A STATISTICAL APPROACH TO 2-7 HR PREDICTION OF CEILING AND VISIBILITY. VOLUME I,

Abstract

The statistical forecast technique Regression Estimation of Event Probabilities (REEP) is applied to predicting ceiling and visibility at Chicago and three West Coast terminals for periods of 2, 3, 5 and 7 hours. The REEP forecast systems developed for each of the 36 predictands are applicable at any hour of the day and any day of the year. The systems were applied to an independent sample of 1460 cases for each predictand. The resulting forecasts were compared to forecasts prepared by the subjective forecaster under routine operational conditions, and to forecasts made by two control techniques, persistence and Climatological Expectancy of Persistence (CEP). The forecasts made by the four techniques are expressed in categorical and/or probabilistic form. The cat- egorical forecasts are evaluated by a score de- vised by the U. S. Weather Bureau and designed to measure skill in forecasting operationally important categories of ceiling and visibility. (Author)

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1964
Accession Number
AD0451123

Entities

People

  • Edna L. Sorenson
  • Isadore Enger
  • John A. Russo Jr.

Tags

DTIC Thesaurus Topics

  • Collaborative Techniques
  • Delphi Method
  • Probability
  • Terminals
  • Visibility

Fields of Study

  • Environmental science

Readers

  • Atmospheric Science/Meteorology
  • Climatology
  • Regression Analysis.