GEOWOW provides improvements for a wide range of weather data users. The increased accessibility of the THORPEX Interactive Grand Global Ensemble (TIGGE) helps scientists and data users in National Weather Services, Universities and various international organisations to access data in a significantly reduced time (this especially improves the ability of users with limited internet bandwidth to access TIGGE data). In addition to this, the new TIGGE LAM (Limited Area Model) extension of the archive targets primarily the European community by providing standardized archival and access facilities to a range of Limited Area Models from Europe. This will help users to compare models and improve their methodologies.
Work on severe weather product development is done in close collaboration with users. Many of these users are from developing or least developed countries and participating in World Meteorological Organization’s program of Severe Weather Forecast Demonstration Project (SWFDP). This program focuses on improving the use of weather forecast data by different decision makers in these countries. Although focusing on severe weather the SWFDPs encompass users in a range of SBAs, including weather, water, agriculture and disaster management. This provides a natural multidisciplinary framework to demonstrate the use of TIGGE data.
The TIGGE-LAM database already contains two of the 10 expected Limited Area Models from Europe (COSMO-LEPS and ALADIN-LAEF). The archival of the other models (MOGREPS, DMI - HIRLAM, GLAMEPS, COSMO DE EPS, PEARP, COSMO-SREPS, SREPS, SRNWP PEPS) is either in progress or expected to happen in the near future. The archive will contain some high priority surface parameters on the models’ original native grids.
The data volumes involved in the TIGGE archive for multi-model Ensemble Prediction System (EPS) modeling are very large, and it is currently impossible to perform research on long time-series without retrieving all the original fields from tape. This makes a retrieval of selected data series through limited internet bandwidth extremely difficult. There was therefore a strong need from users to have access to time-series at selected point locations which would speed up the retrieval process significantly. Typical application locations considered are airports and official monitoring (observation) stations. This part of the project is in a phase where all necessary tools are ready and discussions are in progress with all data providers about the optimal archive system setup. After finalizing the data structure the population of the archive can be started for all TIGGE models.
Following a review of current status of methodologies for calibration and combination of ensemble forecasts, the Weather partners are working on multi-model combination and calibration for TIGGE models. This project has the potential to cover the area of Europe and to use more recent, higher resolution TIGGE forecast data. In addition to precipitation, the work covers several other forecast parameters: surface air temperature and dew point, surface wind speed and mean sea-level pressure. To improve the quality of the TIGGE based products a new ensemble calibration scheme has also been developed which directly targets the local statistical reliability of the ensemble predictions – to minimise any systematic errors in the probabilities of forecast outcomes.
In the area of the TIGGE-based weather forecast product development, work is being done in several directions. A survey was carried out of forecasters participating in the SWFDP to assess requirements for forecast products for severe weather. The responses confirm that there is considerable interest in the use of ensemble-based products to support forecasting of tropical cyclones, heavy precipitation and strong wind and development focuses in these areas. A set of multi-model severe weather products is being developed using a TIGGE-based forecast climatology with four of the main models. The generation of this new product set in real time still requires a set of technical issues to be solved. Similarly several multi-model products have been developed for tropical cyclone tracking. A tracking algorithm was designed and set up running for some of the main TIGGE models and their multi-model version. All these new TIGGE products are going to be provided to the SWFDP users for testing and evaluation.
In addition to the severe weather product development, the TIGGE archive is also used to demonstrate how these models can be used to aid forecasting of high impact weather events. This work is conducted mainly through important case studies. An example of these cases is the study of a high precipitation event in association with the passage of an African Easterly Wave (AEW) over West Africa. It was used to show how the inherent variability within the ensembles can be utilized to gain information on the possible ranges of intensity or evolution that can be supported by the large scale environmental conditions. To assess how much information is present in the TIGGE ensembles, an EOF (Empirical Orthogonal Function) and cluster analysis was performed to identify distinct AEW scenarios that are contained in the ensemble data. Future and ongoing work applies the method (EOF and clustering analysis) to case studies of additional hazards (e.g. tropical cyclones), and other regions as well, as to demonstrate the robustness of the approach in predicting severe weather events.