Laserfiche WebLink
<br />higher rates is demonstrated for the Cleveland and Denver areas in section 5. Therefore, <br />Fisher-Porter gages have little utility for relating radar observations to hourly snowfall <br />observations. <br /> <br />Even if snowboards are used to manually measure snowfall and are set flush with the snow <br />or ground surface, windy sites can result in drifting of additional snow onto them or scouring <br />of snow off of them. Windy sites must be avoided for quantitative snowfall measurements. <br /> <br />Because of the problems noted and others discussed by Groisman and Legates (1994), the <br />existing national precipitation gage network was determined at the onset to be inadequate <br />for the purposes of this study. Most climatological gages are read daily and, therefore, do not <br />have the needed time resolution. Many recording gages with hourly time resolution do not <br />have the desired mass resolution for snowfall (O.OI-inch melted water equivalent or less). <br />Tipping bucket gages usually can resolve 0.01 inch of water, but most are unheated so they <br />cannot measure snowfall with any reasonable accuracy. And even where gages with adequate <br />mass resolution exist, they tend to be located near WFOs, typically at wide-open, windy <br />airports. The undercatch of such gages can be serious and unknown in magnitude. Snowfall <br />measurements from such locations can add considerable variability to attempts to relate <br />snowfall accumulation to radar observations. <br /> <br />Small clearings in widespread conifer forest generally provide excellent snowfall <br />measurement. sites (e.g., Brown and Peck, 1962). Clearings in thick deciduous forest, <br />especially if low brush is common, can also provide well-protected snowfall observing sites. <br />Such forest clearings, together with gage wind shields, can almost eliminate gage undercatch <br />caused by airflow around the gage orifice. Gages installed and operated for this study were <br />placed in forest clearings wherever possible. Of course, most of the U.S. does not have <br />widespread forest, and alternatives needed to be found for protecting gages from the wind in <br />the absence of forest. As will be discussed, different approaches to attempting to solve the <br />snowfall measurement problem were. taken at each of at the three measurement areas <br />(Albany, Cleveland, a~d Denver) during the 1995-96 winter. <br /> <br />It is desirable to locate snow observing sites intended for Ze-S comparisons as near the radar <br />as practical. Such locations minimize the vertical distance between the radar beam and the <br />surface, which reduces un~ertainties caused by wind advection of snow particles. In addition, <br />the volume sampled by the radar increases with range as the beam broadens in width and <br />height. So the representativeness of a surface point observation for the overlying range bin <br />or bins becomes increasingly uncertain at more distant ranges. However, as a practical <br />matter, these factors must be weighed against usually greater ground clutter contamination <br />near the radar and whether suitable surface observing sites exist near the radar. The <br />tradeoffs involved in selecting a snow observing site near a radar are perhaps best illustrated <br />in the discussion of site selection in the Denver area in section 3.3. <br /> <br />3.2. Albany, New York, Observations <br /> <br />Unlike the other two sites, Reclamation had a limited role in data collection within range of <br />the Albany, New York, WSR-88D. Reclamation supplied one Belfort gage with an Alter wind <br />shield. John Quinlan, a forecaster at the Albany WFO, supplied another Belfort gage, and <br />Reclamation provided an Alter shield for that gage as well. These two gages, noted in table <br />1, were installed by Albany WFO personnel. Reclamation also supported computer data entry <br />of volunteer observations by university students.p <br />4 <br />