Intl. J. River Basin Management Vol. 5, No. 3 (2007), pp. 207–221
© 2007 IAHR, INBO & IAHS

The state of the Colorado River ecosystem in Grand Canyon: Lessons from
10 years of adaptive ecosystem management
JEFF LOVICH and THEODORE S. MELIS, United States Geological Survey, Southwest Biological Science Center,
2255 North Gemini Drive, MS-9394, Flagstaff, Arizona 86001, USA
ABSTRACT
The year 2005 marked the 10th anniversary of the completion of the Final Environmental Impact Statement (EIS) on the Operation of Glen Canyon
Dam on the Colorado River, USA. A decade of research and monitoring provides an important milestone to evaluate the effects of dam operations on
resources of concern and determine whether or not the desired outcomes are being achieved, or if they are even compatible with one another or not.
A comprehensive effort was undertaken to assess the scientific state of knowledge of resources of concern, as identified in the EIS. The result was the
first systematic attempt by scientists to conduct an assessment of the changing state of Colorado River ecosystem resources in Grand Canyon over
a decadal timeframe. In the EIS, 30 resource attributes are listed along with predictions for how those resources would respond under the Secretary
of the Interior’s 1996 Record of Decision, an operating prescription based on the preferred alternative of Modified Low-Fluctuating Flows (MLFF).
Because of a lack of data or subsequent analyses to confirm whether some predictions stated in the EIS were correct, or not, 14 or 47 percent of
the outcomes, are essentially unknown. Excluding outcomes that are unclear, then the remaining predictions in the EIS were correct in 7 out of 16
outcomes, or 44 percent of the categories listed. Mixed outcomes occur in 4 out of 16, or 25 percent of the categories, and failed predictions, occur
in 5 out of 16, or 31 percent of the categories. As such, less than 50 percent of the outcomes were predicted correctly, underscoring the uncertainties
associated with working in a large complex system with few to no long-term data sets. Similar uncertainties are faced by all resource managers
charged with ecosystem restoration globally. The acceptability of this kind of uncertainty is influenced by interpretation, societal values, agency
missions and mandates, and other factors. However, failure to correctly predict the future, in and of itself, is not deleterious under the paradigm
of adaptive management where large uncertainties provide opportunities for learning and adjustment through an iterative process of “learning-bydoing” (Walters and Holling, 1990). Although recent science has documented a continued decline of environmental resources of the Colorado River
below Glen Canyon Dam, it has also identified options that might still be implemented by managers to achieved desired future conditions in Grand
Canyon.

Keywords: Adaptive Management; Colorado River; Grand Canyon; Glen Canyon Dam; humpback chub; Grand Canyon Monitoring
and Research Center; sediment; sand.
Introduction

The Colorado River

March, 2005 marked the 10th anniversary since completion of the
Final Environmental Impact Statement (EIS) on the Operation of
Glen Canyon Dam (U.S. Department of the Interior, 1995). The
next year, 1996, saw the implementation of the Glen Canyon
Dam Adaptive Management Program (AMP) as a means of guiding the operation of Glen Canyon Dam to achieve the objectives
outlined in the EIS. Gloss et al. (2005) document the first systematic review of the effects of Glen Canyon Dam operations on
the natural and cultural resources of the Grand Canyon and thus
provide an important milestone for evaluating the success of the
program to date. This paper has four objectives: (1) to provide
an overview of the Colorado River Basin and Glen Canyon Dam
as background to the other objectives; (2) to briefly describe the
roles and functions of the Glen Canyon Dam Adaptive Management Program; (3) to discuss lessons learned over the last decade
and their applicability to dam management in general; and (4) to
discuss opportunities for future research.

The Colorado River is one of the most important rivers in the
western United States providing water for over 25 million people
(Schmit et al., 2005). According to information summarized by
Blinn and Poff (2005) the basin encompasses over 642,000 km2
and ranges from 42◦ N to 32◦ N latitude. Draining nearly 8% of
the land area of the United States, the Colorado River and its 22
major tributaries flow through seven states (Wyoming, Colorado,
Utah, Arizona, Nevada, New Mexico, and California), and two
countries, including a small part of Mexico.
The river heads at over 3,000 m in the Rocky Mountains of
the United States and unregulated flows were driven largely by
snowmelt before the construction of Glen Canyon Dam and other
dams. Flows through the Grand Canyon ranged from a low of
85 m3 /s in the late summer to about 2,300 m3 /s in the late spring.
Flood flows of 8,500 m3 /s were recorded in historical times and
flows of greater than 14,160 m3 /s may have occurred in the last
1,600 years. The annual unit area discharge of 29,800 m3 /km2

Received on May 24, 2006. Accepted on September 1, 2006.

207

208

Jeff Lovich and Theodore S. Melis

or runoff of 2.98 cm/yr gives the Colorado River basin the distinction of being one of the driest in the world (Blinn and Poff,
2005). It is also one of the most regulated and impacted rivers in
the world with over 40 large flow-regulation structures and diversions along its length (Blinn and Poff, 2005; Nilsson et al., 2005).
Most of the runoff (64%) is used for irrigation and another 32%
is lost to evaporation from reservoirs. Additional information on
the river is provided by Blinn and Poff (2005) and Topping et al.
(2003) and the reader is referred there for details beyond those
presented in this paper.
The river flows through Grand Canyon National Park creating
one of the world’s premier river canyons. The park was designated a World Heritage Site by the United Nations Educational,
Scientific and Cultural Organization in 1979. Created in 1919,
the park encompasses almost 500,000 ha. The canyon is approximately 450 km in length, 24 km across at its widest point, and over
1,800 m deep at some points. In addition to its physical attributes,
the canyon has important cultural significance to several Native
American tribes (Fairley, 2005).

photosynthetic activity to occur in the river bed and large reaches
of the river now have energy inputs dominated by autochthonous
instream production (Kennedy and Gloss, 2005). However, tributaries below the dam, such as the Paria and Little Colorado
Rivers, intermittently add high concentrations of fine sediment
to the main channel (Wright et al., 2005). In addition, because the
water enters penstocks deep below the surface of the lake before
being routed through the turbines, discharges are cold year round
ranging from 7◦ to 12◦ C (Vernieu et al., 2005).
Post-dam changes in the river environment have not been
favorable to native fish populations, while fostering the dominance of non-native fish species (Gloss and Coggins, 2005),
especially salmonids (rainbow trout – Onchorhyncus mykiss and
brown trout – Salmo trutta). In addition to these ecological effects
of dam operations there have been beneficial effects on recreational river use (Loomis et al., 2005) and water and power uses
(Harpman and Douglas, 2005). Additional details on these benefits are provided in the sections on hydropower and recreation
below.

Glen Canyon Dam

Glen Canyon Dam Adaptive Management Program

Construction of Glen Canyon Dam began in 1957, and the last
bucket of concrete was poured in 1963. The concrete arch structure is 216 m high and impounds an approximately 300 km long,
65,315 ha reservoir named Lake Powell. The storage capacity of
Lake Powell is about 3.42 × 1010 m3 . Eight generating units are
fitted at the dam with a combined output of about 1.3 million kilowatts (Bureau of Reclamation, 2006). These units are typically
operated to meet “peaking” electrical demand in the southwestern
U.S., with diurnal releases ranging from about 142 to 708 cubic
meters per second. The dam is situated about 24 km upstream
from the eastern boundary of Grand Canyon National Park.
Prior to the construction of the dam the Colorado River in
the Grand Canyon was characterized by highly variable flows
(see above) and high sediment loads. The volatile nature of the
river contributed to regular scouring of the riparian habitat. Water
temperatures ranged from a low of about 0◦ to a high of about
29◦ C. The sediment-laden water prevented significant development of aquatic plants and as such, energy inputs into the system
were dominated by allochthonous sources such as driftwood and
terrestrial leaf litter (Kennedy and Gloss, 2005). This low productivity likely contributed to low species diversity, but the harsh
conditions fostered the development of a significant endemic fish
fauna. Eight fish species occurred in the Grand Canyon prior to
the construction of the dam (Mueller and Marsh, 2002) and four
species are now extirpated (see below).
After the construction of the dam, flows were altered significantly by removing most of the seasonal and annual variation
that occurred under natural conditions (Topping et al., 2003).
Flood frequency was significantly reduced by dam operations.
Most variation in the system now occurs on a daily basis to
enhance hydropower generation. The dam also traps about 84%
of the sand that formerly entered Grand Canyon resulting in
discharges of clear water instead of the turbid water that characterized the pre-dam environment. The increased water clarity allows

Glen Canyon Dam was completed before the passage of the U.S.
National Environmental Policy Act in 1969 or the Endangered
Species Act in 1973. Environmental concerns regarding the dam
initially focused on inundation of Glen Canyon upstream from
Grand Canyon. However, after construction of the dam was completed, concern mounted over the effects of dam operations on
downstream resources in Grand Canyon National Park. Apprehension regarding the environmental effects of the dam led to
a series of environmental studies from 1982 through 1996, passage of the Grand Canyon Protection Act in 1992, and ultimately,
completion of the Final EIS and Record of Decision in 1995 and
1996 (U.S. Department of the Interior, 1995, 1996), respectively
(see Schmit et al., 2005; Stevens and Gold, 2003 for more details
on the evolution of the program).
All of the alternatives considered in the EIS recommended
the implementation of “adaptive management” as a framework
for dam management, and the Record of Decision institutionalized that approach when the preferred alternative (Modified Low
Fluctuating Flows [MLFF]: see Table 1 for defining parameters)
was selected. The EIS (U.S. Department of the Interior, 1995)
states:
“It is intended that the ROD [Record of Decision]
will initiate a process of ‘adaptive management,’
whereby the effects of dam operations on downstream resources would be assessed and the results
of those resource assessments would form the basis
for future modifications of dam operations. Many
uncertainties still exist regarding the downstream
impacts of water releases from Glen Canyon Dam.
The concept of adaptive management is based on
the recognized need for operational flexibility to
respond to future monitoring and research findings
and varying resource conditions.”

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

209

Table 1 Operating limits and general likelihood of occurrence under the preferred alternative (U.S. Department of the Interior, 1995) of Modified Low
Fluctuating Flows. Conditions are based on operating rules constrained by annual hydrology.

Minimum releases 7a.m.–7 p.m.
(cfs)
Minimum releases 7p.m.–7 a.m.
(cfs)
Maximum Peak under diurnal
releases (cfs)

General Range of Hydrologic Conditions for Glen Canyon Dam
Dry (minimum of 8.23–10 Million Normal (10–15 Million Acre Feet
Acre Feet of annual release)
of annual release)
8,000 (likely only during
8,000 (unlikely to occur)
weekends)
5,000 (very likely to occur on
5,000 (unlikely to occur)
weekends)
25,0001 (daily peaks reach about
25,000 (daily peaks reach about
20,000–24,000 cfs, mainly during
18,000–19,000 cfs, mainly during
summer)
summer)

Wet (15–20 Million Acre Feet of
annual release)
8,000 (very unlikely to occur)
5,000 (very unlikely to occur)
25,000 (steady flows at this level
occur for 1.5 MAF monthly
releases)

Daily fluctuations (cfs/24 hrs)

5,000 6,000 (all possible) 8,0002

5,000 (unlikely) 6,000 (possible)
8,0003 (Most Likely)

5,000 (unlikely) 6,000 (unlikely)
8,0004 (most likely)

Ramp rate (cfs/hr)

4,000 up (always) 1,500 down
(always)

4,000 up (always) 1,500 down
(always)

4,000 up (always) 1,500 down
(always)

Monthly volume (Million Acre
Feet)

480,000–900,000

700,000–1,200,000

800,000–2,000,000

1 Exceeded during habitat maintenance flows.
2 Maximums represent normal or routine limits
3 Maximums represent normal or routine limits
4 Maximums represent normal or routine limits

and may necessarily be exceeded during high water years.
and may necessarily be exceeded during high water years.
and may necessarily be exceeded during high water years.

Thus, the Glen Canyon Dam Adaptive Management Program
(AMP) was created. Adaptive management within the context of
Glen Canyon, (1) embraces a relatively new paradigm of natural
resource agencies governing with input from stakeholders; (2)
recognizes the uncertainty associated with agency decisions and
policies; (3) allows stakeholders to make formal recommendations to the U.S. Secretary of the Interior via a Federal Advisory
Committee; and (4) places a premium on stakeholders developing
a shared vision.
The structure of the AMP is shown in Figure 1. Ultimately,
the U.S. Secretary of the Interior has statutory responsibility for
controlling water on the Colorado River. The Federal Advisory
Committee is called the Adaptive Management Work Group

Figure 1 Organizational structure of the Glen Canyon Dam Adaptive
Management Program modified from Stevens and Gold (2003). The
Designee refers to the Chair of the Adaptive Management Work Group
established by the Secretary of the Interior. GCRMC refers to the Grand
Canyon Monitoring and Research Center of the U.S. Geological Survey.
Refer to text for additional details.

(AMWG) and is comprised of 24 members (Table 2) and a
Chair designated by the Secretary of the Interior. AMWG members focus on policy issues and make their recommendations
on dam operations to the Chair who, in turn, communicates
them to the Secretary for final decisions on workplans, funding, experimentation, etc. In turn, each AMWG member has a
technical representative in the Technical Work Group (TWG) that
interfaces with the science components of the program.
The primary science provider for the AMP is the U.S. Geological Survey, Grand Canyon Monitoring and Research Center
(GCMRC). Created in 1995, the Center facilitates research and
monitoring in the program and communicates their findings to
the program in briefings, publications (e.g., Gloss et al., 2005)
and symposia. The mission of the GCMRC is to provide credible, objective scientific information to the Adaptive Management
Program on the effects of operating Glen Canyon Dam on the
downstream resources of the Colorado River ecosystem, utilizing
an ecosystem science approach
Independent Review Panels comprised of academics with
expertise germane to the program, are convened periodically to
provide peer review and quality control as needed. Included under
this Panel is a standing Science Advisory Board that works with
the TWG and GCMRC to bring recommendations and information to the AMWG at regular meetings. Meetings are posted in
the U.S. Federal Register and are open to the public.
Lessons learned in over a decade of research
and monitoring
After more than a decade of research and monitoring a systematic review of the impact of dam operations on natural and
cultural resources in Grand Canyon was published (Gloss et al.,

210

Jeff Lovich and Theodore S. Melis
Table 2 Stakeholders in the Glen Canyon Dam
Adaptive Management Program with membership
on the Federal Advisory Committee known as the
Adaptive Management Work Group.
Cooperating agencies
1. Bureau of Reclamation
2. U.S. Fish and Wildlife Service
3. Department of Energy
4. Bureau of Indian Affairs
5. National Park Service
6. Hopi Tribe
7. Hualapai Tribe
8. Navajo Nation
9. Southern Paiute Consortium
10. Pueblo of Zuni
State agency
11. Arizona Game & Fish Department
Environmental Groups
12. Grand Canyon Wildlands Council
13. Grand Canyon Trust
Recreation Interests
14. Grand Canyon River Guides
15. Federation of Fly Fishers
Basin States
16. Arizona
17. California
18. Colorado
19. Nevada
20. New Mexico
21. Wyoming
22. Utah
Federal Power Purchase Contractors
23. Colorado River Energy Distributors Association
24. Utah Associated Municipal Power

2005). One of the goals of this compilation was a review of the
predictions contained in Table II-7 of the EIS (summarized in
Table 3). The table contains a list of resources and associated
predictions on how those same resources would respond under
the preferred alternative. An expanded list of predictions in a
slightly different time scale is given in Schmidt et al. (1998).
During preparation of the EIS, the best scientific data were
used to generate those prognostications. From the precipice of
time, a decade later, we have significant new information to use
in evaluating the dam’s re-operation relative to objectives of the
1995 EIS, and the 1992 Grand Canyon Protection Act. In this section we will summarize what we have learned about the Colorado
River ecosystem from the perspective of over a decade of scientific inquiry, and discuss challenges facing continued research
and monitoring in support of the Glen Canyon Dam Adaptive
Management Program (AMP).
Integrated Quality-of-water – The presence and design of Glen
Canyon Dam have caused major environmental changes to the
Colorado River ecosystem including: (1) alterations in the timing

and variability of the river’s annual, season and daily flow patterns, (2) drastic reduction of the sand supply to the Glen, Marble
and Grand Canyon reaches and (3) reduced annual variability
in temperature. On the basis of current science information,
the MLFF operating alternative has not effectively mitigated
the influence of regulation with respect to either the thermal
and hydrologic changes, or the sand supply limitation of the
downstream ecosystem.
Detailed monitoring of water quality throughout Lake Powell
spans a period of about 40 years, making this one of the most
robust monitoring efforts conducted under the AMP. These data
coupled with downstream temperature data provide rich opportunities for the limnological modeling of the reservoir; an effort
that is currently underway. Given the importance of Lake Powell
as the major source of water for the Colorado River ecosystem
below, this monitoring program serves as an early warning system
for changes in quality-of-water.
While the information on the reservoir briefly summarized
by Vernieu et al. (2005) has shown that dam operations have
affected some resources downstream from Glen Canyon Dam,
water quality in this large reservoir appears to be largely unaffected by the new dam operations since 1991. However, the
reservoir does have the potential to exert substantial impacts on
downstream resources. For example, the current drought that
started in 2000, reduced the level and volume of Lake Powell
to elevations not seen since the reservoir was initially filling in
the late 1960’s. The lower reservoir storage level has brought
warmer surface (epilimnetic) waters to the penstocks causing
increased water temperatures to be discharged into the canyon.
In addition, continued reservoir drawdown has resulted in the resuspension of large amounts of deltaic sediment in the reservoir.
The organic material contained in this sediment caused a decrease
in oxygen concentrations in the reservoir that could potentially
be discharged downstream.
Overall, the water quality of the reservoir appears to be more
strongly linked to climatic annual to decadal variability governing spring inflow events, protracted swings in upper Colorado
River Basin hydrology and by continuing depletions basin-wide,
than it is by the re-operation of the hydropower plant over the
last decade under the preferred alternative (MLFF). This is true
because the Record of Decision did not change Long-Range
Operating Criteria tied to mandatory transfers from Upper to
Lower Colorado River Basin states. However, periodic release of
short-duration spills that bypass the hydropower plant (controlled
floods, termed “Beach-Habitat Building Flows”) are one element
of the current Record-of-Decision that does have an immediate
influence on the structure of the reservoir. The changes appear
related to the fact that such releases are made from outlet works
at depths beneath the power plant intakes where temperatures
and oxygen levels tend to be lower. Potentially, releases made
from higher elevations in the reservoir (by means of a Selective Withdrawal Structure [SWS]) would also change reservoir
characteristics, but with uncertain influence on the downstream
ecosystem, particularly with warmer discharges.
The effects of warmer water on downstream biological
resources are difficult to predict with certainty and potentially

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

211

Table 3 Natural and cultural resources of the Colorado River ecosystem and predictions from the Final Environmental Impact Statement (EIS) for
the Operation of Glen Canyon Dam (U.S. Department of the Interior, 1995) on how they would respond under the preferred alternative of Modified
Low Fluctuating Flows. Resources are ordered as they appear in the EIS. Symbols are as follows: a plus sign (+) indicates that prediction was correct
or exceeded expectations, a minus sign (−) indicates that the prediction was not entirely correct or did not achieve the desired outcome. A “plus”
and “minus” (+/−) sign together indicate a mixed outcome. Data unavailable may imply a total absence of data or that the data are not available to
the USGS, GCMRC through the scope of their research and monitoring program.
Resource
SEDIMENT & AQUATICS
Fine sediment (sand bars and related
physical habitats linked to native fishes
[backwaters], terrestrial vegetation,
marshes, campsites for recreation and
in-situ preservation of archeological
resources).

Coarse sediment5 (debris flow impacts
from tributaries and their influence on
the navigability of rapids and terrestrial
sand bars).

Prediction

Outcome

Comments

Modest improvement through
implementation of constrained daily power
plant operations and periodic
implementation of Beach-Habitat Building
Flows following accumulation of new
tributary sand supplies in the main channel
of the ecosystem. Sand accumulation was
predicted to occur under average-to-below
average hydrology and associated
hydropower operations.
Inputs of coarse-grained sediment from
tributary debris flows will continue to
accumulate in the main channel under
constrained hydropower operations,
causing rapids to worsen and burying sand
bars under coarse deposits. High flow
releases may partially rework the new
deposits and improve navigation within
rapids.

−

Sand bars continued to erode and new sand
inputs were not accumulated within the
main channel. However, Beach-Habitat
Building Flows were allowed. Flexibility in
the timing and frequency of these
controlled floods limited their effectiveness

+

The influence of ongoing, naturally
occurring debris flows, in terms of
aggradation of rapids and burial of sand
bars, has been partially mitigated by
occasional Beach-Habitat Building and
Habitat-Maintenance Flows. The ability of
high dam releases to rework new debris
flow deposits is related more to peak
discharge and timing after debris-flow
events than it is related to duration of the
high releases

Aquatic food web

“Potential major increase”

±

Apparently increased in Lees Ferry reach,
but not necessarily canyon-wide, as
fine-sediment inputs from tributaries below
this reach are most likely limiting factor in
primary productivity

Native fish

“Potential minor increase”

±

Decrease in adult humpback chub, native
suckers may be stable or slightly increasing

Non-native fish

“Potential minor increase”

+

Rainbow trout population increased
substantially following the operational
change in the Lees Ferry reach and within
Marble Canyon.

Interactions between native and
non-native fish

“Potential minor increase in warm, stable
microhabitats”6

−

Exception owing to drought-lowered Lake
Powell levels not directly related to dam
operations

Trout

“Increased growth potential,
stocking-dependent”

−

Rainbow trout numbers have increased in
the Lees Ferry reach, body condition has
declined, and stocking is not required

VEGETATION
Woody plants

Modest increase7

+

There has been a significant increase,
especially arrowweed and non-native
tamarisk in the riparian zone that was
formerly inundated frequently under the
“No-Action” hydropower operation

Emergent marsh plants

“Same as or less than no action”

−

Reduction in wet marsh species with an
increase in dry marsh species, likely owing
to the accumulation of fine sediment in
these areas through time, but without
periodic rejuvenation through floods.

WILDLIFE
Wintering waterfowl

Potential increase

+

(Need comment here)

212

Jeff Lovich and Theodore S. Melis
Table 3 (Continued)

Resource
Prediction
ENDANGERED AND OTHER SPECIAL STATUS SPECIES

Outcome

Comments

Native fish (humpback chub, razorback
sucker, flannelmouth sucker)

“Potential minor increase”

±

Decrease in adult humpback chub, native
suckers may be stable or slightly increasing

Bald eagle

“Potential increase”

?

Eagle numbers in Arizona have increased
overall

Peregrine falcon

No effect

+

Stable in Grand Canyon since 1988

Kanab ambersnail

“Some incidental take”

±

Snail habitat increased since 1998, but not
snail numbers which are relatively stable

Southwestern Willow Flycatcher

“Undetermined increase”

−

No increase. Uncommon in Grand Canyon.

CULTURAL RESOURCES
Archaeological sites affected

“Moderate (Less than 157)”

?

Subsequent analyses have not been
conducted to fully assess

Traditional cultural properties
affected

“Moderate”

?

Subsequent analyses have not been
conducted to fully assess

Traditional cultural resources
affected

“Increased protection”

?

Subsequent analyses have not been
conducted to fully assess

“Slight reduction”

?

Data unavailable

AIR QUALITY
Effect of emissions on regional air
quality
RECREATION
Angler safety

“Moderate improvement”

?

No long-term monitoring data

Day rafting (navigation past 3-mile
bar)

“Major improvement”

?

Pre-EIS study suggests net
willingness-to-pay values insensitive to
flows. More studies needed

White-water boating safety

“Minor improvement”

?

No long-term monitoring data

White-water boating camping
beaches (average area at normal peak
stage)

“Minor increase”

−

Camping areas have been diminished
owing to vegetation expansion and sand
bar erosion, despite the fact that the new
operating policy has limited daily peaking
release to 25,000 cfs

White-water boating wilderness
values

“Moderate to potential to become major
increase”

?

Potential decrease and decline in campable
area

Economic benefits (not related to
hydropower revenue)

Positive

+

Both local and regional impacts

POWER
Annual economic cost (forgone
hydropower revenue)

Acceptable costs relative to other alternatives

?

Subsequent studies not available to fully
assess

Wholesale rate of power

Acceptable costs relative to other alternatives

?

Subsequent studies not available to fully
assess

Retail rate of power (70% of end
users)

“No change to slight decrease”

?

Subsequent studies not available to fully
assess

Retail rate of power (23% of end
users)

“Slight decrease to moderate increase”

?

Subsequent studies not available to fully
assess

Retail rate of power (7% of end users)

Acceptable costs relative to other alternatives

?

Subsequent studies not available to fully
assess

NON-USE VALUE

“No data”

+

Substantial non-use value ($3-4 billion)
has been demonstrated

5

This resource was not included in Table II.7 of the EIS but predictions regarding the fate of this material were given on pages 104–105 of the document.
It is included in this tabulation for the sake of completeness but the outcome is not included in the percentages discussed in the text.
6
This would favor native fish and non-native warm water fish.
7
Exotic species included (tamarisk, camelthorn).

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

include both positive and negative ecological consequences. Possibilities include allowing for mainstem spawning of native fish,
allowing the invasion and dominance of warmwater fishes from
Lake Mead and the Little Colorado River, possible adverse alteration of the food base, and uncertain effects on the coldwater
fishery in the Lees Ferry reach. The “natural” warming of Glen
Canyon Dam releases owing to falling reservoir levels provides an
important opportunity to test system responses preceding installation of a SWS on the dam. In addition, lower reservoir storage
levels are contributing to increased salinity in discharges. Continued monitoring of both the reservoir and downstream quality
of water in the river ecosystem will be required to mitigate any
adverse impacts associated with dam operations under continued
drought conditions. Linking reservoir processes to downstream
quality of water parameters is a high priority for future research
and monitoring, especially as related to native and non-native
fish interactions, recruitment success in humpback chub and
dynamics associated with the food base of the river. An annual
to decadal-scale synthesis of the forty-year time series for Lake
Powell quality of water is a next essential step in the research
program.
Climate and drought – The preceding section on water quality monitoring in Lake Powell underscores the effect of putative
short-term climate trends on limnological parameters. Current
understanding of global climate drivers provides little ability to
predict the timing or extent of droughts over much of the Colorado
River Basin (Webb et al., 2005). Data from tree rings suggest
that multi-decadal droughts are not unknown in the basin and the
organisms and natural processes that developed in Grand Canyon
evolved in response to such natural variation. While the current
drought may not be a so-called “megadrought” because of the limited duration to date, only time will tell when it ends. Research is
needed to improve the ability to forecast climate in the Colorado
River Basin. Jain and others (2005), have recently reported a late
20th century trend toward increasing annual variation in streamflow within the large river basins of the western U.S., including
the Colorado River. They also report an increasing synchroneity
in the response of these western river basins over the same period
since the 1970’s when water use in the southwestern U.S. began to
accelerate. If such a trend persists, then the challenges associated
with simultaneously managing water supplies and maintaining
environmental resources of the arid southwest will likely only
increase through time under increasing depletions and forecast
uncertainties (Pulwarty and Melis, 2001).
Mitigating the influence of protracted drought was initially
the main objective for constructing Glen Canyon Dam. This was
achieved by locating the dam at a site that maximized the storage capacity of Lake Powell. The vast storage volume of the
reservoir (about 3.42 × 1010 m3 ), provides a time buffer of up
to several years against periods of below average runoff in the
upper basin. This storage capacity is of immense benefit to upper
Colorado River Basin states faced with the challenge of simultaneously pursuing development, while also needing to meet the
water supply needs of downstream users. The strategy of maximizing water storage above Glen Canyon Dam is also of interest
environmentally in that it is effective at mitigating the ecological

213

role that drought plays in the Colorado River ecosystem below
the dam. For instance, operation of the dam without a sediment
bypass system or SWS means that the river’s flow and sedimenttransport capacity are increased on average (Topping et al., 2003).
However, the thermal regime of the river now hovers somewhere
between the conditions that occurred in the pre-dam era between
winter and summer, while the currently diminished sediment
load is representative of a pre-dam mega-drought with respect
to tributary sand production. These limitations in the dam’s operation appear to greatly limit options for achieving environmental
objectives related to restoration and sustainable management of
unique aquatic and terrestrial parts of the Grand Canyon’s riverine
landscape.
Hydropower production – As the preceding two sections have
demonstrated, climatic variability and its affect on hydrologic
conditions can have large, if not overriding, effects on resources
of concern to the AMP. Hydropower is also a highly valued
resource of concern to society and environmental constraints
on dam operations under the MLFF policy have large annual
economic costs (see Harpman and Douglas, 2005). The acceptability of those costs has to be interpreted within the context
of societal values which abruptly shifted away from a focused
policy of resource development toward one where preservation
of natural resources is increasingly valued. This shift in values
occurred only after river regulation was underway. Loomis et al.
(2005) demonstrated that dam operations benefited both local
and regional economies through stabilization of flows relative
to pre-EIS and pre-dam conditions, which in turn facilitated the
development of a significant industry associated with white-water
rafting (see recreational use and non-use section below). Significant non-use values add to the economic value of the Grand
Canyon experience regardless of whether respondents visit the
canyon or not. A detailed ex-post facto cost-benefit analysis of
Glen Canyon Dam operations will ultimately be needed to fully
assess the MLFF operation versus documented environmental
benefits. Such assessments are also needed to fully evaluate any
consideration of additional experimental designs or to eventual
implementation of any alternative, longer-term changes in the
current operating strategy intended to achieve management goals.
Fine sediment – The much hoped for outcome of modest
improvement in sand bar resources, as originally proposed and
predicted in the EIS, has not been realized (Table 3). Detailed
synthesis studies of sand inputs and outputs to the system (sand
mass balance), intensive field monitoring, and remotely sensed
change detection analyses all point to a decrease in fine sediment resources in Glen, Marble and Grand Canyons in the
post-EIS era. These changes have resulted in smaller and coarsergrained sand bar deposits, particularly in the upper reaches of
the ecosystem where sand supply is most limited. Loss of sand
habitats in the ecosystem was documented under the No-Action
era (1964–1990), but has continued since dam operations have
been altered to mitigate sand bar erosion. Sand export appears to
be driven primarily by the annual strategy of patterning monthly
release volumes within MLFF to match seasonally varied regional
water supply in the west, variation in seasonal power demands
and benefits related to economics of daily “load-following.”

214

Jeff Lovich and Theodore S. Melis

Sand-transport data collected from 1999 through 2004, indicate
that whenever the monthly flow regime from the Glen Canyon
Dam forces daily peak discharges above 10,000 cfs for extended
periods, such operations tend to export newly input sand from
downstream tributaries and pre-existing sand storage in the main
channel relatively quickly. Sediment transport calculations during the EIS substantially under-predicted the capacity of such
operations to export new sand inputs and instead predicted that
average to below average MLFF operations would allow new sand
supplies from tributaries below the dam to accumulate through
time over multiple years. Such accumulated inputs were then
the basis for predicting that periodically released Beach-Habitat
Building Flows would result in sustainable sand bar restoration.
Unfortunately, even during years of minimal release hydrology
(8.23 million acre feet volumes released from 2000–2004), such
exporting flow peaks of 18,000 to 20,000 cfs are a frequently
occurring phenomenon in both winter and summer months that
annually limits accumulation of new sand supplies from tributaries (Rubin et al., 2002; Topping et al., 2000a, b; Topping
et al., 2006; Wright et al., 2005).
Although MLFF limitations on the daily allowable peak discharge were intended to reduce sand export and bar erosion, it
appears that the annual pattern of monthly volumes released from
the dam (with peak daily flows at their highest during the summer
sediment input months of July and August) is the greatest factor preventing accumulation of new sand inputs from tributaries
over multi-year time scales. Despite minimum allowable annual
volumes of water released from Glen Canyon Dam during 2000–
2004 (8.23 million acre feet), sand inputs from the Paria and Little
Colorado Rivers were quickly transported downstream during
months in which release schedules forced daily peaks to between
18,000–20,000 cfs. Owing to this, the greatest factor limiting
Beach-Habitat Building Flows in achieving habitat restoration is
tied to condition of sand supply limitation that persists from year
to year, even during continuous, multi-year drought conditions in
the Upper Colorado River Basin and below average MLFF operations. While the seasonally adjusted pattern of monthly releases
during a prolonged drought provides significant economic benefit to some stakeholders in terms of hydropower revenues, these
benefits apparently come at the expense of environmental goals
tied to downstream sand resources and related habitats.
Since closure of Glen Canyon Dam has eliminated about 84%
of the sand that historically entered Grand Canyon, managing the
remaining supply below the dam will apparently require careful
prescription of managed BHBF’s, strategically released immediately following natural sand inputs. It is not yet clear whether
even this strategy will succeed in sustainable restoration of sand
resources throughout Grand Canyon (Topping et al., 2006; Rubin
et al., 2002). In essence, efforts to restore sand bars without
sufficient sand supplies are analogous to attempting to restore
a bankrupt financial system with modest infusions of interest
rather than substantial investments of principal. Since physical
processes related to hydrology and sediment transport are relatively well studied, a logical question is why the writers of the EIS
predicted this outcome incorrectly? Again, the EIS writers used
the best science that was available, but three critical monitoring

programs for measuring suspended-sediment flux throughout the
Canyon were discontinued in the early 1970’s, forcing the EIS
team to evaluate a discontinuous set of post-dam transport data
and then work with flawed assumptions rather than continuous data records of flow and sediment concentration. Clearly,
long-term core monitoring efforts are essential to the success of
adaptive management and accurate predictions.
The EIS assumption that sand would accumulate on the bed
of river over multiple years is now known to be flawed. Our
research suggests that future management of sediment should
involve high flow releases immediately following inputs of sand
and finer sediment from tributaries below the dam. While such
releases are controversial in that they must bypass the hydropower
plant, recent studies also suggest that the duration of such flows
may need to be only a small fraction of what was originally suggested. Such fine tuning in the prescription of controlled floods
used for achieving habitat restoration (Stevens et al., 2001) will
likely reduce the financial impacts and controversy associated
with such management actions. One alternative test of the MLFF
concept for multi-year accumulation of sand supply might be to
equalize monthly volumes during droughts so as to further limit
daily peaks over such periods. If such a test failed to increase
sand supply through multi-year accumulation of tributary sand
inputs, then objectives for sand habitat restoration might have to
be reconsidered, or more proactive strategies, such as sediment
augmentation, might need to be implemented.
Coarse sediment – This resource was not included in Table
II-7 of the EIS but is included because of its importance to the
overall ecological condition of the river in Grand Canyon. In contrast to fine sediment, coarse sediment inputs have largely been
unaffected by the presence or operation of Glen Canyon Dam.
This owes to the presence of hundreds of unregulated tributary
canyons along the length of Grand Canyon that contribute coarse
sediment to the system. However, with the erosion of fine sediment from the system (see above) tied to dam operations under
Modified Low Fluctuating Flows, an overall sediment trend for
the Grand Canyon reach has been coarsening of the substrate in
the river channel, as fines are eroded, leaving gravel and larger
material. The impact of this “coarsening” of the river substrate
has two profound biological implications. First is the creation
of preferred habitat for benthic invertebrates, an important component of the food base. Second, is the creation of spawning
substrate for the non-native rainbow trout. Both of these changes
move the system farther from pre-dam conditions, potentially
benefiting non-native species at the expense of natives.
Inputs of coarse sediment are largely localized and episodic,
triggered by local storm events and rockslides (Webb et al., 2005).
When debris flows occur, they contribute to the existing system of
rapids for which Grand Canyon is famous by restricting channel
width in outwash areas. Under regulated flows associated with
Glen Canyon Dam operations, floods above 45,000 cfs are now
extremely rare in the Canyon. As a result the river no longer
has the power to rework coarse sediment deposits in rapids to
the extent possible before construction of the dam. It is possible
that some rapids will increase in size owing to regulated flows.
Research is needed to determine if high spike flows from the

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

dam would be effective at reworking coarse sediment deposits
periodically to mimic unregulated conditions.
Aquatic food base – Prior to the construction of the dam, the
sediment-laden waters of the Colorado River continually scoured
river bottom habitats and permitted little light penetration, both
of which strongly limited the growth of algae and other aquatic
plants. Most of the nutrients and energy that entered Grand
Canyon arrived as drift from sources outside the river. These
external inputs of energy arrived in the form of leaf litter, coarse
woody debris, and other materials carried down from terrestrial
sources or upstream tributaries. Following completion of Glen
Canyon Dam, water discharged downstream was cold and clear,
as described in the section on Quality of Water. This clear water
allowed for the development of a substantial food base dominated
primarily by algae and aquatic plants, thus changing the dominant energy inputs from external sources to instream production.
The plentiful algae in the post-dam Colorado River support a high
density of invertebrates (Kennedy and Gloss, 2005).
As one goes downstream from the Lees Ferry tailwaters,
tributary inputs of sediment limit light penetration, and hence
photosynthesis, causing abrupt decreases in algae standing mass
and a transition back towards the dominance of external energy
sources. Invertebrates communities also shift with this transition
in energy sources, and filter-feeding invertebrates that capture
fine particles of leaf litter and other detritus appear to dominate downstream reaches. However, the biomass of invertebrates
appears to also decrease with downstream distance, perhaps
because there is less energy available in the system due to the
reduction in algae standing mass. Considerable research is still
needed to better understand the effects that dam operations and
episodic inputs of sediment from tributaries have on the food base
across the entire Colorado River ecosystem.
Dam operations under MLFF probably led to increases in the
standing mass of food base resources (i.e., algae and invertebrates) due to steadier flows and greater minimum discharges of
clearer water (Table 3). However, declines in humpback chub
populations since implementation of MLFF suggest this change
in the aquatic food base may have not necessarily benefited
native fish that evolved in the turbid waters that characterized
the pre-dam environment.
A critical future research need is to develop a better understanding of the linkages between the food base and the actual
prey base of fish, both native and non-native. A large amount of
data have been collected on the food items consumed by nonnative rainbow and brown trout, but comparable data are not
available for the humpback chub due to the endangered status of
this fish. Use of stable isotope analysis will be critical to assessing
the energy sources and trophic pathways that are importance to
fish. Additionally, critical tests of the hypothesis that competition
between non-native trout and humpback chub is negatively affecting humpback chub populations are not available. Eating the same
food items is only the first criteria to establish that competition is
negatively impacting a particular species. For competition to have
a negative impact on a species, overlapping resources must also
be limited and one species must be more efficient at harvesting
the limited resources than another. In the food-rich environment

215

of the post-dam Colorado River, the second criteria may not be
a reasonable assumption. Further research will be required to
determine if this hypothesis is supported by data.
Fish – Eight species of native fish were found in the Grand
Canyon prior to the modern era. Most of these were endemic to
the Colorado River system making this one of the most unique fish
communities in the world (Mueller and Marsh, 2002). Of those,
only four remain including the humpback chub, the bluehead
sucker, the flannelmouth sucker and the speckled dace. Gone are
the razorback sucker, bonytail, roundtail chub and the Colorado
pikeminnow, all of which survive outside of Grand Canyon. The
humpback chub is endangered and its numbers have dropped
dramatically in the last decade. At the same time, non-native fish
have increased in both diversity and abundance.
The reasons for the decline of native fish are commonly cited
to include the construction and operation of numerous dams on
the Colorado River, the introduction of non-native predatory and
competitive fishes, and the introduction of diseases and parasites
(Mueller and Marsh, 2002). The actual mechanisms of decline
and extirpation are poorly known owing to a lack of early data on
population numbers and the fact that declines were well established prior to construction of major dams (but not prior to the
introduction of exotic species).
Our knowledge of the effect of dam operations on native fish is
incomplete, but there is only a limited body of evidence that Modified Low-Fluctuating Flows have been beneficial to the survival
and recruitment of humpback chub, to support the prediction of
the EIS (Table 3). However, we do not know if the overall adult
decline of chub, and the recent evidence of recruitment and flattening of the population trend during the decade of operations
under this regime is attributable to Modified Low-Fluctuating
Flows per se (Gloss and Coggins, 2005; Coggins et al., 2006a;
Melis and Coggins, in review). Mechanical removal of non-native
fish, especially rainbow trout is currently in the fourth year of
a four year implementation strategy to test the hypothesis that
reduction of predatory and competitive fish species will result
in an increase in survival and recruitment of humpback chub.
While the GCMRC has demonstrated that mechanical removal is
an effective way to significantly reduce the number of non-native
trout in the removal reach (Gloss and Coggins, 2005), fishery scientists are still collecting the data required (2003 and beyond) to
evaluate the hypothesized response in humpback chub. Given our
knowledge of the life history of the chub, a signal is not expected
for four or five years after the initiating of mechanical removal
efforts.
In contrast to recent decline in the abundance of adult chub,
populations of both bluehead and flannelmouth suckers appear
to have remained relatively stable under the current flow regime
of Modified Low Fluctuating Flows. The reasons for this persistence are currently unknown. On the other hand, the relatively
stable habitat conditions created under Modified Low Fluctuating Flows, coupled with a coarsening of substrate in the river
channel (see above section on fine sediment), appear to have
greatly favored rainbow trout, particularly in the Lees Ferry
reach, as reflected in their increasing numbers during the last
decade.

216

Jeff Lovich and Theodore S. Melis

In addition to establishing better linkages between our knowledge of the food base, dam operations, and fish populations,
we need to focus future research on the effects of warming
discharges of water from Glen Canyon Dam. An SWS is anticipated to be installed on the dam sometime later in this decade
and substantial questions remain unanswered on its effects.
Will the SWS warm water temperatures to a point where main
channel spawning of humpback chub is possible? Will young
native fish emerging from natal habitat in the Little Colorado
River be able to survive and grow in the warmer waters? Will
the tailwater trout fishery below the dam remain healthy and
viable? Will new diseases, parasites and invasive species become
established in warmer habitats? Will warmwater predatory fish
from downstream establish themselves as dominant species in
the Grand Canyon fish community? Will non-native trout in
the Little Colorado River/main channel confluence decrease
in abundance as a result of increasing temperatures? These
are all critical questions that must be answered in the near
future.
Terrestrial vegetation – Dam operations under Modified Low
Fluctuating Flows greatly stabilized antecedent and pre-EIS flow
conditions in the Grand Canyon with significant impacts on riparian vegetation (Ralston, 2005). The EIS predicted a modest
increase in woody vegetation (Table 3) and that prediction is
largely correct, if not understated. The EIS also predicted that
marsh communities would be the same as or less than expected
under the No Action alternative, but that prediction was largely
incorrect. Since implementation of Modified Low Fluctuating
Flows there has been a decrease in wet marsh vegetation and an
increase in dry marsh vegetation.
The pre-dam environment was subject to large variations in
flow that regularly scoured shoreline habitats of vegetation, resetting the system after major floods. Scouring flows are now rare
adding an element of stability to the system and ultimately affecting plant communities. For example, since daily peaking releases
under Modified Low Fluctuating Flows are limited to no more
than 25,000 cfs, the old high water vegetation zone is now
dependent entirely on incident precipitation for water. This has
resulted in some changes to species composition, particularly
those species that don’t have deep tap roots to reach the water
table associated with river flows. Removing access to surface
water from desert plant communities, forcing them to rely on incident precipitation only, has been documented to lead to decreases
in plant biomass over time (Schlesinger et al., 1989).
The stability of flows has also encouraged an increase in vegetation density in and near the wetted zone. While an increase
in vegetation may appear to be desirable, one of the impacts is a
decrease in available camping area for recreationists (Kaplinski
et al., 2005). Furthermore, the increase is largely attributable to
expansion of non-native saltcedar or tamarisk into the riparian
zone, often resulting in a monoculture that is not always beneficial to wildlife (Lovich and de Gouvenain, 1998). However,
recent research suggests that moving toward more natural river
flow regimes with the use of pulse floods allows native plants
species to become established despite the presence of tamarisk
(Glenn and Nagler, 2005).

Overall, interim (1991–1996) and Modified Low Fluctuating Flows (1996–2005) have caused a contraction in the extent
of the riparian plant community, comprised of more homogeneous (including non-native species) plant assemblages, with
more mature individuals. The export of fine sediments from
the system is further reducing available substrate for vegetation
and marsh vegetation. The effects of large-scale restructuring
of riparian plant communities in the Grand Canyon is expected
to have cascading impacts on co-adapted animal species, especially neotropical migrant birds. Future research should focus on
identifying the responses of animals to this fundamental change
in habitat structure, the relationship between riparian vegetation
and insects as related to the food web of the river, and examining
the affects of human-mediated removal of exotic vegetation vs.
natural disturbance.
Birds – Bird monitoring in the Grand Canyon related to the
Glen Canyon Dam Adaptive Management Program has been
driven by two factors: 1) concern regarding the possible effects
of dam operations on sensitive species (e.g., Bald Eagles –
Haliaeetus leucocephalus, Peregrine Falcons – Falco peregrinus anatum, and Southwestern Willow Flycatchers – Empidonax
traillii extimus), and 2) the perception that birds are good ecological indicators of short-term (post-dam) changes in the canyon.
Overall, it appears that birds are neither driving the ecological
system in the Grand Canyon, as they do in other ecosystems
(Croll et al., 2005), nor are their regional population dynamics strongly driven by factors within the canyon. Research has
demonstrated only a minimal direct effect of dam operations on
most birds. An exception to this generalization is shown for waterfowl that appear to have benefited from the stability of post-dam
flows (Table 3) (Holmes et al., 2005).
However, at the local level, changes in riparian vegetation,
mentioned above, have resulted in increased bird density and
diversity in the Grand Canyon. As such, post-dam flows have
had indirect effects on the bird community via their more direct
effects on plant community structure. Beyond that, it is difficult to
relate dam operations to bird dynamics, especially for neotropical
migrants. These hemispherical migrants are affected by factors
throughout both their breeding and wintering ranges. For example, a decline in Southwestern Willow Flycatcher numbers might
be caused by mortality on the wintering range in Central or South
American and have nothing to do with impacts in the Grand
Canyon, or elsewhere, in the breeding range.
Future bird research and monitoring in the Grand Canyon
requires answering some fundamental questions. First, should
birds continue to be monitored under the Glen Canyon Dam
Adaptive Management Program given the indirect relationship
between dam operations and most bird population dynamics?
Second, how exactly do dam operations affect plant community dynamics and ultimately birds? Finally, how do vegetation
characteristics affect individual bird species?
Special status species – Many of the research and monitoring
efforts in the Glen Canyon Dam Adaptive Management Program
are necessitated by compliance requirements from various laws,
especially the U.S. Endangered Species Act (Ralston, 2005). One
species, the Kanab ambersnail, is assumed to be found naturally

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

at only two locations, one in the Grand Canyon and another in
southern Utah. Genetic analyses are underway to assess the level
of differentiation between Grand Canyon populations and snail
populations with similar morphology north of the canyon.
The largest population at Vaseys Paradise is found in the varial
zone of dam operations and is thus susceptible to being washed
away during rapid changes in flows. In the past, significant efforts
have been undertaken to protect snail populations during experimental floods, including temporarily uprooting snail habitat (with
snails) immediately before a flood and then returning the habitat
at the conclusion of the experimental flow. Despite the vulnerability of such a small species, in such a spatially-limited distribution,
snail numbers appear to be relatively stable post-EIS.
With only two breeding pairs of Southwestern Willow Flycatchers present in the Grand Canyon in recent years, efforts to
monitor the effects of dam operations on this species are hampered by small sample sizes and issues identified in the section
on Birds above. Monitoring efforts should be continued for other
rare and special status species to prevent them from declining
further and being listed as threatened or endangered species.
Cultural resources – Mass balance studies examining inputs
and exports of sand and other fine sediments in the Colorado
River ecosystem demonstrate a net export of this resource as
a result of dam operations. The sand that is being exported is
coming not only from new tributary inputs, but also existing
beaches and river terraces that contain archaeological sites. Many
archaeological sites in Grand Canyon have been covered with
windborne (aeolian) sand for centuries. This sand was transported
from lower-elevation beaches that were frequently re-supplied
with new sand sources derived from annual floods during the
pre-dam era. Following dam closure in 1963, operations were
optimized for maximum water storage and power revenue (within
the constraints of existing law and policies), rather than for strategically conserving limited downstream remaining sand supplies
for restoration of sand bars (and presumably, long-term preservation of cultural sites). With more sand continually leaving the
ecosystem than being supplied, more and more of these preservation sites are being exposed to the ravages of erosion. As sites
are eroded, artifacts and structures are increasingly exposed,
where they become susceptible to visitor impacts and destabilization due to loss of the surrounding sedimentary matrix. Such
changes make it difficult, if not impossible for archaeologists
to reconstruct and interpret the historical and cultural information contained within these important settings in Grand Canyon
National Park (Fairley, 2005).
Future research and monitoring of cultural resources needs
to focus on the effects of all components of the MLFF dam
operations, most specifically, how the monthly pattern of annual
releases might be revised to more effectively influence conservation of new sand inputs from tributaries so as to mitigate the
current, continuous fine-sediment deficit.
Recreational use and non-use values – One of the resources
that appear to have benefited most from the stabilizing influence
of Modified Low Fluctuating Flows, relative to more variable dam
operations, is recreational use of the river. By eliminating very
high and very low discharges conditions, flows that favor year

217

round recreational boating and fishing are fostered. Although the
most comprehensive regional economic study of recreation is
now 10 years old (Douglas and Harpman, 1995), the figures are
impressive: over $46 million in non-resident total expenditures
and maintenance of 586 jobs, with 438 jobs in commercial rafting
alone. Presumably those figures are even higher now (Loomis
et al., 2005).
Based on data collected over 10 years ago, non-use values,
or estimates of the “existence” value of Grand Canyon National
Park, Glen Canyon National Recreation Area, tribal lands, and
the resources therein, to citizens who may never actually engage
in recreational activities in the area, are estimated at $3–4 billion
(Loomis et al., 2005). The public at large is willing to pay to
have flows and other management actions that benefit the canyon
and its resources. While some may question the utility of nonuse valuation of Grand Canyon resources, the estimates reflect
the iconic values that make Grand Canyon National Park famous
throughout the world.
Future research should focus on detailed analyses of how Glen
Canyon National Recreation Area fishing use, catch rates and
fish condition are related to flows. Economic data on fishing and
recreational rafting also need to be updated to establish current
baseline data before new flow regimes are initiated. Finally, studies to quantify the wilderness experience of recreationists need to
be initiated so that the benefit of eventually achieving ecosystem
restoration can be fully evaluated.
Campsites – Since the implementation of Modified LowFluctuating Flows, there has been about a 15% per year reduction
in campable area in the Grand Canyon (Kaplinski et al., 2005).
Losses occurred in both critical (campsite-limited) and noncritical reaches, but the largest losses have occurred in critical
areas. Losses are thought to be attributable to both net sediment
exports under current dam operations (see section above on fine
sediment), as well as encroachment of woody vegetation (see
section above on terrestrial vegetation). However, while daily
river stage restrictions under the MLFF provide some benefits to
campsite area by limiting the varial zone to below 25,000 cfs,
camping areas between this upper stage and the former daily
limit under the No Action policy (33,000 cfs) have become less
accessible owing to vegetation expansion. The exact relationship
and interaction between these two factors and dam operations
is unknown and provides a challenge for future researchers. For
example, increased vegetation in sand bar areas may also provide greater substrate stability and shade, both limiting campsite
erosion rates and enhancing camping areas from an aesthetic
perspective. Tradeoffs between vegetation expansion and sand
bar stability must also be considered from the perspective that
increased vegetation might also limit the potential for wind processes to blow sand deposits upslope into cultural preservation
sites subject to rainfall and runoff erosion?

How accurate were the 1995 EIS predictions?
The outcomes summarized in Table 3, and presented in the discussions above, present a kind of “report card” to assess the accuracy

218

Jeff Lovich and Theodore S. Melis

of predictions that formed the foundation of the Glen Canyon
Dam Adaptive Management program, implemented under the
EIS with the preferred alternative of Modified Low-Fluctuating
Flows. It is important to emphasize that the outcomes listed are
based on the best available scientific information and judgment,
and are subject to revision as additional information becomes
accessible. We attempted to generalize our evaluations in a conservative fashion; details are contained in the original works cited
in this paper. With these caveats, we offer the following observations on the ability of current technologies to predict ecosystem,
societal and economic responses in Grand Canyon National Park
and Glen Canyon National Recreation Area and associated tribal
land areas of the ecosystem beyond those boundaries.
Because of a lack of data or subsequent analyses to confirm
whether the prediction stated in the EIS was correct, or not, 14
of 30 (excluding coarse sediment: see above), or 46 percent of
the outcomes, are essentially unknown at the time of this writing,
despite ten years since completion of the EIS. These unresolved
outcomes constitute an important list for further study in the program. Correct predications accounted for only 5 outcomes, or
17 percent of the categories listed. Mixed outcomes occur in 6
or 20 percent of the categories, and failed predictions, scored as
negative outcomes, occur in 5 or 17 percent of the categories.
Excluding predictions where the outcome is still unknown
(n = 14), the percentage of outcomes that were correctly predicted (5 out of 16 remaining) is only 31 percent, underscoring
the uncertainties associated with working in a large complex system with few to no data sets. Assuming only three categories
of outcomes (+, −, or +/−), and that the outcomes are correctly assigned, then the proportion of each outcome mentioned
above is not statistically different from an equal probability of
any outcome (χ2 = 0.125, P = 0.94). In other words, the
ability of the EIS writers to predict the outcome of Modified
Low-Fluctuating Flows 10 years latter was, it appears, no better
than random chance. A more optimistic approach would be to
assign at least a partially successful rating to mixed outcomes
(+/−). Under that scenario, 11 out of 16, or 69 percent of the
outcomes might be deemed correctly predicted, at least in part.
However, this proportion too is not statistically different from
random chance (χ2 = 2.25, P = 0.13).
Does this mean that the program is a failure? Not necessarily.
The acceptability of this kind of uncertainty is influenced by interpretation, societal values, agency missions and mandates, and
other factors (Clark, 2002). However, failure to correctly predict
the future, in and of itself, is not deleterious under the paradigm of
adaptive management where large uncertainties provide opportunities for learning and adjustment through an iterative process
of “learning-by-doing” (Figure 2).

Future challenges
Sustainable river management involves interrelationships among
assessments of ecological risk (Nilsson et al., 2005), sustainable
management, dealing with uncertainty, adaptive management
and decision support (Clark, 2002). While adaptive management

Figure 2 Conceptual illustration of the Glen Canyon Dam Adaptive
Management Program process. GCMRC refers to the Grand Canyon
Monitoring and Research Center of the U.S. Geological Survey, established as the science provider for the Program. The SCORE report refers
to the State of the Colorado River Ecosystem in Grand Canyon publication synthesizing the state of scientific knowledge for the Program
(Gloss et al., 2005). The next step in the process is knowledge assessment and policy deliberations after which the cycle will continue. Refer
to text for additional details.

with stakeholder involvement is the stated framework for the
Final Environmental Impact Statement (U.S. Department of the
Interior, 1995) for the Operation of Glen Canyon Dam, it is not
without challenges. One of the constraints of the Glen Canyon
Dam Adaptive Management Program (AMP) is the artificial prescription of boundaries that extend from the forebay of Glen
Canyon Dam to the backbay of Lake Mead, and from the old high
water mark on river left, to its counterpart on river right. The definition of the scope of the program is thus largely derived from
a human or political perspective. This constraint imposes two
important challenges to scientists. First, is the fact that this definition of a boundary does not fully circumscribe natural patterns
or processes that affect the Colorado River ecosystem. Others
have attempted to define ecosystems on the basis of hydrographic,
physiographic, floristic and ethnographic boundaries (Grayson,
1993), all of which present problems in the case of the Grand
Canyon. For example, there is no clear floristic definition of the
Colorado River ecosystem owing to extremes in elevation and
precipitation. By imposing artificial limits on the boundary of
the ecosystem, we impose limits on our ability to understand the
larger landscape factors that drive physical and thus ecological
processes in Grand Canyon.
The second challenge that an artificial boundary presents to
scientists is that of restricting research and monitoring activities
to the area circumscribed in the definition above, to the exclusion
of comparative sites. In the past, researchers in the AMP have
been unable to conduct comparative research in river systems
less-influenced by high dam operations. For example, Cataract
Canyon, Desolation-Gray Canyons, and Westwater Canyon all
provide examples in the Upper Colorado River of systems that
are more like the pre-dam Grand Canyon than any segment of the
river included in the AMP boundary. While there are no rivers
or river segments that perfectly match the characteristics of the

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

pre-dam Grand Canyon, important analytical opportunities are
lost by not expanding the study to incorporate sites more like a
control.
Spatial scope is not the only constraint that presents a challenge to understanding the form and function of the Colorado
River ecosystem. Temporal issues operate on scales of over 10
orders of magnitude, ranging from billions of years to minutes,
depending on whether one considers the geomorphic framework
of the canyon or the hour-to-hour effects of dam operations on
that same framework. This extreme variability happens simultaneously, confounding efforts to determine cause and effect. An
example is provided by studies directed at understanding the population ecology of the humpback chub. These fish evolved in a
canyon that reflects billions of years of geological history, over
millions of years of natural selection, with a life span of decades, a
time to maturity of 3–5 years, with wide annual pre-dam variation
in flows and temperatures, now complicated with daily and hourly
variations in flow (but not temperature) post-EIS. The complexity of factors and temporal scales present enormous challenges
to effectively determining how population numbers vary in time
and space and underscores the importance of long-term studies
to describe patterns and processes. Long-lived organisms like the
humpback chub require decades to capture population variability
related to their life span (Congdon et al., 1994).
Another problem, identified in the past, has been a need for
greater integration of studies. As the complexity of issues in the
Glen Canyon Dam Adaptive Management Program become more
obvious, so does the need for interdisciplinary, not just multidisciplinary science. Great progress has been achieved since this
issue was highlighted by Meretsky and Melis (1997). Continued
efforts will be required to integrate knowledge across disciplines
and scales, and develop a more robust conceptual model for the
Colorado River ecosystem (Walters et al., 2000; Walters and
Korman, 1999).

Conclusion
The outcomes of many of the predictions contained in the 1995
EIS are unknown or unsatisfactory. The proportions of outcomes
that are correct vs. incorrect are not significantly different from
what would be expected by random chance (Table 3). Like
predicting the weather, there is a substantial amount of uncertainty, even more, in predicting the ecophysical impacts of dam
operations on a large river.
Considerable knowledge has been gained from focused, systematic studies of the Colorado River ecosystem in Grand
Canyon. Research, including conceptual and other models
(Walters et al., 2000; Coggins et al., 2006b) and monitoring conducted by U.S. Geological Survey scientists and their cooperators
have conclusively demonstrated a net loss of sediment from the
system and documented the decline of the federally endangered
humpback chub during the last decade. Both findings are critical
pieces of information to assess conditions and adjust management
actions in the spirit of adaptive management. As in all research,
more questions are generated than answers, but those questions

219

form the basis for future studies, another palpable benefit of the
program to date. It is also important to note that water delivery
requirements continued to be met throughout the decade postEIS, despite increased costs associated with environmental and
experimental regulation of flows.
In an evaluation of a similar Colorado River conservation program, Brower et al. (2001) identified several solutions to the
problems of consensus-based management (it is important to
note that the AMP is not consensus-based): (1) program success
should be judged by species recovery, (2) the Federal government
should retain the ability to impose regulatory sanctions in the
event of continued population declines (see also Meretsky et al.,
2000), and 3) funding should be provided by an agency with clear
species-protection responsibilities to reduce the disproportionate
influence of utilitarian interest groups. In large measure, these
approaches are implemented under the AMP. Just as important,
the success of the approach taken to achieve the desired outcomes
of the management program should be evaluated against whether
or not the actions taken result in ecosystem responses that move
in at least the general direction of those stated goals and objectives (Palmer et al., 2005). A substantial body of science now
exists for the Colorado River ecosystem. The overarching question is, what will society do with the knowledge now available
to move into the next phase of the Glen Canyon Dam Adaptive
Management Program?

References
1. Blinn, D.W. and Poff, N.L. (2005). Colorado River Basin.
in A.C. Burke and Cushing, C.E. (eds.), Rivers of North
America. Academic Press, San Diego, pp. 483–538.
2. Brower, A., Reedy, C. and Yelin-Kefer, J. (2001). “Consensus versus Conservation in the Upper Colorado River
Basin Recovery Implementation Program,” Conservation
Biology, 15, 1001–1007.
3. Bureau of Reclamation. (2006). Glen Canyon Dam Facts
(http://www.usbr.gov/uc/news/gcdfacts.html).
4. Clark, M.J. (2002). “Dealing with Uncertainty: Adaptive
Approaches to Sustainable River Management,” Aquatic
Conservation: Marine and Freshwater Ecosystems, 12,
347–363.
5. Coggins, L.G, Pine, W.E. III, Walters, C.J., Van
Haverbeke, D.R., Ward, D. and Johnstone, H.C.
(2006a). “Abundance Trends and Status of the Little Colorado River Population of Humpback Chub,” N. Amer. J.
Fish. Mngt, 26, 233–245.
6. Coggins, L.G, Pine, W.E. III, Walters, C.J. and Martell,
S.J.D., (2006b). “Age-Structured Mark-recapture Analysis:
A Virtual-Population-Analysis-Based Model for Analyzing
Age-Structured Capture-Recapture Data,” N. Amer. J. Fish.
Mngt, 26, 201–205.
7. Congdon, J.D., Dunham, A.E. and van Loben Sels,
R.C. (1994). “Demographics of Common Snapping Turtles
(Chelydra serpentina): Implications for Conservation and

220

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Jeff Lovich and Theodore S. Melis

Management of Long-Lived Organisms,” Amer. Zool, 34:
397–408.
Croll, D. A., Maron, J.L., Estes, J.A., Danner, E.M.
and Byrd, G.V. (2005). “Introduced Predators Transform
Subarctic Islands from Grassland to Tundra,” Science, 307,
1959–1961.
Douglas, A.J. and Harpman, D.A. (1995). “Estimating Recreation Employment Effects with IMPLAN for
the Glen Canyon Dam Region,” Journal of Environmental
Management, 44, 233–247.
Fairley, H.C. (2005). Cultural resources in the Colorado
River Corridor, in S.P. Gloss, Lovich, J.E. and Melis, T.S.
(eds.), The state of the Colorado River ecosystem in the
Grand Canyon: U.S. Geological Survey Circular 1282. 220
pages, pp. 177–192.
Glenn, E.P. and Nagler, P.L. (2005). “Comparative Ecophysiology of Tamarix Ramosissima and Native Trees in
Western U.S. Riparian Zone,” Journal of Arid Environments,
61, 419–446.
Gloss, S.P. and Coggins, L.G. (2005). Fishes of Grand
Canyon, in S.P. Gloss, Lovich, J.E.: and Melis, T.S. (eds.).
The state of the Colorado River Ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 33–56.
Gloss, S.P., Lovich, J.E. and Melis, T.S. (eds.). (2005).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages.
Grayson, D.K. (1993). The desert’s past: A natural prehistory of the Great Basin, Washington, D.C. Smithsonian
Institution Press. 356 pp.
Harpman, D.A. and Douglas, A.J. (2005). Status and
trends of hydropower production at Glen Canyon Dam,
in S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 165–176.
Holmes, J.A., Spence, J.R. and Sogge, M.K. (2005). Birds
of the Colorado River in Grand Canyon: a synthesis of status,
trends, and dam operation effects, in S.P. Gloss, Lovich,
J.E. and Melis, T.S. (eds.), The state of the Colorado River
Ecosystem in the Grand Canyon: U.S. Geological Survey
Circular 1282. 220 pages, pp. 123–138.
Jain, S., Hoerling M. and Eischeid, J. (2005), “Decreasing
Reliability and Increasing Synchroneity of Western North
American Streamflow,” Journal of Climate, 18, 613–618.
Kaplinski, M., Behan, J., Hazel, J.E., Parnell, R.A. and
Fairley, H.C. (2005). Recreational values and campsites
in the Colorado River ecosystem, in S.P. Gloss, Lovich,
J.E. and Melis, T.S. (eds.), The state of the Colorado
River ecosystem in Grand Canyon: U.S. Geological Survey
Circular 1282. 220 pages, pp. 191–205.
Kennedy, T.A. and Gloss, S.A. (2005). Aquatic ecology: the role of organic matter and invertebrates, in S.P.
Gloss, Lovich, J.E. and Melis, T.S. (eds.), The state of
the Colorado River ecosystem in the Grand Canyon: U.S.
Geological Survey Circular 1282. 220 pages, pp. 87–101.

20. Loomis, J., Douglas, A.J. and Harpman, D.A. (2005).
Recreation use values and nonuse values of Glen and Grand
Canyons, in S.P. Gloss, Lovich, J.E. and Melis, T.S (eds.)
The state of the Colorado River Ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp.153–164.
21. Lovich, J.E. and de Gouvenain, R.G. (1998). Saltcedar
invasion in desert wetlands of the southwestern United
States: ecological and political implications, in S.K.
Majumdar, Miller, E.W. and Brenner, F.J. (eds.). Ecology of Wetlands and Associated Systems. Pennsylvania
Academy of Science, pp. 447–467.
22. Meretsky, V.J., Wegner, D.L. and Stevens, L.E. (2000).
“Balancing Endangered Species and Ecosystems: A Case
Study of Adaptive Management in Grand Canyon,” Environmental Management, 25, 579–586.
23. Meretsky, V.J. and Melis, T.S. (1997). Integration of
Grand Canyon physical and biological information: A
progress report: Proceedings of the Third Biennial Conference of Research on the Colorado Plateau, C. van Riper III
and Deshler, E.T. , (eds.). U.S. Department of the Interior,
National Park Service, p. 193–213.
24. Mueller, G.A. and Marsh, P.C. (2002). Lost, a desert
river and its native fishes: A historical perspective of the
lower Colorado River. Information and Technology Report
USGS/BRD/ITR-2002-0010. U.S. Government Printing
Office, Denver, Colorado. 69 pp.
25. Nilsson, C., Reidy, C.A., Dynesius, M. and Revenga, C.
(2005). “Fragmentation and Flow Regulation of the World’s
Large River Systems,” Science, 308, 405–408.
26. Palmer, M.A., Bernhardt, E.S. , Allan, J.D. , Lake,
P.S., Alexander, G., Brooks, S., Carr, J., Clayton, S.,
Dahm, C.N., Follstad Shah, J., Galat, D.L., Gloss,
S., Goodwin, P., Hart, D.D., Hassett, B., Jenkinson, R.,
Kondolf, G.M., Lave, R., Meyer, J.L. , O’Donnell, T.K.,
Pagano, L. and Sudduth, E. (2005). “Standards for Ecologically Successful River Restoration,” Journal of Applied
Ecology, 42, 208–217.
27. Pulwarty, R.S. and Melis, T.S. (2001). Climate Extremes
and Adaptive Management on the Colorado River: Lessons
from the 1997–1998 ENSO Event,” Journal of Environmental Management, 63, 307–324.
28. Ralston. (2005). Riparian vegetation and associated
wildlife, in S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 103–122.
29. Rubin, D.M., Topping, D.J., Schmidt, J.C., Hazel, J.,
Kaplinski, M. and Melis, T.S. (2002). “Recent Sediment
Studies Refute Glen Canyon Dam Hypothesis,” Eos, Amer.
Geophys. Union, 83(25), 273, 277–278.
30. Schlesinger, W.H., Fonteyn, P.J. and Reiner, W.A.
(1989). “Effects of Overland Flow on Plant Water Relations,
Erosion, and Soil Water Percolation on a Mojave Desert
Landscape,” Soil Science Society of America Journal, 53,
1567–1572.

The state of the Colorado River ecosystem in Grand Canyon: Lessons from 10 years of adaptive ecosystem management

31. Schmidt, J.C., Webb, R.H., Valdez, R.A., Marzolf, R.
and Stevens, L.E. (1998). “Science and Values in River
Restoration in the Grand Canyon,” Bioscience, 48, 735–747.
32. Schmit, L.M., Gloss, S.P. and Updike, C.N. (2005). Introduction, in S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 1–16, .
33. Stevens, L.E. and Gold, B.D. (2003). Monitoring
for adaptive management of the Colorado River ecosystem in Glen and Grand Canyons, in D.E. Busch, and
Trexler, J.C. (eds.). Monitoring ecosystems: interdisciplinary approaches for evaluating ecoregional initiatives.
Island Press, Washington, D.C. 447 pages, pp. 101–134.
34. Stevens, L.E., Ayers, T.J., Bennett, J.B., Christiansen,
K., Kearsley, M.J.C., Meretsky, V.J., A.M. Phillips III,
Parnell, R.A., Spence, J., Sogge, M.K., Springer, A.E.
and Wegner, D.L. (2001). “Planned Flooding and Colorado
River Riparian Trade-Offs Downstream from Glen Canyon
Dam,” Arizona. Ecological Applications, 11, 701–710.
35. Topping, D.J., Rubin, D., Schmidt, J., Hazel, J., Melis,
T., Wright, S., Kaplinski, M., Draut, A., Breedlove, M.,
2006, Comparison of sediment-transport and bar response
results from the 1996 and 2004 controlled-flood experiments
on the Colorado River in Grand Canyon, in Proceedings
of the 8th Federal Interagency Sedimentation Conference,
April 2–6, (2006) Reno, NV, pp. 1–8.
36. Topping, D.J., Schmidt, J.C. and Vierra, L.E., Jr. (2003).
Computation and Analysis of the Instantaneous-Discharge
Record for the Colorado River at Lees Ferry, Arizona –
May 8, 1921, through September 30, 2000. U.S. Geological
Survey Professional Paper 1677. 118, pp.
37. Topping, D.J., Rubin, D.M., Nelson, J.M., Kinzel, P.J. and
Corson, I.C. (2000a). “Colorado River Sediment Transport
1. Natural Sediment Supply Limitation and the Influence
of Glen Canyon Dam,” Water Resources Research, 36(2),
515–542.
38. Topping, D.J., Rubin, D.M. , Nelson, J.M., Kinzel,
P.J. and Corson, I.C. (2000b). “Colorado River Sediment Transport 2. Systematic Bed-Elevation and GrainSize Effects of Sand Supply Limitation,” Water Resources
Research, 36(2), 543–570.
39. U.S. Department of the Interior. (1995). Operation of Glen
Canyon Dam Final Environmental Impact Statement: Salt

40.

41.

42.

43.

44.

45.

46.

47.

48.

221

Lake City, Utah, Bureau of Reclamation, Upper Colorado
Region, 337 p., appendices.
U.S. Department of the Interior. (1996). Record of Decision,
Operation of Glen Canyon Dam: Washington, D.C., Office
of the Secretary of Interior, 13 p.
Vernieu, W.S., Hueftle, S.J. and Gloss, S.P. (2005). Water
quality in Lake Powell and the Colorado River, in S. P.
Gloss, Lovich, J.E. and Melis, T.S. (eds.). The state of
the Colorado River ecosystem in the Grand Canyon: U.S.
Geological Survey Circular 1282. 220 pages, pp. 69–85.
Vernieu, W.S., Hueftle, S.J. and Gloss, S.P. (2005).
Water quality in Lake Powell and the Colorado River, in
S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.). The state
of the Colorado River ecosystem in the Grand Canyon: U.S.
Geological Survey Circular 1282. 220 pages, pp. 69–85.
Walters, C.J., Korman, J., Stevens, L.E. and Gold, B.D.
(2000). “Ecosystem Modeling for Evaluation of Adaptive
Management Policies in the Grand Canyon,” Conserv. Ecol.,
4, 1–65.
Walters, C. and Korman, J. (1999). “Cross-Scale Modeling of Riparian Ecosystem Responses to Hydrologic
Management,” Ecosystems, 2, 411–421.
Walters, C.J. and Holling, C.S. (1990). Large-Scale
Management Experiments and Learning by Doing,” Ecology, 71(6), 2060–2068.
Webb, R.H., Hereford, R. and McCabe, G.J. (2005).
Climatic fluctuations, drought and flow in the Colorado
River, in S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 57–68.
Webb, R.H., Griffiths, P.G., Magirl, C.S. and Hanks,
T.C. (2005). Debris flows in Grand Canyon and the rapids of
the Colorado River, in S.P. Gloss, Lovich, J.E. and Melis,
T.S. (eds.). The state of the Colorado River ecosystem in the
Grand Canyon: U.S. Geological Survey Circular 1282. 220
pages, pp. 139-152.
Wright, S.A, Melis, T.S., Topping, D.J. and Rubin, D.M.
(2005). Influence of Glen Canyon Dam operations on downstream sand resources of the Colorado River in Grand
Canyon, in S.P. Gloss, Lovich, J.E. and Melis, T.S. (eds.).
The state of the Colorado River ecosystem in the Grand
Canyon: U.S. Geological Survey Circular 1282. 220 pages,
pp. 17–31.