Machine learning improves weather front detection with reanalysis data
Detecting Weather Fronts from Satellite Observations: What Does Reanalysis Add?
Machine LearningComputer Vision and Pattern Recognition
Summary
Weather fronts are areas where different air masses meet and often bring storms and rain. The authors tested a machine learning model that identifies these fronts using only satellite images and then with extra weather data called reanalysis. They found that adding reanalysis data greatly improved the model’s accuracy. Some weather features, like wind near the surface, are harder to predict from satellite images alone. This work helps understand how combining data sources can better detect weather fronts.
What this means in practice
- •For meteorological agencies: Improve operational weather front detection by integrating reanalysis data with satellite imagery using machine learning models.
- •For renewable energy operators: Enhance weather risk assessments for wind farms by using more accurate frontal detection informed by combined satellite and reanalysis data.
Authors
Ioannis Dravilas, Miruna Stoicescu, Roope Tervo, Carlos Horn, Oriol Hinojo Comellas, Sina Montazeri, Andrea Meraner, Ana Lucic
Abstract
Weather fronts are transition regions between air masses with contrasting thermodynamic properties and are associated with mid-latitude precipitation and severe weather. We investigate machine learning (ML) methods for detecting weather fronts and quantify the contribution of reanalysis data to performance. We compare two configurations of the same ML model for pixel-wise frontal classification over the North Atlantic and Europe: (1) using only geostationary infrared satellite observations, and (2) additionally using 28 ERA5 reanalysis fields. The satellite-only model achieves a neighbourhood critical success index at a 150 km tolerance (CSI@150) of $0.402$, which lies within the range of agreement between analyses from different meteorological services. The satellite$+$ERA5 model achieves $0.575$, an improvement of $0.173$. Our analysis shows that low-level winds are the least predictable from the satellite representation ($R^2 = 0.23$, compared with $0.88$ for temperature), while removing winds or moisture causes the largest decreases in detection performance.