In Fuzhou section of Minjiang River Basin,multiple linear regression(MLR),machine model and deep learning model were constructed by using Sentinel-2 remote sensing images.The inversion accuracy of different models was compared and analyzed.The results showed that the inversion model based on the gradient boosting decision tree(GBDT)had the highest accuracy,inversion of the test set data resulted in the determination coefficients of 0.846,0.882 and 0.819 for potassium permanganate index(CODMn),total nitrogen(TN)and total phosphorus(TP),respectively.The contents of CODMn,TN and TP in the Fuzhou section of Minjiang River Basin during 2021-2023 were inverted by GBDT model,and the spatiotemporal variations of CODMn,TN and TP were analyzed.
remote sensing imageryinversionwater quality parametersgradient boosting decision tree
NETL
NSTL
万方数据
