6. How do scientists intend to improve the models within the "Year of Polar Prediction"?

Scientists around the world are trying to improve ice and weather forecasts for the polar regions, especially for the Arctic. On the one hand with the melting of sea ice, the Arctic will be of interest as a shipping route, and on the other hand the conditions in the Arctic also influence the weather and climate in the mid-latitudes - for example, in Central Europe and North America.

One problem is that there are far too few measurements from the Arctic. With the international project "Year of Polar Prediction" (YOPP) a lot of new data will be gathered through intensive measurement campaigns that include ship expeditions, aircraft campaigns, and measuring sites on the sea ice. The main goal is to find out which observations are particularly important for ice and weather prediction and climate modelling in the polar regions. The researchers also want to know when and where future measurements should be carried out that are particularly representative of the situation in the polar regions. Meteorologists hope, for example, to be able to make predictions of the future ice cover for a period of one hour to half a year in advance. Researchers such as Helge Gößling expect new insights into the physical processes in the Arctic and Antarctic. "With this knowledge, we can improve the simulation of these processes in our weather and climate models."

Helge Gößling is entering new territory with the project, as short-term predictions do not usually belong to the tasks of a climate modeler. Meteorologists feed weather observations of the past few days into their computers in order to predict the weather for the next days. However, climatic models calculate long periods. Usually they do not provide weather data for a specific day, but rather probabilities - for example, the location and extent of low pressure systems. "For us it is a challenge to enter into the short-term prediction of ice cover in the Arctic," says Helge Gößling.

For meteorologists, it is natural to work with actual observational data. While Climate modellers first have to build a software tool that allows to enter actual ice or weather data into their climate models - to "assimilate" observations. "Data assimilation is an art in itself," says Helge Gößling, who wants to extend the AWI Climate Model during the Year of Polar Prediction with a data assimilation scheme, so that the model can be used to predict the ice cover up to half a year in advance. Ultimately, however, his goal is not to deliver operational predictions using his own model – this remains the task of the operational prediction centres, which are heavily involved with YOPP. "Thanks to the close international cooperation between forecast centres and climate researchers, both can advance their models and prediction systems much further than any single group could do alone," says Helge Gößling.