Development of an Antarctic regional climate system model. Part II: Station validation and surface energy balance

In this, the second part of the analysis of an Antarctic regional climate system model, the model results and analyses are compared to a series of observational data from automated weather stations at a number of Antarctic stations, radiosonde launches, and surface energy balance climatology. The observational analyses show significant biases in comparison with station data, which is attributable in part to the errors in and low resolution of the elevation dataset. This is a factor in model performance also. Further, a dominant factor in the generation of the model biases is the simulation of atmospheric water vapor and cloud. The known 'cold' bias in the clear-sky longwave radiation scheme is amply compensated for by excessive cloudiness in many cases. The strong vertical moisture transport contributes to the excessive cloud formation. The biases explored in detail in this paper are common to regional and global simulations of the Antarctic region and highlight areas in which model development should be concentrated, particularly in coupled models with greater degrees of freedom.