Fields of research

The research group of meterological analysis and prediction, are mainly working in different aspects of numerical weather prediction.

The research unit Meteorological Analysis and Prediction primarily work with development of Numerical Weather Prediction (NWP) models. The forecast models utilize numerical methods in order to simulate the flow of the atmosphere. Our research unit actively works within the following fields all in order to reach a common goal - better weather forecasts.

Data assimilation : When starting a numerical forecast simulation you need to know something about the initial conditions. This information is retrieved from observations. The problem is that observations do not exist at every grid point in the prediction model, therefore, we need to estimate using a technique called Data Assimilation. Our unit has been putting in a lot of effort in order to develop new assimilation techniques, and recent work also includes remote sensing analysis data.

Ensemble Prediction System: The atmosphere is chaotic, therefore, the growth of initial uncertainty during a given forecast period is flow dependent. The evolution from the initial state can sometimes be highly predictable, however, just a slight alteration in the initial state can evolve to a highly unpredictable state (Palmer et al, 2006). Depending on the location of the initial conditions completely different predictions may result. By integrating the forecast model fifty times all with different initial states, achieved by inserting small perturbations into the analysis field, one can better capture this flow dependent predictability.

Physical Parameterisation: Many physical processes in the atmosphere occur on spatial scales smaller than those explicitly resolved by numerical weather prediction models. Therefore these processes need to be parametrically included. The role of physical parameterizations in an atmospheric model is to describe diabatic effects (processes that force a heating or cooling in an air parcel), and sources and sinks of the model (like friction and other effects of mountains).

Observational Systems : Observation systems and the their design and maintenance is of great importance for various aspects of meteorology. The purpose of the observations is to collect information of the past and present atmospheric state in order to create an initial field for the forecast.

Statistical Post Processing : There is still some potential to improve the skill of weather forecasts by using a suitable system to statistically post-process (SPP) the raw forecast fields that come out of an NWP model.