基于Landsat-8 OLI的鄱阳湖叶绿素a浓度定量反演  

Quantitative Retrieval of Chlorophyll a Concentration Based on Landsat-8 OLI in the Lakes

作　　者：黄灵光[1] 方豫[1] 张大文[2] 林联盛 

机构地区：[1]江西省遥感信息系统中心,南昌330046 [2]江西省农业科学院,南昌330200 [3]江西省山江湖开发治理委员会办公室,南昌330046

出　　处：《江西科学》2016年第4期441-444,456,共5页Jiangxi Science

基　　金：国家自然科学基金项目(编号:31360133);江西省重点项目五河流域海量遥感影像数据库平台研究与应用项目(编号:20144BBB70004)

摘　　要：Landsat-8 OLI数据的应用越来广泛、成熟,但尚未见诸在水质监测,尤其是叶绿素a浓度监测方面的应用。以叶绿素a作为参数,结合准同步的实测数据,通过在鄱阳湖的应用案例,研究、探讨利用Landsat-8 OLI卫星数据对湖泊叶绿素a遥感反演的可行性和相关潜力。根据研究,鄱阳湖与实测叶绿素a相关性最高的是波段B4(红波段),Pearson相关系数为0.916,并由此而选择B4作为构建模型的波段组合,建立线性回归、对数、指数、乘幂、二次多项式等五种回归模型,通过R2系数对比,显示二次多项式回归模型为最佳模型。经实测点验证,该模型能较好地反演出鄱阳湖的叶绿素a浓度。The application of Landsat-8 OLI data now is more widely used except in water quality monitoring,especially in the applications of chlorophyll a concentration monitoring. Through the application cases in Poyang Lake,this thesis will take chlorophyll a to combined with the quasi-synchronous measured data from Landsat-8 OLI,and to discuss the potential of remote sensing retrieval of chlorophyll a in lakes. The research shows the most relevant range of Poyang Lake and chlorophyll a is Band B4（ Red） and Pearson Correlation Coefficient is 0. 916. Therefore,this thesis chooses band B4 to establish five regression models-linear regression,logarithmic,index,exponentiation,quadratic polynomial. By comparison with R2 coefficient,the quadratic polynomial regression model is the best model. Verified by the measured points,the quadratic polynomial regression model can well performance the chlorophyll a concentration of Poyang Lake.

关 键 词：鄱阳湖 叶绿素A 模型 Landsat-8 OLI 

分 类 号：X173[环境科学与工程—环境科学]