Statistical Inference of Cloud Thickness from NOAA 4 Scanning Radiometer Data

Abstract

A statistical correlation between cloud thickness and brightness is shown by regression analyses using the least-square method. Cloud thicknesses are obtained from two sources; pilot reports and the three-dimensional nephanalysis program (3DNEPH) for cases of single stratus and stratocumulus layers. Brightness values are obtained from the NOAA 4 satellite scanning radiometer. Regression analyses are performed on both thickness data sources used in conjunction with the scanning radiometer data. The results are shown by the regression curve relating pilot report thicknesses and brightness accounting for 66% of the variance in the cloud thickness, and the regression curve relating 3DNEPH thicknesses and brightness accounting for 46% of the variance in the cloud thickness. Moreover, in view of the effect of cloud compositions on the cloud brightness, regression analyses are performed on both thickness data sources excluding those cases whose origin is an unstable maritime tropical air mass. Results of these regression analyses reveal increases in the correlation between cloud thickness and brightness with 88% and 55% of the variances accounted for by pilot reports and 3DNEPH program data sources, respectively.