A Deep Learning Method Integrating Multisource Data for ECMWF Forecasting Products Correction

Jingming Xia

doi:10.1109/LGRS.2023.3307717

A Deep Learning Method Integrating Multisource Data for ECMWF Forecasting Products Correction

2023-01-01

2

PRE

AI

DOI：10.1109/LGRS.2023.3307717

Original

Original request for help

Save

Abstract

En 中文

Accurate numerical weather prediction is essential for public and commercial meteorological services, but due to the complexity of the meteorological system and the uncertainty of observation data, there may be certain errors in the forecast results. Therefore, this letter proposes a deep learning-based numerical forecast correction network (NFC-Net) that integrates multisource heterogeneous data from FY-4A satellite, digital elevation model (DEM) and ERA5. NFC-Net leverages a spatial resolution alignment module and a spatiotemporal feature extraction module to extract and fuse features from diverse data sources and then applies UNet to correct European Centre for Medium-Range Weather Forecasts (ECMWF) forecast products. The model is evaluated using 2 m Temperature (2 m-T) and 10 m Wind Speed (10 m-WS) data, and compared against anomaly numerical-correction with observations (ANOs), Convlstm, and Fuse-CUnet, as well as ERA5 observations. Results demonstrate that NFC-Net outperforms other methods, reducing root mean square error (RMSE) of 2 m-T and 10 m-WS by 49.71% and 50.86%, respectively, compared to ECMWF forecast products. NFC-Net's success highlights the importance of effectively integrating and processing heterogeneous data.

Keywords:

Attention mechanism

deep learning

error correction

multisource data fusion

numerical weather prediction

A Deep Learning Method Integrating Multisource Data for ECMWF Forecasting Products Correction

Abstract

Journal

Organization