A global cloud cover data set, derived from the USAF 3D NEPH Analysis, was developed for use in climate studies and for Earth viewing applications. This data set contains a single parameter - total sky cover - separated in time by 3 or 6 hour intervals and in space by approximately 50 nautical.miles. Cloud cover amount is recorded for each grid point (of a square grid) by a single alphanumeric character representing each 5 percent increment of sky cover. The data are arranged in both quarterly and monthly formats. The data base currently provides daily, 3-hr observed total sky cover for the Northern Hemisphere from 1972 through 1977 less 1976. For the Southern Hemisphere, there are data at 6-hr intervals for 1976 through 1978 and at 3-hr intervals for 1979 and 1980. More years of data are being added. To validate the data base, the percent frequency of or = 0.3 and or = 0.8 cloud cover was compared with ground observed cloud amounts at several locations with generally good agreement. Mean or other desired cloud amounts can be calculated for any time period and any size area from a single grid point to a hemisphere. The data base is especially useful in evaluating the consequence of cloud cover on Earth viewing space missions. The temporal and spatial frequency of the data allow simulations that closely approximate any projected viewing mission. No adjustments are required to account for cloud continuity.

Using forecast relative humidity (RH) from a global model, several pre-existing diagnostic RH-to-cloud schemes were tested to forecast global fractional cloud cover in a postprocessor format. Since none of the schemes tested provided a superior cloud forecast when compared to Air Force Global Weather Central's (AFGWC) operational 5LAYER cloud forecasts, a new RH-to-cloud scheme was developed by relating cumulative frequencies of forecast RH to cumulative frequencies of analyzed cloud cover from the AFGWC RTNEPH cloud analysis. This scheme creates a series of forecast time-dependent RH-to-cloud curves that can be temporally updated to account for changes in season, cloud analysis, or forecast model, The global model used was a spectral-type developed by the Geophysics Laboratory (GL) using parameterized diabatic physics presently incorporated in the operational GSM (global spectral model) at AFGWC.

A new operational sensor will be on board the next polar-orbiting defense Meteorological Satellite Program spacecraft, which is scheduled to be launched in June 1987. It is called the Special Sensor Microwave/Imager (SSM/I). The SSM/I is a seven channel, four frequency linearly polarized, passive microwave radiometer. The SSM/I will provide estimates of several surface and atmospheric parameters. One of the parameters is cloud amount (percent cloud coverage), which is the topic of this report. SSM/I cloud amount estimates will include some of the situations in which there are difficulties with the Air Force Global Weather Central's Real-Time Nephanalysis automated global cloud analysis using visible and infrared satellite data. Hughes Aircraft Company developed two algorithms for estimating cloud amounts from SSM/I brightness temperatures. One is applicable over snow backgrounds; the other, over land backgrounds. However, it is not possible to obtain cloud amount estimates for land covered with vegetation. No cloud amount estimation algorithms for ocean or oceanic ice backgrounds were required to be developed; even though the potential over both of these backgrounds is good.

The Real-Time Nephanalysis (RTNE PH) generates real-time global analysis of cloud extent, base, height, and type. The RTNE PH makes extensive use of satellite data that are measured in the 10-12 micron infrared (IR) window for thermal mapping of clouds and the earth's surface both day and night. Water vapor absorption effects are significant in such data, especially for high viewing angles and tropical atmospheres where water vapor concentrations are generally high. Water vapor attenuates the emitted ground radiance, reducing its magnitude and with it the apparent temperature of the surface being viewed. Currently, the RTNEPH takes into account such attenuation losses using lookup tables that are functions of viewing geometry, time of day, type of background, and the like. However, these corrections have limited value and leave room for significant improvement. This report outlines and presents two new techniques that are expected to enhance the attenuation correction capabilities of the RTNEPH, multivariate polynomial regression (the Weinreb technique) and the correlated K method.