Guardado en:
Detalles Bibliográficos
Autores principales: Shiraishi, Kenta, Muto, Yuka, Okazaki, Atsushi, Kotsuki, Shunji
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2507.17798
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916021655306240
author Shiraishi, Kenta
Muto, Yuka
Okazaki, Atsushi
Kotsuki, Shunji
author_facet Shiraishi, Kenta
Muto, Yuka
Okazaki, Atsushi
Kotsuki, Shunji
contents High-resolution (HR) precipitation prediction is essential for reducing damage from stationary and localized heavy rainfall; however, HR precipitation forecasts using process-driven numerical weather prediction models remains challenging. This study proposes using Wasserstein Generative Adversarial Network (WGAN) to perform precipitation downscaling with an optimal transport cost. In contrast to a conventional neural network trained with mean squared error, the WGAN generated visually realistic precipitation fields with fine-scale structures even though the WGAN exhibited slightly lower performance on conventional evaluation metrics. The learned critic of WGAN correlated well with human perceptual realism. Case-based analysis revealed that large discrepancies in critic scores can help identify both unrealistic WGAN outputs and potential artifacts in the reference data. These findings suggest that the WGAN framework not only improves perceptual realism in precipitation downscaling but also offers a new perspective for evaluating and quality-controlling precipitation datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2507_17798
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Wasserstein GAN-Based Precipitation Downscaling with Optimal Transport for Enhancing Perceptual Realism
Shiraishi, Kenta
Muto, Yuka
Okazaki, Atsushi
Kotsuki, Shunji
Machine Learning
High-resolution (HR) precipitation prediction is essential for reducing damage from stationary and localized heavy rainfall; however, HR precipitation forecasts using process-driven numerical weather prediction models remains challenging. This study proposes using Wasserstein Generative Adversarial Network (WGAN) to perform precipitation downscaling with an optimal transport cost. In contrast to a conventional neural network trained with mean squared error, the WGAN generated visually realistic precipitation fields with fine-scale structures even though the WGAN exhibited slightly lower performance on conventional evaluation metrics. The learned critic of WGAN correlated well with human perceptual realism. Case-based analysis revealed that large discrepancies in critic scores can help identify both unrealistic WGAN outputs and potential artifacts in the reference data. These findings suggest that the WGAN framework not only improves perceptual realism in precipitation downscaling but also offers a new perspective for evaluating and quality-controlling precipitation datasets.
title Wasserstein GAN-Based Precipitation Downscaling with Optimal Transport for Enhancing Perceptual Realism
topic Machine Learning
url https://arxiv.org/abs/2507.17798