Integrating Redundancy Correction and GIS‐Based Spatial Optimization for Enhanced Lake Eutrophication Assessment

ABSTRACT Traditional comprehensive trophic level index (TLI) methods for lake eutrophication assessment often neglect redundancy among evaluation factors and face challenges with spatial interpolation accuracy using ArcGIS. To address these issues, in this study, we propose an optimized eutrophication evaluation framework by integrating a novel “redundancy correction” weighting method and a refined GIS‐based spatial interpolation optimization strategy. A year‐long water quality monitoring campaign was conducted in 2022 to analyze the spatiotemporal distributions of permanganate index (COD Mn ), total nitrogen (TN), total phosphorus (TP), and chlorophyll a (Chl‐a) concentrations, as well as water transparency (SD). Building upon the conventional TLI framework, we introduce a “redundancy correction” weighting strategy that incorporates correlation matrices to eliminate redundant information, thereby optimizing the weight distribution of Chl‐a, TN, TP, COD Mn , and SD. Through coupling with an enhanced ArcGIS interpolation strategy based on sampling point densification and re‐interpolation, the eutrophication status of Aixi Lake was comprehensively evaluated. We found that TN concentrations ranged from 0.36 to 4.32 mg/L (mean: 1.04 mg/L), with a maximum value 2.16 times the Class V upper limit; Chl‐a varied between 1.56–372.71 μg/L, and TP concentrations ranged from 0.02 to 0.35 mg/L. Moreover, seasonal trends for Chl‐a, TN, TP, and COD Mn concentrations generally followed the pattern: wet season < normal season < dry season, with notably lower SD values occurring during dry period (mostly < 1 m at monitoring sites). For spatial mapping, when only a limited number of monitoring sites are available, preliminary interpolation maps are first generated. Then, additional sampling points were added using appropriate methods (e.g., 200 m × 200 m grid spacing in this study), their TLI values were extracted through ArcGIS, and final interpolation was performed using both the original and newly added points. This approach effectively improved the accuracy of the spatial evaluation. Finally, applying our methodology to observe spatiotemporal patterns, we found that the eutrophication gradients varied across seasons, though nutrient risks consistently intensified from the southern to the central lake regions (peaking at site S4). This integrated methodological framework enhances the precision and reliability of eutrophication assessments, providing practical guidance for lake‐management optimization.