Abstract
This paper examines how three-dimensional radiative effects influence the way cloud fields appear in high-resolution shortwave satellite images. To do so, it uses cloud reflectance fields simulated by a Monte Carlo radiative transfer model. This study examines the influence of the two counteracting three-dimensional mechanisms: the smoothing effect of radiative diffusion, which can reduce brightness variations, and the sharpening effect caused by thick areas intercepting extra radiation through their sides and casting shadows on the thin areas behind them, which can enhance brightness variability. The findings suggest that current high-resolution retrievals of cloud structure can be significantly biased because they do not take these effects into account. For oblique sun, high-resolution retrievals can overestimate both the scene-averaged optical thickness and the magnitude of cloud variability, and yield systematically distorted cloud shapes and artificially anisotropic cloud structures. It is shown that the biases are especially large when cloud-top height variations are present because cloud-top variations can cause a much stronger sharpening effect than internal cloud variability. To prevent erroneous interpretation of retrieval results, an algorithm is proposed to determine whether retrievals based on any satellite image are affected significantly by these biases.
* Current affiliation: University of Maryland, Baltimore County, Baltimore, Maryland.
Corresponding author address: Dr. Tamás Várnai, Climate and Radiation Branch, Code 913, NASA Goddard Space Flight Center, Greenbelt, MD 20771.
Email: varnai@climate.gsfc.nasa.gov