Guest seminar: Patrick Eriksson från Geoscience and Remote Sensing Division, Chalmers University of Technology, Göteborg.

Abstract:
Atmospheric ice mass has been poorly represented in global models and observations alike. The new generation of km-scale models already shows signs of improvement, but satellite-based observations for verification are lacking. We are supporting the development by providing data from novel measurements in the sub-millimetre wavelength region and making better use of historical data using machine learning. In particular, we have developed two versions of CCIC (Chalmers Cloud Ice Climatology) providing quasi-global coverage (60S to 60N) going back 25/45 years with 0.5/3 hour resolution, respectively. The datasets appear robust over the full time spans, showing no or low trends in global mean ice mass but substantial regional trends. The data are currently used to assess tropical diurnal cycles, including identifying a new feedback process, and to show that the tropical high cloud feedback is near-neutral or positive, in contrast to the present assumption of a negative value.

Welcome!