Comparisons of Cloud Heights Derived from Satellite and Atmospheric Radiation Measurement Surface Lidar Data

Cloud heights derived from single-channel, satellite infrared data can be relatively uncertain under certain conditions such as overlapped or optically thin clouds. During the daytime, optical depths derived from 0.63 μm visible (VIS) reflectances are used to adjust the altitude of optically thin clouds. Without the adjustment, the cloud heights are significantly underestimated diminishing the reliability and utility of the data. Therefore, at night, other data are required to determine cloud optical properties. In this paper, a new technique is developed to determine the phase, optical depth, effective particle size, and altitudes of optically thin clouds at night using combinations of the 3.7-μm or solar infrared (SI), the infrared (IR) window at 11 μm, and the 12-μm “split” window (SW) channels. These retrieval methods are tested by To date, cloud and top of atmosphere radiation parameters applying them to multispectral Geostationary Operational have been derived from satellite data for ARM IOPs using the Environmental Satellite (GOES) data and comparing the Layer Bispectral Threshold Method (LBTM) during the day results with Atmospheric Radiation Measurement (ARM) and a simple single IR channel threshold method at night Program surface lidar measurements taken during several (hereafter referred to as IRONLY). Minnis et al. (1995a) seasons over central Oklahoma. Both daytime and nighttime describe an LBTM analysis for the April 1994 IOP. Cloud methods are tested for a variety of cloud conditions. This and radiation parameters are derived for 3 layers: low (cloud analysis provides a more comprehensive evaluation of heights z < 2 km), mid (2 < z < 6 km), and high (z > 6 km). satellite-derived cloud heights than heretofore possible. Optical depths J are derived from visible reflectance