Performance of the snowmelt runoff model when remotely-sensed estimates of snow covered area are not available

TitlePerformance of the snowmelt runoff model when remotely-sensed estimates of snow covered area are not available
Publication TypeConference Paper
Year of Publication2010
AuthorsSteele C, Rango A.
Conference NameAmerican Geophysical Union
Date Published12/2010
PublisherAmerican Geophysical Union
Conference LocationSan Francisco, CA
ARIS Log Number274370
AbstractThe Snowmelt Runoff Model (SRM) is usually run with snow cover depletion data. These daily depletion data can be derived through interpolation of periodic, remotely sensed measurements of basin snow covered area (SCA). It is also possible to run SRM successfully without snow depletion data in “no snow cover” mode. In this mode, accumulated winter precipitation is carried forward into the melt season. SRM will start to melt the accumulated winter precipitation incrementally when a degree-day threshold is passed. It is these data that are used in place of the daily snow cover depletion data.
We compared results from SRM run in three modes of operation: (i) “no snow cover” mode, (ii) with depletion curves interpolated from SCA estimated using fine resolution satellite sensor data (30 m), and (iii) with depletion curves interpolated from SCA estimated using coarse resolution satellite sensor data (500 m). Comparisons were made for an average snow year (2001), a heavy snow year (2008) and a low snow year (2002) over 3 watersheds in the Upper Rio Grande basin. In this paper, we discuss the reasons why estimates of runoff differ between the three modes of operation. Factors that may contribute to the variation in the results include the quantity of precipitation preceding the melt season, basin size, basin elevation, proximity of climate stations and the quality of remotely sensed SCA data.
These results have relevance for water managers. If reasonable estimates of runoff from snowmelt can be obtained in “no snow cover” mode, the usability of SRM is improved for the non-expert in remote sensing or when remotely-sensed SCA data are unreliable (e.g., in years of extreme cloudiness).