Enhancing Deterministic Freezing Level Predictions in the Northern Sierra Nevada Through Deep Neural Networks

Abstract

Accurate prediction of the freezing level is essential for hydrometeorological forecasting systems, with direct implications for runoff generation and reservoir management. In this study, we develop a deep learning based postprocessing framework using the Unet convolutional neural network architecture to refine the FZL forecasts from the West Weather Research and Forecasting West WRF model. The proposed framework leverages reforecast data from West WRF and FZL estimates from the California Nevada River Forecast Center to develop Unet models over the northern Sierra Nevada watersheds, such as the hydrologically critical Yuba Feather watershed. We introduce two Unet model variants, Unet_log and Unet_GMM, that employ specialized loss functions beyond the standard benchmarks to enhance forecast skill. Unet_log utilizes the cosine of Error, and Unet_MM uses Gaussian Mixture Model loss functions, to enhance FZL forecasts. Results show that Unet based postprocessing reduces centered root mean squared errors by up to 20% and increases forecast observation correlation by about 10% compared to raw WestWRF. Evaluation using the continuous ranked probability score for Unet_GMM further demonstrates consistent improvements across lead times. While performance fluctuates with forecast horizon, storm variability, and diurnal forcing, Unet_GMM and Unet_log consistently outperform the baseline. The models capture the spatiotemporal variability of the FZL across different elevations, mitigating biases from the West WRF model. This novel deep learning based postprocessing approach demonstrates a promising pathway for integrating machine learning into hydrometeorological forecasting and decision support within the Forecast Informed Reservoir Operations framework.