A comparison of the BALTIMOS coupled climate model with atmospheric and sea surface parameters derived from AMSR-E

Abstract

The BALTEX Integrated Model System (BALTIMOS) coupled atmosphere ocean model was compared to passive microwave observations of the Advanced Microwave Scanning Radiometer (AMSR-E). Emphasis was put on quantifying the uncertainties associated with the different variables based on data screening both in the model and observations. Monthly means of three atmospheric parameters, as well as sea surface temperature, were compared for a period of 1 year. Sea ice extent was also derived from AMSR-E and compared to the model data on a daily basis. It is shown that the accuracy of the comparisons on a monthly mean basis is limited by precipitation screening. Out of the three atmospheric parameters, surface wind speed and water vapor column amount agree with the model data to within the accuracy of the comparison. The vertically integrated cloud liquid water content diagnosed from BALTIMOS is systematically higher than the liquid water content derived from satellite, even if potential systematic errors are accounted for. In terms of coupling, the two most relevant variables discussed are sea surface temperature and sea ice extent. The temporal extent of sea ice in the investigation area is well represented, as are the periods of the main growing and decay periods. The total sea ice cover appears to be underestimated by BALTIMOS, especially in the peak season between January and the beginning of March. The amplitude of the annual cycle of sea surface temperature in BALTIMOS appears to be too weak compared to the observations, leading to too cold sea surface temperatures in summer and too warm sea surface temperatures in winter. This might also partially explain the underestimation of sea ice cover by BALTIMOS.