The characterization of atmospheric spread functions affecting satellite remote sensing of the earth's surface

Abstract

A Monte Carlo simulation program has been used to compute the signal contribution of surface elements outside the field of view of a satellite borne sensor in both the visible and infrared spectra for realistic atmospheric models. For azimuthally symmetric reflection the effect of an arbitrary surface albedo pattern can be represented by two spread functions, namely, 1) the ground-satellite spread function, and 2) the ground-ground spread function. Examples of computed spread functions are shown together with their Fourier correspondents, the modulation transfer functions. It has been shown that the central portions of the line spread functions behave as the inverse distance while the point spread functions always possess a region behaving as the inverse square of the radial distance. The median distance is shown to relate to the weighted molecular and aerosol scale height, and for large distances the spread functions become independent of the aerosol loading.