3

This study examines how the frequency of data assimilation affects the accuracy of near-term ecological forecasts, specifically for water temperature in a eutrophic reservoir. Using the FLARE forecasting system, researchers tested daily, weekly, fortnightly, and monthly data assimilation to predict water temperature 1 to 35 days ahead. They found that daily assimilation produced the most accurate short-term forecasts (1–7 days), while weekly assimilation performed best for longer-term predictions (8–35 days). Seasonal and depth variations influenced forecast accuracy, with higher-frequency assimilation being especially important during summer stratification. The findings suggest that even lower-frequency data (e.g., weekly) can yield skillful forecasts, broadening forecasting applications beyond ecosystems with high-frequency sensors.

Lake thermal dynamics have been altered around the world as a result of climate change, necessitating a predictive understanding of how climate will continue to alter lakes in the future as well as the associated uncertainty in these predictions. We found that the dominant source of uncertainty varied among the thermal metrics, as thermal metrics associated with the surface waters (surface water temperature, total ice duration) were driven primarily by climate model selection uncertainty, while metrics associated with deeper depths (bottom water temperature, stratification duration) were dominated by lake model selection uncertainty.

We scaled a near-term, iterative water temperature forecasting system to six conterminous NEON lakes. We generated forecasts using a hydrodynamic model. Forecasts were more accurate than a null model. Lake characteristics interact with weather to control the predictability of thermal structure.