Validation in Weather Forecasting
Numerical simulations are the core technique in forecasting the weather. The simulation calculates a weather forecast by use of an atmospheric model, which is implemented on a computer. The model itself can be partitioned into various complexity levels, and these can be associated with respective validation concepts. The proper design and implementation of the ‘dynamical core’ (i.e., partial differential equations and their numerical solver) is tested via comparison to idealized test cases. In a subsequent development step, ‘parameterizations’ are added, and then the simulation is considered a serious attempt to forecast the weather. The quality of the forecast is estimated by the retrospective comparison between simulation output and observed weather. In addition, a day-specific estimate of forecast uncertainty is derived via ‘ensemble forecasting’ on a routine basis.
KeywordsWeather forecasting Dynamical core Parameterizations Ensemble Uncertainty
We thank Felix Fundel and Ulrich Damrath (DWD) for delivering several verification plots. We also thank Nicole J. Saam and Claus Beisbart for inviting Susanne Theis to the Symposium ‘How to Build Trust in Computer Simulations’ (2015), funded by VolkswagenStiftung.
- Adrian, G. (2016). Qualitätsmangement-Handbuch für den Deutschen Wetterdienst (DWD). Deutscher Wetterdienst, Offenbach, Germany. Version 52.Google Scholar
- Baldauf, M., Reinert, D., & Zängl, G. (2014). An analytical solution for linear gravity and sound waves on the sphere as a test for compressible, non-hydrostatic numerical models. Quarterly Journal of the Royal Meteorological Society, 140, 1974–1985.Google Scholar
- Bröcker, J., & Smith, L. A. (2008). From ensemble forecasts to predictive distribution functions. Tellus A: Dynamic Meteorology and Oceanography, 60(4).Google Scholar
- Daley, R. (1994). Atmospheric data analysis. Cambridge, UK: Cambridge University Press.Google Scholar
- DWD (2016). Deutscher Wetterdienst: Jahresbericht 2016. Deutscher Wetterdienst, Offenbach, Germany. http://www.dwd.de.
- Jolliffe, I. T., & Stephenson, D. B. (Eds.). (2011). Forecast verification: A practitioner’s guide in Atmospheric Science (2nd ed.). Wiley.Google Scholar
- Lauritzen, P. H., & Thuburn, J. (2012). Evaluating advection/transport schemes using interrelated tracers, scatter plots and numerical mixing diagnostics. Quarterly Journal of the Royal Meteorological Society, 138, 906–918.Google Scholar
- Stensrud, D. J. (2007). Parameterization schemes: Keys to understanding numerical weather prediction models. Cambridge University Press.Google Scholar
- Straka, J. M., Wilhelmson, R. B., Wicker, L. J., Anderson, J. R., & Droegemeier, K. K. (1993). Numerical solutions of a non-linear density current: A benchmark solution and comparisons. International Journal for Numerical Methods in Fluids, 17, 1–22.Google Scholar
- Wilks, D. (2011). Statistical methods in the atmospheric sciences (3rd ed., Vol. 100). Academic Press.Google Scholar
- Yanai, M. and Johnson, R. H. (1993). Impacts of cumulus convection on thermodynamic fields. In The representation of cumulus convection in numerical models, number 46 in Meteor. Monogr. (pp. 39–62). American Meteorological Society.Google Scholar