Science and Values in River Restoration in the Grand Canyon There is no restoration or rehabilitation strategy that will improve the status of every riverine resource R estoration of riverine ecosystems is often stated as a management objective for regulated rivers, and floods are one of the most effective tools for accomplishing restoration. The National Research Council (NRC 1992) argued that ecological restoration means returning "an ecosystem to a close approximation of its condition prior to disturbance" and that "restoring altered, damaged, Of destroyed lakes, rivers, and wetlands is a high-priority task." Effective restoration must be based on a clear definition of the value of riverine resources to society; on scientific studies that document e~osystem status and provide an understanding of ecosystem processes and resource interactions; on scientific studies that predict, measure, and monitor the effectiveness of restoration techniques; and on engineering and economic studies John C. Schmidt (e-mail: jschmidt@cc.usu. edu) is an associate professor in the Department of Geography and Earth Resources, Dtah State University, Logan, UT 84322. Robert H. Webb (e-mail: rhwebb@ sunlpaztcn.wr.usgs.gov) is a research hydrologist at the Desert Laboratory, US Geological Survey, Tucson, AZ 85745. Richard A. Valdez (e-mail: valdezra@aol. com) is a senior aquatic ecologist at SWCA Inc., Salt Lake City, UT 84101. G. Richard Marzolf (e-mail: rmarzo!f@usgs.gov) is chief of the Branch of Regional Research, Eastern Region, Water Resources Division, US Geological Survey, Reston, VA 21092. Lawrence E. Stevens (e-mail: !stevens@gcrs. uc.usbr.gov) is an ecologist at the Grand Canyon Monitoring and Research Center, Flagstaff, AZ 86001. © 1998 American Institute of Biological Sciences. September 1998 If flooding is crucial to the recovery of flood-adapted species but the absence of floods is crucial to the conservation of terrestrial endangered species in new habitats, then managers face an intractable dilemma that evaluate societal costs and benefits of restoration. In the case of some large rivers, restoration is not a self-evident goal. Indeed, restoration may be impossible; a more feasible goal may be rehabilitation of some ecosystem components and processes in parts of the river (Gore and Shields 1995, Kondolf and Wilcock 1996, Stanford et al. 1996). In other cases, the appropriate decision may be to do nothing. The decision to manipulate ecosystem processes and components involves not only a scientific judgment that a restored or rehabilitated condition is achievable, but also a value judgment that this condition is more desirable than the status quo. These judgments involve prioritizing different river resources, and they should be based on extensive and continuing public debate. In this article, we examine the appropriate role of science in determining whether or not to restore or rehabilitate the Colorado River in the Grand Canyon by summarizing studies carried out by numerous agencies, universities, and consulting firms since 1983. This reach of the Colorado extends 425 km between Glen Canyon Dam and Lake Mead reservoir (Figure 1). Efforts to manipulate ecosystem processes and components in the Grand Canyon have received widespread public attention, such as the 1996 controlled flood released from Glen Canyon Dam and the proposal to drain Lake Powell reservoir. The importance of the river and the dam The Grand Canyon is the most famous and extensive canyon in the world; approximately 5 million people visit Grand Canyon National Park each year. Whitewater recreation on the Colorado River is internationally renowned, and 25,000 people travel the river through the Grand Canyon annually. This segment of the Colorado River is a federally designated critical habitat for two endemic endangered fish: the razorback sucker (Xyrauchen texanus) and the humpback chub (Gila cypha). Riparian vegetation along the Colorado River in the Grand Canyon is a federally proposed critical habitat for the endangered Kanab ambersnail (Oxyloma haydeni kanabensis) and the southwestern willow flycatcher (Empidonax traillii). The 735 Downloaded from https://academic.oup.com/bioscience/article-abstract/48/9/735/259413 by guest on 01 December 2018 John C. Schmidt, Robert H. Webb, Richard A. Valdez, G. Richard Marzolf, Lawrence E. Stevens ·________________~lrI3~·--------r_.i~A~H._-2'¥'2C·~~----~ ....~"""" ARIZONA LAKE 37 r \ GLEN POWELL CANYONDAM J LAKE MEAD / ! o km 50 Cardenas Creek Figure 1. The Grand Canyon region. 25 km reach between Glen Canyon Dam and Lees Ferry is a blue-ribbon trophy fishery for non-native rainbow trout (Oncorhynchus mykiss), The river provides essential water for humans as well. Water from the Colorado River has been diverted to southern California for 90 years and is being increasingly diverted to cities in the Wasatch Front in Utah, the Front Range in Colorado, southern Nevada, and central and southern Arizona. The discharge of the Colorado River is relatively small for the basin's size: The mean annual discharge at Lees Ferry was only 505 mJls (15.9 x 10'm'/yr) between 1912 and 1963, before the dam was built (USGS 1996). Therefore, large reservoirs have been constructed to assure water availability, and the Colorado River has the largest reservoir storage capacity in relation to annual discharge of any major watershed in the United States (Hirsch et al. 1990). The potential for flood control and sediment retention by these reservoirs is nearly complete, and restoration or rehabilitation can be achieved only by changing the dams or their operations. Many aspects of water-release Table 1. Controlling factors and ecological processes of the Colorado River in the Grand Canyon that can, and cannot, be manipulated by Glen Canyon Dam. Relation to the existence or operations of Glen Canyon Dam Controlling factors Ecological processes Unrelated Regional climate; regional geology and geomorphology; human activities (prehistoric settlement, spatial patterns of water use, growth in water demand, regional land-use changes, recreational demand); non-native species invasions Regional land use; tributary floods and debris flows; stagedischarge relations of the Colorado River; solar insolation and downstream rate of water warming; regional expansion of some native and non-native species Related Discharge; water temperature; sediment; nutrients; woody debris; non-native species introductions Lake Powell limnology and mainstem sediment transport; stratification; sediment accumulation on river bed; transfer of sediment to eddies and sediment accumulation in eddies; sandbar stability; ice formation and transport; dissolved load transport; woody debris transport and decomposition; aquatic and terrestrial productivity 736 The Colorado River ecosystem in the Grand Canyon The Colorado River ecosystem in the Grand Canyon is sustained by the flow of water and nutrients released by Glen Canyon Dam, but other controlling factors are unrelated to the BioScience Vol. 48 No.9 Downloaded from https://academic.oup.com/bioscience/article-abstract/48/9/735/259413 by guest on 01 December 2018 3S· policy from Glen Canyon Dam are controlled by statutory or administrative rules that are related, directly or indirectly, to the seven-state Colorado River Compact of 1922 that allocated water use among the states. Water released from Glen Canyon Dam constitutes a delivery of water from the upper basin to the lower basin because the division point between the basins is near Lees Ferry. Glen Canyon Dam is the largest dam of the Colorado River Storage Project (CRSP)j its power plant produces approximately 75% of the total CRSP power for a six-state region. Lake Powell holds 80% of the upper Colorado River basin's stored water supplies. The management of that portion of the Colorado River that flows through the Grand Canyon reflects the interests and values of many management and regulatory agencies. The federal agencies that manage or monitor Glen Canyon Dam and the ecological resources of the Colorado River include the US Bureau of Reclamation, the National Park Service, the US Fish and Wildlife Service, and the Grand Canyon Monitoring and Research Center. The Western Area Power Administration markets the power produced by the CRSP and partly determines the daily releases from each CRSP dam. The Arizona Game and Fish Department manages the sport and native fish populations. Approximately 200 km of the left side of the river (facing downstream) forms the reservation boundaries for the Navajo and Hualapai tribes, and five additional Native American tribes have vested interests in Grand Canyon river management. Other interested parties include numerous municipalities and agricultural organizations that use water and electrical energy, national and regional environmental groups, commercial river-running companies, and professional trout-fishing guides. Figure 2. Hydrograph , t dam, such as regional geology and geomorphology, climate, tributary inflows of water and sediment, and human activities (Table 1). Changes in these factors have caused adjustments in channel geomorphology, alterations in riparian vegetation and fish assemblages, decreases in habitat availability for endangered fish, and changes in water temperature and quality. Deciding whether to restore or rehabilitate the Colorado River ecosystem requires an understanding of the role of Glen Canyon Dam, in relation to other factors, in causing ecosystem change and the potential to reverse these changes. of water year 1996compared to pre-dam and post-dam hydrographs for the Colorado River at Lees Ferry, Arizona. (a) Pre-dam(1922-1962), (b) Post-dam (19631995). The heavy black line is the hydrograph for 1996and is the same on both panels. The dashed, solid, and dotted lines connect the mean daily discharge values for each date below which 90%, 50%, and 10% of the years, respectively, occur. Water discharge and sediment transport. The construction and operation of Glen Canyon Dam reduced the frequency, magnitude, and duration of floods through the Grand Canyon. Before the dam was constructed, peak discharge occurred in late spring following snowmelt in the Rocky Mountains (Figure 2). The magnitude of the two-year recurrence flood for the period 19211962 was 2150 m 3/s, and flows that exceeded 1250 m 3/s were typically sustained for 30 days or more (Table 2). Short-duration floods occurred in September and October. Since the dam's completion in 1963, the magnitude of annual high flows is' determined by the magnitude of inflows and the elevation of the reservoir when these inflows occur (Figure 3). Lake Powell did not fill to capacity until 1980, and dam releases were always less than the capacity of the Glen Canyon Dam power plant, which is approximately 891 m 3/s. Large-magnitude dam releases occurred in 1980, and annually between 1983 and 1986, because the reservoir elevation was high and inflows were large. The twoyear recurrence flood at Lees Ferry was 679 m 3 /s for the period 19631996; flood flows of 1250 m 3/s or more now occur less than 1 % of the time (Garrett and Gellenbeck 1991). The 1996 controlled flood of 1272 m 3/s was much smaller than typical pre-dam floods (Figure 2). Dam releases between 1964 and 1990 were characterized by large, hourly flow fluctuations resulting from load-following hydroelectric power production in response to regional demand. The average susZ500 pended-sediment ~ load of the Colorado Ul zooo 1996 River at Lees Ferry ~ was approximately h! 1500 50th Percentile 6.0 X lOW kg/yr bef fore construction of ~ 1000 ~ 10th Percentile the dam. An average additional 1.8 x 10 lD kg/yr is contributed by tributaries downo 50 100 150 200 2SO 300 350 stream from Lees DAYS, BEGINNING OCTOBER 1 Ferry; 70% of that amount comes from the Paria and volume and thus narrowed adjacent Little Colorado Rivers (Andrews rapids because the magnitude of post1990). The magnitude of this annual dam floods has been too small to sediment resupply varies greatly. In transport the coarse debris delivered 1964 and 1965, after the dam was to the river since the dam was comcompleted, the average annual sus- pleted. As rapids narrow, they popended-sediment load of the Colo- tentially become more difficult to rado River at Lees Ferry was only navigate and pose safety hazards. 0.000013 x lOW kgiyr because Lake Thus, high dam releases might be Powell traps all the sediment trans- used to rework accumulating coarse ported from the upper Colorado debris. Releases at maximum powerRiver basin. plant capacity rework parts of recent debris-flow deposits; larger dam reRiver corridor geomorphology. Res- leases, such as those that occurred toration options are determined in between 1983 and 1986 and during part by the geomorphic attributes of the 1996 controlled flood, caused the river corridor. The width of the substantial debris-fan reworking, but Colorado River in the Grand Can- they still did not entirely reverse the yon is constrained by bedrock, talus, narrowing trend (Kieffer 1985, Webb and debris fans (Howard and Dolan et a!. 1996). 1981). Dehris fans, which are comUnvegetated sandbars were a disposed of coarse debris supplied from tinctive landscape feature of the unsteep tributaries, partially constrict regulated river. Sandbars form in the channel and create rapids (Webb eddies that occur downstream from et al. 1989). Before the completion most debris fans (Schmidt 1990, of Glen Canyon Dam, mainstem Schmidt and Rubin 1995). These floods reworked debris fans and re- eddies have relatively low velocity moved all but the largest boulders and turbulence and are prominent from the rapids (Howard and Dolan sites of sand accumulation. Sand1981, Kieffer 1985, Webb et a!. 1989, bars are dynamic features subject to 1996). Debris fans have increased in deposition during floods and ero~ ~ ~ ~ •• ~ ~ I" zsoo zooo r I ~ SOth Percentile 1S00 1996 -----.. 1000 500 50 100 150 200 250 300 350 DAYS, BEGINNING OCTOBER 1 b 3000 r[~~~~~~~~~~~~"""'""'" ~ a J 1- 737 Downloaded from https://academic.oup.com/bioscience/article-abstract/48/9/735/259413 by guest on 01 December 2018 September 1998 i 3000 Table 2. Comparison of pre- and post-dam Colorado River resources downstream from Glen Canyon Dam. Changes occurred on widely differing time scales. Whereas physical changes (in flow, sediment load, and temperature) occurred rapidly after closure of the dam in 1963, geomorphic (e.g., debris-fan reworking and sandbar erosion and deposition) and biotic (e.g., trophic patterns and non-native species invasion) changes occurred more slowly and are ongoing. Before Glen Canyon Dam After Glen Canyon Dam Hydrologic regime Variable; two-year flood was caused by regional snowmelt that averaged 2150 ml/s between 1921 and 1962' Regulated; two-yeat flood of 679 m'/s is less than the power-plant capacity of 940 m'/s; large hourly fluctuations are associated with load-following power ptoduction Sediment load Variable; mean annual suspended sediment load at Lees Ferry was 6.0 x lO lD kg' Virtually zero in dam releases; mean annual contribution of 1.8 x 1010 kg from tributaries downstream from Lees Ferry··b.c Debris fans All but largest boulders from rapids frequently reworkedo,d., Debris flows continued, with consequent aggradation of rapids d,< River temperature Varied seasonally, from near freezing in winter to 25-30 °C in summerf.~ Nearly constant 8-10 °C because water is drawn from below thermal discontinuity in Lake Powell in summer; there is slight year-to-year variation in the temperature of the winter isothermal periodf.g,h,;,) Unvegetated sandbars Common; distinctive features associated with eddies downstream from debris fansk),m A near-river riparian zone has been established that consists of a matsh zone within the range of river stages regulated by power-plant operationsi,",o.P Trophic structure Thought to be heterotrophic because high sediment loads diminished light availability Autotrophic in dam tailwater and in nearshore or cobble-bars downstream"j Fish assemblage Eight native endemic species; 74% level of endemism is highest among North American rivers; heavily dependent on terrestrial food sources; some species extirpated, others endangeredf.q Warm-water fishes introduced to Lake Mead and ttout ro the tributaries by the 1930s; tailwater trout fishery is highly valued "Andrews (1990). bHoward and Dolan (1981). 'Randle et al. (1993). clWebb et al. (1989). "Webb er a1. (1996). fValdez and Rye! (1997). gMarzoif et a1. (1996). hStanford and Ward (1991). 'Stevens et a1. (1997a). JStevens et al. (1997b). kSchmidt (1990). ISchmidt et a1. (1995). mWebb (1996). "Turner and Karpiscak (1980). QJohnson (1991). pStevens et al. (1995). qMiller (1959). sian after flood recession (Rubin et al. 1990). Before the dam's completion, both the size of the sedimentcomprising eddy bars and the shape of these bars reflected the characteristics of sediment transported during recession from the annual spring peak as well as from lower-magnitude latesummer floods. Thus, pre-dam deposits that still exist in the Grand Canyon are typically 7000 ~ very fine sand mixed 6000 with silt and clay. Figure 3. Annual peak discharge of the Colorado River at Lees Ferry, Arizona. Solid line is a weighted average. The year of completion of Glen Canyon Dam (1963) and the magnitude of the 1996 controlled flood are indicated. 738 A sediment budget calculated for a 141 km reach immediately downstream from Lees Ferry indicates that fine sediment accumulates in the Grand Canyon despite the fact that no sediment is released from Glen Canyon Dam (Randle et al. 1993). Accumulation occurs because the undammed Paria and Little Colorado Rivers continue to contribute 5000 4000 3000 2000 \, ~, ' , ' , , , " , YEAR significant amounts of fine sediment to the Colorado River. This sediment accumulates on the channel bed and in eddies because the bars along the river's margin have typically eroded, not aggraded. At least 30% of all large, high~elevation sandbars in the Colorado River decreased in size between 1965 and 1973; 32% decreased in size between 1973 and 1991 (Kearsley et al. 1994). These decreases were caused by degradation and invasion by riparian vegetation. Although high discharges of water in 1983 caused a number of sandbars to increase in size, almost all of these bars had decreased to pre-1983 sizes by 1991. River runners use large sandbars as campsites, and decreases in sandbar size cause decreases in campsite carrying capacity (Kearsley et a1. 1994). Net post-dam erosion of sandbars may decrease with distance downstream from the dam (Webb BioScience Vol. 48 No.9 Downloaded from https://academic.oup.com/bioscience/article-abstract/48/9/735/259413 by guest on 01 December 2018 Riverine feature u Q) U r: ... ~ •J • 1<: A ....... """" ~~: , ~ u •0 Q. , - .... •