Abstract

Spring snowpack is an important water resource in many river basins in the United States in areas where snowmelt comprises a large part of the annual runoff. Increasing temperatures will likely reduce snowpacks in the future, resulting in more winter runoff and less available water during the summer low-flow season. As part of the National Climate Change Modeling Project by the U.S. Geological Survey, distributed watershed-model output was analyzed to characterize areal extent and water-equivalent volumes of spring snowpack for a warming climate. The output from seven selected watershed models from the mountainous western United States and one model from coastal Maine in the northeastern United States shows a future of declining spring snowpack. Snow-cover area (SCA) and snow-water equivalent (SWE) were used to compare the spring snowpack for current conditions (2006) with three time periods in the future (2030, 2060, and 2090) using three Intergovernmental Panel on Climate Change (IPCC) emission scenarios published in the 2007 Special Report on Emission Scenarios (SRES): A2, B1, and A1B. Distributed SWE and SCA values were sorted into elevation zones in each basin. The change in spring snowpack over time was greater than the change among different emission scenarios, suggesting that, even for a globally reduced carbon emission scenario, large decreases in SWE are likely to occur. The SRES A2 scenario resulted in the greatest decrease in SWE for six of the basins, and the SRES B1 and A1B scenarios resulted in the greatest decrease in one basin each.

1. Introduction

In many watersheds in northern and mountainous regions of the United States, spring snowpack is an important feature of the hydrologic system. Accumulated spring snowpack is a crucial water-supply reservoir for irrigation, municipal demands, and hydroelectric power (Barnett et al. 2005). Many studies have linked future climate change to future shrinking snowpacks as a result of predicted increases in air temperature (Jeton et al. 1996; Hamlet and Lettenmaier 1999; Leung and Wigmosta 1999; Barnett et al. 2005; Hayhoe et al. 2006; Mastin 2008). As temperature increases, a greater fraction of precipitation falls as rain rather than snow and more accumulated snowpack melts, resulting in greater runoff in the winter, less snowpack in the spring, and less runoff in the summer.

Observations have shown trends of decreases in snowpack and earlier snowmelt runoff over the last half century in midlatitude mountain basins in the Sierra Nevada and the Cascade Range, with western Canada and coastal Alaska showing the most change (Stewart et al. 2005; Stewart 2009). Recent studies in the northeastern United States have demonstrated strong and consistent evidence of hydrologic changes over the last 30–150 years that is consistent with warming winter–spring air temperatures, including significant changes toward earlier winter/spring snowmelt runoff, decreasing duration of ice on rivers and lakes, decreasing ratio of snowfall to total precipitation, and denser and thinner late-winter snowpack (Dudley and Hodgkins 2002; Hodgkins et al. 2002; Hodgkins et al. 2003; Hodgkins et al. 2005a; Hodgkins et al. 2005b; Huntington et al. 2003; Huntington et al. 2004; Hodgkins and Dudley 2006a; Hodgkins and Dudley 2006b). The area in the Northeast with the strongest and most consistent trends (Maine, New Hampshire, Vermont, and New Brunswick) drains into the Gulf of Maine. Many of the northeastern coastal streams are home to endangered Atlantic salmon. Snowmelt runoff has been identified as a sensitive hydrologic parameter that responds to climate change in New England (Hodgkins et al. 2009). Managers tasked with the protection and recovery of Atlantic salmon in coastal Maine rivers are concerned about the effects of observed trends in winter–spring runoff timing that occur during salmon migration. In the western United States, concerns are similar for Pacific salmon, including the concern about increasing water temperatures that can be expected with increasing air temperature and decreased summer river flows (ISAB 2007).

In general, studies of the effect of climate change on runoff show relatively few or no trends in total annual runoff, but the timing of runoff may be significantly altered (Gleick 1987; Jeton et al. 1996; Dudley and Hodgkins 2002; Hodgkins and Dudley 2005; Mastin 2008). The scenario of increased winter flows, less spring snowpack, and less spring and summer runoff is being considered in long-term planning for reservoir design and water-management strategies (Adeloye et al. 1999; Draper and Kundell 2007), despite incomplete knowledge about the volumetric change in snowpack that can be expected and how snowpack changes may vary regionally and locally.

A key measure of snowpack condition used by resource managers in the western United States is the snow-water equivalent (SWE) on 1 April (Serreze et al. 1999). Researchers have adopted this measure in climate-change studies to characterize the effect of climate change on snowpack (Cayan 1996; Hamlet et al. 2005; McCabe and Wolock 1999). The spring snowpack, as measured by the 1 April SWE, is a function of winter accumulation and ablation. SWE on 1 April is a key indicator of the amount of water available for summer runoff and, in general, is the time when accumulation is greatest (Cayan 1996). Likewise, in the northeastern United States, 1 March is the date when snow accumulation usually is greatest (Hodgkins and Dudley 2006b). The 1 March SWE and depth has been measured at selected snow course sites in Maine since the early part of the twentieth century to aid in emergency-management flood forecasting and reservoir operation management (Hodgkins et al. 2005b). For the remainder of this paper, references to spring snow-covered area (SCA) and SWE refer to values on 1 April for the western basins and 1 March for the Cathance River basin in Maine.

The spring simulated snowpack is examined for three climate-change emission scenarios at eight different snowmelt-dominated basins that are part of the National Climate Change Modeling Project (Hay et al. 2011; referred to herein as the National Project) of the U.S. Geological Survey Global Change Research and Development Program (http://gcp.usgs.gov/).

The National Project considered five general circulation models (GCMs) for the current analysis (Table 1). The GCM output was from the World Climate Research Programme’s Coupled Model Intercomparison Project phase 3 multimodel dataset archive referenced in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report Special Report on Emission Scenarios (SRES) (Alley 2007). For each GCM, one current and three future scenarios were used (Table 2).

Table 1.

GCMs used in this study (Alley et al. 2007). GCM definitions not expanded in the text: Bjerknes Centre for Climate Research Bergen Climate Model (BCCR-BCM2.0); Commonwealth Scientific and Industrial Research Organisation Mark version 3.0 (CSIRO Mk3.0); Institute of Numerical Mathematics Coupled Model, version 3.0 (INM-CM3.0); and Model for Interdisciplinary Research on Climate 3.2, medium-resolution version [MIROC3.2(medres)].

GCMs used in this study (Alley et al. 2007). GCM definitions not expanded in the text: Bjerknes Centre for Climate Research Bergen Climate Model (BCCR-BCM2.0); Commonwealth Scientific and Industrial Research Organisation Mark version 3.0 (CSIRO Mk3.0); Institute of Numerical Mathematics Coupled Model, version 3.0 (INM-CM3.0); and Model for Interdisciplinary Research on Climate 3.2, medium-resolution version [MIROC3.2(medres)].
GCMs used in this study (Alley et al. 2007). GCM definitions not expanded in the text: Bjerknes Centre for Climate Research Bergen Climate Model (BCCR-BCM2.0); Commonwealth Scientific and Industrial Research Organisation Mark version 3.0 (CSIRO Mk3.0); Institute of Numerical Mathematics Coupled Model, version 3.0 (INM-CM3.0); and Model for Interdisciplinary Research on Climate 3.2, medium-resolution version [MIROC3.2(medres)].
Table 2.

GCM future scenarios used in this study.

GCM future scenarios used in this study.
GCM future scenarios used in this study.

2. Study basins

Of the 14 basins analyzed in the National Project, 8 basins where snowmelt is an important runoff component were selected for the analysis described in this paper (Figure 1; more detailed color relief maps of the basins in Hay et al. 2011): the Cathance River, East River, Feather River, South Fork (SF) Flathead River, Naches River, Sprague River, Yampa River, and Sagehen Creek basins. Basin relief as measured from the mean elevations of hydrologic response units (HRUs; land units in the watershed models) ranged from 1887 m in the Feather River basin, California, to 128 m in the Cathance River basin, Maine. The two basins in Colorado (East and Yampa River basins) have the highest mean HRU elevations, which exceed 3500 m (Table 3).

Figure 1.

Location of the eight study basins in the contiguous United States.

Figure 1.

Location of the eight study basins in the contiguous United States.

Table 3.

Basins selected for analysis in this study.

Basins selected for analysis in this study.
Basins selected for analysis in this study.

3. Data and methods

Runoff and snowpack were simulated for the eight basins using the Precipitation-Runoff Modeling System (PRMS) watershed model (Leavesley et al. 1983; Hay et al. 2011) for all basins. The PRMS simulates hydrologic processes at a daily time step using daily precipitation and daily maximum and minimum temperature as inputs. Spatial distribution of the hydrologic processes is attained by partitioning the basin into HRUs, which are units of land in the basin with similar runoff response to climate forcing. A water and energy balance is computed for each HRU by accounting for surface, soil zone, subsurface, and groundwater storages and fluxes. On each HRU, the model simulates initiation, accumulation, and depletion of a two-layered snowpack. Daily outputs of SWE and SCA for each of the HRUs were stored for each model run.

Climate-change fields (percentage changes in precipitation and degree Fahrenheit changes in temperature) were derived by calculating the change in climate from current to future conditions simulated by each GCM. Current conditions were defined to be the average of conditions during water years 1988–2000. Climate-change fields were computed for 12-yr moving windows for water years 2001–99 and the SRES A2, B1, and A1B scenarios. This approach resulted in 1320 future scenarios [(eighty-eight 12-yr climatologies) × (3 GCM scenarios) × (5 GCMs)], which were represented in PRMS by modifying precipitation and temperature inputs according to the mean monthly climate-change fields derived from the GCMs (Hay et al. 2011).

The analysis was simplified by comparing the spring SCA and SWE for four 12-yr moving windows centered on 2006, 2030, 2060, and 2090 for the three climate-change scenarios (Table 2), with the 2006 moving window as the base period for comparison. For each HRU, spring SCA and SWE from each of the five GCMs (Table 1) were averaged over the 12-yr period, and those five averages were then averaged again to get a single SCA and SWE estimate. Performing these calculations for each HRU provided the spatial distribution of the spring SWE and SCA in each basin. The SCA for an HRU was evaluated as either “1” for a snow-covered HRU or “0” for a bare HRU; after averaging, if the value for SCA was less than 0.50, the HRU was considered to be bare of snow. Totals from all HRUs in a basin provided an areal comparison of SCA and a volumetric comparison of SWE for three time periods within each scenario relative to the 12-yr window centered on 2006.

The SCA and SWE also were aggregated into five elevation zones in each basin to illustrate the spatial differences with respect to elevation. The five elevation zones with equal elevation change between zones were defined based on the range of the mean HRU elevations in each basin. The averaged SCA and SWE values were computed for each elevation zone and for each basin.

4. Results and discussion

An example of the 1 April SCA and SWE results by elevation zones and the entire basin is shown for the East River basin (Table 4). The 12-yr, five GCM average SCA and SWE for the simulated snowpack for the climatology centered on years 2006, 2030, 2060, and 2090 for the three climate-change scenarios are presented. Tables 5 and 6 list the changes in SCA and SWE since 2006 for all basins.

Table 4.

The 1 Apr simulated SCA and SWE for the selected 12-yr moving windows and three climate-change scenarios, East River basin, Colorado. SCA is in hectares; SWE is in thousands of cubic meters; and SCA is simulated in each HRU as snow covered (value = 1) or bare (value = 0). If the average SCA of the five GCM models was less than 0.5, the HRU was considered bare, but some snow could be present.

The 1 Apr simulated SCA and SWE for the selected 12-yr moving windows and three climate-change scenarios, East River basin, Colorado. SCA is in hectares; SWE is in thousands of cubic meters; and SCA is simulated in each HRU as snow covered (value = 1) or bare (value = 0). If the average SCA of the five GCM models was less than 0.5, the HRU was considered bare, but some snow could be present.
The 1 Apr simulated SCA and SWE for the selected 12-yr moving windows and three climate-change scenarios, East River basin, Colorado. SCA is in hectares; SWE is in thousands of cubic meters; and SCA is simulated in each HRU as snow covered (value = 1) or bare (value = 0). If the average SCA of the five GCM models was less than 0.5, the HRU was considered bare, but some snow could be present.
Table 5.

Change in simulated spring SCA (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.

Change in simulated spring SCA (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.
Change in simulated spring SCA (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.
Table 6.

Change in simulated spring SWE (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.

Change in simulated spring SWE (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.
Change in simulated spring SWE (1 Apr for basins in western United States and 1 Mar for Cathance River basin in Maine) since 2006 for three time periods and three climate-change scenarios. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.

Changes in simulated SCA for different climate-change scenarios varied substantially between basins (Table 5). The SCA in the Cathance River basin showed the least amount of change over time, indicating that this coastal basin is projected to remain largely snow covered around 1 March of every year. Two of the Rocky Mountain river basins, East River and South Fork Flathead River, showed similar small changes in SCA over time, and the other high-altitude Rocky Mountain river basin, the Yampa River, showed moderate changes in spring snowpack that were more than double the changes in the East River or South Fork Flathead River basins. The greatest change in SCA was in the Feather River basin in California, the basin with the largest relief. For any of the three scenarios, the Feather River basin is estimated to have about 70%–80% less snow cover by 2090 compared to 2006. The three basins with the largest change in SCA (Feather, Naches, and Sprague River) are all in the Sierra Nevada and Cascade Range near the Pacific Coast. A large percentage of land in each of these basins is in a general midlatitude rain/snow transition zone of 1000–2000-m elevation (Marks and Winstral 2007). The Sagehen Creek basin also is located in the Sierra Nevada but showed only moderate change in SCA. The average elevation of the Sagehen Creek basin is higher than that of the nearby Feather River basin, and most of the basin is above the transition zone.

The large change in SCA for the Naches River basin can be seen in maps of SCA over time (Figure 2). The Naches River basin is on the lee side of the Cascade Range, and the highest elevations of the basin are along its western margin, which continues to maintain a spring snowpack even for warmer 2090 conditions. The lower, arid parts on the eastern side of the basin, as well as the valley bottoms in the center of the basin, did not have a spring snowpack even in 2006; thus, the greatest change in SCA occurs in the middle, mid-elevation portion of the basin on the valley slopes and ridge tops.

Figure 2.

Simulated SCA for the SRES A1B climate-change scenario for four time periods, Naches River basin, Washington.

Figure 2.

Simulated SCA for the SRES A1B climate-change scenario for four time periods, Naches River basin, Washington.

Changes in simulated SWE for the different climate-change scenarios expressed as a percentage typically were much greater than those in SCA (Table 6), indicating a potentially substantial reduction in water available for summer runoff and water-use demands. By 2090, SWE decreased by 12.9% to 81.1% with a median decrease of 47.7%. Although the simulated 2090 SCA for the SRES A1B scenario for the Cathance River basin decreased 5.9% since 2006, the decrease in SWE was 70.6% for the same scenario and 79.1% for the SRES A2 scenario. The high-elevation basins, the East River basin, and the interior South Fork Flathead River basin had some resistance to change as they showed the smallest decrease in SWE by 2090, although the range of decreases from 12.9% to 42.1% was substantial. Most of the change in the SWE volume of the spring snowpack occurred for the SRES A2 scenario (Cathance, East, South Fork Flathead, Yampa, Naches, and Feather River basins). The greatest decrease in SWE for the Sprague River basin was for the SRES B1 scenario, and the greatest decrease in SWE for the Sagehen Creek basin was for the SRES A1B scenario. The Sprague River basin showed the greatest decrease of all the basins, 81.1% by 2090 for the SRES B1 scenario. The SWE decreases for the Sprague River basin were substantial (70.3%–81.1%) for all three scenarios. The evolution of change in the 1 April snow distribution for the three scenarios can be seen in the South Fork Flathead River basin over time (Figure 3). The change is greatest between time periods rather than between scenarios, suggesting that, even for a globally reduced carbon emission scenario such as SRES A1B, large decreases in SWE are likely to occur.

Figure 3.

Patterns of simulated SWE for three climate-change scenarios and four different time periods, SF Flathead River basin, Montana.

Figure 3.

Patterns of simulated SWE for three climate-change scenarios and four different time periods, SF Flathead River basin, Montana.

Although the watershed models indicate large reductions in spring SWE, changes in total mean annual discharge for all of the basins expressed as a percentage of change since 2006 were smaller. In a similar manner as the SWE analysis, discharge was averaged over the 12-yr simulations and averaged again for the five GCMs (Table 7). Changes in mean annual discharge between 2006 and 2090 for the three scenarios ranged from a decrease of 25.0% to an increase of 30.1%. The average change was a decrease of 0.6%, and the average absolute change was 10.1%. The relatively small changes in mean annual discharge in these basins indicate a seasonal redistribution of water in the basins where more discharge is expected to occur throughout the winter with less accumulation of snowpack. With less spring snowpack in the future, we can expect lower spring streamflows associated with snowmelt runoff compared to historical conditions.

Table 7.

Mean annual discharge and change in mean annual discharge for three climate-change scenarios centered at 2006 and 2090. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.

Mean annual discharge and change in mean annual discharge for three climate-change scenarios centered at 2006 and 2090. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.
Mean annual discharge and change in mean annual discharge for three climate-change scenarios centered at 2006 and 2090. Values represent averages based on five GCMs. The year in the table is the central year for a 12-yr moving window.

The changes in SWE since 2006 in the five elevation zones displayed variations in spatial patterns between basins. The Cathance River basin, because of its small relief, had almost no change in SWE across elevation zones (Table 3; Figure 4); from the mid-elevation of the lowest elevation zone (46–50 m) to the mid-elevation of the highest elevation zone (170–174 m) is only 124 m. Simulated SWE for the Yampa River basin was the most sensitive to elevation, with the smallest SWE decreases at the highest elevation zone (4%–22% between 2006 and 2090) and the largest decreases in the lowest elevation zone (82%–95% for 2090). Elevation-related patterns of change partly are a function of the amount of relief in the basin, but the Feather River basin, which has the largest relief (more than 10 times that of the Cathance River basin), only had moderate changes in SWE between elevation zones.

Figure 4.

Changes in simulated SWE for the eight study basins by climate-change scenario, time period, and elevation zone, 1 Apr (1 Mar for Cathance River basin, Maine).

Figure 4.

Changes in simulated SWE for the eight study basins by climate-change scenario, time period, and elevation zone, 1 Apr (1 Mar for Cathance River basin, Maine).

This study reemphasizes some of the current thought on the effects of climate change on snow hydrology previously documented in other studies (Hamlet and Lettenmaier 1999; McCabe and Wolock 1999; Hayhoe et al. 2006; Stewart 2009). In addition, it offers the perspective of an analysis of SCA and SWE projections through 2099 for multiple snow-dominated basins across the contiguous United States that were made using a consistent methodology. Notably, the expected future increases in air temperature will cause some storms to precipitate rain instead of snow compared to current conditions, thereby increasing winter runoff, decreasing spring snowpacks, and decreasing summer flows. The results show that the largest decreases in snowpack are expected to occur in mountain watersheds located in the rain/snow transition zone, as demonstrated by the simulations of selected basins in the Cascade Range and Sierra Nevada. If we think of the spring snowpack as a reservoir for water supply for spring and summer instream flows and irrigation needs, water managers will need to supplement the current storage capacity in affected basins by building new reservoirs and/or decreasing the amount of reservoir storage dedicated to reducing flood peak discharges to meet future water demands at current levels. For example, in the Naches River basin, which is a basin projected to have a moderate decrease in SWE, the estimated decrease in 1 April SWE (average of three emission scenarios) is 29.2% by 2060, which equates to a decrease of 179 000 000 m3 of water or about 62.5% of the current total capacity of the two major reservoirs in the basin. By 2090, the estimated decrease in the 1 April SWE is 112.7% of the current total reservoir capacity.

The three emission scenarios simulated by the GCMs result in significantly different projected temperature increases by the end of the twenty-first century, including a best estimate of 1.8°C for the SRES B1 scenario, 2.8°C for the SRES A1B scenario, and 3.4°C for the SRES A2 scenario (Alley et al. 2007). However, these significantly different temperature projections do not result in corresponding significantly different decreases in spring snowpack accumulations. For example, the largest decrease in spring SWE between 2006 and 2090 for all simulated basins was 81.1% in the Sprague River basin for the SRES B1 scenario, with projected decreases of 70.3% and 74.0% for the other two scenarios. Thus, the maximum difference between the three scenarios in projected decreases for spring SWE in the Sprague River basin by 2090 is 10.8% since 2006. The maximum difference between the three scenarios in projected decreases among all simulated basins and time periods is in the Cathance River basin, where by 2090 the maximum difference in projected decreases in spring SWE is 27.3% since 2006. The pattern of a limited range in projected spring snowpack decreases for different scenarios was consistent for all basins. In addition, no one scenario consistently projected the largest snowpack declines. The analysis described in this paper expresses changes in spring snowpack as a percentage of change since 2006, and even small percentages of change may represent large volumes of water. The results represent the central tendencies of the GCM models and care should be taken in the interpretation of the results in light of the large range of uncertainty. There is uncertainty in the GCMs, the emission scenarios, the downscaling of the GCM results, and the watershed-model representation of the hydrologic processes (for more details on the uncertainty, see Hay et al. 2011).

5. Conclusions

All eight basins that were analyzed showed large decreases in the spring snowpack with snowpack SWE estimates decreasing from 12.1% to 81.1% between 2006 and 2090. The decreases varied by climate-change scenario, basin, and elevation. Output from the distributed watershed models provided the areal distribution and SWE of snowpack by elevation. The low-elevation Cathance River basin in Maine showed almost no change in SCA over time (decreases between 2006 and 2090 were 1.0% for the SRES A2 and B1 scenarios and 5.9% for SRES A1B scenario), and the high-elevation East River basin in Colorado showed only moderate decreases (2.0%–14.4% between 2006 and 2090). The largest decreases in SCA were for the three Cascade Range and Sierra Nevada basins in the mid-elevation range or the rain/snow transition zone (47.2%–81.9% between 2006 and 2090). The decrease in SWE since 2006 calculated for five elevation zones showed varying patterns among basins but consistently indicated the largest decreases (expressed as percent change) at lower elevations. Projected decreases in SWE were substantial in the Cathance River basin, which is located near sea level and has limited relief. The high-elevation Yampa River basin had the largest sensitivity to elevation with relatively small projected decreases in SWE at the highest elevations (4%–22% between 2006 and 2090) and large decreases in the lowest elevation zone (82%–95% between 2006 and 2090).

The largest decrease in SWE by 2090 was projected by the SRES A2 scenario for six of the basins, one basin had the largest decrease projected by the SRES B1 scenario, and one by the SRES A1B scenario. The maximum decrease in SWE among the basins, time periods, and scenarios was 81.1% between 2006 and 2090 for the Sprague River basin in Oregon for the SRES B1 scenario, but the other scenarios also resulted in large decreases in SWE in the basin of 70.3% and 74.0%. Future annual-mean discharge was projected to increase and decrease by relatively small percentages, indicating that changes in discharge will be primarily seasonal redistributions of flows in response to shifts in winter precipitation from snow to rain and smaller spring snowpacks. The large projected decreases in spring snowpack for the three simulated climate-change scenarios may challenge water-resource managers to find alternate sources of water or develop new management strategies to meet the summer–fall water-use demands that these basins have traditionally provided.

Acknowledgments

This paper is one of a series of publications from the National Climate Change Modeling Project that is part of the U.S. Geological Survey Global Change Research and Development Program (http://gcp.usgs.gov/).

References

References
Adeloye
,
A. J.
,
N. R.
Nawaz
, and
M.
Montaseri
,
1999
:
Climate change water resources planning impacts incorporating reservoir surface net evaporation fluxes: A case study
.
Water Resour. Dev.
,
15
,
561
581
.
Alley
,
R. B.
,
2007
:
Summary for policymakers
.
Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 1–18
.
Barnett
,
T. P.
,
J. C.
Adam
, and
D. P.
Lettenmaier
,
2005
:
Potential impacts of a warming climate on water availability in snow-dominated regions
.
Nature
,
438
,
1
7
.
Cayan
,
D. R.
,
1996
:
Interannual climate variability and snowpack in the western United States
.
J. Climate
,
9
,
928
948
.
Draper
,
S. E.
, and
J. E.
Kundell
,
2007
:
Impact of climate change on transboundary water sharing
.
J. Water Resour. Plann. Manage.
,
133
,
405
415
.
Dudley
,
R. W.
, and
G. A.
Hodgkins
,
2002
:
Trends in streamflow, river ice, and snowpack for coastal river basins in Maine during the 20th century
.
U.S. Geological Survey Water-Resources Investigations Rep. 02-4245, 26 pp
.
Gleick
,
P. H.
,
1987
:
Regional hydrologic consequences of increases in atmospheric CO2 and other trace gases
.
Climatic Change
,
10
,
137
161
.
Hamlet
,
A. F.
, and
D. P.
Lettenmaier
,
1999
:
Effects of climate change on hydrology and water resources in the Columbia basin
.
J. Amer. Water Resour. Assoc.
,
35
,
1597
1623
.
Hamlet
,
A. F.
,
P. W.
Mote
,
M. P.
Clark
, and
D. P.
Lettenmaier
,
2005
:
Effects of temperature and precipitation variability on snowpack trends in the western United States
.
J. Climate
,
18
,
4545
4561
.
Hay
,
L. E.
,
S. L.
Markstrom
, and
C.
Ward-Garrison
,
2011
:
Watershed-scale response to climate change through the twenty-first century for selected basins across the United States
.
Earth Interactions
,
in press
.
Hayhoe
,
K.
, and
Coauthors
,
2006
:
Past and future changes in climate and hydrological indicators in the US Northeast
.
Climate Dyn.
,
28
,
381
407
,
doi:10.1007/s00382-006-0187-8
.
Hodgkins
,
G. A.
, and
R. W.
Dudley
,
2005
:
Changes in the magnitude of annual and monthly streamflows in New England, 1902–2002
.
U.S. Geological Survey Scientific Investigations Rep. 2005-5135, 37 pp
.
Hodgkins
,
G. A.
, and
R. W.
Dudley
,
2006a
:
Changes in the timing of winter–spring streamflows in eastern North America, 1913–2002
.
Geophys. Res. Lett.
,
33
,
L06402
,
doi:10.1029/2005GL025593
.
Hodgkins
,
G. A.
, and
R. W.
Dudley
,
2006b
:
Changes in late-winter snowpack depth, water equivalent, and density in Maine, 1926–2004
.
Hydrol. Processes
,
20
,
741
751
.
Hodgkins
,
G. A.
,
I. C.
James
II
, and
T. G.
Huntington
,
2002
:
Historical changes in lake ice-out dates as indicators of climate change in New England, 1850–2000
.
Int. J. Climatol.
,
22
,
1819
1827
.
Hodgkins
,
G. A.
,
R. W.
Dudley
, and
T. G.
Huntington
,
2003
:
Changes in the timing of high river flows in New England over the 20th century
.
J. Hydrol.
,
278
,
242
250
.
Hodgkins
,
G. A.
,
R. W.
Dudley
, and
T. G.
Huntington
,
2005a
:
Changes in the number and timing of days of ice-affected flow on northern New England rivers, 1930–2000
.
Climatic Change
,
71
,
319
340
.
Hodgkins
,
G. A.
,
R. W.
Dudley
, and
M. C.
Loiselle
,
2005b
:
Historical late-winter and spring snowpack depth and equivalent water-content data for Maine
.
U.S. Geological Survey Open-File Rep. 2005-1259, 80 pp
.
Hodgkins
,
G. A.
,
R. M.
Lent
,
R. W.
Dudley
, and
C. W.
Schalk
,
2009
:
A USGS hydrologic climate-response network for Maine
.
U.S. Geological Survey Open-File Rep. 2009-1115, 24 pp
.
Huntington
,
T. G.
,
G. A.
Hodgkins
, and
R. W.
Dudley
,
2003
:
Historical trend in river ice thickness and coherence in hydroclimatological trends in Maine
.
Climatic Change
,
61
,
217
236
.
Huntington
,
T. G.
,
G. A.
Hodgkins
,
B. D.
Keim
, and
R. W.
Dudley
,
2004
:
Changes in the proportion of precipitation occurring as snow in New England (1949–2000)
.
J. Climate
,
17
,
2626
2636
.
ISAB
,
2007
:
Climate change impacts on Columbia River basin fish and wildlife
.
Independent Scientific Advisory Board Climate Change Rep., 136 pp
.
Jeton
,
A. E.
,
M. D.
Dettinger
, and
J. L.
Smith
,
1996
:
Potential effects of climate change on streamflow, eastern and western slopes of the Sierra Nevada, California and Nevada
.
U.S. Geological Survey Water-Resources Investigations Rep. 95-4260, 44 pp
.
Leavesley
,
G. H.
,
R. W.
Lichty
,
B. M.
Troutman
, and
L. G.
Saindon
,
1983
:
Precipitation-Runoff Modeling System: User manual
.
U.S. Geological Survey Water-Resources Investigations Rep. 83-4238, 207 pp
.
Leung
,
R. L.
, and
M. S.
Wigmosta
,
1999
:
Potential climate change impacts on mountain watersheds in the Pacific Northwest
.
J. Amer. Water Resour. Assoc.
,
35
,
1463
1471
.
Marks
,
D. G.
, and
A. H.
Winstral
,
2007
:
Finding the rain/snow transition elevation during storm events in mountain basins
.
Proc. Joint Symp. JHW001: Interactions between Snow, Vegetation and the Atmosphere, Perugia, Italy, IUGG, 35
.
Mastin
,
M. C.
,
2008
:
Effects of potential future warming on runoff in the Yakima River basin, Washington
.
U.S. Geological Survey Scientific Investigations Rep. 2008-5124, 12 pp
.
McCabe
,
G. J.
, and
D. M.
Wolock
,
1999
:
General-circulation-model simulations of future snowpack in the western United States
.
J. Amer. Water Resour. Assoc.
,
35
,
1473
1484
.
Ruosteenoja
,
K.
,
T. R.
Carter
,
K.
Jylha
, and
H.
Tuomenvirta
,
2003
:
Future climate in world regions: An intercomparison of model-based projections for the new IPCC emissions scenarios
.
Finnish Environment Institute Rep. 644, 81 pp
.
Serreze
,
M. C.
,
M. P.
Clark
, and
R. L.
Armstrong
,
1999
:
Characteristics of the western United States snowpack from snowpack telemetry (SNOTEL) data
.
Water Resour. Res.
,
35
,
2145
2160
.
Stewart
,
I. T.
,
2009
:
Changes in snowpack and snowmelt runoff for key mountain regions
.
Hydrol. Processes
,
23
,
78
94
.
Stewart
,
I. T.
,
D. R.
Cayan
, and
M. D.
Dettinger
,
2005
:
Changes toward earlier streamflow timing across western North America
.
J. Climate
,
18
,
1136
1155
.

Footnotes

This article is included in the Integrated Watershed-Scale Response to Climate Change in Selected Basins across the United States special collection.