From Nowcasting to Forecasting: Adapting a Reanalysis-Trained
Abstract
Accurate cloud-cover forecasts are important for temperature prediction, radiation forecasting, and solar-power operations. Short-range forecasting methods can preserve observed cloud placement during the first forecast hours, but their skill decreases when cloud fields evolve through formation, dissipation and deformation. Longer lead times require accounting for atmospheric evolution, but operational numerical weather prediction (NWP) forecasts may not accurately represent the satellite-observed cloud state at initialization. We develop CloudCast v2, a machine-learning model for 12-hour cloud-cover forecasting from observation-based initial conditions. The model is first trained on the Copernicus European Regional Reanalysis (Ridal2024) to learn cloud-evolution dynamics, and is then adapted to satellite-derived cloud fields using conditional flow matching (Lipman2023), a generative method that transforms noise into cloud-cover forecasts conditioned on the observed initial cloud fields and NWP inputs. CloudCast v2 reduces mean absolute error by 10% relative to its predecessor, CloudCast v1 (Partio2025), over the 1-12 h range. It also overtakes CloudCast v1 in fractions skill score, a neighborhood-based measure of spatial agreement, after approximately 3-6 h, depending on the cloudiness category. These results show that observation-initialized machine-learning forecasts can extend beyond the usual 1-3-hour nowcasting range while retaining spatial detail from satellite cloud fields.
Related
- Precipitation nowcasting model based on multi-source data fusion and efficient additive attention
- Diffusion-Based Refinement for Kilometer-Scale Probabilistic Precipitation Nowcasting
- Wet-Biased Data Pipelines Mislead the Training and Evaluation of Generative Precipitation Nowcasting
- Meteorology-driven Causal Nowcasting of Fugitive Landfill Emissions from Measured Coupling Timescales
- GenONet: A Generative operator Network for High-Resolution Precipitation Nowcasting
BibTeX
@misc{partio2026from,
title = {{From Nowcasting to Forecasting: Adapting a Reanalysis-Trained}},
author = {Mikko Partio and Leila Hieta and Ossi Laine},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.03763},
eprint = {2609.03763},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.03763},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - From Nowcasting to Forecasting: Adapting a Reanalysis-Trained
AU - Mikko Partio
AU - Leila Hieta
AU - Ossi Laine
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.03763
UR - https://arxiv.org/abs/2609.03763
AB - Accurate cloud-cover forecasts are important for temperature prediction, radiation forecasting, and solar-power operations. Short-range forecasting methods can preserve observed cloud placement during the first forecast hours, but their skill decreases when cloud fields evolve through formation, dissipation and deformation. Longer lead times require accounting for atmospheric evolution, but operational numerical weather prediction (NWP) forecasts may not accurately represent the satellite-observed cloud state at initialization. We develop CloudCast v2, a machine-learning model for 12-hour cloud-cover forecasting from observation-based initial conditions. The model is first trained on the Copernicus European Regional Reanalysis (Ridal2024) to learn cloud-evolution dynamics, and is then adapted to satellite-derived cloud fields using conditional flow matching (Lipman2023), a generative method that transforms noise into cloud-cover forecasts conditioned on the observed initial cloud fields and NWP inputs. CloudCast v2 reduces mean absolute error by 10% relative to its predecessor, CloudCast v1 (Partio2025), over the 1-12 h range. It also overtakes CloudCast v1 in fractions skill score, a neighborhood-based measure of spatial agreement, after approximately 3-6 h, depending on the cloudiness category. These results show that observation-initialized machine-learning forecasts can extend beyond the usual 1-3-hour nowcasting range while retaining spatial detail from satellite cloud fields.
ER -