Date of Award


Document Type


Degree Name

Master of Science


Department of Engineering Physics

First Advisor

Jason P. Tuell, PhD


This research examined the performance of the Air Weather Service (AWS) Fog Model and the potential for using it in the Southeast United States for predicting fog. This task was accomplished in four separate steps. First, a correlation study was performed by comparing different weather elements in observations that met radiational cooling conditions to the observed visibility. This correlation study showed that the 22 UTC dewpoint depression was correlated (0.60) with early morning fog and no other weather elements that are commonly observed had significant correlation with early morning fog. Second, a verification study was conducted on the Saint Louis University (SLU) version of the fog model. This verification study showed that the fog model has an underforecasting bias in the summer season and an overforecasting bias in the fall season and that persistence forecasts beats fog model forecasts for both seasons. Third, a sensitivity study was conducted on the fog model. The sensitivity study showed that the fog model is sensitive to the value input for wind speed; the fog model predicts more fog events as the wind speed is increased. Finally, the SLU version of the AWS Fog Model was modified to adapt it to the Southeast United States and another verification study was conducted. The fog model was adjusted to remove the summer underforecasting bias and the fall overforecasting bias. After this adjustment, the fog model verification scores showed a slight improvement over the verification of the SLU version of the fog model.

AFIT Designator


DTIC Accession Number


Included in

Meteorology Commons