Record Display for the EPA National Library Catalog


Main Title Three Radiometric Data Analysis Techniques for Measuring Air Temperature Profiles.
Author Moun, W. D. ; Fo, B. R. ; Wic, C. V. ; Malo, C. M. ;
CORP Author Sperry Rand Research Center, Sudbury, Mass.
Year Published 1970
Report Number SRRC-CR-70-27; CPA-22-69-116; 0673;
Stock Number PB-199 412
Additional Subjects ( Radiometers ; Data processing) ; ( Temperature inversions ; Mathematical models) ; Statistical analysis ; Thickness ; Atmospheric physics ; Temperature ; Remote sensing ; Radiometric thermasondes
Library Call Number Additional Info Location Last
NTIS  PB-199 412 Some EPA libraries have a fiche copy filed under the call number shown. 07/26/2022
Collation 111p
The problem of determining the height of the base, the intensity and the thickness of an elevated inversion layer from radiometric thermasonde data is discussed. Three advanced radiometric data analysis techniques emphasizing the detection and measurement of these salient features of elevated inversions are proposed. The techniques include: (1) a statistical procedure for finding the best three-straight-line representation of the true profile from the radiometric data--called the gradient altitude model, (2) a relaxation procedure for finding the ambient temperature at sixteen preselected heights from the surface to 2.54 km, and (3) several analog and empirical techniques for detecting the presence and determining the base heights and other features of an elevated inversion directly from the recorder outputs of thermasonde data. The initial development and evaluation of these techniques using simulated and real data are presented. The results show that the problem of determining the features of an elevated inversion becomes more difficult as the height of its base increases, its intensity decreases, and its thickness decreases. The relaxation technique appears to be capable of providing accurate descriptions of the temperature profile in the case of elevated inversions of up to 5. (Author)