AN ASSESSMENT OF GLOBAL CLIMATE MODEL SIMULATIONS OF ARCTIC AIR TEMPERATURES

Simultaneous of Arctic temperatures by 19 general circulation models a re examined as part of a diagnostic subproject of the Atmospheric Mode l Intercomparison Project (AMIP). The forcing of all the models by obs erved sea surface temperatures and sea ice from a 10-yr period (1979-1 988) permits comparative evaluations of the model biases as well as th e models' simulations of the interannual variations contained in the o bservational data. The models capture the latitudinal and seasonal var iability of surface air temperatures in the Arctic, although a cold bi as of -3.3 degrees C (std dev = 3.4 degrees C) is apparent over northe rn Eurasia during spring, especially in the models that do not include vegetative masking of the high-albedo snow. The 19-model mean bias ov er northern North America is less than 2 degrees C in all seasons. Ove r the Arctic Ocean, the spring temperatures generally have a warm bias that averages 3.0 (std dev = 2.9 degrees C), although the bias is sma ller in the models in which the prescribed albedo of sea ice is highes t. For the summer season, correlations between simulated cloudiness an d surface air temperatures are negative and statistically significant, but the corresponding correlations for the winter months are small an d statistically insignificant. The models without gravity wave drag ar e generally colder than the other models at the Arctic surface, especi ally during autumn.