Laserfiche WebLink
<br />A further complication is that many observing sites have been discontinued in recent years <br />as a cost-savings measure. Unless some special need could be shown, a number of sites were <br />judged to be redundant and eliminated. <br /> <br />These changes in SCS observing sites and method are mentioned because they required the <br />implementation of a tedious search for currently valid data. The most recent map of Snow <br />Survey Data Sites for Idaho was published in 1988. NA WC supplied target area maps, which <br />also indicated individual AgI generator locations. The target areas were traced on the Snow <br />Survey Map. All available snow data measuring sites were then identified, as were potential <br />control sites. <br /> <br />5.2 Choosing Control Sites <br /> <br />Potential control sites were chosen as close as possible to the target areas while remaining <br />generally upwind; that is, to the southwest, west, or northwest of the targets. The approach <br />was to proceed upwind from the targets until several measurement sites were identified, but <br />no further upwind than necessary. In most cases, the degree of association (correlation) <br />between snow measurement sites tends to decrease with increasing distance. However, <br />exceptions exist to this general rule as will be shown. <br /> <br />Choosing control sites is a subjective process in terms of deciding how far upwind is too far. <br />In practice, existing topography determines where the snow measuring sites are. Once all <br />sites have been chosen for the first few mountain ranges upwind in a particular direction, <br />resulting in several sites, little reason exists to search further upwind unless resulting R- <br />values are mediocre. <br /> <br />Well over 100 snow measurement sites were identified for the three seeding projects from the <br />map of Idaho Snow Survey Data Sites in Idaho, Montana, and Wyoming. Sites were either <br />within the target areas (in a few cases just over the divide forming the target basin <br />boundary), or were possible control sites. In the case of the Boise target, four sites were <br />chosen a short distance downwind (east) from the target area for examination of possible <br />downwind effects. Downwind sites were not examined for the other project areas. <br /> <br />Once measurement site names were identified, all existing data for each site was downloaded <br />from the SCS's computerized data base in Portland, Oregon. Many sites on the 1988 map <br />were found to have been discontinued, so such sites were excluded from further consideration. <br />Several snow measurement sites were found that had records extending back to the early <br />1960's but not earlier. Therefore, another criterion for keeping a site was that it had to have <br />records for April 1 SWE from at least 1963 through 1993. This criterion provided a minimum <br />of 30 nonseeded winters. At a few snow measurement sites which had only one winter of <br />missing data, the missing point was estimated using a regression equation calculated for <br />other years with a nearby station. <br /> <br />8 <br />