Laserfiche WebLink
<br />OOH82 <br /> <br />wetter years. is not surprising; as discussed previously, however, <br />variation in the intensity or duration of the storms was not taken into <br />account. Probably at least some of the storms in the wet winters were <br />more intense or lasted longer than those in the dry winters (Hindman and <br />McKee, 1981). <br /> <br />Validating Assumptions <br /> <br />The use of interpolated soundings to identify storms requires <br />assumptions discussed in preceding sections. How did these assumptions <br />affect the results? Specifically, how well did st;orms identified from <br />interpolated sourXIings coincide with storms that actually occurred during <br />the 13 winters? -The data required for a direct answer are not available. <br />We attempted an indirect answer. <br /> <br />This entailed comparing the number of storms each winter to total <br />seasonal snowfall at. each basin estimated by maximum spring snowcourse <br />water content. An association between the number of storms each winter <br />and observed water content would lend credence to and validate our use of <br />interpolated soundings. <br /> <br />Actually, we refined our validation procedure by using a computer <br />model designed to 'calculate orographic precipitation over Colorado (Rhea, <br />1978). The model, developed by Dr. Owen Rhea as part of his dissertation <br />in atmospheric sciences at Colorado State University, is used by the <br />National Weather Service and U.S. Forest Service Avalanche Warning Center <br />to predict mountain snowfall. It employs a topographic map of Colorado <br />and the physics and meteorology of airflow over mountain barriers to <br />"predict" snowfall in various catchment areas. Rhea tested his model by <br />correlating seasonally summed model precipitation with April 1 snowcourse <br />" <br />water content. Correlations ranged mainly between .77 and .92 (Rhea, <br />1978: 90) . <br /> <br />23 <br />