ASSESSMENT OF EUTROPHICATION RISKS IN BLACK-ODOR WATER BODIES BASED ON SENTINEL-2 IMAGERY AND KDEVINE COPULA: A CASE STUDY OF CHANGCHUN, CHINA

Fast industrialization in developing countries has deteriorated urban water quality, and the assessment of aquatic ecological environment quality is becoming challenging.In this study, separability and thresholds (SEaTH), integrated with Vector Machine (SVM) was applied based on Sentinel-2 imagery for the identification of black-odor water bodies in several typical water bodies in Changchun City, China.A eutrophication risk assessment model for black-odor water bodies based on KDE (Kernel Density Estimation, KDE)-Vine Copula was proposed, integrated with Geographic Detector analysis.The results showed that from 2017 to 2019, the maximum value of Chlorophyll-a (Chl-a) in the typical water bodies decreased from 42.60 g/L to 38.07 g/L, and the Total Suspended Solids (TSS) content decreased from 208.20 mg/L to 198.27 mg/L, indicating a general reduction in degree of eutrophication.The highest risk probabilities in 2017-2019 were 44.41%, 34.96%, and 51.73%, respectively.Significant progress in the restoration of water ecological functions was observed in 2018, while the water ecological quality declined in 2019.According to the Geographic Detector results, TSS had a stronger explanatory power on eutrophication.However, the interaction between TSS and factors with weaker explanatory power, such as Chl-a and Secchi Depth (SD), did not significantly enhance the explanatory effect.