Improvements to the post-processing of weather forecasts using machine learning and feature selection

TL;DR

This paper enhances weather forecast post-processing by applying machine learning with feature selection, resulting in models that outperform traditional methods in many locations and lead times.

Contribution

It introduces a LightGBM-based post-processing approach with feature selection that improves forecast accuracy over existing models and raw data.

Findings

LightGBM models achieved lower RMSE than neural network baselines.

Models generally outperformed raw MSM forecasts and MSM Guidance.

Event-weighted training improved precipitation forecast performance at higher rainfall thresholds.

Abstract

This study aims to develop and improve machine learning-based post-processing models for precipitation, temperature, and wind speed predictions using the Mesoscale Model (MSM) dataset provided by the Japan Meteorological Agency (JMA) for 18 locations across Japan, including plains, mountainous regions, and islands. By incorporating meteorological variables from grid points surrounding the target locations as input features and applying feature selection based on correlation analysis, we found that, in our experimental setting, the LightGBM-based models achieved lower RMSE than the specific neural-network baselines tested in this study, including a reproduced CNN baseline, and also generally achieved lower RMSE than both the raw MSM forecasts and the JMA post-processing product, MSM Guidance (MSMG), across many locations and forecast lead times. Because precipitation has a highly skewed distribution with many zero cases, we additionally examined Tweedie-based loss functions and event-weighted training strategies for precipitation forecasting. These improved event-oriented performance relative to the original LightGBM model, especially at higher rainfall thresholds, although the gains were site dependent and overall performance remained slightly below MSMG.