Date of Award

3-2001

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Engineering Physics

First Advisor

Michael K. Walters, PhD

Abstract

Cloud-free line-of-sight probabilities were calculated using two separate methods. The first was a variation of a method developed by the Rand Corporation in 1972. In it, CFLOS probabilities were calculated using empirical data based on five years of photograms taken over Columbia, Missouri and forecasted cloud amounts rather than climatological values. The second was a new approach using the Cloud Scene Simulation Model developed by Phillips Laboratory. Cloud scenes were generated using forecasted cloud fields, meteorological inputs, and thirty random numbers. Water content files were produced and processed through a follow-on program to determine the extinction coefficients at each grid point in the working domain. A reiterative routine was written to integrate the extinction coefficients along a view angle from the top of the domain down to the surface at separate points within the horizontal domain. The values of each point were summed and averaged over the working domain to determine the CFLOS probability for the target area. The nadir look angle was then examined for both methods. Stratus, stratocumulus, cumulus, and altocumulus cloud types were independently examined with the CSSM generated cloud scenes. Each method and cloud type were compared against the known CFLOS probability for nadir. Results indicate the method developed in 1972 underestimates CFLOS probabilities by as much as twelve per cent with horizontal cloud coverage ranging from 30 to 80 per cent. CSSM generated cloud scenes varied depending on the cloud type analyzed, with stratocumulus clouds measuring up the best against the known probabilities.

AFIT Designator

AFIT-GM-ENP-01M-03

DTIC Accession Number

ADA392629

Download

Share

COinS