It is clear that the concentration of pollutants into a narrow band of latitudes (i.e., a “corridor effect”) will depend on both the injection and dispersion rates. In this paper, criteria will be developed to assess the likelihood of a detectable corridor effect being caused by the long-term deposition of pollutants at a single latitude. This paper expands the brief discussion of the corridor effect contained in a companion paper by Whitten et al. (1) and identifies the quantities that control the magnitude of the corridor effect. If the magnitude of the corridor effect were large enough, the buildup of water vapor or nitric oxide could conceivably lead to a low-latitude band of noctilucent clouds (2) or to an increase in the ionospheric D-layer electron concentration (3). The effects that will be considered here are not those of individual launches or flights but are, instead, statistical measures of the half-intensity width taken over season-long time periods and over large, zonally averaged areas. The sources will be assumed to operate continuously and at a uniform rate for many years. EFFECTS OF TRANSPORT PROCESSES During the summer and winter months, strong zonal jets appear in the stratosphere and mesosphere (4). Because the wind shear surrounding these jets often reaches values of 10-3 s_], the near-vertical plumes deposited by the launch vehicles will be dispersed into nearly horizontal ribbons in the east-west direction. Atmospheric tidal motions will cause not only zonal motions but meridional dispersion as well (5). Depending on the altitude and latitude, meridional wind speeds due to tidal motions can reach 20 m/s for a few hours each day before the wind direction reverses. Hence the launch plume could be carried as much as 5° north or south of the latitude of injection. The superposition of the atmospheric motions and the 30 or more launches per month imply that a reasonable approximation to the pollutant injection and mixing is to consider a constant rate of injection and instantaneous mixing in a zonal ring located at the latitude of injection. During periods of time when the zonal winds are replaced by wave number 1 flow, the corridor effect will be destroyed. When the zonal winds cease, local concentrations of pollutants might build up somewhat more rapidly than that calculated with the assumption of instantaneous mixing in a zonal ring. However, the dispersal by atmospheric tides and mixing by gravity waves will still provide appreciable dilution of the pollutant. For simplicity, it is assumed that the motions and mixing of a pollutant along a latitude circle is so rapid that the long-term abundance can be calculated by assuming constant injection of the pollutant and instantaneous mixing into a zonal ring. It is also assumed that the only loss of pollutant is by meridional spreading and by chemical transformation. In order that the parameters that control the prediction of the corridor can be identified, and so that the uncertainty in the magnitude of the predicted corridor effect can be estimated, scaling arguments rather than a numerical model will be emphasized whenever possible. In any case, scale analysis must precede calculations by the numerical model in order to establish that the north-south length scale, L, is large compared to the grid spacing in the numerical model. If L is otherwise, the numerical model will necessarily predict too large a value for the corridor width. A related problem in using numerical models to calculate the corridor width arises when it is desired to simulate the deposition of a pollutant in a narrow band of latitudes. When the advection term in the model operates on the narrow deposition profile, it introduces artificial diffusion or dispersion. Because numerical models are subject to these difficulties, it is important to verify that their results are physically
RkJQdWJsaXNoZXIy MTU5NjU0Mg==