Research

Hydrology and Water Resources

Key Findings

The Climate Impacts Group’s (CIG) research on the relationship between climate and Pacific Northwest (PNW) hydrology and water resources has made significant contributions to our understanding of climate influences on PNW water supplies, decision-support for water resource management, and the role of institutional arrangements in shaping vulnerability to climate variability and change. Key findings from this research include the following.

20th Century Hydrologic Trends

In characterizing trends in 20th century PNW hydrologic conditions, CIG has:

• Showed that springtime snow water equivalent in the mountains of the PNW (including British Columbia) has declined since the mid-20th century, a direct result of regional warming (Figure 1). The largest percentage declines occurred at lower elevations (Mote 2003).
• Created extended records of summer and winter streamflow in the Columbia River to 1858 using in situ observations of river stage and gaged records. The extended records have been used to create paleo-reconstructions of Columbia River streamflow based on tree ring records (Gedalof et al., in review).


Figure 1 Trends in April 1 snow water equivalent based on data from 260 snow course collections sites. Most stations show a decline in snow water equivalent. The fact that trends are highest at low elevation sites implicates warming as a cause of the trend.

Impacts of Climate Variability on PNW Hydrology and Water Resources

The CIG has assessed the seasonal and interannual consequences of climate variability for PNW hydrology and water resources, finding that:

• Dry winter weather and warm spring temperatures – a more common occurrence during warm phase El Niño/Southern Oscillation (ENSO) events or a warm phase Pacific Decadal Oscillation (PDO) - lead to lower springtime snowpack and streamflow during spring and summer in snowmelt-driven rivers. As a result, flooding is less likely and drought more likely during warm phase ENSO (El Niño) and PDO. The opposite is true for cool phase ENSO (La Niña) and PDO. PNW winters tend to be cooler and wetter during cool phase ENSO and PDO, resulting in higher than average winter snowpack and spring and summer streamflow in snowmelt-driven rivers. ENSO and PDO also can act in combination, with the largest changes observed when the two are “in-phase,” i.e., El Niño combined with warm PDO or La Niña combined with cool PDO. These features are evident in Figure 2 (Hamlet and Lettenmaier 1999; Miles et al. 2000; Mote et al. 2003; Hamlet et al., in review).
• There is a clear overall trend from low to high probability of flooding across the climate categories from warm PDO/warm ENSO to cool PDO/cool ENSO conditions when considering all river basin types together (Figure 3). Snow-dominated basins are most sensitive to both ENSO and PDO, with the greatest separation between categories where ENSO and PDO are in phase (see figure). Transient snow basins show some sensitivity to ENSO but less to PDO (Hamlet et al., in review).

click images to enlarge



Figure 2 Effects of ENSO and PDO on Columbia River flows at The Dalles.

click image to enlarge



Figure 3 Probability of flooding in 26 PNW rivers based on ENSO/PDO phase categories. There is a clear overall trend from low to high probability of flooding across the climate categories from warm PDO/warm ENSO (El Nino) to cool PDO/cool ENSO (La Nina) conditions. A value of 0.80 is equal to an 80% probability. For probability based on river type, click here.

Climate-based Seasonal Streamflow Scenarios

In its efforts to support the use of climate information in PNW water resource management, the CIG has:

• Developed a long-lead streamflow forecasting approach that uses global and regional climate information made available as early as summer to classify the likely climate state (ENSO and PDO) for the following winter and make forecasts at lead times as long as a year. These forecasts typically reduce the spread of traditional streamflow projections based on climatology alone by two-thirds (Hamlet and Lettenmaier 1999, 2000; Hamlet et al. 2002; Huppert et al., in review). CIG routinely makes these forecasts available for free via the web and at the CIG's annual fall water resources planning workshops.
• Established a real-time monthly to seasonal hydrologic forecasting system for the western U.S. Forecasts are updated every two weeks.

Impacts of Climate Change on PNW Hydrology and Water Resources

The CIG has investigated the projected impacts of climate change for PNW hydrology and water resources, finding that:

• Even acknowledging the potential for future wetter winters, the warmer temperatures projected for the PNW by global climate models for the mid-21st century would:
  • Cause more winter precipitation to fall as rain instead of snow, increasing winter streamflow;
  • Elevate the typical winter snowline in the PNW;
  • Decrease the snow covered area in the mountains and total winter snowpack (Figure 4);
  • Result in earlier snow melt early in the season, moving spring peak flows earlier in the year and increasing the time between snowmelt and fall rains (Figure 4). In the transient snow zone, where mid-winter temperatures are currently close to freezing, streamflow timing would shift even more for the same warming;
  • Decrease summer streamflow, increasing the frequency of significant low flow events (even with projected increases in winter precipitation); and
  • Result in significant water resources impacts within the next few decades in watersheds at moderate elevations in the Cascade Mountains and in the southern interior of the Columbia River basin (e.g., the Snake River basin) (Hamlet and Lettenmaier 1999; Mote et al. 1999; Miles et al. 2000; Mote et al. 2003, Snover et al. 2003; Hamlet et al., in review).
• In the Columbia River basin:
  • The impacts of climate change on streamflow timing would result in a decreased ability of the reservoir system to meet minimum streamflow requirements for fish, a slight reduction in firm power production, and improved compliance with flood control targets (Figure 5) (Hamlet and Lettenmaier 1999; Mote et al. 1999; Miles et al. 2000; Hamlet et al., in review).
  • Related work funded by the Accelerated Climate Prediction Initiative showed that instream fish flow targets would suffer under the range of future climate conditions considered, even with changes in flood operation specifically designed to mitigate the effects of climate change (e.g., reduced flood storage, earlier refill) (Payne et al., in press).
• In municipal water supply systems, projected climate change would affect the ability of municipal water suppliers to provide reliable water to their customers as a result of:
  • Reduced supply, i.e., decreased reservoir storage of water during the summer months due to decreased winter snowpack;
  • Increased demand for water, due to warmer summer temperatures; and
  • An increase in the length of the average summer reservoir drawdown period as a result of reduced supply and increased demand (Palmer and Hahn 2002, Palmer et al. 2004).

  Key findings from specific studies on urban water supplies include:

• For municipal water supply in Portland (Oregon), climate change impacts on supply and demand are expected to have a considerable cumulative impact on water supply availability in the 2040s (i.e., about 50% of the impact of population growth alone in that same period) (Figure 6) (Palmer and Hahn 2002).
• For municipal and agricultural water supply in the Tualatin River Basin (Oregon), climate change is expected to reduce system yield by 1.5% per decade over the next 40 years even as demand increases. This reduction advances the need for system expansion by 5 to 8 years. The Tualatin Basin will also see decreases in summer streamflow on the order of 10-20%, stressing the ability to meet streamflow temperature requirements and maintain instream flows. The impact of extended drought may also be much more signficiant (Palmer et al. 2004).

Climate Change Streamflow Scenarios

CIG removed the bias that typically results from hydrologic simulation models to create climate change streamflow scenarios for use in water resources planning. The streamflow scenarios are numerically consistent with and, for purposes of evaluating system performance, directly comparable to observed streamflow time series. The development of these scenarios, which are used in place of historic streamflow records, gives water resources managers the ability to easily and inexpensively evaluate system vulnerabilities to climate change by allowing for direct incorporation of the streamflow scenarios into existing water resources planning methods (such as critical period analysis). The climate change streamflow scenarios are available via the CIG’s web site (Snover et al. 2003).

Institutional Sources of Climate Vulnerability

The CIG has analyzed the institutional context of regional water resources management, including institutional sources of vulnerability to climate, finding that:

• The vulnerability of the PNW to changes in precipitation is greatest at the extremes, i.e., droughts and floods, but adaptive capacity is highest relative to floods and lowest relative to droughts, especially multi-year droughts. Adaptive capacity depends largely on high levels of institutional integration as demonstrated by the centralization of authority under the U.S. Army Corps of Engineers with respect to flood control. By contrast, a high level of institutional fragmentation, as exists for droughts, severely inhibits effective response capacity on a regional basis (Callahan 1997; Callahan et al. 1999; Miles et al. 2000; Mote et al. 2003; Hamlet et al., in review; Miles and Snover, in review).
• While drought occurrence in Washington’s Yakima Valley is strongly tied to warm PDO conditions, water management in the Yakima basin has not yet accounted for the cyclical nature of drought probability. This limitation, combined with recent trends in irrigation and other agricultural practices, has significantly increased the Valley’s vulnerability to drought over the past thirty years (Gray 1999; Hamlet et al., in review).

Climate Variability, Climate Change, and Evolving Water Policy

The CIG has evaluated the implications of climate variability and change for water resources development and evolving water policy, finding that climate change is likely to lead to transboundary tensions between Canada and the United States given the projected effects of climate change on snowpack and the timing of spring runoff in the U.S. portion of the Columbia River Basin. This potential is further increased by:

• the dependence on the U.S. portion of the basin on snowmelt from the smaller but higher elevation Canadian portion of the basin for natural summer streamflow (while only 30% of the Columbia River Basin is located in Canada, 50% of the Columbia’s natural late summer streamflow originates in Canada on average); and
• the lack of explicit provisions in transboundary agreements such as the Columbia River Treaty (1964) for the transfer of water from Canada to the U.S. for provision of instream flows on the U.S. portion of the main stem Columbia as required by the U.S. Endangered Species Act and other agreements.

Despite these problems, the agreements within the CRT and related transboundary agreements present potential opportunities for the mutually beneficial release of water from Canadian storage in summer to help meet U.S. instream flow objectives. One opportunity comes from the increased marketability of Canadian hydropower as a result of growing summer energy demand outside the PNW (Hamlet 2003).

Climate Change Impacts on the PNW Ski Industry

For ski areas at moderate elevation, the CIG has found that even modest increases in PNW temperature and precipitation as a result of climate change could significantly decrease revenues by shortening the length of the ski season and reducing patronage due to undesirable ski conditions (as a result of increased winter rain). Snow model simulations show that average ski conditions at Snoqualmie Pass (Washington) ski area, whose base elevation is about 3000 ft, could change dramatically by 2025. The simulations suggest that the likelihood of opening by Dec. 1 could decline by 50%, average season length could decline by 28%, and the likelihood of rain when the ski area is open could increase by 25%. The changes in snow conditions by 2025 are less pronounced for Stevens Pass (Washington), whose base is at about 4050 ft. The simulations for Stevens Pass suggest that the likelihood of opening by Dec. 1 could decline by 25%, average season length could decline by 14%, and the likelihood of rain when the ski area is open could increase by 50%.

Research Awards

CIG publications on hydrology and water resources have been recognized nationally by peer-reviewed journals. CIG research publications have won the:

• Boggess Award for best paper in the Journal of the American Water Resources Association (2000) on integrated analysis of the impacts of climate variability and change on Columbia basin water resources (Miles et al. 2000).
• ASCE Journal of Water Resources Planning and Management 2002 Award for Best Practice Oriented Paper for an analysis of the economic value of long-lead streamflow forecasts for Columbia River hydropower production (Hamlet et al. 2002).