An Experimental Determination of the Atmospheric Temperature Profile by Indirect Means

On the statement and solution of inverse problems in satellite meteorology

“…the spectral-angular method. Then we have n functionals Note that the problems for φ*, are solved just once for the assigned functions f t (i/) and gt(p) since the functions $1 (2), xu(z) and the values Ψ,·, Φ<, φ{ are expressed only in terms of the solutions of unperturbed adjoint equations. The values θη ν /ΘΤ and de v /dv can be also tabulated for the fixed values of the standard temperature and wind velocity.…”

“…Therefore the measurements in Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 6/8/15 4:35 AM the C 2 band should be supplemented with the measurement in one of the transparencŷ windows' in the 8-12 mem band. Spectral intervals in the CO 2 band may be selected, as recomended in [2], at frequencies with centers in 677.5, 691, 697, 703, and 709cm" 1 and in the frequency Svindow* 899cm" 1 with a resolution 5-7 cm" 1 . Notice that it is possible to select various vision angles instead of various frequencies or choose an optimal combination of measurements at various frequencies and various vision angles.…”

“…A large number of works are dedicated to particular cases of determination of the temperature field in the system underlying surface-atmosphere from satellites in the IR-spectrum (see, for example, [1][2][3][4][6][7][8]16,20]). However, these works deal with the problems of determination of either vertical temperature profiles in atmosphere or sea surface temperature or vertical humidity profiles.…”

On application of the statistical method for determination of atmospheric temperature profiles from satellites

“…For practical application it is necessary to consider the influence of errors of the experimental data on the solution of the inverse problem. This question is very closely connected with the problem of regularizing the integral equation (1) considered by Tykhonov (1963) and Twomey (1963).…”

“…It is clear that the form (2) permits a linearization of the integral equation (1) without any assumption of the spectral interval of the band under consideration (in contradistiction to the method of Yamamoto, 1961). To test the method of optimal parametrization the radiances 1" in eight spectral intervals were computed for a set of temperature profiles T ( p , ) and for a climatological profile T ( p i ) obtained by averaging T ( p , ) a t every level.…”