STIPP: Space-time in situ postprocessing over the French Alps using proper scoring rules

TL;DR

STIPP is a novel machine learning approach that produces joint spatio-temporal weather forecasts, improving accuracy and correlation structures over traditional methods by leveraging proper scoring rules.

Contribution

It introduces a joint spatio-temporal forecasting model that enhances weather prediction accuracy and correlation structures using a multivariate proper scoring rule.

Findings

Improved accuracy for temperature, wind, humidity, and precipitation forecasts.

Maintains spatio-temporal correlation structures better than baseline methods.

Generates hourly ensemble predictions from six-hourly forecasts.

Abstract

We propose Space-time in situ postprocessing (STIPP), a machine learning model that generates spatio-temporally consistent weather forecasts for a network of station locations. Gridded forecasts from classical numerical weather prediction or data-driven models often lack the necessary precision due to unresolved local effects. Typical statistical postprocessing methods correct these biases, but often degrade spatio-temporal correlation structures in doing so. Recent works based on generative modeling successfully improve spatial correlation structures but have to forecast every lead time independently. In contrast, STIPP makes joint spatio-temporal forecasts which have increased accuracy for surface temperature, wind, relative humidity and precipitation when compared to baseline methods. It makes hourly ensemble predictions given only a six-hourly deterministic forecast, blending the boundaries of postprocessing and temporal interpolation. By leveraging a multivariate proper scoring rule for training, STIPP contributes to ongoing work data-driven atmospheric models supervised only with distribution marginals.