Deep Learning-Enhanced Background Error Covariance Estimation for Massive-Ensemble Kalman Filter in Sea Surface Temperature Forecasting

Deep Learning-Based Statistical Downscaling of Sea Surface Temperature Using a Residual Corrective Neural Network

This study proposes a novel deep learning framework that uses a U-Net to generate an initial high-resolution SST estimate, which is subsequently refined using a residual corrective approach, and progressively refines initial U-Net predictions by incorporating dynamically scaled residuals at each step, enabling accurate...

An Ensemble Kalman Smoother for Online Paleoclimate Data Assimilation

An integrated hybrid ensemble Kalman smoother (IHEnKS) is proposed to optimally utilize proxy data from the past to the future for paleoclimate data assimilation (PDA). As an extension of the integrated hybrid ensemble Kalman filter, IHEnKS assimilates future proxies through cross‐time error covariances, which are es...

Deep Learning-Augmented Soil Moisture Data Assimilation under Temporally Sparse and Irregular Observations: A Case Study over the Tibetan Plateau

Ensemble Forecast Updates without Model Re-integration Based on Ultra-rapid Data Assimilation: Idealized Experiments with a Heavy Rainfall Case

The results show that the threat score for hourly precipitation is generally improved in the forecasts updated by URDA relative to the pre-existing baseline forecasts, through the ensemble-based error covariance.

Drift Field Net: Learning Ocean Lagrangian advection fields from in-situ and satellite observations

DFN is a deep neural network that predicts ocean surface flow fields from operational satellite observations that reduces the mean positioning error after a 7-day forecast compared with the operational model and demonstrates the potential of deep learning for ocean surface flow prediction.

DLESyM-Ocean: A Deep Learning Probabilistic Global Model for Simulating Present-Day Upper Ocean and Sea Ice

DLESyM-Ocean is stable when autoregressively run for multi-year simulations and produces a climatology and variability with minimal bias compared with reanalysis, and the computational efficiency of DLESyM-Ocean makes it a promising tool for subseasonal to seasonal forecasting.