Columbia River Basin 2011
Estimated Volume (acre-feet)
2030 New Irrigation Demand*
WSU Integrated Model
2030 New Municipal and Domestic Demand (including municipally-supplied commercial)
WSU Integrated Model
Unmet Columbia River Instream Flows**
Ecology data, McNary Dam, 2001
Unmet Tributary Instream Flows***
Ecology data, tributaries with adopted instream flows, 2001 drought year
2030 New Hydropower Demand
WSU Surveys and Planning Forecast
Alternate Supply for Odessa
Odessa Draft Environmental Impact Statement (October 2010)
Yakima Basin Water Supply (pro-ratables, municipal/domestic and fish)
Yakima Integrated Water Resource
Unmet Columbia River Interruptibles
40,000 to 310,000
Ecology Water Right Database
* Additional irrigation demands were modeled assuming an equivalent land base for irrigated agriculture,
under a scenario of medium growth in the domestic economy, and medium growth in international trade.
Acreage currently irrigated by groundwater in the Odessa was assumed to be new surface water demand
in 2030, and thus is not reflected in changes in total demand, which includes both surface and groundwater.
Increases in total demand are thus due to the combined impacts of climate change, and changes in crop mix
driven by growth in the domestic economy and international trade.
** Unmet Columbia River instream flows are the calculated deficit between instream flows specified in Washington Administrative Code (WAC) and 2001 (drought condition) actual flows at McNary Dam.
*** Unmet tributary instream flows are the combined deficits between current instream flows specified in WAC and 2001 actual flows at Walla Walla River near Touchet, Wenatchee River at Monitor, Entiat River near Entiat, Methow River near Pateros, Okanogan River at Malott, Little Spokane River near Dartford, and Colville River at Kettle Falls.
The Forecast anticipates:
• 170,000 (±18,000) ac-ft per year of additional total (ground and surface) water agricultural irrigation demand. This number assumes no change in irrigated acreage, and no additional water supply development. This number represents demands for surface and ground water as applied to crops, plus the additional water needed to account for irrigation application inefficiencies.
•430,000 (±14,000) ac-ft per year of additional surface water agricultural demand. This number includes new demands that will be met only by surface waters, and assumes that historical groundwater irrigation demands in the Odessa area will be new surface water demands in the future.
• 117,500 acre-feet per year in additional total diversion demands for municipal and domestic water.
• 500,000 acre-feet per year of unmet tributary instream flows, and 13.4 million acre-feet per year of unmet Columbia River mainstem instream flows, based on observed deficits during the 2001 drought year.
• No demand for new water storage for hydropower generation purposes.
New irrigation and municipal demands do not include improvements in conservation, which could decrease the new demands that need to be met, but might also have complex impacts on return flows. For example, if all municipal and domestic users were able to conserve 10% of their water supplies by 2030, then new municipal demand might drop from 117,500 acre-feet to about 105,000 acre-feet. However, many municipal conservation techniques are nonconsumptive in nature. For example, fixing leaky pipes and installing low flow showers and toilets reduce diversions, but with a corresponding reduction in water returned (via wastewater treatment plants or underground). Alternatively, some conservation measures, such as reducing lawn size, do reduce consumptive use. In addition, conservation is often less expensive than new water supply development.
In addition to these new demands by sector, other studies suggest several areas of unmet demand, some of which are not reflected in these totals. These other studies used different methods of calculating demand, and thus, should not be directly compared to the totals above.
• The draft Environmental Impact Statement for Odessa suggests a preferred alternative of supplying 164,000 acre-feet per year of surface water to current groundwater users in this area. This amount is not included in the total irrigation demands above, which shows changes in total (combined groundwater and surface water) demand between the historical period (which includes Odessa) and 2030.
• The Yakima Integrated Water Resource Management Plan suggests that 450,000 acre-feet per year will be needed for proratable, municipal-domestic and fish needs. These demands overlap partially with the demands shown above.
• The Ecology Water Right Database indicates that in years in which the Mainstem Drought Program is run, there are 40,000 to 310,000 acre-feet per year of unmet needs by interruptible water users, depending on the drought year conditions. These amounts are currently unmet, so are not reflected in the numbers above.
Together, these current and new demands are likely to exacerbate water supply issues in some locations, particularly during the summer.
The agricultural portion of the Forecast focused on irrigation water demands. The 2030 forecast of demand for irrigation water across the entire Columbia River Basin (seven U.S. States and British Columbia) was 13.6 million acre-feet under average flow conditions, assuming an equivalent land base for irrigated agriculture in the future (see table below). The range of estimates was from 13.1–14.1 million acre-feet during wet and dry years, respectively (20th and 80th percentile). This irrigation demand was roughly 2.5% above modeled historic levels under average flow conditions. Conveyance losses, that occur as water is transported through irrigation ditches and canals, were estimated separately.
million acre-feet per year
million acre-feet per year
|Entire Columbia River Basin||13.3 (12.6–13.9)||13.6 (13.1–14.1)||2%|
|Washington Portion of the Columbia River Basin||6.3 (6.0–6.5)||6.5 (6.2–6.6)||2%|
Top of crop agricultural demands under the baseline economic scenario (medium domestic economic growth and medium growth in international trade), excluding conveyance losses, in the Columbia River Basin in the historical and 2030 forecast period. Estimates are presented for average years, with range in parentheses representing wet (80th percentile) and dry (20th percentile) years.
Seasonal timing of forecasted water supply and irrigation water demand is shown in the figure below, with irrigation demands taking a larger proportion of water supplies in summer months by 2030. Instream, hydropower and municipal water demands will also need to be met from these water supplies.
Comparison of regulated surface water supply and surface water irrigation demands for the historical (top) and 2030 forecast (bottom) periods under the medium-growth, medium-trade economic scenario across the entire Columbia River Basin, including portions of the basin outside of Washington State. Wet (80th percentile), dry (20th percentile), and average (50th percentile) flow conditions are shown for both supply and demand.
Within the Washington State portion of the Columbia River Basin, results were similar (see table above):
• Forecast increases in irrigation water demand were an average of 170,000 (±18,000) acre-feet per year, roughly 1.9% above historical conditions, assuming an equivalent land base for irrigated agriculture, and a crop mix influenced by medium growth in the domestic economy and international trade.
• Considering only the climate impacts of temperature and precipitation variations on the irrigation demand, there would be a 3.7% increase in demand. When economic impacts resulting in a new crop mix are considered in addition to the climate impacts, the increase in demand reduces to 1.9%.
Modeling under alternate economic scenarios was used to give information about the potential range of future water demands from irrigated agriculture, if growth in the domestic economy and international trade were higher or lower than anticipated.* Higher income growth leads to an expansion of high value crops like fruits and vegetables at the expense of low value crops. Similarly, stronger growth in exports has a disproportionate impact on higher value crops, although wheat and alfalfa are also sensitive to fluctuations in trade. Production patterns were generally more sensitive to assumptions about trade than to assumptions about economic growth. One exception was wine grapes where most of the growth in demand is expected to come from domestic consumers rather than international exports.
• The low, medium and high economic scenarios forecasted increases of 200,000 (±17,000) acre feet, 170,000 (±18,000) acre feet and 140,000 (±18,000) acre feet over historical demands under average flow conditions within the Washington portion of the Columbia River Basin.
• These estimates assumed no change in the land base for irrigated agriculture, thus differences in the agricultural water demand between different scenarios were due to changes in crop mix and crop water demands under future climate conditions.
Additional scenarios considered the potential impacts of additional water capacity in specific locations corresponding to projects proposed by OCR. Under some scenarios, new water was provided at no cost to users, while in other scenarios, users were charged per unit fees to recover some development costs.
• The development of roughly 200,000 acre-feet of annual water capacity (the medium scenario considered) caused demand for irrigation water to increase by 46,400 (±640) acre-feet per year over baseline 2030 demands (under the medium economic scenario) in the Washington portion of the basin.**
* Domestic economic growth was 1.3-1.8% under low and high scenarios, while international trade included scenarios of low and high growth in trade for specific crop groups (e.g. vegetables, wheat, etc.).
** Under this water capacity scenario, 164,000 acre-feet was developed to meet current agricultural demand in the Odessa, with the rest serving new demands.
Municipal demands, including domestic and municipally-supplied industrial, are likely to increase throughout the entire Columbia River Basin over the next 20 years. By 2030, U.S. Census estimates show population growth in Idaho (25.6%), Oregon (26.2%), and Montana (5.6%). Although some new municipal demands will likely be met by deep groundwater supplies, others will likely come from shallow groundwater or surface water. These additional demands will likely reduce inflows into some parts of Washington. For example, an Idaho study of the Spokane River Basin projected an additional demand on the river of 31 cfs by 2060.*
Within eastern Washington, the Forecast found that:
• Domestic and industrial diversion demands in rural and urban areas (excluding self-supplied industries) were forecasted to be 569,000 acre-feet per year in 2030, an estimated 26% increase over 2010. Consumptive demands are approximately 51% of this amount.
• Per capita water demands varied considerably throughout eastern Washington, with an average total demand (including system losses) of approximately 277 gallons per capita per day (gpcd).**
Across the Columbia River Basin, the Forecast found that:
• Decreases in surface water supplies in summer and early fall may increase the challenge of meeting water needs for fish across the Columbia River Basin by 2030.
• Re-negotiation of the international Columbia River Treaty could change the amounts and timing of water available to meet instream needs in the Columbia River mainstem.
• Quantification of tribal water rights, while outside the scope of this Forecast, could also change surface water supplies for meeting instream demands in unpredictable ways.
Within eastern Washington, the forecast of demand for water to support instream flows found the following:
• In many rivers in eastern Washington, stream flows are below state or federal instream flow targets on a regular basis, particularly in late summer. Surplus water exists in many of these same rivers at other times of year.
• Decreases in surface water supplies in tributaries in summer and early fall may lead to more weeks when instream flows are not met by 2030. This may result in a higher frequency of curtailment of interruptible water right holders in basins with adopted instream flow rules.
• An evaluation of fish, flows, and habitat in eight fish critical basins, available in the Columbia River Instream Atlas (Ecology Publication 11-12-015), will help target investments to maximize the positive impact on fish populations.
Across the Columbia River Basin, the forecast of hydropower demands found the following:
• Demand for water storage to supply hydropower facilities is anticipated to remain unchanged in 2030. Utilities expect to be able to meet projected steady growth in peak winter and summer energy demands through conservation and integration of other energy sources, including those required under Washington’s passage of Initiative 937.
• Several power entities are concerned that climate change and the possible renegotiation of the international Columbia River Treaty will affect hydropower generation capacity.
* 31 cfs = 22,443 acre-feet/year (back)
** 277 gallons per day = 0.429 cfs = 311 acre-feet/year (back)
Surface water supplies and water demands were forecasted for each Water Resource Inventory Area (WRIA) in eastern Washington. Major results for each WRIA are presented in "Overview of Washington's Watersheds: Tier II Results." Cumulatively, the following results were found:
• The greatest concentration of current and future agricultural irrigation and municipal water demands are in the southern and central Columbia Basin, including Lower Yakima (37), Lower Crab (41), and Esquatzel Coulee (36), as well as Rock-Glade (WRIA 31), Walla Walla (32), Lower Snake (33), Naches (38), Upper Yakima (39), and Okanogan (49). Irrigation dominates the demand for water in these WRIAs.
• Unmet demand due to curtailment of interruptible and proratable water rights or insufficient water at the watershed scale was forecasted for Walla Walla (WRIA 32), Yakima (37, 38, & 39), Wenatchee (45), Methow (48), Okanogan (49), Little Spokane (55), and Colville (59).
• Unmet demand for surface water was forecasted for the Odessa due to existing groundwater declines in Palouse (WRIA 34), Esquatzel Coulee (36), Lower Crab (41), Grand Coulee (42), and Upper Crab (43).
Modeled historical and 2030 forecasted surface water supplies were compared to state-level instream flow targets and the Federal Columbia River Power System Biological Opinion (FCRPS BiOp)*.
• Under normal flow conditions, modeled regulated surface water supplies prior to meeting cumulative demands were close to Washington state instream flow regulations in fall/early winter at Priest Rapids Dam (both historical and 2030 forecast), and in July and August at Priest Rapids Dam and McNary Dam (for the 2030 forecast).
• Under normal flow conditions, modeled regulated surface water supplies prior to meeting cumulative demands were not sufficient to meet target flows under the FCRPS BiOp in April, July, and August at McNary Dam, and from November through January at Bonneville Dam. Imbalances were smaller in the 2030 forecast than the historical case for the late winter/spring months, and larger for the late summer.
• Along the mainstem, there are 379 interruptible water rights, the majority of which are agricultural surface water rights. These water users are particularly vulnerable to the potential impacts of water shortages.
Complete results are presented in "Washington's Columbia River Mainstem: Tier III Results."
Collectively, these results suggest that meeting water demands will be more challenging by 2030 as increased demands are placed on limited supplies. Solutions will require combinations of conservation, water banking/ marketing, and new supplies based on groundwater and/or storage of water in peak runoff seasons.
For solutions requiring additional investment in water supply infrastructure, the Forecast's results suggest that at prices in the range of those currently being charged by the Office of Columbia River for new water it may be feasible to recover some or all water supply costs from new users without significantly decreasing the quantity of water demanded by users.
Projects associated with the medium water capacity scenario of an additional 200,000 acre-feet per year for out-of-stream uses were estimated to lead to total employment impacts (including indirect and induced effects) of 6,600 jobs. State and local tax impacts were estimated at about $37 million. These estimates do not subtract the jobs and taxes associated with production if land associated with the new capacity was previously under dryland cultivation. These estimates include economic activity generated from post-farmgate processing of agricultural products that occurs within Washington. While not quantified, it is recognized that maintenance of and improvement to instream flows would also have positive economic impacts on tourism and recreation, generating additional jobs and tax revenues.
This Forecast improves our understanding of future surface water supplies and instream and out-of-stream demands, and will serve as a capital investment planning tool to maintain and enhance the region's economic, environmental, and cultural prosperity. Future forecasts will build upon and expand this knowledge to include assessments of groundwater supplies, the Columbia River Treaty and other pertinent issues.
The 2011 Forecast is available in written and web-based formats. In addition to this Legislative Report, WDFW’s “Columbia River Instream Atlas” (Ecology Publication 11-12-015) includes detailed assessment of 189 stream reaches in fish-critical WRIAs, and WSU’s technical report (Ecology Publication 12-12-001) includes detailed methodology and complete results.
Stakeholder input was essential to the development of the Forecast. WSU researchers presented initial modeling methods to the Columbia River Policy Advisory Group (PAG). This group represents a range of stakeholder interests, and helps OCR identify and evaluate policy issues. Feedback from the PAG and watershed planning unit representatives was used to adapt WSU forecasting methods. To ensure that comprehensive and scientifically valid methods were utilized, an external peer review panel comprised of four national experts in economics, modeling, and regional water issues periodically reviewed and commented on WSU's work.
Preliminary results of the Forecast were presented to the interested public at three public stakeholder events in Wenatchee, Spokane and the Tri-Cities in early September 2011. A draft report was released at the end of September, with public comment accepted for 30 days. Based on feedback received at workshops, through on-line forums, and through the draft comment process, economic and biophysical modeling assumptions were fine-tuned and results were finalized. Comments received, and the responses to comments, are described in the “Summary of Responses to the Draft 2011 Legislative Report for the Columbia River Basin Long-Term Water Supply and Demand Forecast” (Ecology Publication 12-12-004).
Watershed Planning Unit representatives and OCR’s Policy Advisory Group provided input on the development of the 2011 Forecast. National experts in economic, modeling, and regional water issues peer reviewed the integrated modeling methods.
Supply and demand was forecasted at three tiers: the entire Columbia River Basin, the watershed level within Washington State, and the Columbia River mainstem in Washington State (see figures below). Specific objectives at each tier included the following:
Long-term water supply and demand was forecasted at three tiers.
The Columbia River Basin is the fourth largest watershed in North America in terms of average annual flow, encompassing all or parts of seven western states and British Columbia. For thousands of years, the river has shaped the economy and lives of those who lived near it. Over the past two hundred years, the basin has been extensively developed for hydropower generation, irrigation, navigation, and flood control. The river is also managed for the protection of salmonid species listed under the Endangered Species Act, municipal and industrial supplies, maintenance of water supplies in accordance with tribal treaties, and recreation. This creates a myriad of competing demands. Reliable access to water is essential for existing and future regional economic growth and environmental and cultural enhancement. Seasonal variations in water supply and demand have resulted in localized shortages with increasing regularity due to population growth, climate variability and change, and increased implementation of regulatory flow requirements. The competing demands on the region's fresh water resources will only increase in the future, particularly in summer months when demands are high.
Recognizing that development of new water supplies for eastern Washington is a priority concern, the Legislature passed Chapter 90.90 RCW, directing the Department of Ecology (Ecology) to aggressively develop water supplies for instream and out-of-stream uses. The Office of Columbia River (OCR), formed as a result of this legislation, has a mission to develop water supplies for the following purposes:
• Addressing aquifer decline in the Odessa Subarea by replacing groundwater sources with surface water sources.
• Permitting new water rights.
• Securing water for drought relief.
• Providing water for instream flows to benefit fish.
Water supplies developed under this program are to support both instream and out-of-stream uses. For new storage projects, two-thirds of the supply developed must be allocated for out-of-stream uses and one-third for in stream uses. Since 2006, OCR has funded a variety of water supply projects consistent with the four legislative directives (OCR Project Map). With approximately 150,000 acre-feet of water supply already developed and another 200,000 acre-feet in near-term development, OCR is rapidly improving water supply for eastern Washington.* OCR is developing a portfolio of diverse projects including modification of existing storage (e.g. Lake Roosevelt and Sullivan Lake), new storage facilities (e.g. Kennewick, Boise and White Salmon aquifer storage projects), conservation piping and canal lining projects (e.g. Red Mountain AVA (American Viticultural Area), Barker Ranch, Manastash, and Columbia Basin Irrigation District projects), transmission piping projects (e.g. Potholes Supplemental Feed Route and Weber Siphon), and water right acquisitions.
Every five years, OCR develops a long-term water supply and demand forecast (Forecast) and submits it to the Legislature. The Forecast provides OCR with a better understanding of where additional water supply is currently needed, and where it will be needed in the future. OCR uses the Forecast as a capital investment planning tool. The primary purposes of the Forecast are to provide a generalized, system-wide assessment (not project-specific) of
• How future environmental and economic conditions are likely to change water supply and demand.
• Where OCR can invest in water supply projects that have the greatest chance of meeting new demand and improving flows for fish.
In 2006, OCR contracted with Golder Associates and Anchor Environmental to conduct the first Forecast, with WSU researchers providing a forecast of future agricultural demand. Based on 2004 U.S. Geological Survey estimates and estimates of public water system use provided by the Washington State Department of Health, estimates of water use in 2000 for eastern Washington were 467,432 acre-feet per year for domestic and industrial (public and self-supplied), and 3,288,740 acre-feet per year for crop irrigation and golf courses.**
Estimates of future agricultural demand carried out by Golder and Anchor that were based on an analysis of water rights applications suggested a nine percent growth in annual irrigation water demand of about 211,323 acre-feet over the twenty years from 2005-2025. WSU used vector autoregression (a method that captures changes and relationships between variable, time-based data sets) and a survey of expert opinion of future crop prediction and water use for major crops. WSU's Forecast suggested a largely stable picture for future agricultural acreage, though with a large expected range, from nearly one million acres to a decrease of 750,000 acres. The differences between Golder/Anchor and WSU results were a result of the different underlying data and the large amount of uncertainty in both estimates. Projected growth in domestic and industrial demand (public and self-supplied) was projected to be approximately 94,500–109,400 acre-feet per year over the twenty years from 2005 to 2025, depending on the methods used.
The 2011 Forecast updates and expands the 2006 Forecast by delving more deeply into water supply and demand issues. To develop the 2011 Forecast, OCR partnered with Washington State University (WSU) to conduct the agricultural, municipal, and hydropower components of the Forecast, and the Washington Department of Fish and Wildlife (WDFW) to conduct the instream demand component of the Forecast. This 2011 Forecast, described more fully in the "Overview of the 2011 Forecast," uses state of the art biophysical modeling techniques incorporating the impacts of climate change, future regional and global economic conditions, and state level water management actions.
The 2011 Forecast represents an initial effort to employ computer-based modeling to forecast water supply and demand. As such, it represents a major endeavor that OCR will use as a foundation for future forecasts. Improvements being considered for 2016 include the following:
• Incorporation of deep groundwater dynamics into water supply forecast. (Shallow subsurface/surface dynamics are captured in this 2011 report.)
• Adoption of new (AR5) climate model predictions.
• Full integration of economic and biophysical forecasting.
• Extension of economic analysis to cover the portions of the Columbia River Basin outside of Washington State.
• Development of non-agricultural demands.
• Development of economic modeling to include producer responses to water shortages beyond deficit irrigation.
• Extension of economic impacts analysis to include augmentation of streamflow.
• Expansion and update of the 2011 Columbia River Instream Atlas (Ecology Publication 11-12-015).
• Inclusion of water supply and demand issues resulting from changes to the international Columbia River Treaty.
The characteristics of the Columbia River Basin make it particularly sensitive to small changes in overall temperatures. Surface water flows in the Columbia River Basin are dominated by the temperature-sensitive cycle of snow accumulation and melting. During the winter, when the majority of precipitation occurs, snow accumulates in upper elevations of the basin, forming a "natural reservoir" that stores water during times when demands are relatively low. Melting snow subsequently provides peak yearly flows in the spring and early summer, with nearly 60% of the unregulated surface water availability occurring during May, June, and July. For most regions, this is followed by a low flow period in the late summer and early fall, until late fall flows increase due to rainfall. Operations of major reservoirs have attenuated the seasonal nature of the natural hydrograph, shifting a significant amount of water availability from the winter months to the drier summer months and reducing the seasonal pattern.
The climate in the Pacific Northwest is already changing. Average temperatures are about 1.5° F higher than they were a century ago, with more warming during the winter than at other times of year. Regional climate change projections suggest that these trends will intensify, with projected temperature changes in the range of 1 to 5° F over the next 50 years, and a best estimate of about 2.5° F.* This seemingly small amount of warming could fundamentally change the patterns of rain and snowfall in the Columbia River Basin. With more precipitation falling as rain during the winter, and earlier snowmelt, peak flows will likely be earlier, with longer and lower periods of low flows during the summer, when out-of-stream demands are highest and instream demands for hydroelectricity generation and fish are important. Reservoir management can compensate for some timing changes in areas of the basin with storage, though the overall level of storage in the Columbia River Basin is lower (as a percentage of annual runoff) than some other major river systems in the U.S.
Simultaneously, higher summer temperatures under climate change could change out-of-stream demands for water in complex ways. Irrigated crops and natural vegetation are likely to have higher evapotranspiration (loss of water through evaporation and plant transpiration) rates and thus need more water. Decreases in summer precipitation could also increase irrigation demand because irrigation demand is the crop water requirement beyond what is provided by rainfall. Some harvested crops may be planted earlier and reach maturity earlier, which could increase demands for some crops earlier in the season, but reduce demands later in the season. Meanwhile, higher summer temperatures could also cause an increase in domestic water demands.
These temperature-driven changes in water supply and demand have the potential to seriously stress the Columbia River Basin water supply system, which was built to reliably deliver water under historical conditions. Climate change is thus incorporated as an important feature of this Forecast, to provide information that will help legislators, water managers, and agency professionals begin to plan for future conditions that will likely be different than what we have experienced in the past.
Water supply and demand impact each other. Out-of-stream diversions reduce supply downstream, while water that is diverted, but that is not consumptively used (such as water that is lost through leaks in municipal systems), may return to the system and provide water supply downstream. WSU researchers thus simulated surface water supply and out-of-stream demands with an integrated computer model that simulated the relationships between water supply, climate, hydrology, irrigation water demand, crop productivity, economics, municipal water demand and water management at all three geographic tiers. The Forecast's model integrated and built upon three existing models:
1. VIC: Variable Infiltration Capacity, a land surface hydrology model.
2. CropSyst: Cropping Systems Simulation, a cropping system model.
3. ColSim: Columbia Simulator, a reservoir operations model.
Biophysical modeling framework for forecasting surface water supply and irrigation water demand.
Each of these models has been used independently many times to simulate conditions in the Columbia River Basin. What is novel about WSU's approach is that VIC and CropSyst were integrated to exchange hydrologic and crop production information. For example, VIC informed CropSyst of daily weather and water supply; and CropSyst informed VIC of crop water needs and whether or not a particular crop was water stressed on any given day. This new model, termed VIC-CropSyst, used daily precipitation and temperature observations from across the basin for 1977-2006 to generate baseline simulations of present conditions for each location. To forecast future conditions, the model used daily weather information for the 2030s decade (referred to in this report as 2030) from five different climate change scenarios, representing a range of future greenhouse gas emissions and adapted for our region by the Climate Impacts Group at the University of Washington.*
An integrated overview of the modeling structure is shown in the figure below. Instream demands were not determined within modeling, but were represented through the adopted state and federal instream flows which were assumed to be the same in the historical and future periods. Historical and forecasted municipal demands were included in the modeling framework by withdrawing the consumptive use portions from surface water availability.
The models were able to forecast a variety of potential impacts on a spatially distributed basis, including predicted surface water supply, total irrigation demand, unmet irrigation demand due to curtailment, and decreases in crop yield due to curtailment.
Integration of biophysical modeling (surface water supply, crop dynamics and climate) with economic and policy (human decision-making) modeling.
For the supply analysis, the Forecast focuses on surface waters and shallow subsurface/surface hydrologic interactions, and does not analyze deep groundwater dynamics. It is recognized that deep groundwater supplies play a significant role in many parts of eastern Washington, and due to time, resource, and data constraints, deep groundwater supplies will be addressed in future forecasts.
Surface water supplies for our region reflect the current management of the existing reservoir system. The integrated VIC-CropSyst model was thus linked to reservoir and water use curtailment models that enabled evaluation of how a changing water supply might impact future reservoir storages and releases, irrigation application amounts, crop yields, and how frequently some groups of water users might be interrupted. The project did not model all dams in the Columbia River Basin, as there are more than 400 dams (both storage and run-of-the-river) operated to meet a variety of purposes. Reservoir modeling captured operations of the dams shown in the figure below, includingincluding the major storage dams on the Columbia and Snake Rivers, and the five major reservoirs in the Yakima Basin (Keechelus, Kachess, Cle Elum, Tieton and Bumping Lake). Dam management captured within ColSim included operations for power generation, flood control, instream flow targets, water storage, and stream flow regulation.
The modeling effort assumed that dam management does not change going into the future. To better understand how changes in infrastructure and management could change the water supplies entering Washington State in the future, and to help interpret the modeling results, WSU surveyed basin water managers about water supply planning, project development, and water management, using a 29-question survey developed in collaboration with OCR.
VIC-CropSyst focused on agricultural irrigation demands, as irrigation represents the majority of out-of-stream water use in the Columbia River Basin and is a prominent driver of Washington's economy.* Agricultural water uses other than irrigation, such as stock water, were not estimated for this Forecast. While these uses are important within some WRIAs, the magnitude of these uses basin-wide is small relative to consumptive use for crops. The U.S. Geological Service estimated that in 2005, within eastern Washington, stock water uses represented approximately 0.4% of out-of-stream water use considering domestic, irrigation, stock water, aquaculture, industrial, and mining.** If stock water represents a significant proportion of water use in the future, it may merit additional attention in future Forecasts.
To accurately simulate surface water supply and demand, the combined model needed accurate land use information for the entire Columbia River Basin, including upstream areas in other states and British Columbia. The historical simulation (1977-2006) used recent crop mix information from the United States Department of Agriculture (USDA) for areas outside of Washington, and from the Washington State Department of Agriculture (WSDA) for areas inside the state. The WSDA data were used in Washington because they were found to be slightly more precise for the Washington crop mix when evaluated against the USDA data layer. To capture the diversity of agriculture across Washington, nearly 40 groups of field and pasture crops, tree fruit and other perennials were simulated. Because of the status of the Odessa groundwater area, all irrigated agriculture in this area that was served by groundwater in the historical period was assumed to need surface water in the 2030 forecast to grow irrigated crops.
Evaluation of the VIC-CropSyst irrigation water demand simulations was primarily based on observed diversion data at Banks Lake (serving the Columbia Basin Project irrigated area in central Washington). Based on 2008, 2009 and 2010 data, observed irrigation diversions from Banks Lake were in the range of 2.5 to 2.7 million acre-feet per year. The VIC-CropSyst simulated "top of the crop" demand for the period 1977 to 2006 for this area was on average about 2.2 million acre-feet. The difference of 14-22% between the simulation results and observed diversions could be attributed to conveyance losses (which are included in the observed data, but not in the VIC-CropSyst values, which measure "top of the crop" demand). These values are within a reasonable range of expected losses. The WSDA based irrigated acreage extent used by the model for this region (730,000) also agreed reasonably well with 670,000 irrigated acres that the Columbia Basin Project serves, though it may be a bit on the high side.
Lack of high quality metered diversion data was an impediment to doing similar evaluations of modeling results at the watershed scale. Some crop acreage and irrigation demand estimates are indicated in the watershed plans of individual WRIAs, but these numbers have large uncertainties associated with them and are not appropriate for model result evaluation. This data gap needs to be addressed in the future.
Economic analysis was used to analyze historical changes in production and emerging trends within Washington, allowing for a forecast of how the crop mix is likely to change in the future in response to shifting economic and non-economic factors. Land use changes to predict movement of acreage into and out of agriculture were beyond the scope of this Forecast.
Within Washington, modeling captured the fact that over time, producers will respond to changes in the profitability of various crops resulting from changes in domestic economic growth and international trade flows. For example, over the last 20 years, Washington producers have begun to export increasing amounts of hay to meet a demand for hay in Asia, resulting from the growth in Asian meat and milk production to meet demand there. To carry out this analysis, the Forecast used low, medium, and high scenarios for domestic economic growth and international trade. These scenarios were based on statistical projections so that the medium scenario for domestic growth and international trade can be interpreted as the most likely future condition, while the low and high scenarios provide lower and upper bounds on what is likely to happen.
Domestic economic growth captured variation in the growth of the domestic economy and population, which impacts the amount of money that households have to spend on goods. International trade captured variation in imports and exports of agricultural goods, which are an important source of demand for many crops in Washington. Approximately one third ($2.6 billion) of Washington's agricultural production is exported internationally. The trade analysis was based primarily on historical trends in international imports and exports at the state level for broad crop categories, including fruits, vegetables, and wheat, using data provided by the USDA. A detailed analysis was performed for specific crops such as alfalfa and wine grapes that were deemed to be particularly sensitive to assumptions made about changes in trade flows.
Due to resource limitations, it was not possible to model all the ways in which producers could adapt to a reduction in water availability. For example, some producers may switch into less water-intensive crops, particularly if curtailment becomes more regular in the future. In the long run, they may also increase irrigation efficiency by investing in more efficient irrigation infrastructure, or by investing in improved irrigation timing.
Our more simple approach was to try to capture how producers attempt to mitigate water shortages within a growing season by allowing for selective deficit irrigation of less profitable crops. This provides an upper bound on the negative impacts of reduced water availability on production and profitability. A more complex representation of producer decision-making is expected to be a point of emphasis for the 2016 Forecast.
A set of water management scenarios were developed to assess how increasing water availability would affect agricultural production and water use. Working from the baseline scenario of no added capacity, the Forecast examined the following possible water management changes:
• Three different scenarios for water capacity enhancement,corresponding to approximately 100,000, 200,000, and 500,000 acre-feet of additional capacity at specific sites (at no cost to users for new water).
• Recovering direct costs of additional water capacity development at $25, $100 or $200 per acre-feet per year.
The consideration of additional water capacity was based on a list of specific conservation and storage projects currently being considered by OCR that would make additional water available for instream and out-of-stream uses. Details of the projects considered are provided in WSU's technical report (Ecology Publication 12-12-001). One important constraint relevant to the water capacity analysis was that most of the projects OCR is considering would provide water for drought relief or new permits. WSU assumed that any newly irrigated land would have approximately the same mix of crops as is present on nearby farmland, based on the fact that the extent of irrigated production in the Columbia River Basin is primarily constrained by water availability.
In addition to considering the impacts of additional capacity on water demand, WSU analyzed the economic impacts of additional capacity in terms of additional output, employment and tax revenue. The analysis used IMPLAN® data and software, a standard input/output model that captures the interlinkages between industries in our region. This specific package was chosen because it delineates between agriculture sectors by general crop types such as fruits, vegetables, and grains. Out-of-stream water allocated for newly irrigated land was accounted for on a project specific basis at the county level. New water was allocated to new irrigated crops based on the baseline future county-level crop mix for irrigated crops. The conversion of water into land was based on yields under future climate conditions.
The exploration of cost-recovery for the direct costs of developing water was structured to provide information about the potential feasibility of cost recovery strategies for supporting development of new water capacity. The analysis thus considered whether increases in prices would decrease the amount of water demanded by users or impact the total amount of cost recovery that could be expected. Potential changes in the costs of new water were considered on a crop specific basis. The analysis captured the fact that increased costs for water may prompt farmers to adopt new business practices. For example, they may choose to invest in more efficient watering systems, change their crop production choices, or make other changes to use less water.
Three possible prices that could be charged for cost recovery were explored. Existing projects in the region that have attempted to recover some development costs have charged in the neighborhood of $35 per acre-feet. The low price of $25 was considered to approximate this price point. The medium price, $100, was chosen to represent the high end of what has been observed in actual market transactions for agriculture in the region, while $200 was meant to represent a possible high price in the future. The total amount of cost recovery funds that could be expected was determined by discounting the stream of payments received over time into a single present value. Because this Forecast does not consider costs of specific projects it was not necessary (or possible) to directly deal with whether the prices would allow for complete recovery of costs, whether supply costs or economic costs.***
* The U.S. Geological Survey estimated that agriculture represented 61% of out-of-stream water use statewide, considering municipal, domestic, irrigation, stock water, aquaculture, industrial, mining, and thermoelectric uses. Within eastern Washington, irrigation represented 82% of all uses except thermoelectric (which could not be separated regionally due to limitations in data presentation). Lane R.C. 2009. Estimated water use in Washington, 2005. U.S. Geological Survey Scientific Investigations Report 2009-5128, 30 p. (back)
** Ibid. (back)
*** Supply costs normally include capital charges as well as operation and management costs, while economic costs also include opportunity costs.
Municipal use represents a much smaller portion of water use than agriculture in the Columbia River Basin, but one that is important for supporting the continued prosperity of the region.* For areas of the Columbia River Basin outside Washington State, WSU reviewed existing municipal projections. Within Washington, municipal demand, including self-supplied domestic use and municipally-supplied industrial use, was forecasted and integrated with modeling.
Municipal forecasting in Washington State relied on data from water system plans submitted to the Washington State Department of Health from the one to three largest public water systems in each WRIA, scaled to a common analytical base year of 2000. This generally captured a majority of residents in a WRIA. For those municipalities where data allowed, municipally-supplied industrial growth was also included, and was assumed to occur at the same rate as population growth, based on the difficulty of accurately forecasting industrial use using other methods.** Self-supplied industries were outside the scope of this Forecast. These figures were used to compute an Average Daily Demand (ADD) in terms of gallons per capita per day (gpcd). In some instances, diversions were much higher because of system leaks.
Using county-level population estimates obtained from the Washington State Office of Financial Management, city populations were counted in their primary WRIA, while projected county-level population growth outside of cities was distributed evenly by WRIA. Calculations of total WRIA water demand assumed that all people in the WRIA would use the average demand of nearby municipalities. Growth in rural demand will likely be met by groundwater supplies, but it was assumed that domestic wells would be shallow enough to impact surface water flows. Because municipal systems account for only about 10% of consumptive water use in the Columbia River Basin, economic scenario analysis (to explore the impacts of variations in economic growth and trade on water demand) was not carried out for the municipal forecasting.
Consumptive use was estimated by examining the difference between water diversions and discharges at corresponding wastewater treatment plants, while recognizing the potential for significant discrepancies due to municipal inflow and infiltration. Evidence from other western locations shows that loss or addition of flow due to groundwater exchanges in aging wastewater collection systems can be significant. The Utah Division of Water Resources has traditionally estimated the fraction between winter (indoor) water diversions and wastewater discharges to be approximately 0.90 (Oregon uses 0.80-0.90),*** but a study of 52 municipal systems in Utah found great variability in this ratio.**** In our analysis, 28 of 34 WRIAs produced values where wastewater treatment plant discharges were less than diverted amounts, producing positive consumptive use values. The average of the 28 positive values was substituted for the six negative values when calculating consumptive uses.
Municipal demands were incorporated into modeling of water supply and agricultural water demand by withdrawing consumptive demands from the surface water system when water system plans or other evidence confirmed that municipal systems were supplied by surface water, or by groundwater in close hydraulic continuity with surface water supplies.
* The U.S. Geological Survey estimated that domestic uses (including public and self-supplied) represented 11% of out-of-stream water use statewide, considering domestic, irrigation, stock water, aquaculture, industrial, mining, and thermoelectric uses. Within eastern Washington, domestic uses represented 13% of all uses except thermoelectric (which could not be separated regionally due to limitations in data presentation). Lane 2009, op. cit. (back)
** Not all water supply plans include industrial use information; therefore, this could not be included for all WRIAs. (back)
*** Cooper, RM. 2002. Determining surface water availability in Oregon. Open File Report SW 02-002. Oregon Water Resources Department, Salem, OR. http://www.oregon.gov/OWRD/WR/docs/SW02-002.pdf?ga=t (back)
**** Among the 52 municipal systems 63% suffered from excess infiltration or exfiltration, with 17 ratios greater than 1.0 and 16 ratios less than 0.70. The remaining systems averaged a supply/effluent ratio of 0.83 during the winter. Similar analysis of summer flows revealed a return flow ratio of 0.51 indicating nearly half the flow is used for outside irrigation. Hughes, TC. 1996. Consumptive use of municipal water supply. Utah Water Research Laboratory, Logan, UT. http://www.cachecounty.org/docs/water/docs//Consumptive%20Use%20of%20Municipal%20Water%20Supply.pdf (back)
The waters of the Columbia River Basin support a variety of fish and other wildlife important to maintaining cultural, environmental, and recreational opportunities, including several ESA-listed threatened and endangered fish stocks (see table below). Wildlife and fish (including both listed and non-listed species) help support a vibrant tourism, recreation, and fishing industry in the Columbia River Basin, one that plays a vital role in maintaining the rural economy. Recreational spending associated with fishing, hunting, and wildlife viewing was estimated to be $3.1 billion statewide in 2006, according to a study by the U.S. Department of Fish and Wildlife.*
While Ecology recognizes the value of all fish and wildlife, Chapter 90.90 RCW directs OCR to focus on salmonids. Across the Washington portion of the Columbia River Basin, OCR developed a comprehensive database of available historic flow data for each major tributary to the Columbia River. Using this data, OCR compared historic low, average, and high flow water years to state and federal minimum instream flow targets. This work was intended to improve understanding of
• How often minimum flow targets in fish critical basins are being met.
• How often water users subject to minimum flow targets are curtailed.
• Whether trends exist in the historic data relative to water availability, the shape of the hydrograph, or drought severity.
• Where opportunities exist to improve stream conditions by re-timing or re-locating water.
WSU's modeling also integrated quantitative instream flow requirements in the Washington portion of the Columbia River Basin. Within WRIAs, the highest adopted state and federal instream flows for each month were used to express current minimum flows for fish in both the historical and 2030 forecast. State and federal instream flows along the mainstem were also compared to historical and future supplies.
In addition to this work that covered the Washington portion of the Columbia River Basin, OCR contracted with the WDFW to provide information on instream water demands for eastern Washington's eight fish and low flow critical basins:
The Columbia River Instream Atlas (Atlas, Ecology Publication 11-12-015) presents WDFW's analysis of existing data, best professional knowledge, and new data for 189 stream reaches. Each reach was scored on three critical components: fish stock status and habitat utilization, fish habitat condition, and stream flow. This allowed for comparisons of stream reaches within each of the WRIAs. WDFW's results were at a finer geographic scale than WSU's modeling analysis, and were qualitative rather than quantitative. Thus they are presented independently in the Atlas. OCR will use the information in the Atlas, and consultations with WDFW staff, to identify and prioritize projects that benefit stream flows.
ESA Listing Unit by region
|Lower Columbia River|
|Southwest Washington/Columbia River Coastal Cutthroat||Candidate|
|Columbia River Chum||Threatened|
|Lower Columbia River Bull Trout||Threatened|
|Lower Columbia River Chinook||Threatened|
|Lower Columbia River Coho||Threatened|
|Lower Columbia River Steelhead||Threatened|
|Mid-Columbia River Spring Run Chinook||Not Warranted|
|Middle Columbia River Bull Trout||Threatened|
|Middle Columbia Steelhead||Threatened|
|Touchet/Walla Walla (Oregon Recovery Unit) Bull Trout||Threatened|
|Snake River Basin|
|Snake River Sockeye||Endangered|
|Snake River Basin Steelhead||Threatened|
|Snake River Bull Trout||Threatened|
|Snake River Fall Run Chinook||Threatened|
|Snake River Spring and Summer Run Chinook||Threatened|
|Upper Columbia River|
|Upper Columbia River Bull Trout||Threatened|
|Upper Columbia River Spring Run Chinook||Endangered|
|Upper Columbia River Summer and Fall Run Chinook||Not Warranted|
|Upper Columbia Steelhead||Threatened|
|Lake Wenatchee Sockeye||Not Warranted|
|Okanogan River Sockeye||Not Warranted|
|Northeast Washington Bull Trout||Threatened|
*Numbers for eastern WA were not available. U.S. Department of the Interior, Fish and Wildlife Service, and U.S. Department of Commerce, U.S. Census Bureau. 2006. National survey of fishing, hunting, and wildlife-associated recreation. http://www.census.gov/prod/www/abs/fishing.html (back)
According to the Northwest Power and Conservation Council, the more than 75 major federal and nonfederal hydroelectric dams in the Columbia River Basin produce upwards of 15,000 annual average megawatts (MWa) of energy.* This relatively inexpensive source of power accounts for approximately 55% of the power generating capacity in the Pacific Northwest and on average provides about three quarters of the region’s electricity. From a power generation perspective, the most significant of these dams are on the mainstem.
Power entities in the northwest regularly carry out extensive forecasting of electricity demand and power-generating capacity. For this Forecast, WSU reviewed existing projections across the Columbia River Basin with two specific objectives in mind:
Available reports that were reviewed included those carried out by the Bonneville Power Administration (BPA), Northwest Power and Conservation Council (NWPCC), Avista, Idaho Power, Portland General Electric (PGE), and Grant County PUD. BC Hydro documentation was also reviewed, though long-term planning documents were general in nature. Reviews were supported with conversations with staff at public utility districts in Washington State and Avista Utilities.
*NWPCC. 2010. 6th Northwest Conservation and Electric Power Plan. Northwest Power and Conservation Council. http://www.nwcouncil.org/energy/powerplan/6/default.htm
Office of Columbia River
15 W. Yakima Ave., Suite 200
Yakima, WA 98902
For copies of the 2011 Columbia River Basin Long-Term Water Supply and Demand Forecast,
please refer to Publication No. 11-12-011.
For copies of the Columbia River Instream Atlas, please refer to Publication No. 11-12-015.
Copyright © Washington State Department of Ecology. See http://www.ecy.wa.gov/copyright.html.