Relating the Distributional Character of Numerical Model Output Parameters to the Occurrence of Fog Over the North Atlantic Ocean.

Abstract

This report describes an investigation into the statistical distributional of six model output parameters from Fleet Numerical Oceanographic Center's Navy Operational Global Atmospheric Prediction System, as a function of the occurrence of Fog and No Fog for a climatologically-homogeneous area of the North Atlantic Ocean in the summer season. Beta, Normal and Gamma distributions were fitted to these parameters and forecasts of Fog and No Fog were made using Baye's Law. Intercomparisons were made of these forecasts, using various categorical scoring systems (Threat Score, Percentage Correct, False Alarm, Forecast Reliability and Power of Detection), as well as a probabilistic scoring system (Penalty-Reward Score). The forecast results were examined for significant differences using an Anova analysis. It is confirmed that predictor whose distributions are roughly bell-shaped, it is indicated that the Beta distribution can be generally used as a proxy for the Normal, as can the Gamma also. However, in some cases the Normal distribution results in better forecast scores, and the decision on use of a proxy would depend on which of the scores is to be emphasized.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1986
Accession Number
ADA174543

Entities

People

  • Oliver J. Muldoon

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Energy and Power Technologies
  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Air Force
  • Atlantic Ocean
  • Computer Programs
  • Detection
  • False Alarms
  • Information Science
  • Meteorology
  • Normal Distribution
  • North Atlantic Ocean
  • Oceanography
  • Oceans
  • Reliability
  • Research Facilities
  • Statistical Analysis
  • Statistics
  • United States
  • Weather Forecasting

Fields of Study

  • Environmental science

Readers

  • Atmospheric Science/Meteorology
  • Regression Analysis.
  • Statistical inference.