WMOD00511 - Laserfiche WebLink

Home Browse Search

3. A snowboard was periodically placed on the snow surface within 10 ft of the gauges. A 3-13/16-in. I.D. snow coring tube with sharpened edge was pushed vertically through the snow to obtain samples from the board. The resulting snow samples were placed in preweighed plastic bags and later weighed on a balance with resolution of 0.1 gram. Ten snowboard observations were collected during the test period. In addition, a sensitive R.M. Young propeller anemometer/wind vane sensor was located near the gauges at the height of their orifices. Average wind speed and direction were recorded each 15 min on the same data logger used to record the ETI gauge data. RELIABILITY OF ETI PRECIPITATION GAUGE The ETI precipitation gauge proved to be reliable from time of installation in late November 1992 until removal in late March 1993. The only problem encountered had nothing to do with the gauge. The limited solar radiation in the small clearing was not sufficient to keep the battery charged which powered the data logger and radio. The solution was to move the solar panel to a location which received more sunlight. Otherwise, the gauge functioned as it was designed to. The only servicing required was periodic replenishment of the antifreeze solution and removal of the old precipitation/antifreeze mix. The frequency of servicing depended upon snowfall accumulation, but the reservoirs were adequate so that several storms could be sampled before servicing was required. ANALYSIS Precipitation gauge data were totaled and wind speed data were averaged for each hour. Wind direction observations are of little interest because of the light and variable nature of the wind in the test clearing. The highest hourly mean wind speed observed during the 179 h with snowfall measured by one or both gauges was 3.1 mph. Only 6 h had average wind speeds exceeding 1 mph. The average -for most hours was calm, so wind-related gauge losses should have been negligible. Forty-three of the 179 h with snowfall detected by one or both gauges had no snowfall detected by the ETI gauge but some detected by the Belfort. During 36 of these hours the Belfort detected 0.005 in., its resolution limit, but less than the 0.01 in. resolution of the ETI gauge. The maximum detected by the Belfort when no snowfall was detected by the ETI gauge was 0.015 in. The higher resolution Belfort did not detect snowfall during 28 h when some was recorded by the ETI gauge. The ETI gauge indicated 0.03 in. during one of these hours, 0.02 in. during another and 0.01 during the remaining 26 h. Twenty-three of these 28 h were between 0900-1500, when diurnal warming is expected. The absence of snowfall during several of these hours was verified by the field technicians' weather observations. This observation suggests that the temperature compensation was not quite adequate for this load cell gauge, allowing it to falsely indicate minor snowfall amounts during some periods. Load cells are known to be temperature sensitive. Figure 1 shows a plot of all hourly precipitation observations with snowfall detected by either gauge. Agreement is quite good with a linear correlation coefficient (R) of 0.97. The slope of the linear regression equation indicates that overall the ETI gauge tended to indicate slightly lower values than the Belfort. 2