Improving regional weather forecasts with neural interpolation

TL;DR

This paper introduces a neural interpolation method using CNNs and residual networks to enhance boundary data in regional weather models, aiming to better capture multi-scale atmospheric dynamics.

Contribution

It presents a novel neural interpolation approach that combines super-resolution techniques with atmospheric flow modeling for improved weather forecast boundary data.

Findings

Effective boundary data enhancement demonstrated

Potential for generalizing to complex weather models

Improved multi-scale dynamic mapping

Abstract

In this paper we design a neural interpolation operator to improve the boundary data for regional weather models, which is a challenging problem as we are required to map multi-scale dynamics between grid resolutions. In particular, we expose a methodology for approaching the problem through the study of a simplified model, with a view to generalise the results in this work to the dynamical core of regional weather models. Our approach will exploit a combination of techniques from image super-resolution with convolutional neural networks (CNNs) and residual networks, in addition to building the flow of atmospheric dynamics into the neural network