Modeling visible and near-infrared snow surface reflectance-simulation and validation
基本信息来源于合作网站,原文需代理用户跳转至来源网站获取
摘要:
Retrieving snow surface reflectance is difficult in optical remote sensing.Hence,this letter evaluates five surface reflectance models,including the Ross-Li,Roujean,Walthall,modified Rahman and Staylor models,in terms of their capacities to capture snow reflectance signatures using ground measurements in Antarctica.The biases of all the models are less than 0.0003 in both visible and near-infrared regions.Moreover,with the exception of the Staylor model,all models have root-mean-square errors of around 0.02,indicating that they can simulate the reflectance magnitude well.The R2 performances of the Ross-Li and Roujean models are higher than those of the others,indicating that these two models can capture the angle distribution of snow surface reflectance better.The bidirectional reflectance distribution flmction (BRDF) characterizes the angular distribution of surface reflection[1,2].It plays an important role in performing atmospheric correction,detecting land cover types,and calculating other biophysical parameters[3].Howcver,the retrieval of snow BRDF/albedo is always a difficult issue in the application of remotely sensed information.