The unexplained solar absorption and atmospheric H2O:: a direct test using clear-sky data

This paper is motivated by several recent studies that have shown that observations of atmospheric solar absorption systematically and significantly exceed model estimates. This paper tests whether uncertainties in the physics of water vapor absorption in clear skies are directly responsible for this unexplained excess absorption. Radiative transfer calculations of clear-sky solar fluxes are compared to measurements in the tropical Pacific at the surface and the tropopause. We find that the atmospheric absorption in excess of that predicted by radiative transfer models, if it exists, is less than the experimental uncertainty of 7 W m(-2) (diurnally averaged). Furthermore, the difference between observed and modeled absorption is essentially independent of water vapor amounts between 35 and 51 kg m(-2). A more direct test of the accuracy of modeled water vapor absorption is conducted with two independent multi-spectral radiometers at the Atmospheric Radiation Measurement site in Oklahoma, each providing over 16000 surface measurements of direct solar radiation in the 0.94 mu m region. These spectral data confirm state-of-the-art radiation model computations of water vapor line absorption to within 5% for the wavelength region tested. The model-observation agreement for both tropical and Oklahoma data strongly suggests that uncertainties in the physics of water vapor absorption in clear skies are not a source for any significant excess solar absorption, thus narrowing the search to other atmospheric constituents or water vapor in cloudy skies.