Institut für Meteorologie und Klimaforschung

Drei spannende Thesisthemen suchen interessierte Studierende!


Die Forschungsgruppe „Großskalige Dynamik und Vorhersagbarkeit“ bietet derzeit drei Thesisthemen an. Zwei von diesen werden sich auf die am schnellsten aufsteigenden Luftströme in Tiefdruckgebieten mittlerer Breite konzentrieren - sogenannte Warm Conveyor Belts (WCBs) - und ihre Auswirkungen auf die großräumige Strömung und ihre Vorhersagbarkeit. Alle Themen sind für Bachelor- oder Masterarbeiten geeignet.

Wir sind auch offen für andere Themen und über eure eigenen Ideen und Interessen zu diskutieren. Kontaktiere einfach ein Mitglied unserer Arbeitsgruppe.


Project 1: What is the effect of the WCB structure on downstream forecast uncertainty?

WCBs are efficient communicators and amplifiers of forecast errors and uncertainties from the small- to the large-scale flow. Thus, an adequate representation of WCBs in numerical weather prediction models is desirable. In this project, you will analyse the effect of parametrized convection embedded in WCBs on the forecast uncertainty in the operational ECMWF ensemble forecasting system. To this regard you have the chance to analyse a unique data set of forecast data that has been assembled at KIT over the last two years. Ideally, the results of your work will help to understand why some WCBs dilute forecast skill and others do not.


Project 2: Do incorrect sea surface temperatures over the Atlantic explain WCB biases in numerical weather prediction models?

Our current research reveals that state-of-the-art numerical weather prediction models have difficulties in accurately representing the frequency of WCBs over the North Atlantic. The overarching hypothesis is that this bias is due to incorrect sea surface temperatures over the western North Atlantic. To test the hypothesis you will analyse numerical experiments for which biases in sea surface temperature were removed durind the model integration (Vitart and Alonso-Balmasaeda 2018). You will identify WCBs by applying a novel framework based on convolutional neural networks which was recently developed in our group at KIT.


Project 3: How are large-scale weather regimes and European surface weather linked?

Weather forecasts on sub-seasonal time scales cannot predict day-to-day weather (this is simply not possible due to the chaotic behavior of the atmosphere) but rather so-called weather regimes. These describe the large-scale flow pattern on multi-day to weekly time scales. To make such weather regime forecasts useful for end users such as national weather services, the energy industry, agriculture, or tourism, they first have to be “translated” into surface weather on regional scales, which is not a trivial thing to do. For instance, a European blocking weather regime in summer does not always cause a heat wave, and if it does, it may affect only certain regions of Europe. In this project, you will thus systematically investigate how surface weather and its extremes in different European regions are linked to the 7 Atlantic-European weather regimes. Based on this relationship, you will then analyze whether the use of a weather regime index provides more information than simply using the surface weather parameters directly provided by the model. So, if you are interested in weather forecasting, this is the project to choose.



Weather regimes