[Effects of Landscape Patterns at Different Spatial Scales on Dissolved Organic Matter in the Danjiang River Basin]

Landscape patterns can affect the biogeochemical cycles of river ecosystems. The Danjiang River Basin is an important water source for the central line project of the South-to-North Water Diversion. Investigating the influence of landscape patterns on runoff dissolved organic matter (DOM) is of great significance for the ecological protection of the basin. Based on field sampling data collected during the normal season (November 2023), dry season (March 2024), and wet season (July 2024), this study analyzed the spatial-temporal variation characteristics of runoff DOM in the Danjiang River Basin using excitation-emission matrix-parallel factor analysis (EEM-PARAFAC). The impacts of landscape patterns and hydrochemistry on dissolved organic matter (DOM) were assessed using the random forest model, redundancy analysis (RDA), and variance partitioning analysis (VPA). The results showed that the average concentration of dissolved organic carbon (DOC) in runoff in the Danjiang River Basin was 5.03 mg·L^-1, with a variation range of 2.59-9.96 mg·L^-1. The DOC concentration was highest during the wet season (5.80 mg·L^-1), followed by that during the dry season (5.28 mg·L^-1), and lowest during the normal season (4.02 mg·L^-1). Humic-like DOM accounted for 67.07% and 79.45% of the DOM during the normal and wet seasons, respectively. During the dry season, the relative abundances of humic-like and protein-like DOM were 49.21% and 50.79%, respectively. During the wet and normal seasons, dissolved organic matter (DOM) in the water exhibited stronger exogenous characteristics, while during the dry season, DOM showed more pronounced autochthonous characteristics. The spatial and temporal scales with the strongest influence of landscape pattern on DOC concentration and DOM composition were the 500 m circular buffer during the dry season and the 200 m circular buffer during the wet season, respectively. Urban land, ammonia nitrogen, and landscape patch cohesion index were the main factors driving the change in DOC. Water temperature, electrical conductivity, and total phosphorus (hydrochemistry factors), as well as forest land, grassland, and cropland (land use types), were significant influencing factors on the variation of runoff DOM components in the Danjiang River Basin. Additionally, the aggregation index and continuity index (landscape configuration metrics) also played important roles. These findings can provide reliable data for rational land use planning and serve as a reference for water environment management in the Danjiang River Basin.