Reduction of rain-induced errors for wind speed estimation on SAR observations using convolutional neural networks

TL;DR

This paper presents a convolutional neural network approach that significantly reduces rain-induced errors in SAR-based wind speed estimates, improving accuracy especially during rainfall events.

Contribution

It introduces a deep learning model trained on a large SAR dataset to correct rain-related errors in wind speed estimation, outperforming traditional methods.

Findings

Root mean square error reduced by 27% under rain >1 mm/h

Error reduction of 45% during heavy rainfall (>3 mm/h)

Deep learning effectively corrects rain-induced errors in SAR wind estimates

Abstract

Synthetic Aperture Radar is known to be able to provide high-resolution estimates of surface wind speed. These estimates usually rely on a Geophysical Model Function (GMF) that has difficulties accounting for non-wind processes such as rain events. Convolutional neural network, on the other hand, have the capacity to use contextual information and have demonstrated their ability to delimit rainfall areas. By carefully building a large dataset of SAR observations from the Copernicus Sentinel-1 mission, collocated with both GMF and atmospheric model wind speeds as well as rainfall estimates, we were able to train a wind speed estimator with reduced errors under rain. Collocations with in-situ wind speed measurements from buoys show a root mean square error that is reduced by 27% (resp. 45%) under rainfall estimated at more than 1 mm/h (resp. 3 mm/h). These results demonstrate the capacity of deep learning models to correct rain-related errors in SAR products.