Publication Details
Radar-Based Volumetric Precipitation Nowcasting: A 3D Convolutional Neural Network with U-Net Architecture
Rozinajová Věra, Doc., Ph.D. (DIFS)
Bou Ezzeddine Anna, Doc., Ph.D. (DIFS)
precipitation nowcasting, radar imaging, U-Net
In recent years like in many other domains deep learning models have found their place in the domain of precipitation nowcasting. Many of these models are based on the U-Net architecture, which was originally developed for biomedical segmentation, but is also useful for the generation of short-term forecasts and therefore applicable in the weather nowcasting domain. The existing U-Net-based models use sequential radar data mapped into a 2-dimensional Cartesian grid as input and output. We propose to incorporate a third - vertical - dimension to better predict precipitation phenomena such as convective rainfall and present our results here. We compare the nowcasting performance of two comparable U-Net models trained on two-dimensional and three-dimensional radar observation data. We show that using volumetric data results in a small, but significant reduction in prediction error.
@inproceedings{BUT179604,
author="Peter {Pavlík} and Věra {Rozinajová} and Anna {Bou Ezzeddine}",
title="Radar-Based Volumetric Precipitation Nowcasting: A 3D Convolutional Neural Network with U-Net Architecture",
booktitle="Proceedings of the Second Workshop on Complex Data Challenges in Earth Observation (CDCEO 2022)",
year="2022",
journal="CEUR Workshop Proceedings",
volume="3207",
number="2022",
pages="65--72",
publisher="CEUR-WS.org",
address="Vienna",
issn="1613-0073",
url="http://ceur-ws.org/Vol-3207/paper10.pdf"
}