Laserfiche WebLink
<br />~ -~ <br /> <br />2. THE FIRST ANNUAL REPORT <br /> <br />Reclamation provided a research report to the OSF, R-96-04, dated June 1996. The 133-page report is <br />titled, "Snow Accumulation Algorithm for the WSR -88D Radar , Version 1." The report was written by <br />Arlin B. Super and Edmond W. Holroyd of Reclamation, except for appendix A, which was written by <br />James A. Heimbach of the University of North Carolina at Asheville. For general information, the <br />report's abstract is noted below: <br /> <br />This annual Bureau of Reclamation report describes the initial year of a 3-year effort to develop <br />a Snow Accumulation Algorithm for the new network of WSR-88D (NEXRAD) radars.. <br />Snowfall measurements were made during the 1995-96 winter/spring within range ofWSR-88Ds <br />at Albany, NY, Cleveland, OH, and Denver, CO. Observations of S (snow water equivalent) <br />from the latter two locations were related to Ze (effective reflectivity factor) measurements to <br />determine "best fit" << and f3 coefficients for the commonly-used equation Ze = aSp. <br />Recommended <<and f3 yalues are 318 and 1.5 for Cleveland and 155 and 1.6 for Denver. <br /> <br />Observations near Lake Erie revealed a significant range effect. The Cleveland radar <br />underestimated snowfall beyond about 60 km. Radar-estimated snowfall accumulations were <br />about 85, 61, 31, and 22percent of gage observations at 61,87, 115, and 146 km, respectively. <br />A scheme should be developed to use the vertical profile of Ze to adjust for mid- and far-range <br />snowfall underestimates by WSR-88Ds. <br /> <br />Initial snow accumulation algorithm code is described and presented. The code development <br />borrows from the current NEXRAD rainfall algorithm but makes several important modifications, <br />including an advection scheme, because snowflakes can be transported tens of kilometers from <br />the lowest tilt radar beam to the ground. <br /> <br />3. DETERMINATION OF Ze-S RELATIONSHIPS <br /> <br />3.1 General <br /> <br />Two fundamentally different approaches have been used over the years to relate radar measurements of <br />returned power to precipitation as further discussed by Super and Holroyd (1996). One approach uses <br />the summation of the sixth power of drop diameters (equivalent melted drop diameters for snowfall) to <br />calculate radar reflectivity f!ictor, Z, for a given volume. The quantity Z is then related to precipitation <br />intensity, which can be calculated from the same drop observations or otherwise measured. The other <br />approach relates. the equiv~lent reflectivity factor, Ze, measured by radar to precipitation observed at <br />ground level. <br /> <br />Both approaches have uncertainties. Steiner and Houze (1997) note that the relation between Z and <br />precipitation rate may vary significantly between and within storms. They point out that size sorting and <br />growth or evaporation between the radar beam and the ground can create significant differences between <br />what a radar measures, Ze' and calculations of Z. Estimating Z is especially challenging with snow <br />because of uncertainties in fall speeds and masses of large, fragile snowflakes and aggregates. <br /> <br />Reclamation is using the second approach of relating Ze measured by WSR-88D radars to observed <br />snowfall in its SAA development. Part of this work involves deciding upon the appropriate empirically- <br />determined coefficient <<and exponent f3 for the commonly used power law model: <br /> <br />3 <br />