Precipitation Nowcasting Using Physics Informed Discriminator Generative Models

TL;DR

This paper introduces a physics-informed GAN model for short-term precipitation forecasting, integrating meteorological physics into deep learning to improve accuracy over existing methods.

Contribution

It proposes a novel physics-informed discriminator GAN architecture with VQ-GAN and Transformer components for enhanced precipitation nowcasting.

Findings

Outperforms existing deep generative models in nowcasting accuracy

Effectively incorporates physics-based supervision into adversarial training

Demonstrates improved prediction of extreme weather events

Abstract

Nowcasting leverages real-time atmospheric conditions to forecast weather over short periods. State-of-the-art models, including PySTEPS, encounter difficulties in accurately forecasting extreme weather events because of their unpredictable distribution patterns. In this study, we design a physics-informed neural network to perform precipitation nowcasting using the precipitation and meteorological data from the Royal Netherlands Meteorological Institute (KNMI). This model draws inspiration from the novel Physics-Informed Discriminator GAN (PID-GAN) formulation, directly integrating physics-based supervision within the adversarial learning framework. The proposed model adopts a GAN structure, featuring a Vector Quantization Generative Adversarial Network (VQ-GAN) and a Transformer as the generator, with a temporal discriminator serving as the discriminator. Our findings demonstrate that the PID-GAN model outperforms numerical and SOTA deep generative models in terms of precipitation nowcasting downstream metrics.