The search result changed since you submitted your search request. Documents might be displayed in a different sort order.
  • search hit 1 of 2
Back to Result List

RainNet v1.0

  • In this study, we present RainNet, a deep convolutional neural network for radar-based precipitation nowcasting. Its design was inspired by the U-Net and SegNet families of deep learning models, which were originally designed for binary segmentation tasks. RainNet was trained to predict continuous precipitation intensities at a lead time of 5min, using several years of quality-controlled weather radar composites provided by the German Weather Service (DWD). That data set covers Germany with a spatial domain of 900km × 900km and has a resolution of 1km in space and 5min in time. Independent verification experiments were carried out on 11 summer precipitation events from 2016 to 2017. In order to achieve a lead time of 1h, a recursive approach was implemented by using RainNet predictions at 5min lead times as model inputs for longer lead times. In the verification experiments, trivial Eulerian persistence and a conventional model based on optical flow served as benchmarks. The latter is available in the rainymotion library and hadIn this study, we present RainNet, a deep convolutional neural network for radar-based precipitation nowcasting. Its design was inspired by the U-Net and SegNet families of deep learning models, which were originally designed for binary segmentation tasks. RainNet was trained to predict continuous precipitation intensities at a lead time of 5min, using several years of quality-controlled weather radar composites provided by the German Weather Service (DWD). That data set covers Germany with a spatial domain of 900km × 900km and has a resolution of 1km in space and 5min in time. Independent verification experiments were carried out on 11 summer precipitation events from 2016 to 2017. In order to achieve a lead time of 1h, a recursive approach was implemented by using RainNet predictions at 5min lead times as model inputs for longer lead times. In the verification experiments, trivial Eulerian persistence and a conventional model based on optical flow served as benchmarks. The latter is available in the rainymotion library and had previously been shown to outperform DWD's operational nowcasting model for the same set of verification events. RainNet significantly outperforms the benchmark models at all lead times up to 60min for the routine verification metrics mean absolute error (MAE) and the critical success index (CSI) at intensity thresholds of 0.125, 1, and 5mm h⁻¹. However, rainymotion turned out to be superior in predicting the exceedance of higher intensity thresholds (here 10 and 15mm h⁻¹). The limited ability of RainNet to predict heavy rainfall intensities is an undesirable property which we attribute to a high level of spatial smoothing introduced by the model. At a lead time of 5min, an analysis of power spectral density confirmed a significant loss of spectral power at length scales of 16km and below. Obviously, RainNet had learned an optimal level of smoothing to produce a nowcast at 5min lead time. In that sense, the loss of spectral power at small scales is informative, too, as it reflects the limits of predictability as a function of spatial scale. Beyond the lead time of 5min, however, the increasing level of smoothing is a mere artifact – an analogue to numerical diffusion – that is not a property of RainNet itself but of its recursive application. In the context of early warning, the smoothing is particularly unfavorable since pronounced features of intense precipitation tend to get lost over longer lead times. Hence, we propose several options to address this issue in prospective research, including an adjustment of the loss function for model training, model training for longer lead times, and the prediction of threshold exceedance in terms of a binary segmentation task. Furthermore, we suggest additional input data that could help to better identify situations with imminent precipitation dynamics. The model code, pretrained weights, and training data are provided in open repositories as an input for such future studies.show moreshow less

Export metadata

Additional Services

Search Google Scholar Statistics
Metadaten
Author details:Georgy AyzelORCiD, Tobias SchefferORCiD, Maik HeistermannORCiDGND
DOI:https://doi.org/10.5194/gmd-13-2631-2020
ISSN:1991-959X
ISSN:1991-9603
Title of parent work (English):Geoscientific Model Development
Subtitle (English):a convolutional neural network for radar-based precipitation nowcasting
Publisher:Copernicus Publ.
Place of publishing:Göttingen
Publication type:Article
Language:English
Date of first publication:2020/06/11
Publication year:2020
Release date:2020/07/08
Tag:models; skill; weather
Volume:13
Issue:6
Number of pages:14
First page:2631
Last Page:2644
Funding institution:Universität Potsdam
Funding number:PA 2020_054
Organizational units:Mathematisch-Naturwissenschaftliche Fakultät / Institut für Geowissenschaften
DDC classification:5 Naturwissenschaften und Mathematik / 55 Geowissenschaften, Geologie / 550 Geowissenschaften
Peer review:Referiert
Grantor:Publikationsfonds der Universität Potsdam
Publishing method:Open Access
Institution name at the time of the publication:Mathematisch-Naturwissenschaftliche Fakultät / Institut für Erd- und Umweltwissenschaften
License (German):License LogoCC-BY - Namensnennung 4.0 International
External remark:Zweitveröffentlichung in der Schriftenreihe Postprints der Universität Potsdam : Mathematisch-Naturwissenschaftliche Reihe ; 964
Accept ✔
This website uses technically necessary session cookies. By continuing to use the website, you agree to this. You can find our privacy policy here.