| Summary: | Nutrient transport in seasonally ice-covered lakes is an important factor affecting spring algal blooms in eutrophic waters; because phase changes during the ice growth process redistribute the nutrients. In this study, nutrient transport under static conditions was simulated by using two ice thickness models in combination with an indoor freezing experiment under different segregation coefficient conditions for nutrients. A real-time prediction model for nutrient and pollutant concentrations in ice-covered lakes was established to explore the impact of the ice-on period in eutrophic shallow lakes. The results demonstrated that the empirical degree-day model and the high-resolution thermodynamic snow and sea-ice model (HIGHTSI) could both be used to simulate lake ice thickness. The empirical degree-day model performed better at predicting the maximum ice thickness (measured thickness 0.22-0.55 m; simulated thickness 0.48 m), whereas the HIGHTSI model was more accurate when estimating the mean thickness (5-6% error). When simulating ice growth, the HIGHTSI model considered more meteorological factors impacting ice cover ablation; hence, it performed better during the ablation stage relative to the empirical degree-day model. Two non-dynamic nutrient transport models were developed by combining the segregation coefficient model and the ice thickness prediction model. The HIGHTSI nutrient transport model can be used to predict real-time changes in nutrient concentrations under ice cover, and the degree-day model can be used to predict changes in the lake water ecosystem. Peer reviewed
|
|---|