LARS Tech Report Number
Already during this course you have heard occasional references to the effect of the atmosphere upon radiation measurements obtained by remote sensing techniques. Usually the indication was that in some way the atmosphere acted to limit the quantity and quality of information available to the measuring system.
In certain respects this is a quite valid assessment of atmospheric processes. In fact, it is possible to envision the atmosphere as performing a dual role: 1) by acting as a generator it can introduce a spurious signal into the measured quantities, and 2) by acting as an unwanted filter the atmosphere can remove portions of the signal from the system. The effect of each of these actions is to disguise the information being sought and, as a result, techniques must be devised to remove such unwelcome effects.
Consider, for example, the photographer who places a filter in front of his normal lens system. His hope is to remove an excessive amount of scattered light brought about by hazy atmospheric conditions. This is an example of having to combat item 1 above. In addition, we might consider the effect of clouds upon the instrument system. Most certainly clouds act as a filter by reducing the overall level of the signal measured by the instrument. But more than that, clouds are not uniform in their radiative properties across the wavelength spectrum. Consequently, the presence of clouds will alter the relationships between signals obtained in different spectral intervals. Until such time as information about the radiative properties of clouds -- and of various types of clouds -- can be included in a data analysis procedure, methods of remote identification of surface properties from spectral signatures are severely hampered. At the present time, aircraft measurement programs can avoid this problem by not flying on cloudy days -- an effective but unsatisfactory solution. Satellite programs simply accumulate large quantities of unusable data with this constraint.
We shall briefly consider the physical properties of the atmosphere which produce these effects. Firstly, let us develop a means of describing the transfer of radiative energy through a medium. Then, we shall apply this general treatment to cases of special interest to remote sensing technology.
Date of this Version