Abstract:
A high-order discrete-ordinate approximation is utilized to solve the radiative transfer equation for both solar and terrestrial spectra. The solutions have been compared with other methods and found to be reliable and efficient. These solutions have been used to construct a complete and comprehensive radiation model for the arctic atmosphere. The bulk radiative properties (e.g. fluxes and heating/cooling rates) as well as the angular distribution of intensity can be computed as functions of wavelength at various levels in vertically inhomogeneous atmospheres. The radiation model treats Rayleigh scattering, gaseous absorption/emission, scattering and absorption/emission by cloud droplets and haze particles. Snow conditions of the arctic region are simulated by snow grains and soot contamination in the surface layers. A unified treatment of shortwave and longwave radiative transfer is achieved. Use has been made of the five McClatchey atmospheres and of data from the Arctic Stratus Clouds Experiment collected in 1980. Results are compared among broad-band, narrow-band and line-by-line (restricted to gases) computations. We find that at the expense of accuracy by a few watts.m('-2) for flux or a few tenth (DEGREES)C/day for heating/cooling rate computations, the broad-band models are very fast and suitable for certain types of climate modelling. During the arctic summer, stratus clouds are a persistent feature and decrease largely the downward flux at the surface. Arctic haze is important if it is above the cloud layer or in air with low relative humidity and also decreases the downward flux at the surface. The greenhouse effect of doubling the CO(,2) amount can be offset by the haze condition or by the increase in cloudiness of about 4%. In late June, we find that a clear sky condition results in more available downward flux for snow melt than does a cloudy sky condition. This is because the increase of infrared radiation diffused back to surface by the cloud can not compensate the reduction of solar radiation. If the snow starts to melt, the decreasing snow albedo further accelerates the melting process.
