其决定系数分别为0.804 0和0.760 9,均方根误差分别为0.191 5和0.315 2,利用光谱特征参数可以有效地监测棉花整个生育期的长势变化。Two field experiments were conducted with different densities and waters,the relationship was analyzed between spectral reflectance and biomass and LAI（leaf area index） and the estimating models were established for biomass and LAI in the whole growth stages of cotton.The results indicated that there were significantly correlations among the LAI and biomass with all the vegetation indexes.NDVI（890,670） and Height（920） were higher correlated with LAI and biomass,theirs correlation coefficient（r） were 0.896 7 and 0.872 3,respectively.NDVI（890,670） and the LAI and height（920） and biomass model were established for cotton growth estimation,the determination coefficient（R2） was 0.804 0 and 0.760 9,root mean square error（RMSE） was 0.191 5 and 0.315 2,respectively.The overall results suggested that spectral parameters can be effectively used to estimate cotton growth changes during growth stages of cotton.

关 键 词：棉花 LAI 生物量 植被指数 监测 

分 类 号：S562[农业科学—作物学]