Performance Comparison between Deep Learning and Optical Flow-Based Techniques for Nowcast Precipitation from Radar Images

In this article, a nowcasting technique for meteorological radar images based on a generative neural network is presented. This technique’s performance is compared with state-of-the-art optical flow procedures. Both methods have been validated using a public domain data set of radar images, covering an area of about 104 km2 over Japan, and a period of five years with a sampling frequency of five minutes. The performance of the neural network, trained with three of the five years of data, forecasts with a time horizon of up to one hour, evaluated over one year of the data, proved to be significantly better than those obtained with the techniques currently in use.

[1]  T. Scheytt,et al.  Using radar-based quantitative precipitation data with coupled soil- and groundwater balance models for stream flow simulation in a karst area , 2020 .

[2]  I. Zawadzki,et al.  Scale-Dependence of the Predictability of Precipitation from Continental Radar Images. Part I: Description of the Methodology , 2002 .

[3]  Henk A. Dijkstra,et al.  Probabilistic Forecasting of El Niño Using Neural Network Models , 2020, Geophysical Research Letters.

[4]  Sunmin Kim,et al.  Improvement of rainfall and flood forecasts by blending ensemble NWP rainfall with radar prediction considering orographic rainfall , 2015 .

[5]  Daniele Nerini,et al.  Pysteps: an open-source Python library for probabilistic precipitation nowcasting (v1.0) , 2019 .

[6]  Tobias Scheffer,et al.  RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting , 2020, Geoscientific Model Development.

[7]  A. Seed.  A Dynamic and Spatial Scaling Approach to Advection Forecasting , 2001 .

[8]  D. Bae,et al.  Correcting mean areal precipitation forecasts to improve urban flooding predictions by using long short-term memory network , 2020 .

[9]  Maik Heistermann,et al.  Optical flow models as an open benchmark for radar-based precipitation nowcasting (rainymotion v0.1) , 2019, Geoscientific Model Development.

[10]  Oleg R. Nikitin,et al.  All convolutional neural networks for radar-based precipitation nowcasting , 2019, Procedia Computer Science.

[11]  Sa-Kwang Song,et al.  Computer Vision in Precipitation Nowcasting: Applying Image Quality Assessment Metrics for Training Deep Neural Networks , 2019, Atmosphere.

[12]  Stéphane Laroche,et al.  Retrievals of Horizontal Winds from Single-Doppler Clear-Air Data by Methods of Cross Correlation and Variational Analysis , 1995 .

[13]  V. Chandrasekar,et al.  An Enhanced Optical Flow Technique for Radar Nowcasting of Precipitation and Winds , 2017 .

[14]  C. Collier.  On the propagation of uncertainty in weather radar estimates of rainfall through hydrological models , 2009 .

[15]  Christoph Schär,et al.  Probabilistic Flood Forecasting with a Limited-Area Ensemble Prediction System: Selected Case Studies , 2007 .

[16]  V. Chandrasekar,et al.  Fully Spectral Method for Radar-Based Precipitation Nowcasting , 2019, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[17]  Wang-chun Woo,et al.  Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting , 2017 .

[18]  V. Chandrasekar,et al.  The CASA Nowcasting System , 2011 .

[19]  D. Sempere-Torres,et al.  A semi-distributed hydrological model using rainfall estimates by radar. Application to Mediterranean basins , 2000 .

[20]  Todini,et al.  Coupling meteorological and hydrological models for flood forecasting , 2005 .