Effects of satellite data resolution on measuring the space/time variations of surfaces and clouds

Abstract

The correlated distributions of satellite-measured visible and infrared radiances, caused by spatial and temporal variations in clouds and surfaces, have been found to be characteristic of the major climate regimes and can be described by the attributes of bidimensional and monodimensional histograms and time-composite images. Most of the variability of both the surfaces and clouds is found to occur at scales larger than the minimum resolved by satellite imagery. Since satellite imaging data sets are difficult to analyze because of their large volumes, many studies reduce the volume by various sampling or averaging schemes. The effects of data resolution and sampling on the radiance histogram statistics and on the time-composite image characteristics are examined. In particular, the sampling strategy used by the International Satellite Cloud Climatology Project is tested. This sampling strategy is found to preserve the statistics of smaller cloud variations for most regions, with the exception of very rare events, if they are accumulated over large enough areas (at least 500 km in dimension) and long enough time periods (at least one month).