熱赤外3バンド画像による大気中の放射吸収物質の空間分布推定

抄録

Described is a method for extracting the spatial pattern of the amount of radiation absorbers in IR region (substantially CO₂ and H₂O) from remotely sensed data. The radiance estimated from the data is determined by three independent factors. The factors are the transmittance of the atmosphere determined by the amount of the absorbers, the terrestrial surface temperature, and the atmospheric effective temperature. The transmittance varies with wavelength, and the others are assumed to be independent of wavelength. Hence three IR bands, (4.3-5.5μm, 4.5-4.9μm, and 10.5-12.5μm are used here), are necessary to obtain the pattern.
By computer simulation and field experiments, it is confirmed that the ratio of the difference of radiance between two spectral bands (W(10.5-12.5μm)-W(4.5-4.9μm))/(W(10.5-12.5μm)-W(4.3-5.5μm)) is an increasing function of the amount of the radiation absorbers. The band selection algorithm is also proposed.
Strictly speaking, the atmospheric effective temperature depends on a spectral band used for observation. Thus three spectral bands mean three effective temperatures. This causes the variance in the results. The possible variance is estimated with computer simulation by using LOWTRAN 5, and is shown to be comparable to the variance seen in the results of the field experiments.