Associations between urbanization-driven watershed landscape pattern shifts and seasonal riverine environmental-ecological responses

Rapid urbanization reshapes land use, water quality, and sediment conditions in river basins, yet their combined effects on phytoplankton remain inadequately understood. In particular, whether intensified landscape fragmentation during flood periods can decouple phytoplankton communities from their usual dependence on water quality remains an unresolved question. This study investigated spatio-temporal variations of land use, water quality, sediment, and phytoplankton (structure, biomass, diversity, resource use efficiency) in the Tuojiang River's Chengdu section to clarify their relationships and impact pathways. Redundancy analysis, variance partitioning analysis, Mantel test, and Partial least squares path model were used to quantify relationships and impact pathways. Results showed water quality was the dominant driver of phytoplankton, while land use (e.g., cropland proportion, patch density) became significantly more important during floods periods (explanation rate increased from 6.0% to 11.5%). Spatially, from upstream (forest-dominated, low fragmentation) to downstream (cropland-dominated, high fragmentation), phytoplankton richness and diversity rose, but evenness and RUE declined potentially associated with the increased dominance of Bacillariophyta (diatoms, >90% proportion), which may have altered community balance under eutrophic and hydrologically disturbed conditions. PLS-PM revealed period-specific pathways: land use exhibited stronger statistical associations with phytoplankton communities during the flood period, potentially decoupling communities from their usual dependence on water quality, whereas the phytoplankton community during the dry period was primarily associated with changes in water quality. Thus, priority should be given to internal pollution mitigation during dry periods, while controlling land use types and runoff during flood periods. These findings suggest a seasonal shift from internal river regulation toward watershed-scale landscape correlational modulation under hydrological disturbance, though further validation is needed to indicate causal mechanisms.