214
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
10.1 INTRODUCTION
Snow comprises only about 5% of all precipitation reaching the earth’s surface (Hall
and Martinec 1985), but has a great impact on the earth’s energy balance due to
its high albedo and low thermal conductivity (Hall and Riggs 2007). According to
Brooks et al. (2003), a third of the water used for irrigation in the world comes from
snowpack and its subsequent melt (p. 373), 50%–90% of the yearly precipitation and
runoff in Arctic regions comes from snow fall (König 2001), and much of the water
supply used for domestic purposes originates as snowpack, particularly in mountainous areas throughout the world (Hall and Riggs 2007). Water accumulated and
stored as snow, therefore, forms an important component of the hydrologic cycle in
many regions of the world (Parajka and Blöschl 2008b).
Notwithstanding the importance of snow and the resulting melt, several studies
have indicated a reduction in annual snowpack accumulation that has been occurring
over the past half-century. For example, snowpack in the mountains of the western
United States has declined as a result of a warming climate from its value in 1950
(Mote 2003, 2006; Cayan et al. 2008; Barnett et al. 2005; Day 2009). A decline in
snow cover extent has been indicated by satellite measurements since 1966 (Robinson
1999). Some have attributed the reduction in snowpack accumulation of the western
United States to a shift in winter precipitation from snow toward rain (Van Kirk and
Naman 2008; Mote 2006; Regonda et al. 2005); this is supported by the reports of
others who have observed a shift in the timing of snowmelt runoff toward earlier in
the water year (Gillan et al. 2010; Stewart et al. 2004; Van Kirk and Naman 2008).
Extrapolating the trend of warming climate, Barnett et al. (2005) projected that the
western U.S. spring stream-flow maximum will come about 1 month earlier by the
year 2050.
The reduction in snowpack accumulation, acceleration of melt, and the observed
and projected shift to earlier timing of spring runoff have led to an increased interest
in the use of available snow cover information to model snowmelt runoff processes
associated with climate change scenarios. In response to this heightened interest, the
aims of this chapter are to (1) provide a review of the present state of snowmelt runoff
modeling and remote sensing of snow for use in snowmelt runoff models (SRMs),
(2) present a technique for combining ground-based snow data with remote sensing
to generate snow-covered area (SCA) depletion curves, and (3) describe a case study
of snowmelt runoff modeling for climate change scenarios with respect to changing
temperature and precipitation patterns.
10.2 SNOWMELT MODELING
Most models of snowmelt runoff consist of two components: a snowmelt model,
which simulates the process of snow accumulation and melting, and a transformation model, which takes the snowmelt or the rainfall as input data and yields the
basin runoff as output (WMO 1986). Undoubtedly, spatially distributed hydrologic
models of snow-dominated areas must incorporate a snowmelt component owing to
the significance of snow to the hydrologic cycle of those areas (Garen and Marks
2005). Because snowmelt is a primary water input to the soil and stream system, melt
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
10.1 INTRODUCTION
Snow comprises only about 5% of all precipitation reaching the earth’s surface (Hall
and Martinec 1985), but has a great impact on the earth’s energy balance due to
its high albedo and low thermal conductivity (Hall and Riggs 2007). According to
Brooks et al. (2003), a third of the water used for irrigation in the world comes from
snowpack and its subsequent melt (p. 373), 50%–90% of the yearly precipitation and
runoff in Arctic regions comes from snow fall (König 2001), and much of the water
supply used for domestic purposes originates as snowpack, particularly in mountainous areas throughout the world (Hall and Riggs 2007). Water accumulated and
stored as snow, therefore, forms an important component of the hydrologic cycle in
many regions of the world (Parajka and Blöschl 2008b).
Notwithstanding the importance of snow and the resulting melt, several studies
have indicated a reduction in annual snowpack accumulation that has been occurring
over the past half-century. For example, snowpack in the mountains of the western
United States has declined as a result of a warming climate from its value in 1950
(Mote 2003, 2006; Cayan et al. 2008; Barnett et al. 2005; Day 2009). A decline in
snow cover extent has been indicated by satellite measurements since 1966 (Robinson
1999). Some have attributed the reduction in snowpack accumulation of the western
United States to a shift in winter precipitation from snow toward rain (Van Kirk and
Naman 2008; Mote 2006; Regonda et al. 2005); this is supported by the reports of
others who have observed a shift in the timing of snowmelt runoff toward earlier in
the water year (Gillan et al. 2010; Stewart et al. 2004; Van Kirk and Naman 2008).
Extrapolating the trend of warming climate, Barnett et al. (2005) projected that the
western U.S. spring stream-flow maximum will come about 1 month earlier by the
year 2050.
The reduction in snowpack accumulation, acceleration of melt, and the observed
and projected shift to earlier timing of spring runoff have led to an increased interest
in the use of available snow cover information to model snowmelt runoff processes
associated with climate change scenarios. In response to this heightened interest, the
aims of this chapter are to (1) provide a review of the present state of snowmelt runoff
modeling and remote sensing of snow for use in snowmelt runoff models (SRMs),
(2) present a technique for combining ground-based snow data with remote sensing
to generate snow-covered area (SCA) depletion curves, and (3) describe a case study
of snowmelt runoff modeling for climate change scenarios with respect to changing
temperature and precipitation patterns.
10.2 SNOWMELT MODELING
Most models of snowmelt runoff consist of two components: a snowmelt model,
which simulates the process of snow accumulation and melting, and a transformation model, which takes the snowmelt or the rainfall as input data and yields the
basin runoff as output (WMO 1986). Undoubtedly, spatially distributed hydrologic
models of snow-dominated areas must incorporate a snowmelt component owing to
the significance of snow to the hydrologic cycle of those areas (Garen and Marks
2005). Because snowmelt is a primary water input to the soil and stream system, melt