An innovative methodology for analyzing digital visibility images in an urban environment.

A novel methodology combining digital imaging, conventional fixed visibility monitors, and solar radiation monitors has been developed to characterize the visual air quality of the El Paso and Ciudad Juarez urban vista. The authors have found that the digital image quality is reproducible and useful for quantitative analysis of visibility conditions. Regions of interest were selected in images along view paths of interest and values for a contrast variable of interest, typically the coefficient of variation or contrast ratio (CR) for discrete targets, were computed. Both of these indices are bounded at 0 and 1 and are scaled to the "clean day" maxima for a given date, time, and selected view paths. This produces a relative visibility index for various view paths. With the siting of a Belfort (6230A) visibility monitor at a central location, it has been possible to initiate contrast analysis of various targets in current and archived camera images obtained near this monitor. For uniformly "clean" days, as indicated by fine particulate matter observations and visual inspection, the authors have been able to use the extinction coefficient (Bext) derived from the 6230A to put the relative visibility index, based on CR, on an absolute basis in terms of an ideal target located at a given distance. This permits the generation of contrast extinction, Bext/C, for each view path that is independent of the actual target intrinsic contrast (within limits) and allows the comparison of Bext/C along different view paths with other air quality indices. Multiple linear regression was used to derive a relationship between the CR-based Bext/C value and air quality parameters. Visibility attenuation because of sulfate particles was found to have the highest correlation with Bext/C. In addition, solar radiation was observed to be a significant predictor of visibility in the urban region.