Abstract
Assessing compound weather risks requires forecasts representing dependence between variables. CLARA (Calibrated Advection-Routing Attention) learns joint Gaussian predictive distributions of five surface variables from station observations alone, without numerical weather prediction or reanalysis; the approximately 28,000-parameter model supports CPU training and prediction. Across six multi-year folds on 96 stations, its lead-mean energy score is 4.9% lower than that of a learned comparator with matched temporal inputs (4.7% with a similar parameter count) and 11-65% lower than those of statistical baselines. Holding marginal variances fixed, removing learned correlations worsens joint negative log-likelihood by 1.0-2.8 nats per station. A covariance-scale estimator, proved consistent under stated assumptions, improves short-lead calibration but over-corrects at long leads. Synthetic interventions show an attention-bias coefficient alone does not measure forecast influence. Retrained in ten regions on six continents, CLARA outperforms persistence in all 60 multi-year region-lead comparisons and a similarly sized learned model in 57 of 60.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Yi, C., & Seo, Y. A. (2026). Learning joint probabilistic weather forecasts from station observations alone. https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone
MLA 9
Yi, Chaeyeon, and Yun Am Seo. "Learning joint probabilistic weather forecasts from station observations alone." https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone.
Chicago (author–date)
Yi, Chaeyeon, and Yun Am Seo. 2026. "Learning joint probabilistic weather forecasts from station observations alone." https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone.
Harvard
Yi, C. and Seo, Y. A. (2026) 'Learning joint probabilistic weather forecasts from station observations alone', Available at: https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone.
Vancouver
Yi C, Seo YA. Learning joint probabilistic weather forecasts from station observations alone. https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone
IEEE
C. Yi, and Y. A. Seo, "Learning joint probabilistic weather forecasts from station observations alone," https://omanscience.com/en/articles/learning-joint-probabilistic-weather-forecasts-from-station-observations-alone.