WMOD00429 - Laserfiche WebLink

Home Browse Search

di recti.on and ai r temperature were a1 so obtai ned for compari son. The corre lati on coeffi c i ents showed vr~ry good agreement among all the sensors. However, regression eqwitions indicated significant differences between sensors with the "'truell snowfall assumed to be the average snow board measurement. The snow pi 11 owreadi ngs averaged 5 pl~rcent 1 ess than measured by the snow boards. The Sl10W pillow also appeared to 1I1ag" the snow boards when compared on a daily basis during a storm lasting several days. The heated tipping bucket also caught considerably less snowfall than the snow boards. Three possible reasons were given: (1) wind errors common to all gages above ground level, (2) evapor'a- tion and sublimation losses due to heating and melting of snow, and (3) losses due to snowflakes being deflected by the warm air rising above the gage orifice. Excellent agreement was found between the snow boards and the shielded weighing bucket gage fitted ~,ith a 287-mm (11-5/16-in) diameter orifice [double the orifice Cilrea of a standard 200-mm (8-in) gage] as shown in figure 3. Although the orifice size and shape were not hi gh1y important in ter'ms of absolute measurement, the larger orifice did increase the re!solution of precipitation measurements and prevented gage capping. A network of 28 to 33 large-orifice, shielded weighing bucket precipitation gages was operated on and downwi nd of the Sri dge!r Range duri ng the 1969 through 1972 field programs. The gage's worked reliably and during the course of the program, about 96 percent of all possible precipitation data were obtained. A 3-cm weather radar was also used on the Bridger Range Program for several purposes, one of which was to obtain an estimate of precipita- tion. The info'rmation necessary for the measurement of snowfall by radar are the measurement of range and returned power, a computed reflectivity factor, and a relationship linking reflectiVity to rate of snowfall. The calculated regression equation was Z = 2490 S1.44 in which Z is the reflectivity factor and S is the precipitation rate. Several studies were conducted to relate snowfall to the storm environment. They indicated that radar-estimated snowfall amounts were reasonable, but the natural variability of snowfall in time and space is considerable even in simple orographic storms. The attempts to directly detect ground-based seeding effects were not successful. However, a conclusion of the Montana State University report was that the added capabil i ty of a di gi tal data logger mi ght increase the opportunity for success. 3