Methodological Changes to the Attention ResUNet Hourly Precipitation Postprocessor
Abstract
This note is a technical companion to a previously published preprint describing an Attention Residual U-Net that postprocesses deterministic forecasts from The Weather Company's Global and Regional Atmospheric Forecast (GRAF) model into probabilistic hourly precipitation forecasts. It documents what has changed in that method since publication. Feature-wise Linear Modulation conditioning on calendar season and forecast lead time is used to produce a single trained model for each season, replacing 192 separately trained per-month, per-lead checkpoints. Lead time is extended from 48 to 72 h. Two new input channels are used, per-pixel local solar hour and a static, monthly-varying precipitation climatology. During verification, the climatological reference against which the Brier Skill Score is computed now has an added diurnal dimension, on top of the monthly resolution it already had. Brier Skill Score and reliability are compared between the new vs. the previous training. Forecasts generated with the new training show a modest, consistent improvement of the current training over the original.
Related
- Precipitation nowcasting model based on multi-source data fusion and efficient additive attention
- Global Changes in Precipitation Concentration Through Consecutive Event Organization
- All that changes is not shift: methodological choices influence niche shift detection in freshwater invasive species
- RainFlow: Flow Matching for Infrared-Only Quantitative Precipitation Estimation
- Tail Skill and Ensemble-Configuration Sensitivity in Six-Hour Precipitation Postprocessing
BibTeX
@misc{hamill2026methodological,
title = {{Methodological Changes to the Attention ResUNet Hourly Precipitation Postprocessor}},
author = {Thomas M. Hamill},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.38609},
eprint = {2609.38609},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.38609},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Methodological Changes to the Attention ResUNet Hourly Precipitation Postprocessor
AU - Thomas M. Hamill
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.38609
UR - https://arxiv.org/abs/2609.38609
AB - This note is a technical companion to a previously published preprint describing an Attention Residual U-Net that postprocesses deterministic forecasts from The Weather Company's Global and Regional Atmospheric Forecast (GRAF) model into probabilistic hourly precipitation forecasts. It documents what has changed in that method since publication. Feature-wise Linear Modulation conditioning on calendar season and forecast lead time is used to produce a single trained model for each season, replacing 192 separately trained per-month, per-lead checkpoints. Lead time is extended from 48 to 72 h. Two new input channels are used, per-pixel local solar hour and a static, monthly-varying precipitation climatology. During verification, the climatological reference against which the Brier Skill Score is computed now has an added diurnal dimension, on top of the monthly resolution it already had. Brier Skill Score and reliability are compared between the new vs. the previous training. Forecasts generated with the new training show a modest, consistent improvement of the current training over the original.
ER -