Meta-heuristic CRPS minimization for the calibration of short-range probabilistic forecasts

• Published in: Meteorology and atmospheric physics Vol. 128; no. 4; pp. 429 - 440
• Main Authors: Mohammadi, Seyedeh Atefeh, Rahmani, Morteza, Azadi, Majid
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.08.2016; Springer Nature B.V
• Subjects: Aquatic Pollution; Atmospheric models; Atmospheric Sciences; Earth and Environmental Science; Earth Sciences; Heuristic; Math. Appl. in Environmental Science; Mathematical models; Meteorology; Optimization; Original Paper; Probabilistic methods; Probability density functions; Probability distribution; Probability theory; Swarm intelligence; Terrestrial Pollution; Variance; Waste Water Technology; Water Management; Water Pollution Control; Weather forecasting; ISW, Iran; Probabilistic Forecast; Brier Score; Ensemble Member; Particle Swarm Optimization; Numerical Weather Prediction Model
• ISSN: 0177-7971; 1436-5065
• DOI: 10.1007/s00703-015-0426-9

This paper deals with the probabilistic short-range temperature forecasts over synoptic meteorological stations across Iran using non-homogeneous Gaussian regression (NGR). NGR creates a Gaussian forecast probability density function (PDF) from the ensemble output. The mean of the normal predictive PDF is a bias-corrected weighted average of the ensemble members and its variance is a linear function of the raw ensemble variance. The coefficients for the mean and variance are estimated by minimizing the continuous ranked probability score (CRPS) during a training period. CRPS is a scoring rule for distributional forecasts. In the paper of Gneiting et al. (Mon Weather Rev 133:1098–1118, 2005) , Broyden–Fletcher–Goldfarb–Shanno (BFGS) method is used to minimize the CRPS. Since BFGS is a conventional optimization method with its own limitations, we suggest using the particle swarm optimization (PSO), a robust meta-heuristic method, to minimize the CRPS. The ensemble prediction system used in this study consists of nine different configurations of the weather research and forecasting model for 48-h forecasts of temperature during autumn and winter 2011 and 2012. The probabilistic forecasts were evaluated using several common verification scores including Brier score, attribute diagram and rank histogram. Results show that both BFGS and PSO find the optimal solution and show the same evaluation scores, but PSO can do this with a feasible random first guess and much less computational complexity.