Controlling Weather Field Synthesis Using Variational Autoencoders

TL;DR

This paper explores how variational autoencoders can be used to control the synthesis of weather fields, enabling generation of more or less extreme climate scenarios based on learned data distributions.

Contribution

It introduces a novel approach using variational autoencoders to manipulate weather data synthesis towards desired climate extremities.

Findings

Mapping weather data to known distributions enables control over scenario extremity.

The method effectively generates weather fields with varying degrees of extremity.

Results demonstrate improved bias control in climate scenario generation.

Abstract

One of the consequences of climate change is anobserved increase in the frequency of extreme cli-mate events. That poses a challenge for weatherforecast and generation algorithms, which learnfrom historical data but should embed an often un-certain bias to create correct scenarios. This paperinvestigates how mapping climate data to a knowndistribution using variational autoencoders mighthelp explore such biases and control the synthesisof weather fields towards more extreme climatescenarios. We experimented using a monsoon-affected precipitation dataset from southwest In-dia, which should give a roughly stable pattern ofrainy days and ease our investigation. We reportcompelling results showing that mapping complexweather data to a known distribution implementsan efficient control for weather field synthesis to-wards more (or less) extreme scenarios.