|
a }
<br /> It seems unlikely that two years of SNODAS an in-depth examination, however, in a recent Colo-
<br /> data would convey the long-term variance of precipi- rado seeding feasibility study (Super and Heimbach
<br /> tation across the Colorado River Basin,even if those 2005).
<br /> two years exhibited a large variation about the mean
<br /> in precipitation amounts. Nevertheless, SNODAS 4. CONCLUSIONS
<br /> SWEs at the 16 Basin SNOTEL locations(see previ-
<br /> ous section) were compared to the 30-year (1971- Within the limitations of the SNODAS data set
<br /> 2000) SNOTEL SWE averages. The 2004 SNODAS and stated uncertainties of our calculations we esti-
<br /> SWE mean for all sites was 12.5 inches and the 2005 mate that, for an average precipitation year in the
<br /> mean was 20.5 inches.The 30-year average SNOTEL Colorado River Basin, cloud seeding could generate
<br /> SWE for the sites is 17.9 inches, intermediate to the an additional one million acre feet of water storage
<br /> SNODAS SWE for the two years, and close to their in the basin-wide snowpack. In drought years, seed-
<br /> mean of 16.5 inches. This calculation lends confi- ing might produce about half that amount,or 500,000
<br /> dence that the target area-integrated means of the acre feet. In wet years, more than one million acre
<br /> two-year SNODAS data represent a climatologically feet could be produced,but how much more would
<br /> average year. These means are presented in the right- be limited by seeding suspension criteria. These es-
<br /> most column of Table 4. timates are close to those of two older studies. There-
<br /> fore, application of a modern, sophisticated snow
<br /> It is instructive to compare these means with modeling and assimilation system has produced simi-
<br /> those of Table 1. The wet year(2005) is close to the lar water yields as the older studies and gives confi-
<br /> Stanford Research and Twelve Basin liberal figures, dence that such yields are representative.
<br /> whereas the mean is very close to the Twelve Basin
<br /> conservative value.Using half of the Stanford figures Acknowledgments. Many thanks to Ra Aman and
<br /> again,that report states that"One year out of three,it Steffen Meyer for programming and GIS support.
<br /> [the yield] might be either lower than 550,000 or This project was sponsored by a grant from the Colo-
<br /> higher than 1,800,000 acre-feet." The dry snow ac- rado Water Conservation Board(CWCB) entitled "A
<br /> cumulation year ending April 1,2004 is very close to Winter Storm Climatology,"and was administered by
<br /> the low figure of the Stanford study. The high figure Reclamation Technical Service Center account num
<br /> of that study seems optimistic, especially since both ber A1R175893350010000. We thank Joe Busto of
<br /> Colorado and Utah suspend seeding operations when the CWCB for his technical assistance and guidance.
<br /> snowpack SWE exceeds certain percentages of nor- Don Griffith of North American Weather Consultants
<br /> mal.If one attempts to account for seasonal snowmelt and Larry Hjermstad of Western Weather Consult-
<br /> (see previous section), the mean yield at lower right ants provided valuable insights, as did Drs. Arlin
<br /> of Table 4 might be near 1 million acre-feet. This Super and Jim Heimbach. Thanks to Todd Adams of
<br /> figure is close to that from the Stanford study and is Utah Dept.of Water Resources and HDR Engineer-
<br /> intermediate to the conservative and liberal Twelve ing of Denver for GIS data of existing seeding target
<br /> Basin values. This amount,of water is significant for areas.
<br /> the Colorado River basin water balance; for example,
<br /> it is two-thirds of the 1.5 million acre-feet of the 5. REFERENCES
<br /> river's annual flow that is legally obligated to Mex-
<br /> ico. American Meteorological Society(AMS), 1998:Pol-
<br /> These values should be considered approximate. icy Statement,Planned and Inadvertent Weather
<br /> Since they compare favorably to the estimates of two Modification.Bulletin American Meteorological So-
<br /> earlier studies,however,we have more confidence in ciety, 79, 2771-2772. Online at
<br /> them. There are many variables in determining the hqp://www.ametsoc.orgZpolicy/wxmod98.html
<br /> effectiveness of seeding,which could lead to substan-
<br /> tial deviations from the assumed 10% augmentation Carroll,Tom,Don Cline,Greg Fall,Anders Nilsson,
<br /> used herein. Besides the choice of target areas, there Long Li,and Andy Rost,2001:NOHRSC Operations
<br /> are natural hydrologic and meteorological variables. And The Simulation Of Snow Cover Properties For
<br /> Then there are those associated with seeding meth- The Coterminous U.S. 69th Annual Meeting of the
<br /> ods.For those methods to be effective, seeding mate- Western Snow Conference. Online at
<br /> rials must be dispersed in sufficient concentration in http://www.nohrsc.noaa.gov/technology/pdf/wsc200l
<br /> cloud regions with adequate supercooled liquid water pddf
<br /> and temperatures cold enough for the seeding materi-
<br /> als to function as intended. An examination of these Cline,Don,Long Li,and Nick Rutter,2004: 1D
<br /> variables is beyond the scope of this study. There is Validation of SNODAS Model Physics.National
<br />
|