Accurate streamflow forecasting is vital for sustainable water resource management but remains challenging due to pronounced spatiotemporal variability. This study evaluates two process-based models, the SWAT (comprehensive) and the GWLF (parsimonious), and a data-driven random forest (RF) model for monthly streamflow simulations in two contrasting Chinese basins: the humid southern basin (SSB) and the semi-arid northern basin (SRB). Using four statistical metrics (NSE, R², MAE, RMSE), we assess model accuracy, robustness in capturing extremes, and sensitivity to hydrological characteristics and data availability. The results reveal consistently superior performance in the SSB across all models, with SWAT demonstrating the highest overall accuracy—especially for peak flows—due to its physically based structure. The GWLF provides acceptable simulations with minimal data requirements, offering a practical alternative in data-limited regions, like the SRB. RF performs well in the SSB under zero-lag conditions but requires hydrologically informed lag structures in the SRB. However, it consistently underestimates high flows due to its lack of physical constraints. The findings underscore that model selection must, therefore, be guided not only by predictive performance but also by the underlying hydrological context, data availability, and the need for physical realism in decision-making.

Accurate streamflow forecasting is vital for sustainable water resource management but remains challenging due to pronounced spatiotemporal variability. This study evaluates two process-based models SWAT (comprehensive) and GWLF (parsimonious)—and a data-driven Random Forest (RF) model for monthly streamflow simulation in two contrasting Chinese basins: the humid southern basin (SSB) and the semi-arid northern basin (SRB). Using four statistical metrics (NSE, R2, MAE, RMSE), we assess model accuracy, robustness in capturing extremes, and sensitivity to hydrological characteristics and data availability. Results reveal consistently superior performance in the SSB across all models, with SWAT demonstrating the highest overall accuracy—especially for peak flows—due to its physically based structure. GWLF provides acceptable simulations with minimal data requirements, offering a practical alternative in data-limited regions like the SRB. RF performs well in the SSB under zero-lag conditions but requires hydrologically informed lag structures in the SRB. However, it consistently underestimates high flows due to its lack of physical constraints. The findings underscore that model selection must therefore be guided not only by predictive performance, but by the underlying hydrological context, data availability, and the need for physical realism in decision-making.

Ecological water replenishment is crucial for alleviating ecological degradation of lakes in cold and arid areas. However, especially if high-nitrogen water is used as a water source for replenishment, the water volume in lakes effectively increases, resulting in new environmental pressures, posing potential threats to aquatic ecosystems. Clarifying the response characteristics of plankton communities to high nitrogen replenishment is crucial for lake management. Under high nitrogen replenishment conditions, we focused on the ecological niche characteristics (including niche width and niche overlap), interspecies relationships and interaction networks of dominant phytoplankton and zooplankton species in Ulansuhai Lake in Inner Mongolia, with the aim of exploring the potential associations and response patterns of the plankton community. The results indicated that under high nitrogen replenishment conditions, the structure of plankton communities generally remained stable, with no significant fluctuations in species composition, density, biomass, or diversity indices. The niche width values of the dominant species were mostly above 0.6, suggesting that these species adopted niche-broadening strategies to enhance their environmental adaptability. Within the phytoplankton community, species exhibited a significant positive correlation, reflecting strong co-occurrence and niche differentiation. In contrast, interspecific relationships among zooplankton were not significant, reflecting asynchronous dynamics and different developmental trends under fluctuating environments. Co-occurrence network analysis further indicates that both phytoplankton and zooplankton form complex networks characterized by a high degree of modularity. Nevertheless, the phytoplankton and zooplankton networks in 2024 exhibited a reduction in size, an increase in negative cohesion (indicating heightened competition), and a decline in network robustness compared to 2023. The findings of this study provide a theoretical basis for the scientific implementation of ecological water replenishment and ecosystem management in cold and arid lakes under high-nitrogen conditions.