DEVELOPMENT OF STATISTICAL OPERATORS FOR PREDICTION OF LOW CLOUDS
Abstract
A study was made of the statistical prediction of low-cloud amounts and cloudbase heights. Cloud data and other atmospheric parameters over the central and eastern United States were analyzed on a grid mesh of approximately 52 mi (1/4-NWP grid). Predictability of low-cloud amount was evaluated by using the screening regression method and testing the significance of the selected predictors. Predictors considered were low-cloud amount, empirically normalized cloud height, pressure, 850-mb height, surface and 850-mb temperature and dew- point spread, 850-mb geostrophic wind, and derived terms such as vorticity, divergence, and advection. The regression equations were tested on independent data. The equations may be useful for short-period prediction because they provide a better cloud forecast than persistence. They would probably be improved by including other predictors and by extending the area from which the predictors are chosen.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1963
- Accession Number
- AD0414449
Entities
People
- Abraham M. Pavlowitz
- Duane S. Cooley