Fog Occurrence, Distribution, and Coverage through GIS Analysis

The project develops, tests, and implements a method for the improved fog prediction in New Jersey and the northern Mid-Atlantic region using GIS analysis. This is accomplished through a partnership with external collaborators and the creation of a Fog-GIS database at Kean University to identify and relate point and regional occurrences of fog for each season as a function of surface and upper air atmospheric parameters and patterns and their interaction with, and relationship to, the local physiographic features. The intensity of fog is noted according to visibility in order to separate “fog” events from “dense fog” events and these are compared to null cases in order to improve the probability of detection while reducing the rate of false alarms. GIS mapping and statistical summation provide for a description of the fog occurrence population characteristics, features, and behaviors by season as well as the leading causative factors. These are based upon layer information (e.g., soil, land use, population, et cetera) that provides ‘local feature' parameters to be used in conjunction with atmospheric parameters. Summary results are presented for a prototype season (winter) and are considered with regard to operational forecast support and guidance. The findings will also be used to develop an inference algorithm for predicting fog occurrence at other locations and data void/sparse regions in order to derive a more precise plot of the spatial coverage patterns of fog that may be verified through satellite and other techniques.