Laserfiche WebLink
<br />I <br /> <br />I <br />I <br /> <br />important targeting issues. Their rediscovery permitted that to be done in support of the 2003/04 <br />randomized experiment. Results are discussed in this section. <br /> <br />I <br /> <br />The National Oceanic and Atmospheric Administration (NOAA) Atmospheric Modification Program, in <br />cooperation with the Utah Division of Water Resources and Bureau of Reclamation, conducted extensive <br />experimentation on the Wasatch Plateau during the 1990s. A considerable body of publications and <br />papers resulted from these investigations as summarized and cited by Super (1999). Most data sets and <br />field notes from those experiments were discarded over the years as offices were moved, people retired, <br />etc. However, data relevant to the 2003/04 experimentation are available. These measurements are from <br />the mid-January to mid-March 1994 (hereafter early 1994) major field program, and more limited pulsed <br />seeding experiments during the winters of 1994/95 and 1995/96. These data have now been analyzed in <br />detail to address three important objectives: <br /> <br />I <br /> <br />I <br /> <br />1) To better define the wind directions associated with successful targeting in order to consider <br />the usefulness of wind direction in partitioning statistical experiment units. <br /> <br />I <br /> <br />2) To better test the expectation that the TAR and nearby GTR gauge would usually be targeted <br />by HAS-released plumes. <br /> <br />I <br /> <br />3) To examine the suitability of two plateau-top precipitation gauges as control gauges for the <br />statistical experiment. They were located south (GSO) and north (GNO) of the instrumented <br />TAR station and nearby primary target gauge, GTR. There was special concern about the south <br />control gauge, located only 1.6 km southeast, and 1.07 km due south (crosswind) of the TAR <br />which might sometimes be "contaminated" with seeded snowfall. Gauge GSO was even closer to <br />the primary target gauge GTR, located only 1.3 km southeast and 1.0 km due south of it. <br /> <br />I <br /> <br />I <br /> <br />The related questions of potential partitioning by wind direction and routine targeting will first be <br />addressed. The control gauge suitability issue will then be considered. <br /> <br />I <br />I <br /> <br />Partitioning by wind direction would be expected to "sharpen" statistical analyses since transport and <br />dispersion of seeded ice crystal plumes to the target can only occur within some finite wind direction <br />range. An obvious example is that easterly airflow between the HAS and TAR would carry any seeded <br />crystals away from the intended target area. It would not make physical sense to include such cases in <br />statistical analyses. Examination of a large-scale detailed topographical map of the experimental area <br />strongly suggests that HAS plumes would be transported south of the TAR and its nearby precipitation <br />gauge if winds had any northerly component. The situation can also be visualized using Fig. 1. The <br />memory of the lead field technician for the 1994/95 and 1995/96 experiments was that HAS-released <br />plumes would no longer be detected at the TAR once the HAS wind veered north of due west. Similar <br />recollections exist from field personnel involved in plateau-top tracking of plumes released from near the <br />HAS during the experiments of mid-January to mid-March 1991, and from the HAS during mid-January <br />to mid-March 1994, hereafter referred to as the early 1991 and early 1994 experiments. But memories <br />can be faulty and basing wind direction partitioning on actual observations clearly is a superior approach. <br />Therefore, it was attempted to determine whether particular ranges of wind direction could be used to <br />predict successful targeting from earlier observations. Conversely, if other ranges could be shown to be <br />highly associated with targeting failures, such cases should be rejected from statistical analyses. <br /> <br />I <br /> <br />I <br /> <br />I <br /> <br />I <br /> <br />Two wind observation sites used for the 2003/04 field program can be directly tested with earlier data. <br />Winds at the HAS and TAR were observed by robust heated wind vanes and anemometers on towers (see <br />Fig. 1). Similar observations were made during prior field seasons with the exception that unheated wind <br />sensors were used at the HAS where currently existing commercial power was unavailable during the <br />1990s. <br /> <br />I <br />I <br />I <br /> <br />8 <br />