Syllabus
Introduction to atmospheric modeling
Models of general circulation of the atmosphere as an historical example, water and energy cycles of the atmosphere, dimensional analysis, basis of the numerical analysis, simple modeling problems.
Dynamics
Basic types of waves, scales of atmospheric motions (spatial and temporal), basic hypothesis and useful equations, discretization, types of numerical instabilities, methods of horizontal and vertical discretization, time marching schemes, numerical filters vs filtered equations, potential vorticity and examples of use (baroklinic instability, jet streams, fronts).
Physical parameterizations
Role of parameterizations and their historical evolution. Parameterization of radiative transfer, water phase changes, microphysics of clouds and precipitation, subgrid-scale orography effects, exchange with surface and turbulence, non precipitating convection, deep moist convection.
Analysis and data assimilation
Aplication of the optimal estimation theory in meteorology on the analysis problem, BLUE equation in multiscale, observation errors, modeling of first guess errors. Analysis methods, tangent linear and adjoint models, basis of data assimilation. Construction of variational data assimilation, example of 3DVAR with extention to 4DVAR. Curent use of observations in data assimilation, monitoring, quality control impact on analysis and forecast quality.
As main application examples of these methods, up-to-date numerical prediction models are used: IFS/ARPEGE (a common numerical weather prediction system of the European Centre and Météo-France) and its limited area fine-mesh version ALADIN (a system developed in international collaboration and used operationally at the Czech Hydrometeorological Institute).
Annotation
Basis of the atmospheric modeling, dynamics and instabilities in the atmosphere, physics, data assimilation, synthesis. The main goal of the seminar is to demonstrate that learning models' behaviour is as instructive as the comparison of their results with observations.