【摘要】 以松花江哈尔滨江段高锰酸钾指数为研究对象,于2012年7月14～15日,测定了11个断面31个采样点的高锰酸钾指数数据,同步获取了采样点水质高光谱反射率数据,使用相关性统计分析和多元非线性回归分析方法,用其中21个样点数据构建反演模型,得到模型决定系数(R2)为0.890,用10个样点数据进行验证,得到预测值的均方根误差(RMSE)为0.439 mg·L-1,平均绝对偏差(MAD)为0.365 mg·L-1,平均绝对偏差率(MADR)为8.43%.更多还原

【Abstract】 In this paper,the research object is potassium permanganate index of the Songhua River in Harbin section.During July 14th～15th 2012,31 sampling points based on 11 sections of the river were surveyed and evaluated.At the same time,high spectral reflectance data of the water quality of the sampling points was acquired by correlation analysis and multivariate nonlinear regression analysis.Inversion model was built with the data on 21 sampling points among all the 31 sampling points.Coefficient of determination(R2) for the model was 0.890.The result was verified by data of 10 sampling points,and the root-mean-square error of predictive value(RMSE) was 0.439 mg·L-1,and the mean absolute deviation(MAD) was 0.365 mg·L-1.The mean absolute deviation rate(MADR) was 8.43%.The results of this study could provide a theoretical basis for monitoring potassium permanganate index of the Songhua River.更多还原