Preprints
https://doi.org/10.5194/gmd-2022-265
https://doi.org/10.5194/gmd-2022-265
Submitted as: model description paper
16 Jan 2023
Submitted as: model description paper | 16 Jan 2023
Status: this preprint is currently under review for the journal GMD.

GAN-argcPredNet v2.0: A Radar Echo Extrapolation Model based on Spatiotemporal Process Intensification

Kun Zheng1,, Qiya Tan1,, Huihua Ruan2, Jinbiao Zhang2, Cong Luo3, Siyu Tang3, Yunlei Yi4, Yugang Tian1, and Jianmei Cheng5 Kun Zheng et al.
  • 1School of Geography and Information Engineering, China University of Geosciences, Wuhan, 430074, China
  • 2Guangdong Meteorological Observation Data Center, Guangzhou, 510080, China
  • 3Guangdong Meteorological Observatory, Guangzhou, 510080, China
  • 4Wuhan Zhaotu Technology Co. Ltd., Wuhan, 430074, China
  • 5School of Environmental Studies, China University of Geosciences, Wuhan, 430074, China
  • These authors contributed equally to this work.

Abstract. Precipitation nowcasting has important implications for urban operation and flood prevention. Radar echo extrapolation is the common method in precipitation nowcasting. Using deep learning models to extrapolate radar echo data has great potential. The increase of lead time leads to a weaker correlation between real rainfall evolution and generated images. The evolution information is easily lost during extrapolation, which is reflected as echoes attenuation. Existing models, including Generative Adversarial Network (GAN)-based models, are all difficult to reduce loss and curb attenuation, which results in insufficient rainfall prediction accuracy. Aim to the problem, a Spatiotemporal Process Intensification Network (GAN-argcPredNet v2.0) based on GAN-argcPredNet v1.0 is designed. GAN-argcPredNet v2.0 reduces the loss by intensifying the influence of the previously input evolution information. A Spatiotemporal Information Changes Prediction (STIC-Prediction) network is designed as generator. With the intensification of echo feature sequence, the generator focuses on the spatiotemporal variation and generates more accurate images. Furthermore, discriminator is a Channel-Spatial Convolution (CS-Convolution) network. The discriminator intensifies the discrimination of echoes information by strengthening spatial information of single image. Identification results are fed back to the generator, which reduces the loss of important evolutionary information. The experiments are based on the radar dataset of South China. The results show that GAN-argcPredNet v2.0 performs better than other models. In heavy rainfall prediction, compared with baseline, the Probability of Detection (POD), the Critical Success Index (CSI), and the Heidke Skill Score (HSS) increase by 24.8 %, 22.2 % and 21.5 % respectively. The False Alarm Ratio (FAR) decreases by 3.76 %.

Kun Zheng et al.

Status: open (until 13 Mar 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Kun Zheng et al.

Kun Zheng et al.

Viewed

Total article views: 158 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
137 18 3 158 0 0
  • HTML: 137
  • PDF: 18
  • XML: 3
  • Total: 158
  • BibTeX: 0
  • EndNote: 0
Views and downloads (calculated since 16 Jan 2023)
Cumulative views and downloads (calculated since 16 Jan 2023)

Viewed (geographical distribution)

Total article views: 139 (including HTML, PDF, and XML) Thereof 139 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 28 Jan 2023
Download
Short summary
Radar echo extrapolation is the common method in precipitation nowcasting. Deep learning has potential in extrapolation. However, the existing models have low prediction accuracy for heavy rainfall. In this study, the prediction accuracy is improved by enhancing the influence of real rainfall information on the extrapolation, especially for heavy rainfall. The results show that our model has better performance, which is useful for urban operation and flood prevention.