Computing the ensemble spread from deterministic weather predictions using conditional generative adversarial networks

TL;DR

This paper introduces deep learning models to accurately estimate ensemble weather forecast spread from a single deterministic forecast, reducing computational costs and bypassing traditional ensemble generation.

Contribution

The study adapts and applies conditional generative adversarial networks to predict ensemble spread from deterministic forecasts, a novel approach in weather prediction post-processing.

Findings

Models accurately predict ensemble spread from control forecasts.

Deep learning reduces the need for multiple ensemble simulations.

The approach is validated on operational weather forecast data.

Abstract

Ensemble prediction systems are an invaluable tool for weather forecasting. Practically, ensemble predictions are obtained by running several perturbations of the deterministic control forecast. However, ensemble prediction is associated with a high computational cost and often involves statistical post-processing steps to improve its quality. Here we propose to use deep-learning-based algorithms to learn the statistical properties of an ensemble prediction system, the ensemble spread, given only the deterministic control forecast. Thus, once trained, the costly ensemble prediction system will not be needed anymore to obtain future ensemble forecasts, and the statistical properties of the ensemble can be derived from a single deterministic forecast. We adapt the classical pix2pix architecture to a three-dimensional model and also experiment with a shared latent space encoder-decoder model, and train them against several years of operational (ensemble) weather forecasts for the 500 hPa geopotential height. The results demonstrate that the trained models indeed allow obtaining a highly accurate ensemble spread from the control forecast only.