Remote Sensing Evaluation of Trophic Status in the Daihai Lake Based on Fuzzy Classification

Trophic state index (TSI) is a critical ecological and environmental issue in water resource management that has garnered significant attention. Given the complexity of optical characteristics in aquatic environments, this study employs fuzzy classification methods (FCM) and composite nutrient status indices to meticulously classify in-situ remote sensing reflectance data, aiming to develop evaluation models for different nutrient status categories to facilitate the assessment of the Daihai River in Inner Mongolia, China. Subsequently, we applied this model to MSI data to analyze the nutrient status of Daihai Lake from 2016 to 2021. Furthermore, a structural equation model (SEM) was utilized to explore the primary driving factors influencing nutrient status. The results indicated that the water bodies in Daihai Lake can be broadly classified into three categories, with the nutrient status models demonstrating robust performance for each category (R2 = 0.80, R2 = 0.83, and R2 = 0.74). Comparisons were made between nutrient status accuracies obtained through the NCM and FCM based on measured data, yielding R2 values of 0.74 and 0.85, respectively. Furthermore, the TSI results derived from MSI inversion were validated, with NCM achieving an R2 of 0.49, RMSE of 6.88, and MAPE of 10.36%, while FCM exhibited an R2 of 0.55, RMSE of 8.89, and MAPE of 13.18%. An SEM–based analysis revealed that over the long term, human activities exerted a more substantial impact on eutrophication in Daihai Lake, while climatic factors played an accelerating and reinforcing role. These results are consistent with prior research in the Daihai area, indicating a state of mild eutrophication and the potential of the fuzzy classification method and comprehensive trophic status index method in eutrophication assessment.