The Arctic climate has experienced large changes over recent decades, the largest for any region on Earth. To understand the underlying reasons for this climate sensitivity, we need both models and observations. Unfortunately, due to the rough climate and the inaccessibility to Polar Regions, there is a lack of meteorological observations over the Arctic Ocean. However, the few times series that are available are frequently used to evaluate, confirm and challenge our existing weather forecast and climate models and of course to increase our understanding of the Arctic climate system. For an additional contribution to this knowledge, reanalyses are valuable tools, but these reanalyses also need to be evaluated.
The Arctic System Reanalysis (ASR) is a regional reanalysis, forced by the global ERA-Interim reanalysis at the lateral boundaries, incorporating model physics adapted to Arctic conditions. ASR was developed to serve as a state-of-the-art high-resolution synthesis tool for assessing Arctic climate variability and monitoring Arctic climate change. In this thesis, data from the Arctic Summer Cloud-Ocean Study (ASCOS) field experiment, conducted in August and early September 2008, have been used to evaluate the performance of ASR and ERA-Interim for the Arctic Ocean.
It was found that the reanalyses captures basic meteorological variations, coupled to the synoptic scale systems, but have difficulties in estimating clouds and atmospheric moisture. While ERA-Interim has a systematic warm bias in the lowest troposphere, ASR has on average a cold bias of about the same magnitude. The roots to this are suggested to lie within the cloud description in ASR. The results indicate that more sophisticated descriptions of cloud microphysics in ASR did not significantly improve the modeling of cloud properties, especially for mixed-phase clouds, compared to ERA-Interim.
These conclusions motivate a more detailed study of the characteristics of Arctic clouds, where some research questions are: how frequent are the clouds that are difficult to model? Can the satellite observations tell us more about the cloud phase and the vertical and horizontal structure of the clouds? How do the clouds influence the radiative balance? These questions can hopefully be answered using both ASCOS data and satellite observations from the remote sensing instrument MODIS onboard the Terra and Aqua satellites.