| Abstrakt: |
Water turbidity is one of the most important parameters of water quality, which represents the transparency of water and is effective in eutrophication. This research was done to estimate the amount of water turbidity using remote sensing data and the random forest technique. For this purpose, the water quality monitoring data of Chitgar Lake in Tehran were used, which is an artificial shallow lake with recreational and urban scenery usage. The Landsat 8 OLI/TIRS and Sentinel 2 MSI satellite images were extracted after matching the date of field observation data and satellite images from 2016 to 2021. Data were divided into calibration and validation datasets. After performing pre-processing processes on satellite images, important bands were recognized using the random forest method. Afterward, appropriate band composition and algorithms were selected and regression models were fitted and validated. The optimum model was able to estimate water turbidity with Adj.R2=0.6, RMSE=1.07 NTU, and NRMSE=12% for Landsat-8 as well as with Adj.R2=0.73, RMSE=1.23 NTU and NRMSE=9% for Sentinel-2 satellite and estimated with a power of 80% for Chitgar Lake. Consequently, the optimal predictive model in Sentinel-2 was chosen with the assistance of the random forest. Moreover, the predictive model was able to estimate the water turbidity in Chitgar Lake with acceptable accuracy. [ABSTRACT FROM AUTHOR] |
