The U.S. Air ForceWeather Agency’s mesoscale ensemble: scientific description and performance results

Abstract

A B S T R A C TThis work evaluates several techniques to account for mesoscale initial-condition (IC) and model uncertainty in a short-range ensemble prediction system based on the Weather Research and Forecast (WRF) model. A scientific description and verification of several candidate methods for implementation in the U.S. Air Force Weather Agency mesoscale ensemble is presented. Model perturbation methods tested include multiple parametrization suites, landsurface property perturbations, perturbations to parameters within physics schemes and stochastic 'backscatter' streamfunction perturbations. IC perturbations considered include perturbed observations in 10 independent WRF-3DVar cycles and the ensemble-transform Kalman filter (ETKF). A hybrid of ETKF (for IC perturbations) and WRF-3DVar (to update the ensemble mean) is also tested. Results show that all of the model and IC perturbation methods examined are more skilful than direct dynamical downscaling of the global ensemble. IC perturbations are most helpful during the first 12 h of the forecasts. Physical parametrization diversity appears critical for boundary-layer forecasts. In an effort to reduce system complexity by reducing the number of suites of physical parametrizations, a smaller set of parametrization suites was combined with perturbed parameters and stochastic backscatter, resulting in the most skilful and statistically consistent ensemble predictions.

Selection of polynomial chaos bases via Bayesian model uncertainty methods with applications to sparse approximation of PDEs with stochastic inputs

“…This is because they indicate the contribution of each individual model in the mixture for the representation of the real system output. Posterior estimates for the weight functions t k p¨qu can be computed asˆ 15) along with the associated standard errors according to the Markov chain CLT [40]. Moreover, the weight functions allow the determination of a reasonable input space partition tX k u K k"1 where each sub-region X k " tx P X | k pxq " maxp 1 pxq, ..., K pxqqu includes the input values that model S pkq is more preferable to be used than the rest.…”

“…Moreover, higher grid spacing does not necessarily lead to more accurate simulations because WRF is sensitive to other physical parametrizations which is uncertain how they are affected by the grid spacing. Combination of physics variability is expected to result better predictions in climate models [15]; hence interest lies in combining suitably these computer models in order to integrate the associated physics and fidelity variations.…”

“…Therefore, in such cases, the resulting calibrated computer model mixture is able to produce more accurate simulations than the single ones. This covers a large range of important real world applications [15], such as the WRF one analyzed here.…”

Exchange Processes in the Atmospheric Boundary Layer Over Mountainous Terrain

“…One way of accounting for model errors is to introduce stochastic perturbations in some aspects of the parameterization schemes adopted by the ensemble members (multi-physics ensembles [239] offer another approach to the problem).…”

Some issues in uncertainty quantification and parameter tuning: a case study of convective parameterization scheme in the WRF regional climate model

“…Even if the calibration produces realistic regional means, important spatial variability may not be reproduced if observed spatial patterns from high-resolution measurements are not utilized in the global tuning. Hacker et al (2011) evaluated the impacts of initial condition and model parameterization uncertainties on a WRF-based ensemble prediction system and found that different combinations of parameterization schemes associated with perturbed parameters could generate the most skillful ensemble prediction.…”