Estimation of CODMn in Tai lake basin using Landsat-8 satellite

COD_Mn is employed as a kind of water quality parameter indicating eutrophication. This paper focuses on the estimation of COD_Mn in Tai lake basin, China, and has proposed an Advanced COD_Mn Forecast Index (ACFI) with 3 sub-indices X₁, X₂ and X₃, based on empirical and semi-empirical methods. Landsat-8 has been involved in the research for the purpose of exploring its band features in water quality monitoring. Comparing in-situ data and Landsat-8 image on July 19th 2013, we selected band 8 for X₁, combination of band 1 and band 7 for X₂ and the ratio of band 4 to band3 for X₃. ACFI model is proved to work well on the estimation of COD_Mn with the smallest average absolute relative error 7.8% and the biggest average absolute relative error 17.1% from June and March respectively using the fitting coefficient of July. What is more, the research also shows that the ACFI model is suitable when COD_Mn is in the range of 3-5mg/L.