Atmospheric visibility

Abstract

The presence of atmospheric particles always causes a reduction of visibility. When looking towards a distant target, the appearance of the target is altered in such a way, that it looks more similar to the horizon, and normally its contrast against the background becomes less with increasing distance. The contrast change can be calculated, if the illumination, the scattering function, the intrinsic brightness of the target and the extinction coefficient of the atmosphere are known. At the distance, where the contrast of the target equals the contrast threshold of the human eye, the target just is visible (i.e., this distance is the visibility). The first theoretical considerations of contrast reduction have used a simple homogeneous atmosphere with an absolutely black target seen against the horizon. The result was a formula, giving an inverse relation between the extinction coefficient and the visibility. For more complicated situations extensions of this method can be used. Unfortunately in most practical applications the necessary parameters of the atmosphere, target, and illumination are only known incompletely, since they may vary as a function of distance from the observer. Therefore visibilities calculated from the data available at the location of the observer may be different from an observed visibility, as well as visibilities using different targets in different directions. Observations performed under controlled conditions in the laboratory have shown, that visibilities can be measured with accuracies of a few per cent, and all conclusions which can be drawn from the usual theories have been verified; therefore an estimation of possible errors or deviation from the ideal case (black target, homogeneous illumination and aerosol, contrast threshold of 0.02) can be performed easily. Most targets are not ideally black. Coniferous forests have luminances relative to the horizon of mostly less than 0.2 and thus the visibility observed maybe up to 4 % smaller than using an ideal black target. When the targets are in their own shadow, they are dark enough, that a visibility comparable to a black target is observed, brighter targets normally are less visible than black targets; the extent of visibility decrease can be calculated if the intrinsic brightness of the target is known, which can be obtained with a high enough accuracy by using nearby model surfaces. When visibility observations are to be performed in an inhomogeneous atmosphere, under most conditions the average extinction coefficient between target and observer determines the visibility. Therefore the visibility has the advantage of averaging. Inhomogeneous illumination can cause the visibility to increase or decrease. Generally an increase in visibility is attained, when the atmosphere is less illuminated between the target and the observer. Under most infavourable conditions, such as clouds between the observer and the target and sunshine behind the target, the visibility is 20 % larger than with homogeneous illumination. With cloud spacings more likely to exist in the atmosphere, the deviations are in the order of 5%; for some cloud spacings the visibility equals the visibility under homogeneous conditions independently of the position of the clouds. The contrast threshold of the human can influence the observed visibility, larger contrast thresholds give smaller visibilities. Under normal daylight illumination conditions the threshold has a constant value, which increases only at illumination levels after sunset or before sunrise. If the targets have a small angular size (less than a few minutes of an arc) the contrast threshold also increases, thus the visibility of this target decreases. Therefore small targets should not be used at large distances. Mostly the atmospheric aerosol has an extinction coefficient decreasing in magnitude with increasing wavelength. Therefore targets are better seen in the red part of the spectrum than in the green or blue part, thus the most important wavelength for vision is at 580–600 μm. As laboratory experiments and observations in the atmosphere have shown, visibilities can be measured quite accurately. Since variations in visibilities due to changing observation conditions can occur, the determination of a standard visibility is desirable. It can be obtained from the measured visibility, by inclusion of correction factors for the inherent contrast and possible deviations due to inhomogeneous illuminations.