WMOD00500 - Laserfiche WebLink

Home Browse Search

OCTOBER 1988 HOLROYD, McPARTLAND AND SUPER 1133 . . FIG, 2, Part 3, Plume locations and wind vectors for a southwesterly flow, c 10 i 20 :'n~. 111 I mary, when, under clear morning conditions, AgI was released into northeast flow from the north slope. The aspect of that snow-covered slope with respect to the sun minimized solar heating, especially compared to the south-facing slopes. The other clear-air experiments were conducted near sunset in order to minimize solar heating. A close examination of the data did not dis- close demonstrable effects of solar heating on the be- havior of the AgI plumes during these experiments. A few times the tethersonde appeared to record bubbles of air up to about 20C warmer than ambient, appar- ently rising in an otherwise dry adiabatic environment, but such temperature variations were similar to those found farther away from the surface, thereby rendering their effects on the plume difficult to detect. a, Horizontal extents For the period of each experiment the winds at each station were analyzed to produce a mean speed and mean direction along with the standard deviations of those parameters. The shortest time resolution data available from surface and tower wind stations were 5 min means. The standard deviations, therefore, are not totally compatible with those often used in air pollution investigations, which are based on fast-response sensors recorded near 1 Hz. The winds are plotted in Fig. 2 as wind vector wedges, with the apex at the station and the broad end in the direction from which the wind was coming. The mean wind vector is plotted as a cross. The arcs in each wedge show the magnitude of the average speed plus and minus the standard deviation; the sides indicate the average direction plus and minus the standard de- viation. The effect is to create what look like arrowheads that show which way the air was flowing; the length of the arrowhead indicates speed and the width, direc- tional variation. The scale at the bottom of Fig. 2 dou- bles for distance and wind speed. The scatter of indi- vidual wind vectors, not illustrated, was nearly always within about two standard deviations of the averages. If the scatter were Gaussian, then about 68% and 95% of the observations should be within one and two stan- dard deviations, respectively. In some parts of Fig. 2 the wedges are formed with dotted lines, which indicate either the winds aloft as determined by the acoustic sounder in the center of the mesa (site A in Fig. 1) or the upper-level winds measured by the tethersonde at the seeding site. The levels of those average winds aloft, given in km in Table 1 under "Acou" and "Teth," varied with each exper- iment because of the limits imposed by the conditions associated with their recording. The winds within 150 m of the surface, as determined by the tethersonde, are plotted with solid lines like the surface and tower data. In one case (Fig. 2k), the drainage winds from the northeast at the end of the experiment are ignored. In another (Fig. 2g), the tower winds adequately described the air movement below the inversion, while the winds determined by the acoustic sounder above the inversion were not relevant for the figure. In a third (Fig. 2i), the tethersonde measured only the near-surface winds and could not rise into the excessively strong winds at the higher levels. In each part of Fig. 2, the locations of the detected AgI plume are indicated by bold lines. The solid inner lines represent the average locations of the edges of the plume as determined by either the acoustical ice nu- cleus counter or by a measurement of the resulting ice particle plume. Table 1 shows which average was used. The dotted lines indicate the extremities ofthe plumes by either detection system. The solid lines, therefore, can be considered to represent an instantaneous plume width that meanders within the span of the dotted lines.