-
extracted text
-
FREEING THE DESCHUTES: ASSESSING THE IMPLICATIONS
OF SEDIMENT TRANSPORT IN DAM REMOVAL:
A CASE STUDY OF THE 5TH AVENUE DAM,
OLYMPIA, WASHINGTON
by
Jennifer Garlesky
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2015
©2015 by Jennifer Garlesky. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Jennifer Garlesky
has been approved for
The Evergreen State College
by
________________________
Kathleen Saul-B.A., B.S., M.A., M.E.S.
Member of the Faculty
________________________
Date
ABSTRACT
Freeing the Deschutes. Assessing the Implications
of Sediment Transport in Dam Removal Projects:
A Case Study of the 5th Avenue Dam,
Olympia, Washington
Jennifer Garlesky
Dam removal is an emerging restoration measure to return river systems back to their
natural states. Managing the sediment that is released when a dam is removed requires
extensive planning to determine the volume and magnitude of sediment that will be
released once the structure is removed. In the case of the 5th Avenue Dam in Olympia,
Washington, the removal of this concrete earthen dam will restore 260 acres of urban
estuary habitat. In 2006, the Deschutes Estuary Hydrodynamic and Sedimentation Report
used the Delft 3D model to predict where sediment erosion and deposition will occur.
Since the completion of this study, the number of dam removal has increased, and various
models have been developed to assess and predict how to manage the release of
sediment. Results from the research completed highlight that estuary restoration is
feasible based upon predictions determined in the 2009 Deschutes Estuary Feasibility
Study. The methods used to manage the sediment release are featured in this thesis, based
on three different management methods.
Table of Contents
Chapter 1 Introduction………………………………………………………………...…..1
Chapter 2 Methodology……………………………………………………………….....15
Chapter 3 Dam Removal……...…………………………………………………………17
3.1 Dam Removal………………………………………………………………..17
3.2 Return of the River…………………………………………………………...33
3.3 Influences on River Restoration……………………………………………...35
3.4 Regulations and Management………………………………………………..37
3.5 Inventory of Dams…………………………………………………………...38
3.6 Size of Dams…………………………………………………………………39
Chapter 4 The 5th Avenue Dam.…………………………………………………………43
4.1 Deschutes River and Estuary………………………………………………...43
4.2 Capitol Lake………………………………………………………………….43
4.3 Sediment……………………………………………………………………..45
4.4 Monitoring…………………………………………………………………...48
4.5 Sediment Composition……………………………………………………….49
4.6 Modeling……………………………………………………………………..52
Chapter 5 Research Analysis………………………………………………………….…54
Chapter 6 Discussion ……………………………………………………………………62
Chapter 7 Conclusion…………………………………………………………………....72
Bibliography…….……………………………………………………………………….74
v
List of Figures
Figure 1.1 Map of all dam removal projects in the United States………………………...4
Figure 1.2 Map of the 5th Avenue Dam and the Deschutes Estuary………………………6
Figure 1.3 Numerical Model Simulation of predicted sediment transport and deposition
in the removal of the 5th Avenue Dam…………………………………………...10
Figure 3.1 Detailed design of the 5th Avenue Dam………………………………………21
Figure 5.1 Map of the Elwha River Near Shore Zone…………………………………...55
Figure 5.2 Map of pre-dam removal morphology and elevation changes on the
Elwha River……………………………………………………………………...57
Figure 5.3 Map of predicted areas of erosion and deposition in the
Deschutes Estuary………………………………………………………………..59
Figure 5.4 Map of potential sediment transport scenario post removal of the
5th Avenue Dam………………………………………………………………….61
Figure 6.1Illustration of sill basins………………………………………………………63
Figure 6.2 Illustration of tidal marshes…………………………………………………..64
Figure 6.3 Illustration of matting…………………………………………………….......65
Figure 6.4 Illustration of pile dikes………………………………………………………66
Figure 6.5 Illustration of the 5th Avenue Dam Ogee Weir……………………………….67
Figure 6.6 Illustration of culvert placement for sediment deflection…………………….68
vi
Acknowledgements
I would first like to thank my thesis reader Kathleen Saul, who has been a wonderful
mentor to me during my final year of graduate school. I would also like to thank Michael
Radelich with the Evergreen State College Writing Center for his support and editing
guidance. I would like to thank all of the organizations I worked with throughout my
thesis process. I would also like to thank Julian Close, Pam Abreu and Pah-tu Pitt for
their support over the past two years. Finally, I would like to thank my friends and family
who supported my decision to quit my job and move across the country to pursue my
Master’s Degree in Environmental Studies.
vii
CHAPTER 1
INTRODUCTION
On September 17, 2011, a barge carrying earth-moving equipment started the
four-year process of the removal of the Glines Canyon Dam on the Elwha River in
Washington State. Citizens, environmental organizations, tribal members, and state and
federal agency officials watched a live broadcast of the demolition of this structure
(Olympia National Park Service, 2011). That day is considered groundbreaking in river
restoration management: The removal of the Glines Canyon and the Elwha Dams, two
hydroelectric facilities located in Olympic National Park, set the groundwork for future
dam removal projects. People involved with the removal of the Elwha River Dams
incorporated numerous management strategies in deciding on how these two structures
(and others like them) should be removed (Gowan, Stephenson, Shabman, 2006). Those
choices rest on years of data collection to establish a baseline of the river’s current status
as well as projections about what the river would look like once free of its impediments.
Field crews collected an array of pre-removal data to compare to after the dams were
removed. Information gathered included fish and wildlife populations, vegetation patterns
and dynamics, sediment transport and storage in reservoirs, river channel and coastal
evolution downstream of the dam site, the hydrological processes, near shore bathymetry,
coastal habitats, and beach erosion (Duda, Warrick, Magirl, 2011).
The removal of a dam from any river system can create a number of positive and
negative feedback loops that scientists must factor in when developing their restoration
strategy. Dam removal changes the river’s physical, chemical, and biological processes.
Parameters such as flow, discharge, sediment grain size, sediment load, level of cohesion,
1
deposition, channel morphology, and erosion are a few of the physical properties that
researchers record in order to determine how the river might respond to removal. In the
case of the Elwha River dams, natural resource managers determined that pre-and postmonitoring measurements were needed in order to assess the river’s response to the
removal. Since the fall of 2011, scientists have been recording the restoration progress of
the Elwha River. Field technicians continue to monitor the effects of the dams’ removal.
The data collected illustrates how the ecosystem is restoring itself (Gowan, Stephenson,
Shabman, 2006). To date, the ecological response to the dam removal has been highly
significant in terms of the return of salmon runs and restoration of the river’s hydrologic
flow regime (Gowan, Stephenson, Shabman, 2006). In a one-day field survey in the fall
of 2013, biologists surveyed 1,741 adult Chinook salmon (Oncorhynchus tshawytscha)
and mapped 763 reds (salmon eggs) in the newly created river habitat (The Seattle Times,
2013). The biologists emphasized that 75 percent of the salmon were spotted upstream
from the dam’s site (The Seattle Time, 2013). The increase in salmon habitat is just one
of the many positive ecological feedback loops that occur when a dam is removed.
Even though National Park Service, and The United State Geological Survey
consider the Elwha River restoration project a groundbreaking case in river restoration
history, the idea of dams being removed from our river systems is not a new concept. The
removal of dams from the United States river systems has been occurring since the early
1900s: One of the first documented dams to be removed was the Russell Dam in
California back in 1922 (Pohl, 2002). The number of dams removed has increased since
the early 1920s and has been one of the top river restoration strategies implemented.
Historical data shows that during the 1960s and 1970s, fewer than 20 dams were removed
2
(Stanley & Doyle, 2003). This number increased during the 1980s to the
decommissioning of approximately 100 dams, and continued to rise in the 1990s to 160
dams removed (Stanley & Doyle, 2003). In 2002 alone, reports show that 63 dams were
removed from river systems in the United States (Stanley & Doyle, 2003). More recently
in 2014, 72 dams were removed, which restored approximately 730 miles of river habitat,
according to a report released by American Rivers, a national nonprofit based in
Washington, D.C. In total, over 500 dams have been removed in the past two decades
(Stanley & Doyle, 2003). Figure 1.1 is a map of dams removed from the United States
River Systems. A brief evaluation of these numbers demonstrates that dam removal is
increasing as a management tool to restore river systems.
3
Figure 1.1 A map of all dam removal projects in the United States. (American Rivers)
4
In this thesis I am using the removal of the 5th Avenue Dam, located in Olympia,
Washington, as a case study for assessing the implications of sediment transport in smallscale dam removal projects. The 5th Avenue Dam, built in 1951, creates Capitol Lake, an
iconic public recreation site reflecting the Washington State Capital building in
downtown Olympia. Figure 1.2 is a map of the 5th Avenue Dam and the Deschutes
Estuary.
5
Figure 1.2. Map showing Capitol Lake in 2004. The four distinct sub-basins are South
Basin, Middle Basin, Percival Cove, and North Basin and are connected through the
labeled features. The Deschutes River enters South Basin from the southwest. The Port of
Olympia and municipal marina reside north of the 5th Avenue Dam and Bridge in Budd
Inlet. (George, Gelfenbaum, & Stevens, 2012).
6
Over the years, sediment deposition has decreased the lake’s volume by 25
percent, thus creating some negative environmental impacts to the reservoir (Washington
Department of Enterprise Services, 2006). The Washington Department of Enterprise
Services has closed Capitol Lake to water recreation activities for two primary reasons.
First, the New Zealand Mud Snail (Potamopyrgus antipodarum), an invasive freshwater
mud snail that spreads rapidly due to its lack of predators, has populated the lakebed,
causing state officials to close the reservoir to prevent spread of the species to
surrounding water bodies. Additionally, Capitol Lake’s dissolved oxygen levels are
producing harmful algae blooms. The appearance of algae shows the negative feedback
loop created by the dam. Poor circulation and the decomposition of the algae blooms
depletes the reservoir’s oxygen content. If the levels continue to drop, the water body can
become hypoxic, a very low oxygen condition that can kill aquatic species. These two
negative consequences of having the 5th Ave Dam have spurred state officials to look at
new ways to manage the reservoir.
In 1999, the Washington State Department of Ecology, the Washington State
Department of Fish and Wildlife, and the Washington State Department of Enterprise
Services jointly decided to explore various restoration options to manage the sediment in
Capitol Lake. The “Deschutes Estuary Feasibility Study” (referred to as the Study) was
completed in 2009, and four restoration scenarios, listed below, were examined:
1) Lake/River/Wetland,
2) Lake,
3) Estuary, and
4) Lake/Estuary.
7
The Study determined the estuary restoration was feasible and could be restored
through the removal of the 5th Avenue Dam. In order to predict how to manage the
sediment released under that scenario, state officials developed a technical committee to
focus on restoration efforts. The technical committee dedicated a portion of the study to
examining the Deschutes River flow and tidal processes. The interaction between these
two processes ultimately determines how sediment gets deposited and transported into
the estuary and river system. The subsequent “Deschutes Estuary Hydrodynamic and
Sedimentation Report” used the Delft3D model (explained below on page 9) to predict
how the mixing of freshwater and tidal waves will distribute sediment through the estuary
and river habitat.
However, even the most sophisticated modeling has some limitations, creating a
level of uncertainty that researchers must take into account when analyzing the results.
“There are a lot of natural processes that are occurring at once. For example, where the
channel forms, where the sediment deposits, you are trying to predict what this might
look like,” Guy Gelfenbaum, Researcher with U.S. Geological Survey, Pacific Coastal
and Marine Science Center said (Personal Communications, January 27, 2015).
“Modeling is very useful in predicting how these natural processes might occur, but there
are some limitations. It’s like if you blur your eyes [while looking at the model], you will
see what the general behavior is going to be- that is what our prediction is from the
model,” he explained.
According to Downs et al (2009), resource managers have little guidance on how
to manage the sediment after dam removal. The researchers claim that data input is one of
the most critical elements for a model to predict sediment transport scenarios. Although
8
the ultimate goal in dam removal is to restore the natural pulses of flow and sediment into
the downstream reaches, reconstructing a river channel’s morphology can be difficult and
the outcomes can be hard to predict. In order to project how sediment could be
transported and deposited into a river, researchers rely upon numerical simulations to
predict the channel response to the removal (Downs et al, 2009). Suspended sediment
load, stream flow or sediment load characteristics are measurements scientists use in the
various models to help predict the river channel’s evolution.
The 2009 Deschutes Estuary Feasibility Study utilized numerical modeling to
predict how the sediment would move throughout the Capitol Lake region and into the
estuary, Gelfenbaum explained (Personal Communication, Jan. 27, 2015). By using this
type of model, the report authors could determine how and where the sediment might
deposit throughout the lake’s system, and estimate the amount of time it will take to
return the estuary back to pre-dam conditions. Based on the model simulations, the
researchers developed a timeframe of ten years for the estuary to reach pre-dam
conditions. Figure 1.3 shows one of the many numerical simulations created from the
Delft3D model used in the 2009 Deschutes Estuary Feasibility Study.
9
Figure 1.3. An example of a numerical model simulation created by the Delft3D model used in the
2009 Deschutes Estuary Feasibility Study. (2006 Deschutes Estuary Hydrodynamic and Sedimentation Report)
10
Once field samples have been collected, researchers enter the data into a
numerical model that predicts how the river channel might change after the demolition of
the dam. For example, a one-dimensional sediment transport model can be used in
watersheds to predict large-scale spatial and temporal movements of sediment (Downs et
al, 2009). The model can provide insight on how the rapid release of sediment might be
deposited or suspended in the river’s channel during high and low flow events. A
sediment core is a method of analysis that allows researches to examine the composition
of river and lakebed. Core samples can help researchers determine the interaction
between sediment grain size, volume, and transport. This information is important part in
the process of removing a dam because the models provide a roadmap of where the
reservoir sediment could be deposited and offers predictions to the potential changes in
the river’s channel morphology (Downs et al, 2009). These can then be used to support
restoration efforts and policy formulation.
When the “Deschutes Estuary Feasibility Study” was completed in 2009, a lack of
peer-reviewed studies on dam removal, thus highlighting a weakness in the literature and
illustrates the information gap in dam removal projects to date. Scientists have noted this
gap in the literature and are developing new sediment transport scenarios (Konrad, C. P.
(2009); Mussman, E. K., Zabowski, D., & Acker, S. A. (2008); Pizzuto, J. (2002); Poff,
N. L., & Hart, D. D. (2002); Sawaske, S. R., & Freyberg, D. L. (2012)). For example, the
2006 “Deschutes Estuary Sedimentation Transport Report” used the Delft3D computer
program, a numerical hydrodynamic and morphological model developed by Delft
Hydraulics, the Netherlands. This software was developed in the 1980s and is still
considered to be one of the premier models for sediment transport and morphology
11
investigations, according to the Deschutes Estuary Feasibility Report. The Delft3D model
factors in sediment grain size, deposition, flow, and discharge, but fails to include
parameters for the dam removal timeline or sediment cohesion, or to explain how all of
these variables can create various negative ecosystem responses. A negative feedback
loop created from dam removal is the deposition of fine-grained sediment on flood plains.
The influx of non-cohesive material could bury plants and influence the establishment of
invasive vegetation, such as Reed Canarygrass (Phalaris arundinacea). The report’s
primary focus was not on the broader ecosystem but on determining which restoration
alternative was the best scenario. The report determined that estuary restoration was
feasible and would occur after the 5th Avenue Dam removal.
The Deschutes Estuary Feasibility Study used the Delft3D model to assess the
four restoration scenarios (listed on page 7) to predict the volume of sediment transported
into the estuary based on high and low erosion levels that could occur during four
different flow events. Each scenario provided an estimated amount of sediment that
would accumulate at the mouth of the Deschutes River and could be transported into
Budd Inlet. However, this model relied on a limited amount of empirical data available
for the modeling program, thus creating a level of uncertainty about the impacts of stored
sediment upon the estuary’s ecosystem. The inconsistency in sampling methods used also
created a high level of variability in the Sedimentation Transport Report; the results
might not have provided an accurate portrayal of sediment conditions of the lake or river
channel. Finally, the Deschutes model failed to incorporate any best management
techniques for the future management of Budd Inlet.
12
In 2005, various state and federal agencies collected baseline data for the
“Deschutes Estuary Feasibility Study.” They broke the monitoring of the lake and river
into several sections. Monitoring sites were located through the North, Middle and South
Basins of the river and estuary (see Figure 1.2 on page 6). The Department of Ecology
surveyed the South Basin and Percival Cove. The Department of Fish and Wildlife
surveyed the area under the Interstate 5 Bridge. The United States Geological Survey
completed a bathymetric survey between the 5th Avenue Dam and the Port of Olympia.
The United State Army Corps of Engineers surveyed the Port of Olympia and sections of
Budd Inlet. The Thurston County Regional Planning Department provided the
topographic information of the area. The information collected by field staff provided
five quantitative data sets used in the Delft3D model to determine sedimentation and
hydrodynamics scenarios for the proposed four restoration strategies. The field data
collected included surface sediment samples, flow, and cross-section profiles.
Unfortunately, the data used in the report were lacking in several areas, such as the
various levels of soil erosion. According to the report, the authors decided to determine
the level of erosion that could occur in the watershed by looking at two levels of
erodibility1: a low level and high level. The authors then analyzed how these two
parameters would react when critical sheer stress from events, such as floods and the rate
of erosion, occurred.
After researching “Deschutes Estuary Feasibility Study” and “Deschutes Estuary
Hydrodynamic and Sedimentation Report” and scrutinizing the data supporting their
1
The level of erosion in the Deschutes Estuary Sedimentation Report is based on two parameters: a high
and low level of erosion. These two parameters can impact how sediment is transported and accumulated
at the mouth of the river.
13
conclusions for the 5th Avenue Dam, I decided to focus my thesis research on
supplementing the work completed in the sediment transport report. More generally, I am
researching the implications of sediment transport in small-scale dam removal projects.
This research will answer the following questions: 1) What variables should be
considered in the development of a comprehensive sediment transport model when
removing a small dam? and 2) What restoration methods should be considered for longterm maintenance of the area?
Research gathered for my thesis will assess how natural resource managers
address sediment transport in dam removal projects. The thesis will be organized as
follows. Chapter 2 features the methodology used to complete my research and
documents my research process and how I selected candidates to interview to gain
background information on the Deschutes River and Estuary. Chapter 3 is an overview
of dam removal projects, including the influences on river restoration, regulations and
management in dam removal projects, inventory of dams, and the size of dams. In
Chapter 4, I highlight the issues surrounding the restoration of the Deschutes River and
Estuary and the management methods used to address sediment transport in the removal
of the 5th Avenue Dam. This chapter also has background information on the Delft3D
model, which was used to determine sedimentation scenarios for the removal of the dam.
Chapter 5 contains my research analysis. Finally, Chapters 6 and 7 feature a discussion
section and a conclusion for this project.
14
CHAPTER 2
METHODOLOGY
The research completed for my thesis was based primarily on the “Future Areas
of Study” section of the 2009 “Deschutes Estuary Feasibility Study.” That section listed
four areas of study that required future research: 1) contamination and pollution
dispersion; 2) examination of water quality and seasonal river flows; 3) sea level rise; and
4) improved modeling techniques. After reading the report, I chose to focus on improved
modeling techniques and water quality impairments that occurs when a dam is removed.
Once I narrowed down my topic I contacted Sue Patnude, Executive Director of the
Deschutes Estuary Restoration Team, to discuss my research and develop a list of
technical experts who participated in the feasibility study. Ms. Patnude provided me with
a list of potential contacts who could provide the history of the study might provide
technical advice on the restoration effort. I contacted and interviewed the following
people during the months of January and February 2015:
1) Curtis Tanner, Division Manager of the Environmental Assessment and
Restoration Division of The U.S. Fish and Wildlife Service, Washington Fish and
Wildlife Office;
2) Lance Whitica, Executive Director of the South Sound Salmon Enhancement
Group;
3) Guy Gelfenbaum, Researcher with U.S. Geological Survey, Pacific Coastal and
Marine Science Center;
4) Carrie Martin, Assets Manager, Washington State Department of Enterprise
Services;
5) Elizabeth Grossman, Author, Watershed, The Undamming of America;
6) Padraic Smith, Environmental Engineer, Restoration Division,
WashingtonDepartment of Fish and Wildlife.
15
In addition to the interviews, I read journal articles that pertained to the topics of
sediment transport, modeling, dam removal, and river restoration practices from the
following publications: BioScience, Geomorphology, Water Resources Research, Nature,
Journal of Hydraulic Engineering, Journal of the American Water Resources
Association, Ecological Engineering and BioOne.
While writing my literature review, I realized that my research topic required
sections that highlighted the historical information of how downtown Olympia converted
its estuary habitat to a managed lake. Furthermore, as I began conducting my interviews,
I soon realized that I needed to incorporate the evolution of dam removal as a restoration
practice, and the way in which researchers are implementing this new strategy to restore
rivers and estuary systems.
After reviewing the data and comparing the information provided to methods used
in various case studies, and after a site visit to the 5th Avenue Dam, I decided to focus on
results on two parameters: volume of sediment released and the timing of how the
sediment can be transported and deposited into the system.
16
CHAPTER 3
DAM REMOVAL
3.1 DAM REMOVAL
Dam removal is a viable restoration practice that returns river systems back to
their free-flowing state. Dams can create economic opportunities when installed into a
river’s system. Rivers provided a seemingly endless supply of water and energy for
purposes such as irrigation or hydropower, and create reservoirs for recreation use and
water storage. According to the National Dam Inventory Database, 75,000 dams exist in
the United States today. The majority of these structures were built during 1900-1949,
when our nation’s decision makers viewed natural resources as a fuel to support the war
effort and for industrialization (Shuman, 1995). Engineers and architects viewed the
dams as self-sustaining structures that would only require periodic maintenance and
would serve society through their infinitely long lifespan (Shuman, 1995).
However, architects, engineers, and builders failed to factor in how a run-of-ariver dam or the construction of large dams with reservoirs would interact with or impact
the rivers’ physical, chemical, and biological characteristics and processes (Hart,
Johnson, Bushaw-newton, et al., 2002). A river’s system changes immediately with dam
construction; the impoundment directly effects alterations to the river’s floodplain,
channel development, and sediment supply (Pizzuto, 2002). (An impoundment is a body
of water, a reservoir or lake, created when a dam is constructed in a river system.) Dams
also prevent flooding and stop sediment from flushing down into the lower reaches of the
stream channel. Weather events, such as snowmelt and heavy rainstorms, can produce
high flows, causing rivers to exceed their banks and flood. A dam can help reduce
17
flooding in the downstream reaches of a river system (Shuman, 1995) when its
impoundment captures the excess runoff during high flow events. Also, the stored water
can be released during scheduled times and used during periods of droughts for irrigation
purposes.
Nonetheless, flooding is important in the development of a river’s ecological
processes. The movement of sediment and periodic high water flows in a river’s system
help create and form a stream’s hydrologic regime (Pizzuto, 2002). River systems benefit
from flooding because these natural events transport silt into the floodplains. A flood
event helps replenish the soil along a river’s riparian area. The soil deposited after a flood
has nutrients that aid in crop production. When a dam is constructed in a river, it prevents
flooding from occurring and can starve the riverbanks from soil deposition. Over time,
the decline in these natural processes can lead to water quality impairment and the loss of
aquatic habitat (Gottgens & Evans, 2007).
Stanley and Doyle (2002) argue that the effects of nutrient dynamics, such as the
movement of nitrogen and phosphorous in a river system, must be a priority in research
and restoration plans. The retention of nutrients in a reservoir or lake is a concern for
natural resource managers because the imbalance of nutrients can cause negative effects
on the downstream reach. Impairments to the stream’s water quality and sediment
deposition can be harmful to migratory taxa, such as salmon and other aquatic species.
Unfortunately, the ability to predict how stored nutrients move throughout the river
channel after dam removal is a complex problem. Some scientists, Stanley and Doyle
(2002) and Poff and Hart (2002), for example, claim that restoration efforts should be
examined from a watershed scale perspective so that factors, such as land use change, can
18
be incorporated into restoration plans. Furthermore, Stanley and Doyle (2002) advocate
for dam removal to be used as an experiment to test predictions and to gain insight in
how nutrients can be transported and retained in a river’s reach.
The Deschutes Estuary Sedimentation Model did not include a comprehensive set
of sediment samples but instead focused on bathymetry, boundary conditions, and timevarying processes, such as river discharge. In recent dam removals, such as the Elwha
and Glines Canyon Dam Removal projects, scientists used a landscape evolution model
to determine the delta’s erosion by integrating field monitoring data and modeling
software. A river’s delta, located at the mouth of a river system, is at a low elevation and
its formation results from the transport of sediment carried from the river’s source,
commonly found at higher elevations. Hydrologic and hydrodynamic processes influence
the particular morphology adopted by a river’s delta (Gelfenbaum, et al, 2015). Field
studies of river delta morphology have allowed researchers to classify river deltas based
on tidal processes, waves, and river discharge (Gelfenbaum, et al 2015). In addition to
these processes, the formation of the delta is dependent up the effect of the sediment
grain size and how this will interact with the delta’s morphology (Gelfenbaum, et al,
2015). The type of sediment deposited in a delta can be transferred from erosion or
through the transport processes that can vary with different flow events (Gelfenbaum, et
al, 2015.). In addition, the sediment transferred into the delta can be deposited in
adjacent beaches and near shore zones (Gelfenbaum, et al, 2015).
The installation of the 5th Avenue Dam has starved Budd Inlet from receiving
sediment from the Deschutes River. Sediment transported from the Deschutes River is
accumulated in Capitol Lake because of the 5th Avenue Dam. The dam has disconnected
19
the river and estuary habitats. The 5th Avenue Dam, operated by the Washington
Department of Enterprise Services, prevents the Deschutes River from reaching Budd
Inlet in a natural fashion. Composed of concrete, the 5th Avenue Dam rises height 45 feet
and has an 82-foot wide rectangular spillway. Figure 3.1 shows the design of the 5th
Avenue Dam. The dam features a fishway channel and gates driven by water level
sensors in the lake and in Budd Inlet (Carrie Martin, Asset Manager, Washington
Department of Enterprise Services explained (Personal Communication, March 3, 2015)).
20
Figure 3.1 A detailed schematic design of the 5th Avenue Dam.
(Washington Department of Enterprise Services)
21
Dam operators monitor the lake’s water level to ensure that when the dam’s gates
open, tidal water does not enter Capitol Lake. The 5th Avenue Dam’s gates are schedule
to release water based upon the Deschutes River flow. Capitol Lake’s depth is
approximately 6.2 feet in the summer and approximately 5.2 feet during the winter
season; therefore, water levels fluctuate during the various seasons and weather events.
Additionally, the inflow rate from the Deschutes River affects the lake level. When high
flow events, such as winter rain storms, occur in the Deschutes River basin, the dam
operators will release the additional flow into Budd Inlet (Martin, Personnel
Communication).
Dam operators rely upon the tide cycle to determine when they will open and
close the gates. For example, there are two high tides and two low tides each day, and the
field staff monitoring these cycles will open the gates twice a day. During the scheduled
release events, some of the sediment that has been deposited into the lake will enter Budd
Inlet. State officials do not have a clear understanding the behavior of how and where the
sediment gets deposited because there has not been a completed field survey of how the
sediment might be transported. To date, the only research that has been completed on
sediment transport was completed during the Study, when officials examined if it was
feasible to restore the estuary, Sue Patnude, Executive Director of the Deschutes Estuary
Restoration Team said (Personal Communication, February 12, 2015). Even though
estuary restoration is feasible, state officials have not made a decision to restore the
estuary and Capitol Lake has been just operating at the status quo.
All of the sediment transported from the Deschutes River’s source gets deposited
in Capitol Lake’s three basins. State officials have been debating over various methods to
22
manage the sediment deposited in Capitol Lake. To date, the only method used to manage
the accumulated sediment has been the dredging of lake in 1979 and 1986.
When the 5th Avenue Dam is removed, the sediment that has deposited in these
regions will become part of the re-formation of the estuary. However, that will require a
strategic plan to address how the sediment will be transported into the Budd Inlet, and at
what rate. The Study determined that approximately 60 percent of the existing sediment
would need to be removed, Curtis Tanner, Division Manager of the Environmental
Assessment and Restoration Division of The U.S. Fish and Wildlife Service, Washington
Fish and Wildlife Office, explained. The removal of this sediment would be
accomplished by dredging the fill material and transporting the debris to an off site
location. The remaining material would then be allowed to flow into Budd Inlet.
Determining how the sediment and tides will interact in influencing the deposition of the
soil will require extensive study, modeling, and planning. One of the methods to predict
how these interactions will occur is through the use of models, such as the Delft3D
model. In the discussion portion of the Deschutes Estuary Feasibility Study, the authors
stated that further research should explore what new modeling techniques have been
developed since the publishing of the report to provide a more accurate prediction of
erosion and deposition.
In the last decade, scientists have started viewing dam removal as a method to
restore river systems back to their pre-disturbed states. Over time, the dam structures and
the surrounding environment have begun to deteriorate, causing increases in the dams’
operational and maintenance costs. Dam operators must address potential stressors, such
as harmful algal blooms and the crumbling or cracking of concrete due to pressure from
23
the collection of silt. Dams severed salmon migration, which is a concern for
communities in particular those located in the Pacific Northwest that thrive on fish
populations to support the local economy. The installation of fish ladders, dredging of the
reservoir, and scheduled water releases are a few best management practices
implemented to address these issues.
However, these mitigation strategies are only temporary measures to the longterm operational and maintenance of a structure (Pohl, 2002). According to Babbitt
(2002) the old view of dams, the “build now, ask questions later” approach, has been
replaced by river restoration efforts, largely due to the high level of uncertainty in the
impacts that impoundments have upon environmental and public health. For example, if
anyone proposes a dam project today, either building or removing one; they must
complete a series of environmental and social impact studies, such as how the installation
or removal of a dam might affect area businesses. “The efficacy of dams is being
scrutinized in new comprehensive analyses of ecology, economics, energy efficiencies,
water conservation and public safety,” Elizabeth Grossman wrote in her 2002 book
Watershed: The Undamming of America. Completing an assessment of the environmental
and social impacts before the installation of a new dam or upgrade to an existing dam is
important because it allows all parties involved to examine the alterations to the
surrounding ecosystem. In some dam construction projects, the structure can impede fish
passage, or fragment habitat that is utilized by a variety of terrestrial and aquatic species.
By completing these studies, researchers and policymakers are able to gain a better
understanding of the impact that removal, installation, or upgrade will have upon the
entire ecosystem and what ecosystem services might be lost or altered. Babbitt says local
24
and national governments rely upon the precautionary principle 2 to determine if the
benefits of removal outweigh the continued use and benefits of the dam.
Federal and state agencies inspect and monitor dams to ensure that the structures
meet federal and state codes. In Washington State, the Washington Department of
Ecology (also known as Ecology) regulates dams that capture and store at least 3.2
million gallons of water (Washington Department of Ecology, 2014). Ecology’s Dam
Safety Office currently oversees 1,019 of 1,141 dams across the state and monitors these
structures through inspections (Washington Department of Ecology, 2014). These
periodic visits give regulatory staff the ability to assess the structures’ integrity. In the
case of the 5th Avenue Dam, different types of inspections for the dam occur at different
intervals, Martin explained (Personal Communication, March 3, 2015). Because the dam
also serves as a bridge, the Washington Department of Transportation checks it for
structural soundness. Safety technicians inspect the accessible areas of the structure once
every two years and perform an underwater structure inspection every five years.
Additionally, the Department of Enterprise Services maintains the mechanical systems,
exterior sensors, hydraulic system, and electronic control components. Moffatt and
Nichol, an engineering firm based in Seattle, Washington, completed the last full
assessment of the dam in 2008.
Pohl (2002) breaks down the rationale behind dam removal into three main
categories: economics, safety, and the environment. The economic costs of repairing
defects found during an inspection of a dam often drives the removal of these structures
(Babbitt, 2002). The economic analysis of the Edwards Dam on the Kennebec River
2
Policymakers use the precautionary principle as a decision-making tool to determine if a particular course
of action should be made on an issue, in this case the removal of a dam. This principle is used when
extensive scientific knowledge is limited on a subject matter
25
showed that dam removal would cost $2.7 million, compared to $10 million for dam
modifications (Doyle, Harbor, Stanley, 2003). Even though the Edwards Dam still
produced electricity, it generated only a small amount, an estimated 0.1 percent of
Maine’s total electric supply (Doyle, Harbor, Stanley, 2003). The cost of upgrades to the
structure warranted total dam removal rather than the necessary modifications. The
Woolen Mills Dam, located in Wisconsin, was an abandoned structure that was deemed a
public safety hazard and required repair (Doyle, Harbor, Stanley, 2003). The
modifications to this structure would have cost an estimated at $3.3 million, compared to
the $80,000 to remove the structure (Doyle, Harbor, Stanley, 2003). Again, the high costs
of upgrades led to the removal of the dam in 1988.
The financial burden to repair or upgrade a dam is prompting dam owners to use
dam removal as a tool to return rivers back to their natural states. This change in
awareness, and a change in attitude of many from pro-dam to pro-removal, has opened up
new avenues in river restoration methods, adding dam removal to the list of potential
strategies to restore ecologically degraded rivers (Babbitt, 2002). Additionally, dam
removal taps into a social component of environmental stewardship. Downs et al (2009)
says that society’s shifting ethos toward this new trend allows the public to “feel good”
about removing a dam.
Returning to the case of the 5th Avenue Dam in Olympia, state and federal
officials have come to a roadblock in the removal of this structure. Tanner, Division
Manager of the Environmental Assessment and Restoration Division of The U.S. Fish
and Wildlife Service, Washington Fish and Wildlife Office, says that society’s opinion on
whether or not the dam should be removed plays a critical role in whether or not the river
26
and estuary are restored: “What do you want it to look like?” he asked (Personal
Communication, Jan. 23, 2014). One of the biggest hurdles in the removal of the
Edwards Dam in Maine was explaining to people what the river system would look like
after dam removal, Grossman, author of Watershed: The Undamming of America, said
(Personal Communication, February 10, 2015). “There were no photographs of what the
river was like before the dam went in. They thought the impounded river was the way a
river was suppose to be. Some thought that if you removed the dam, it was going to be an
empty bathtub.” Gelfenbaum shares a similar opinion about restoration ecology in
general: “Restoration is a little about the science and a lot about what people want,”
(Personal Communication, Jan. 27, 2014).
The Milltown Dam, located on the Clark Fork River in Montana, illustrates how
arduous dam removal can be for a community. The Milltown Reservoir, 7 miles east of
Missoula, Montana, was found to be contaminated when Montana Department of
Environmental Quality discovered arsenic in the groundwater. This likely resulted from a
massive flood in 1908, which washed millions of tons of mine waste into the Clark Fork
River (Grossman, 2002). Because of the dam, the pollutants remained trapped in the
reservoir. In 1983, United States Department of Environmental Protection added the
Milltown Dam to the National Priorities List for removal due to groundwater
contamination. Only in 2001 did state and federal agencies start the process of removing
more than 2 million cubic yards of sediment; in 2006 the dam was demolished
(Grossman, 2002). One of the struggles with the removal of the Milltown Dam, explained
Grossman, was the disposal of that contaminated sediment. “The fate of this toxic
material was the biggest issue,” she said. State officials decided to haul the sediment
27
upstream to the town of Opportunity to deposit the material. There was some opposition
about the removal, but overall, people did understand the need for the dam removal.
According to Grossman, this came from the efforts of the Clark Fork Coalition, a local
watershed group based in Missoula, which attempted to educate the communities on the
problems with the sediment and the importance of its removal.
Now that dam removal is an option, ecologists are engaging in natural
experiments with dam removal by documenting the return of ecosystem processes lost or
altered by dams. Prior to the removal of the Glines Canyon and Elwha Dams on the
Elwha River in Washington, the majority of the research literature primarily focused on
how dams alter the river’s ecological processes (Gregory, Li, & Li, 2002). Engineers and
geomorphologists studied river characteristics such as incision, floodplain formation, and
channel development--all functions critical in assessing the health of a river’s hydrologic
regime. Water temperature, channel morphology, and habitat diversity were considered to
be “master variables” in river restoration ( Poff & Allan, 1997). These parameters do
play an important role in determining the distribution and abundance of riverine species.
Yet historically, environmental regulations have been limited in scope, mainly focusing
on one aspect of water quality: minimum flow. Poff and Allen (1997) argued that by
returning a river back to its natural flow regime, the biodiversity and ecosystem processes
of the river would be restored.
The scientific framework for reducing the impact of dam removal upon the
upstream and downstream ecosystems is emerging in the field of restoration ecology
(Hart, Johnson, Bushaw-newton, et al., 2002). Although dams have been removed from
rivers throughout history, restoration ecologists have noted the lack of documentation on
28
the size of the dam, and methodology utilized in past removal projects (Stanley & Doyle,
2002; Pizzuto, 2002). Only a few peer-reviewed articles on the different approaches used
to remove dams appear in a handful of scientific journals prior to the 1980s (Gottgens &
Evans, 2007). This small number of peer-reviewed studies indicates a significant lack of
information on how ecosystems respond to dam removal (Hart, Johnson, Bushawnewton, et al., 2002).
More recently, ecologists have been filling in this information gap by
incorporating the response and recovery process into their restoration plans and
publishing this information in journals. After completing an extensive review of literature
for this project, I found a trend in the publication of research results. From my research, I
have discovered that the majority of articles were published in late-1990s, 2000, and
2002. This number has risen since dam removal projects have increased, in particular
with the removal of large-scale hydroelectric facilities. For example, the monitoring of
the Elwha and Glines Canyon Dam removals alone has increased the amount of literature
on dam removal projects. Researchers are closely monitoring various terrestrial and
aquatic parameters, such as vegetation establishment and morphological response time,
which were typically not documented before.
For example, in January 2015, the journal Geomorphology featured a series of
articles about the Elwha River removal project. The United States Geological Survey, the
Bureau of Reclamation, the National Park Service, the National Oceanic and
Atmospheric Administration, and the University of Washington collaborated on reporting
the most recent findings. The group published five separate journal articles on the
following topics: sediment budget, erosion of reservoir sediment, fluvial sediment load,
29
river channel and floodplain geomorphic change, and coastal geomorphic changes. The
scale of analysis and documentation of the both positive and negative results that are
occurring from this specific removal project is changing the scientific community
perception on river and coastal restoration efforts. Scientists are evaluating dam removal
and river restoration practices in a more holistic method by observing parameters not
only occurring on the river, but also occurring in the river’s delta and near shore area,
connecting the effects that are appearing in the river and coastal zones. The prior
methods, used were typically reductionist, where only the parameters of water quality and
channel morphology were documented (Bednarek, 2001). In the article featuring the
Elwha River delta coastal geomorphic changes, the journal authors wrote the following
about the removal project: “ The removal project provided an unprecedented opportunity
to examine the geomorphic response of a coastal delta to these increases [the sediment
released]” (Gelfenbaum et al, 2015).
Furthermore, researchers have been examining different methods of how to
remove a variety of dams, from large hydroelectric facilities to a low-head dam on a
small stream. In the case of the Elwha River, the project managers determined that a
staging was the best option in the removal of these two structures. (Staging is the process
of removing the dam in several phases to allow the ecosystem to slowly respond to the
increased movement of sediment and flow.) For the removal of the Marmot Dam, located
on the Sandy River in Oregon, the researchers decided to use the “blow and go method.”
(The blow and go method utilizes blasting equipment to remove the dam from the river
all at once.) By studying the different removal processes, researchers are monitoring how
the ecosystem is changing. In Coastal Geomorphic Changes, the researchers spent the
30
first two years after the dam removal observing and documenting the alternations to the
Elwha River delta. Weekly observations and repeated beach survey results were
combined with digital elevation models to determine the amount of accumulation of
sediment in the river’s delta (Gelfenbaum et al, 2015). The monitoring efforts provided
unprecedented data about the beach and seafloor grain size changes. By documenting
these parameters, the researchers could calculate a sediment budget based upon the
volume and grain size observed in the delta (Gelfenbaum et al, 2015). Thus, by
combining field observations and the use of modeling results, the researchers were able to
develop and validate the increases in sediment.
The combination of pre- and post-monitoring of a river’s ecosystem represents a
shift in the strategies to restore river systems. In the case of the Elwha and Glines Canyon
Dam removals, ecologists started gathering data at the site several years before the
barriers were slated for removal. Water quality, established channel width, and the
presence of aquatic and terrestrial species were a few of the parameters documented in
the year prior to the removal. The data collection allowed them to establish a baseline of
information to assess the stressor-response relationship that existed between the dams and
the river. The monitoring also investigated potential ecological responses that might be
manifest when the dams were breached (Hart, Johnson, Bushaw-newton et al., 2002).
Researchers measured parameters such as river cross-sections, sediment cores, flow
velocity, sediment grain size, color, composition, and structure. The incorporation of
temporal and spatial components, such as time and recovery of the river channel, and the
correlation of this information to the watershed characteristics, has helped inform the
adaptive management and restoration techniques of dam removal (Gregory et al., 2002).
31
Now, factors such as a river’s longitudinal profile, channel position, suspended
sediment load, and water levels are collected for several months to a year before a dam is
scheduled for demolition. Additionally, the timing of the removal and the estimated
volume of sediment distributed once the barrier is removed must also be considered.
Most dam removal projects typically occur during the late spring and early fall seasons as
a result of concerns about stream flows--the natural resource manager’s target for dry
temperature and consistent stream flows to transport the material downstream.
Researchers now actively document flow and sediment parameters prior to the removal
so that models can predict the outcome of such actions and guide which restoration
techniques should be applied (Konrad, 2009).
In 2005, Cui et al developed a two-part model called Dam Removal Express
Assessment Model (DREAM). The DREAM model simulates non-cohesive sediment
transport after dam removal. The DREAM-1 model documents fine sediment transport
(sand or finer material), while the DREAM-2 model is able to simulate both coarse and
fine sediment. This model was pivotal in the removal of the Marmot Dam because the
dam by allowing researchers to predict the scale of the sediment flume that would be
transported downstream once the structure was demolished. However, this model also has
limitations. Cui et al was unable to provide a detailed channel response at a
morphological unit scale (Cui et al, 2006). With extensive monitoring, scientists are using
this data to track the river’s ecological process, as well as to advance the science of
ecology by viewing restoration efforts from a systematic approach (Hart, Johnson,
Bushaw-Newton, et al., 2002).
32
3.2 RETURN OF A RIVER
The term “river restoration” generates controversy because it suggests that a
river’s fluvial system can be returned to a pre-disturbed state (Pizzuto, 2002). Returning
a river back to anything resembling the previous natural hydrologic regime requires
extensive planning. The use of historic aerial photography, paleohydrologic studies of
debris left by floods, and studies of historical damage to living trees are several methods
used to assess how a river system’s floodplain and riparian vegetation responds during
high flow events (Schmitz et al, 2009). For the Deschutes Estuary, historical
photographs have provided researchers and me with a glimpse of the conditions existing
prior to the installation of the 5th Avenue Dam.
The current scientific knowledge about predicting the ecosystem response to dam
removal is relatively limited (Poff & Hart, 2002). A river’s own fluvial processes adds
levels of uncertainty to the process (Pizzuto, 2002). As a result, the ability to fully
understand the complexity of a river’s hydrologic structure may require extensive
monitoring for decades, even centuries. Poff and Hart (2002) argue that in order to have
an effective restoration plan, scientists must identify the full range of potential stressors
to the river, including the volume and sediment grain size that could be released once the
dam is breached. To predict these outcomes, scientists must gather baseline data, often
information not available due to the lack of pre-removal monitoring efforts. Since in the
past many viewed dams as having infinite lifespans, the idea of removing the structures
was not something planners’ or politicians’ considered and they did not contemplate the
need to collect data on how the system operated prior to removal. For example, when the
state legislature approved the installation of the 5th Avenue Dam in 1938, state officials
33
did not consider in 50 years the state legislature would contemplate the structure’s
removal.
More recently, scientists have started using spatial distribution models to predict
the volume and magnitude of discharge and suspended sediment load once the
impoundment is removed (Granata, Cheng, & Nechvatal, 2008). For example, scientists
utilize stream channel cross-section surveys to generate a potential sediment discharged
rate. Parameters such as channel depth and width provide researchers with data that can
be plugged into a Manning’s equation3 to create a probable discharge rate that could
occur when the barrier is removed (Granata et al., 2008). Also, researchers can tap
archival data, such as maximum daily turbidity, and correlate the turbidity levels to
discharge rates by assessing trends seen in a regression model (Granata et al., 2008). The
two parameters determine how much total suspended solids exist in the water. Also,
evaluating these parameters helps determine how much existing sediment is being
transported by the river’s system.
Parameters such as dissolved oxygen, total suspended solids, daily stream flows,
and so forth, monitored by various state and federal agencies and private companies, give
scientists just enough information to predict ecological outcomes. Poff and Hart (2002)
claim that monitoring all potential parameters may be labor-intensive and increase
removal budgets, but this information is necessary in order to develop a comprehensive
restoration plan. Some researchers argue that ecological responses can be predicted based
only on the use of existing data (Chang, 2008). This conflict has prompted scientists to
3
A Manning’s equation is one of the most commonly used equations to determine stream channel flows.
Robert Manning, an Irish Engineer, first introduced the equation in 1889 as al alternative to the Chezy
equation. The Manning’s equation is an empirical equation that is applied to uniform stream flow by
utilizing the channel’s velocity, flow area and channel slope.
34
develop conceptual models such as the DELF3D to determine which restoration scenario
is the most feasible and best suited for the ecosystem (Hart, Johnson, Bushaw-Newton, et
al., 2002). Various models provide various outcomes. Spreadsheet-based models
calculate sediment yield, accounting for trapping of sediment by upstream reservoirs and
changing trap efficiency with time (Cui et al, 2006). The landscape evolution model
(LEM) simulates topographic evolution at a wide variety of spatial and temporal scales
(Cui et al, 2006). Hydraulic flow and sediment transport models such as SRH-2D
produce different results from LEMs because the simulations of the flow range are less
detailed (Cui et al, 2006). Each one of these various types of models predict an outcome,
but the level of detail of the results can vary based upon its application.
3.3 INFLUENCES ON RIVER RESTORATION
Scientists continue to face controversy over the notion of returning our rivers back
to a “natural state” because of impacts historical and current land use practices have upon
rivers and river systems. Many of our rivers have been impacted not only by the
impoundments themselves but also by various anthropogenic factors such as nonpoint
source pollution (Stanley & Doyle, 2002). Oil leaks, garbage, pet waste, and lawn
fertilizer are several sources of nonpoint source pollution that impair the Deschutes River
and Estuary. On the other hand, environmentalists and some politicians see dam removal
as a method to compensate for the negative effects of human activities on river
ecosystems. However, dam removal should not be considered as a solution that will
automatically return a river back to its natural state. The ability to totally remove the
impact of humans from the ecosystem is often impossible (Hart, Johnson, BushawNewton, et al., 2002). Even after impoundment removal and restoration efforts, the river
35
system may not be restored fully due to its proximity to urbanized areas (Roberts,
Gottgens, Spongberg, Evans, & Levine, 2007). Stanley and Doyle (2002) and Hart et al
(2002) state that restoration plans must examine both the local and regional environment
to determine the dynamics between the barrier and the river. By looking at the dam from
a variety of perspectives in the restoration planning process, scientists and restoration
ecologists can develop a broad understanding of the magnitude and range of physical,
chemical and biological responses to dam removal.
Indeed, Hart and Poff (2002) view dam removal as an experimental practice that
can provide a better understanding of the intricate relationship between the barrier and
the river. Hart and Poff claim that by assessing existing dams, reviewing case studies of
previous dam removals, and incorporating current monitoring and historical data, models
such as the erosion prediction model can prevent extensive damage from occurring when
the dam is removed. Researchers on the Elwha River project, for example, used tools
such as the Universal Soil Loss Equation, a formula that helps determine the annual
amount of soil loss that occurs in river systems due to erosion. Team gathered sediment
core samples from the dam’s reservoirs and created an experimental model that examined
the application of three different treatment methods to reduce further erosion from
occurring in the river’s channel post removal (Mussman et al., 2008). By considering the
potential outcomes, researchers were able to develop a multi-stage restoration plan that
incorporated various best management practices to help reduce runoff and erosion from
occurring in the up and downstream reaches.
36
3.4 REGULATIONS AND MANAGEMENT
The environmental trade-off of changing a river’s flow regime from a lotic4 to a
lentic system was rarely considered (Graf, 1999) until the late 1980s, when federal
agencies, such as the Federal Energy Regulatory Commission (FERC), began cracking
down on deadbeat dams by enforcing environmental regulations. Mandates set forth by
the Clean Water Act, the Endangered Species Act, and the Federal Power Act are driving
dams to be removed from rivers (Babbitt, 2002), as outlined below. These specific
legislative acts require FERC and other federal agencies to examine the environmental,
economic and social tradeoffs occurring in the surrounding area since the installation of
the barrier.
The Federal Power Act of 1920 created FERC, which oversees the construction
and maintenance of dams in the United States. FERC determines the operational uses of a
dam by considering the environmental impacts that the structure has upon the ecosystem.
In the 1960s, FERC began regulating all non-federal hydro projects across the country
(Winter & Cain, 2008). FERC issues permits to dam operators for either 30 or 60 years in
order for the dam to generate power; once the permit has expired, the power company
must apply for a new permit. During this process, FERC can determine if the dam should
be maintained, upgraded, or removed. In the case of the Condit Dam on the White
Salmon River in Washington, FERC required the power company to install a fish passage
system in order to receive its re-licensing agreement to generate power. However, the
power company decided to remove the dam because the demolition costs were
significantly lower than the mandated upgrades.
4
A lotic system describes fast moving water, and an example of this would be a river system. A lentic
system is a still body of water, and an example of this would be a lake or reservoir.
37
The majority of the dams have been removed from United States river systems
because of the high costs of upgrading the structure to meet environmental regulations.
Because those costs far exceed those of demolishing the structure, dam owners typically
decided to remove the barrier. Researchers estimate that by the year 2020, 85 percent of
the dams in the United States will be near the end of their operational lives (Doyle,
Harbor & Stanley, 2003). This statistic indicates that dam removal most likely will
continue to occur and that methods on how to address the ecosystem response to the
breaching need to be addressed.
3.5 INVENTORY OF DAMS
State and federal agencies created databases to monitor the number of dams and
their potential safety hazards. The Army Corps of Engineers manages the National
Inventory of Dams (NID); this system classifies all dams and tracks the age and repair
status of the barriers. The National Inventory of Dams records all dams that are potential
safety hazards. According to the National Inventory of Dams’ website, the dams listed in
the database meet at least one of the following criteria:
1) High hazard classification - loss of one human life is likely if the dam fails,
2) Significant hazard classification - possible loss of human life and likely
significant property or environmental destruction,
3) Equal or exceed 25 feet in height and exceed 15 acre-feet in storage,
4) Equal or exceed 50 acre-feet storage and exceed 6 feet in height.
However, this database is not an accurate representation of all the dams that have
been installed in US river basins. The majority of the dams located throughout the US
river systems are categorized as small-scale structures—having a height of 6 feet or less.
Many of the small dams constructed in river drainage basins do not meet the criteria and
are not listed in the NID database due to the height requirement. This highlights the major
38
drawback of the NID: the list is not comprehensive and lacks a significant portion of
barriers impeding river systems. Stanley & Doyle (2003) report that small dams are being
removed from river systems but their removal was never well-documented because of
their categorization as a small-scale structure. Many researchers have focused on removal
of large-scale hydro projects and viewed small dam removal as not having a significant
impact upon the ecosystem, but this perspective is changing. Regardless of a dam’s size,
the structure impacts the stream’s physical, chemical, and biological processes.
3.6 SIZE OF DAMS
Small-scale dam removal is not well studied; there is a deficiency in the amount
of published literature on their removal (Stanley & Doyle, 2002). However, because there
are so many small-scale dams in our river systems, researchers are now looking at
various methods to remove these structures. Some river systems may have several smallscale dams in their stream reaches, and removal of one or all of the dams could have a
significant ecological impact. Still, we know little about the magnitude of potential
positive or negative effects because of the absence of peer-reviewed studies that
specifically focus on small-scale dams (Stanley & Doyle, 2002).
The size of the dam in question was ignored by scientists in previous restoration
efforts. Scientists now compare the effects of removal of human-made, small-scale dams
to that of natural barriers like beaver ponds and waterfalls. Researchers are drawing
parallels between the physical or biological impacts of small-scale dams and natural
barriers in a stream’s channel design. Some parameters, such as the effects on nutrient
cycling, habitat and biotic migration, could be monitored to develop a conceptual
framework that reveals what ecological effects might be reversible, post-dam removal.
39
The criteria set forth by the National Inventory of Dams and the state’s dam safety
office provides a set of guidelines for local, state, and federal agencies to monitor dams.
In the case of the 5th Avenue Dam, the Department of Enterprise Services regularly
monitors this structure. The 5th Avenue Dam, with a height of 45 feet, falls into the
category of a large-scale dam. In addition, the extent of sediment deposition in the lake is
severe. To remove the structure, natural resource managers will have to develop a
restoration plan that addresses the volume and magnitude of sediment that would be
released. This would require an examination of the volume of sediment that will be
flowing into Budd Inlet. Padraic Smith, Environmental Engineer, Restoration Division,
Habitat Program at Washington Department of Fish and Wildlife, suggested that the
restoration of the Deschutes Estuary and the removal of the 5th Avenue Dam should be
compared to a natural disaster in terms of how to manage the sediment released. Smith,
who works closely with the ArmyCorps of Engineers on managing and upgrading the
sediment retention structures on the Toutle River, suggested that some of the methods
applied to the Toutle system could be applied to the restoration of the Deschutes Estuary.
“You have two perspectives on how you should manage this area. We [engineers/natural
resource managers] could trample down the grass three times a year, or should we create
a large event (natural disturbance) and let the channel equalize itself,” he explained.
Additionally, since the Deschutes Estuary has been altered since 1951 when the dam was
constructed, the sediment that will be dispersed into the system will provide new habitat
for aquatic and terrestrial species. However, some pro-lake supporters don’t quite
understand the importance of sediment and view it as a nuisance and not as an important
resource. “Sediment isn’t a bad thing. It’s part of the system,” Smith said.
40
Applying these theories and concepts to the 5th Avenue Dam may be difficult
because the dam is located at the mouth of the Deschutes River, which creates the
Deschutes Estuary. I have spent countless hours researching case studies trying to find a
scenario that is similar to the 5th Avenue Dam. I have discovered that the 5th Avenue Dam
is a unique situation. The majority of dams installed into a river system are located
several miles upstream in the upper estuary or river environments, not directly located in
the estuary habitat.
One example with some attributes similar to those at the 5th Avenue Dam is the
Chambers Creek Dam, located two miles from Steilacoom in Pierce County, Washington,
in the upper estuary environment. The Puget Sound Near Shore Ecosystem Restoration
Project has developed a restoration plan for this area, featuring two restoration options:
full or partial restoration. Because the Chambers Creek Dam is located in the upper
reaches of the estuary and in a former industrial, now urbanized, area, the restoration of
this creek and estuary requires extensive modifications to the surrounding infrastructure.
For full restoration of the estuary to occur, the plan calls for the removal of the dam, the
removal of culverts to daylight two streams, the relocation of a roadway, and the
extension of the railroad trestle to increase the width of the inlet to the Puget Sound.
Even though the Deschutes Estuary and the Chambers Bay Estuary feature similar
attributes to the restoration of the freshwater and tidal environments, the projects are very
different. The restoration of the Deschutes Estuary has been an ongoing topic of
conversation by local and state leaders; however, movement to restore this urban habitat
has not progressed. Restoration of this area has been halted largely due to the fact that the
41
dam creates Capitol Lake, an iconic public recreation area that is heavily used by the
public for recreation purposes.
Additionally, the removal of the 5th Avenue Dam and the restoration of this area is
relatively in its infancy. The Deschutes Estuary Feasibility Study’s primary goal was to
examine the four restoration scenarios and determine, which option was best suited for
the area. Beyond this study, there has not been any further development of a restoration
plan, such as the one drafted for the restoration of Chambers Bay. Chapter 6 below
develops a list of sediment management scenarios for the removal of the 5th Avenue
Dam.
42
CHAPTER 4
THE 5TH AVENUE DAM
4.1 DESCHUTES RIVER AND ESTUARY
The Deschutes Estuary Restoration Team (DERT) aims to remove the 5th Ave
Dam at the mouth of the Deschutes River in Olympia, Washington. The dam, built in
1951, creates a reflecting pond, known as Capitol Lake. The lake, maintained by
Washington Department of Enterprise Services, violates the Clean Water Act because the
dissolved oxygen levels are not meeting federal standards. As outlined above, the dam
also traps sediment and keeps it from reaching Budd Inlet to provide the necessary habitat
for aquatic species.
An environmental assessment completed by the Capitol Lake Adaptive
Management Plan (CLAMP) Steering Committee identified dam removal as the best
option for the Deschutes River watershed restoration. CLAMP commissioned a six-year
study of four scenarios for the future of the watershed. But progress to move forward in
the removal process has stopped. In 2010 the state government disbanded CLAMP and
movement to remove the 5th Avenue Dam turned from a state-funded project into a local
grassroots campaign.
4.2 CAPITOL LAKE
The Washington Department of Enterprise Services (DES) manages Capitol Lake,
and oversees operation and maintenance of the reservoir. Facilities staff monitors flow
conditions and conducts schedule water releases to minimize flood risks from high tides
and rain events, Carrie Martin, Assets Manager of Washington Department of Enterprise
Services, explained (Personal Communication, Feb. 2, 2015).
43
Additionally, DES manages the trapped sediment in Capitol Lake. To date,
Capitol Lake has been dredged twice since the dam’s installation. Historical records show
that from 1952 to 1974 an estimated 660,000 cubic feet of sediment had accumulated in
the lake. DES first dredged the lake in 1979. Approximately 250,000 cubic feet of
sediment was removed from the South and Middle Basins and was used to create
Tumwater Historical Park (Washington Department of Enterprise Services, 2006). The
second dredging event occurred in 1986, when approximately 57,000 cubic feet of
material was removed from the Middle Basin. After this dredging event, state officials
decided to form a committee to develop a restoration plan for maintenance dredging.
During this process the Capitol Lake Management Plan (CLAMP) Steering Committee
was formed in 1997. In 1999, CLAMP started to explore four restoration scenarios:
1) Lake/River/Wetland
2) Lake
3) Estuary
4) Lake/Estuary
Based upon the report’s findings, the best alternative was complete estuary
restoration. However, there has been no progress to move forward with the removal of
the 5th Avenue Dam and the restoration of the 260 acres of urban estuary habitat. “One of
the main issues with restoring the estuary is that people don’t know what the watershed is
suppose too look like,” Sue Patnude, Executive Director of the Deschutes Estuary
Restoration Team said (personal communications, Feb. 2, 2015). Since 2011, the
restoration team has been working to educate the public on the importance of restoring
this urban watershed. The group has also been creating a bipartisan relationship with the
44
Capitol Lake Improvement and Protection Association (CLIPA), a pro-lake non-profit
group located in downtown Olympia. Both organizations have been meeting on a
monthly basis to develop a community forum to present their arguments and help educate
the downtown community about various proposed methods on the way in which Capitol
Lake should be managed. Even though the estuary scenario was selected as the most
economic and effective option, state officials have not moved forward with the
development of a restoration plan. “To date, no permits or management alternatives have
been selected for Capitol Lake,” Martin explained (personal communication, Feb. 4,
2015).
4.3 SEDIMENT
As indicated earlier, dams trap sediment. The sediment stored behind the dam is
sculpted by the river’s hydrologic regime. The amount of sediment deposited behind the
barrier depends on flow of the river and the size of the dam. Large-scale hydropower
dams will have a greater amount of sediment deposited due to the large reservoirs
created. Small dams or valley dams have a lower amount of sediment deposition because
they are smaller in scale (Sawaske & Freyberg, 2012). Ecologists and environmentalists
worry about the amount of sediment stored behind the reservoir because the volume of
sediment movement could exceed the river’s established bankfull width5 (Gregory et al.,
2002). If bankfull width is exceeded, events like flooding or a large wave of sediment
can cause extensive damage to the downstream aquatic and terrestrial environments.
Researchers are trying to minimize the potential negative impacts by using models to
estimate the river and barrier stressor-response relationship. Researchers are starting to
5
Bankfull width is the point in a stream channel where water reaches a point in elevation on the riverbank
before it overflows into the floodplain.
45
track a reservoir’s storage capacity to determine the capacity loss that can occur from
upstream sedimentation events.
The development of tools, such as sedimentation models, allows researcher’s to
predict sedimentation rates and identify reservoirs that might be subject to rapid
sedimentation (Minear & Kondolf, 2009). Sedimentation models operate on small
temporal and spatial scales and utilize data, such as daily flows, reservoir bathymetry and
sediment grain size (Minear & Kondolf, 2009). However, these existing models fail to
include two important factors: the effects of trapping by upstream reservoirs, and changes
in the rate of sediment retention, which is know as the trap efficiency, every time as a
reservoir fills (Minear & Kondolf, 2009). Minear and Kondolf (2009) argue that
sedimentation models could benefit natural resource managers if they included a tool that
would allow researchers to expand the scope of the model to a regional level so that
reservoirs located within a large-scale watershed could be identified based on land
disturbances that are occurring throughout the watershed area. For example, if a landslide
would occur in the upper reaches of the watershed, releasing massive amounts of
sediment into the river, the downstream reservoir could capture this excess material. In an
event, utilizing a model to factor in these potential stressors could alert managers to apply
best management practices to manage the influx of the sediment. Three possible
countermeasures are the installation of upstream sediment traps, scheduling water
releases, or dredging of the material. Additionally, by increasing the model’s scope based
upon influx of regional sediment yields, there is the potential increase for sediment
management practices or the dam’s removal (Minear & Kondolf, 2009).
46
A river needs inputs and outputs of sediment into its system. The flux of sediment
movement throughout a river system is a natural process that occurs as a result of
disturbances, such as landslides or high water events. Natural disasters can recharge a
river’s system by distributing important nutrients like nitrogen and phosphorus and
dispensing fine and coarse substrate that create habitat for aquatic species. The removal
of a dam may create an ecological response similar to a natural disaster. Researchers are
trying to mimic these natural processes in the dam removal process.
The accumulation of sediment creates a number of issues that ecologists and
others must factor into their dam removal and restoration plans. It can be difficult to
predicting the rate, volume of erosion, and pattern of sediment transport (Hart, Johnson,
Bushaw-newton, et al., 2002). A method of doing so requires monitoring of the river’s
channel, observing and documenting parameters such as the grain size, level of cohesion,
channel slope, and flow. However, a lack of time and funding often prevents an effective
monitoring plan from being developed.
Poff and Allen (1997) argue that current management approaches fail to recognize
the fundamental scientific principle that the integrity of flowing water systems depend
largely on their natural dynamic character. A river’s natural flow regime plays a critical
role in sustaining native biodiversity and ecosystem integrity in rivers (Poff & Hart,
1997). This argument is the reason that watersheds require managers to develop protocols
that can not only incorporate the economic services of the natural environment but also
protect the ecosystems functions (Poff & Hart, 1997).
47
4.4 MONITORING
A comprehensive monitoring plan offers researchers the ability to develop a
sediment budget for a river system. According to Minear and Kondolf (2009), reservoir
sedimentation is a particularly serious problem in many regions with high sediment yield,
particularly in geologically active regions. Some experts, such as Lance Whitica,
Executive Director of the South Sound Salmon Enhancement Group, shares a similar
opinion about the Deschutes watershed: “The Deschutes is a geologically young river that
is trying to reach equilibrium.” Whitica and his team of restoration ecologists are
implementing several projects to reduce erosion from occurring throughout the river’s
basin. Installing structures along selected vertical sheer banks to prevent future erosion
from occurring, and promoting riparian corridors are two examples of projects that
Whitica and his group are implementing.
Sediment budgets for downstream reaches may need to be reconsidered
depending on whether incision or floodplain development is expected to occur in the
river’s channel (Sawaske & Freyberg, 2012). Channel width and depth upstream from the
dam will provide researchers with an existing channel form and can allow researchers to
predict how the channel may establish itself once the reservoir is removed. Field
technicians usually monitor longitudinal profiles and cross-section surveys to determine a
river’s sediment budget. The monitoring of these variables allows scientists to forecast
potential channel development. Even so, the consequences in the downstream reaches
can be hard to predict in sediment transport models, because of the high degree of
uncertainty of the extent of sediment deposition, erosion, and flooding impacts in the
48
downstream channel. Also, the recovery response in a river’s channel is hard to predict
and is highly dependent on the purpose the dam served in the ecosystem.
Stream flow gauges measure a river’s flow, timing, variability, and natural flow
regime. These data should be taken over time to reflect extreme high and low flows, or a
range of flows, often expressed as a daily average discharge. The lack of long-term
stream flow data can be supplemented statistically from gauged streams in the same
geographic area (Poff & Hart, 1997).
4.5 SEDIMENT COMPOSTION
The sediment particulate size is a function in the river system that provides habitat
in the downstream reaches. The larger sized gravel and cobbles are stopped by dams,
which negatively impacts aquatic habitat that might rely on those sediments, especially in
coastal management zones. Dams close to estuary habitats change water flow and
temperature, and decrease the flux of fine sediment, which creates a critical habitat for
near shore species. The Elwha River Dam Removal project has provided 160,000 m3 in
sediment to the mouth of the river, which has provided a vital habitat for clam beds
(Gregory et al., 2002). The use of one-dimensional numerical simulations allows
researchers to predict a river’s reach-averaged channel response, the most practical tool
for determining the evolution of non-cohesive reservoir sediment deposits (Downs et al,
2009). One-dimensional sediment transport models are best utilized over large spatial and
temporal scales. Field samples featured in the Downs et al (2009) journal article
“Managing Reservoir Sediment Release in Dam Removal Projects: An Approach
Informed By Physical and Numerical Modeling of Non-cohesive Sediment” reveal that
subsurface material taken from rivers that have not experienced a large scale influx of
49
fine sediment. The report has shown that gravel-bedded rivers often contain a substantial
fraction of fine sediment as a background condition. The journal article also states that in
rivers composed of predominately sand-sized sediment, the pulse will only infiltrate
immobile gravel deposits to a few gravel diameters in depth. Furthermore, “the rapid
release and transporting of fine sediment pulses over a gravel deposit, rather than the
transportation of fine sediment over a gravel deposit, rather than transporting the same
volume of sediment at a slower rate over an extended period will not result in increased
infiltration and may even limit infiltration” (Downs et al, 2009).
Dams capture all but the finest sediments moving down a river (Poff & Hart,
1997). When dam operators decide to conduct scheduled water releases, the sedimentdepleted water released can erode finer sediments from the downstream reach. This can
leave a coarse downstream streambed, which reduces habitat for many aquatic species.
Furthermore, channels may erode, or down-cut and trigger rejuvenation of tributaries,
which themselves can become subjected to erosion (Poff & Hart, 1997). The increased
release of fine sediment can be problematic in the newly created pilot channels or
tributaries because this new material can be re-deposited in between coarse material
located throughout the streambed. During high and low flow events; the distribution of
fine or coarse sediment can create stresses for aquatic species by reducing habitat
creation or function. For species to survive, there is a range of flows that are necessary to
scour and revitalize gravel beds, to import woody debris, and to organize matter from the
floodplain and provide access to productive riparian wetlands. Additionally, the interannual variation in these flow peaks is critical for maintaining channel and riparian
dynamics. Poff and Hart (1997) state that virtually all rivers are inherently variable, and
50
that this variability is critical to ecosystem function and native biodiversity. The greatest
challenge that faces natural resource managers and the restoration of rivers is that they
are also used to satisfying human needs. Paradoxically, for restoration to occur, it will be
depend upon the extent of human intervention and flow alteration affecting a particular
river.
Another issue the researchers must consider is the presence or absence of heavy
metals in the reservoir and stream channel. Toxins can be absorbed physically through
sediment particles or through biota attached to the sediment. If a dam is located in an
urban watershed, there could be high concentrations of toxins due to anthropogenic
influences, such as nonpoint source pollution. The build up of these toxins in the
reservoir and behind the dam is a concern because once the impoundment is breached
these substances will be dispersed into the downstream channel and can cause adverse
impacts in water quality and wildlife habitat standards. Stanley and Doyle (2002) state
that there is a lack of studies that examine the removal of contaminated soils from a dam
reservoir. The most noted case study seen in literature is that of the removal of the Fort
Edwards Dam on the Hudson River in New York. In 1973, a environmental contracting
firm demolished this barrier, releasing sediment that contained high levels of oil and
polychlorinated biphenyl’s (PCBs). The sediment wave left a long legacy of negative
feedback loops in the river system and required extensive cleanup efforts and
contaminated the food web for several years (Stanley & Doyle, 2003).
51
4.6 MODELING
In Chapter 3 of this thesis I discuss the evolution of dam removal and highlight
how the application of various restoration techniques are evolving since large-scale dam
removal projects, such as the Elwha and Glines Canyon Dam on the Elwha River, and the
Marmot Dam on the White Salmon River. According to Poff and Hart (2002), there has
not been a quantitative geomorphic study that has continued long enough to document the
establishment of an equilibrium channel prior to dam removal. Both researchers
advocate in favor of the development of a theoretical framework that will help ecologists
predict sediment movement after removal. Many researchers are trying to solve this
challenge through the use of computer models and statistical analysis (Peck, Mullen,
Moore, & Rumschlag, 2007).
Recent attempts to understand river dynamics have ranged from simple analysis
of pre-dam channel geometry to data intensive, three-dimensional numerical models
(Sawaske & Freyberg, 2012). Sawaske and Freyberg (2012) are pioneering this field by
developing a tool to predict the rate and volume of sediment deposited by analyzing
sediment, discharge, deposit, removal timeline, channel, and watershed data. Their model
examines the evolution of the deposited sediments once the dam is breached and the
system begins to restore itself. Both researchers saw a lack in the comparative analysis of
post-removal monitoring data, particularly in the evolution of sediment filled reservoirs
following dam removal (Sawaske & Freyberg, 2012). Sawaske and Freyberg claim that
by comparing amount and quality of post-removal monitoring data, there is a
considerable amount of variation in studies. The researchers plan to fill in the gaps by
52
developing this modeling tool to estimate ecological response factors for river systems
that span over temporal and spatial scales (Sawaske & Freyberg, 2012).
Models are limited when it comes to being able to describe site-specific aspects of
the ecosystem process if a comprehensive set of monitoring data is not available for use.
Models cannot predict the full range of potential stressor-responses. To date, there has
not been one specific model that is able to incorporate all the elements of dam removal.
53
CHAPTER 5
RESEARCH ANALYSIS
After reviewing the literature and case studies, and conducting interviews, I
concluded that two parameters play an important role in the decision to manage sediment
transport in dam removal projects: the volume of sediment that would be released, and
the application of methods that are used to manage sediment transport. The recently
published literature documenting the habitat modifications that are occurring throughout
the Elwha River watershed after dam removal has provided pivotal findings that have
supported the claims developed throughout my research process. The Elwha and Glines
Canyon Dam Removal project has provided unprecedented information on sediment
transport in freshwater and tidal environments. The use of models and field data has
allowed the researchers to document the changes occurring throughout the watershed
area. Figure 5.1 shows some of the parameters that field crews monitored on the Elwha
River Near Shore Zone.
54
Figure 5.1.Map showing locations of beach topographic and near shore bathymetric data from surveys completed in September 2013.
Also shown are locations of instrumented tripods E and W, geodetic control monuments for GPS base stations (green triangles), sound
velocity profiles (white squares) used in speed of sound corrections to bathymetric soundings, and biweekly beach topography and
grain size profiles (blue circles). (Gelfenbaum et al, 2015).
55
The Delft3D model simulated the water motion from tides, waves, wind, and
buoyancy effects (Gelfenbaum et al, 2015). Additional modeling software could predict
the effects of wave motion on the sediment and near shore zone dynamics. The
Simulating Waves Nearshore (SWAN) model, for example, which simulates wave
propagation in time and space by solving the spectral action balance equation, was used
to simulate the interaction of the Pacific Ocean and the near shore zone development.
Data was also incorporated into Landscape Evolution Models (LEMs) to determine
where sediment was depositing, and to document how the elevation was changing after
removal. Figure 5.2 depicts of pre-dam removal morphology and elevation actual
changes at the river mouth and near-shore-zones of the Elwha.
56
Figure 5.2. Map showing the geomorphic evolution of the active delta before and during
the first two years of dam removal. [A] August 2011, [B] May 2012, [C] August 2012,
[D] March 2013, [E] September 2013. (Gelfenbaum et al, 2015).
57
Unlike the Elwha project, the 2009 Deschutes Estuary Feasibility Study used only
the Delft3D model to determine how sediment would be distributed throughout Budd
Inlet. This model provided results that researchers and policymakers used to determine
that estuary restoration could be an option. But based the findings of this thesis research,
this model should be coupled with other fluvial and wave models to develop further
analysis of the way in which sediment could be transported and the development of
sediment deflection design scenarios. Figure 5.3 is a map produced from the 2009
Deschutes Estuary Feasibility Study that predicts where sediment will deposit and what
areas might erode.
58
Figure 5.3. a. Erosion and deposition for the restored estuary 10 years after dam removal.
Blues indicate erosion, and reds show deposition. Middle Basin experiences the most
widespread erosion, while North Basin and the region outside of the estuary accumulate
sediment. b Volume change through different segments of the estuary. (George,
Gelfenbaum, & Stevens, 2012).
59
The information compiled on the Deschutes Estuary can be applied to various
models to help guide natural resource managers on sediment transport. However, as the
restoration efforts move forward, project leaders there should factor in how the volume of
sediment released will interact with the tidal and fresheater dyanmics in the watershed
area. Figure 5.4 is a map of the potential image of how the sediment might be dispersed
in the estuary. However, estuary evolution will be dependent upon the length of time
allowed for the natural freshwater and tidal processes to occur to establish equilibrum.
Thus, the methods used to remove, manage, and restore the estuary are all interconnected
and should be modeled and conceptualized that way.
60
Figure 5.4 is a map of the potential image of how the sediment might be dispersed in the
Deshutes Estuary and Budd Inlet. (Geleysnse et al, 2010).
61
CHAPTER 6
DISCUSSION
I have developed a list of sediment management scenarios that can be used in the
removal of the 5th Avenue Dam located in Olympia, Washington. I base my
recommendations on the literature I have reviewed and interviews I have conducted with
experts in the field. After a site visit to the 5th Avenue Dam with Padraic Smith,
Environmental Engineer, Restoration Division, Habitat Program of Washington State
Department of Fish and Wildlife, we discussed several sediment management scenarios
for the removal of the 5th Avenue Dam.
The sediment management scenarios are grouped into three approaches: direct,
indirect, and a hybrid approach. Each of these management scenarios features several
different restoration techniques that can be used to mange the sediment. The direct
approach would require direct management of the sediment with the use of equipment.
These scenarios features two methods:
1) Dredging of the basin, the technique presented in the 2006 Deschutes Estuary
Feasibility Study; and
2) Blow and go of the dam, a method used in dam removal cases studies, where
the dam is completely removed all at once. This scenario utilizes the river and
estuary physical processes to transport the sediment and allow equilibrium to
be reached naturally.
The second group employs an indirect approach, which would allow for the
application of restoration techniques that guide where the sediment is transported.
62
1) Install sill basins to allow the tide to use its natural hydrodynamics to capture the sediment in the selected basins
Figure 6.1 Sill basins are a method that can be used to help capture sediment. These structures can be located
throughout the Deschutes River and river right behind the railroad tracks to help build up the sediment.
(Mount St. Helens National Park)
63
2) Create low, medium, and high tidal marshes to trap sediment;
Figure 6.2 Tidal marshes can be created from the sediment released in the removal of the 5th
Avenue Dam, Olympia, Washington. (Institute of Applied Ecology)
64
3) Lockdown the sediment in areas to allow erosion to occur in selected sections of the basin
as equilibrium is reached; and
Figure 6.3 Matting can prevent erosion from occurring in projected
locations predicted from the 2006 Hydrodynamic and Sedimentation
Report, part of the 2009 Deschutes Estuary Feasibility Study.
(Ridges of Restoration)
65
4) Install pile dikes to form pilot channels throughout the lower basin to create sinuosity
for the sediment to be captured by the dikes.
Figure 6.4 An example of a pile dike that can be placed throughout the
Deschutes River to drop out sediment.
(Mount St. Helens National Park)
66
The final group takes a hybrid approach. This approach can occur in multiple phases to manage the sediment.
1) Make modifications to the ogee structure in the dam and change how the dam is operated. By altering the water level and
gate release times, sediment can slowly be introduced into the estuary. Additionally, the dam’s existing ogee structure,
which is an artificial slope that directs sediment and flow into Budd Inlet, can be modified to redirect the sediment.
Figure 6.5 Modifying the ogee weir would allow the sediment transport into the estuary by making changes
to this structure and dam operations. (Washington Department of Enterprise)
67
2) Install a structure (i.e. culvert) on river right to divert flow into the basin behind the railroad tracks. (See arrow in Figure
6.6 for further illustration). As the basin fills up and reaches its natural capacity, modifications to the dam and its
operations can occur during this time. By using both of these techniques, the sediment will be managed and distributed
throughout Budd Inlet, and the estuary.
Figure 6.6 Divert sediment by installing a culvert and creating a channel to allow flow to enter basin behind railroad track.
Allow the basin to fill and while this process is occurring, modify the dam’s operations and the structure
so that sediment can be deflected from the Port of Olympia. (Aerial Images NW)
68
In addition to these sediment management scenarios, I am recommending that
state officials develop a project team that focuses solely on sediment transport and an
exploration of various restoration efforts that are cost effective for the removal of the 5th
Avenue Dam. State officials will have to secure all the proper permits to begin
construction on these efforts. Most importantly, the installations of these structures are
not an exact science and will require state agencies to have an adaptive management plan
that can be implemented when changes occur due to natural processes.
Furthermore, the Washington State Legislature is looking at the long-term
maintenance of Capitol Lake during the writing of this thesis. In the proposed state
budget for 2015-2017, the state listed the following:
Capitol Lake Long-term Management Planning (30000740)
The appropriation in this section is subject to the following conditions and limitations:
(1) The appropriation is provided solely for the development of a conceptual plan for the
future of Capitol Lake and the Deschutes Estuary that is designed to meet multiple
objectives, including achieving broad community support and preliminary commitments
from state and local funding sources to share costs. The appropriation must be used to
develop a financially feasible conceptual plan, including general cost estimates, which
incorporate, and achieve compromise between key features of the most widely discussed
concepts.
(2) The plan must address these multiple objectives:
(a) Some improvement of estuary functions and fish habitat; (b) Retention of
portions of the northern portion of the lake, in accordance with the historic features of the
Capitol campus design; (c) Improvement of water quality of the lake sufficient to expand
69
water-related recreation opportunities, which improvement strategies shall take into
account information gathered to date through the department of ecology's Deschutes river
TMDL study, storm water runoff from Interstate 5 and State Route No. 101, and from
Olympia and Tumwater and Thurston county sources; (d) A conceptual plan for shared
financing of the plan between state and local agencies, based on both benefits received
and liabilities contributed, potentially using the state's lake management district
legislation as a model, together with an assessment of whether federal funds might be
available; and (e) A conceptual plan for shared governance.
(3) Public input must be sought as the plan is developed.
(4) The plan must be submitted to the state capitol committee and appropriate committees
of the legislature by November 1, 2017.
Even though a feasibility study was completed in 2006, there has not been any
progress on developing a restoration plan or securing funding. The potential allocation of
funds is the first movement towards a long-term management plan. However, the
language used in the state budget is vague and their interpretation will be based upon the
goals and objectives set forth by the legislature and the public. The sediment management
scenarios that I developed are based on improving habitat conditions. By examining the
restoration of this area from this perspective there are many ecosystem functions that
could be utilized by various aquatic and terrestrial species. Some functions that could be
restored include:
Restore tidal wetlands and estuary habitat
Improved water flow and quality
70
Improved habitat connectivity between the near-shore zone, freshwater
environment, and adjacent areas
Restored natural formation of tidal channels
Unrestricted flow of freshwater sources
Accumulation and retention of organic material from plants and aquatic
species
Unrestricted movement and migration of fish and wildlife
71
CHAPTER 7
CONCLUSION
My research on the removal of the 5th Avenue Dam in Olympia, Washington has
led me to conclude that estuary restoration is possible based on the modeling results
produced during the 2009 Deschutes Estuary Feasibility Study. Managing the transport of
sediment in dam removal is a difficult task, but as research literature has proven, models
have provided some insight on how rivers and estuary systems might function after
removal. In the case of the Deschutes Estuary, the researchers’ main focus was to
examine the four possible restoration scenarios. The scope of their work was limited;
therefore they did not go into explicit detail in managing the volume of sediment released
and what methods should be used in order to manage erosion and deposition.
The research I completed for this project has illustrated the paradigm shift
occurring in river restoration and watershed management efforts. The increase in dam
removal projects has provided a foundation for natural resource managers to consider
dam removal as a restoration measure. However, my research has led me to believe that
in order for removal to occur, there are many factors that must be considered beyond the
impacts to the physical environment. In the case of the 5th Avenue Dam in Olympia,
Washington, the structure is located in the heart of downtown, and much of the city’s
infrastructure is dependent upon the dam. As I highlighted in Chapter 3, one of the major
uncertainties in dam removal is the ability to visualize the restored environment.
Managers have started to incorporate technology, such as the use of one- or threedimensional models to predict the volume of sediment released and the amount of time it
will take for the river or estuary system to reach equilibrium. Yet it is still hard for
72
citizens and policymakers to visualize what the final outcome will look like. I believe this
is one of the main struggles in dam removal projects. The science behind removing dams
is growing, but the public’s perspective on restoration efforts can impede the restoration
process.
However, the practice of river restoration is changing as efforts to monitor dam
removal projects continue to emerge and the changes are documented and published. The
Elwha and Glines Canyon Dam removal projects demonstrate how dam removal can
achieve full-scale restoration. Efforts to accomplish the removal of these two dams
started in 1986, with complete restoration occurring in 2013. Based on the years it took to
remove the Elwha River dams, it will most likely take years to complete the restoration of
the Deschutes Estuary. The grassroots efforts that the Deschutes Estuary Restoration
Team is completing can help guide restoration measures to transpire by educating the
community on the positive feedback loops that will occur by having a restored estuary.
Imagine that twenty years after the 5th Avenue Dam has been removed, the
estuary reaches equilibrium. Kayakers and stand-up paddle boarders recreate in the newly
created habitat. Salmon, seals, and shorebirds can be seen around the estuary while you
walk your dog. During low tide, the newly created tidal marshes and mud flats show the
pilot channels that have formed since the dam has been removed.
And at high tide, the Capitol Building is reflected in the waters of the restored
estuary.
73
Bibliography
Babbitt, B. (2002). What goes up, may come down. BioScience, 52(8), 656–658.
Bednarek, A. T. (2001). Undamming rivers: A review of the ecological impacts of dam
removal. Environmental Management, 27(6), 803–814.
Chang, H. H. (2008). Case study of fluvial modeling of river responses to dam removal.
Journal of Hydraulic Engineering, 134(3), 295–302.
Cui Y, Braudrick, C., Dietrich, W.E. (2006). Dam Removal Express Assessment Models
(DREAM). Journal of Hydraulic Research 44:308-323.
Downs, P. W., Yantao, C., Wooster, J. K., Dusterhoff, S. R., Booth, D. B., Dietrich, W.
E., & Sklar, L. S. (2009). Managing reservoir sediment release in dam removal
projects: An approach informed by physical and numerical modeling of noncohesive sediment. International Journal of River Basin Management, 7(4), 433452.
Doyle, M. W., Harbor, J. M., & Stanley, E. H. (2003). Toward policies and decisionmaking for dam removal. Environmental Management, 31(4), 0453-0465.
Duda, J.J., Warrick, J.A., & Magirl, C.S., (2011). Elwha River Dam Removal-Rebirth of
river: U.S. Geological Survey Fact Sheet 2011-3097, 4.
Gelfenbaum, G., Stevens, A. W., Miller, I., Warrick, J. A., Ogston, A. S., & Eidam, E.
(2015). Large-scale dam removal on the Elwha River, Washington, USA: Coastal
geomorphic change. Geomorphology (Article in Press).
Geleynse, N., Storms, J. E., Walstra, D., R., Jagers, H. R., Wang, Z. B., & Stive, M. J.
(2010) Controls on river delta formation; insights from numerical modeling.
Earth and Planetary Science Letters, 202(1-2): 217-226.
George, D. A., Gelfenbaum, G., & Stevens, A. (2012). Modeling the hydrodynamic and
morphologic response of an estuary restoration. Estuaries and Coast, 35(6):
1510-1529.
Gottgens, J. F., & Evans, J. E. (2007). Dam removals and river channel changes in
Northern Ohio: Implications for Lake Erie sediment budgets and water quality.
Journal of Great Lakes Research, 33(2), 87–89.
Gowan, C., Stephenson, K., Shabman, L. (2006). The role of ecosystem valuation in
environmental decision-making: Hydropower relicensing and dam removal on the
Elwha River. Ecological Economics. 56(4). 508-523.
74
Graf, W. (1999). Dam nation: A geographic census of American dams and their largescale hydrologic impacts. Water Resources Research, 35(4), 1305–1311.
Granata, T., Cheng, F., & Nechvatal, M. (2008). Discharge and suspended sediment
transport during deconstruction of a low-head dam. Journal of Hydraulic
Engineering, 134(5), 652–657.
Gregory, S., Li, H., & Li, J. (2002). The conceptual basis for ecological responses to dam
removal. BioScience, 52(8), 713–723.
Grossman, E. (2002). Watershed: The undamming of America. New York: Counterpoint.
Hart, D. D., Johnson, T. E., Bushaw-Newton, K. L., Horwitz, R. J., Bednarek, A. T.,
Charles, D. F., Velinsky, D. J. (2002). Dam removal: Challenges and
opportunities for ecological research and river restoration. BioScience, 52(8),
669–682.
Konrad, C. P. (2009). Simulating the recovery of suspended sediment transport and riverbed stability in response to dam removal on the Elwha River, Washington.
Ecological Engineering, 35(7), 1104–1115.
Minear, J. T., & Kondolf, G. M. (2009). Estimating reservoir sedimentation rates at large
spatial and temporal scales: A case study of California. Water Resources
Research, 45 (12), 302-312.
Mussman, E. K., Zabowski, D., & Acker, S. A. (2008). Predicting secondary reservoir
sediment erosion and stabilization following dam removal. Northwest Science,
82(1), 236–245.
Peck, J. A., Mullen, A., Moore, A., & Rumschlag, J. H. (2007). The legacy sediment
record within the Munroe Falls dam pool, Cuyahoga River, Summit County,
Ohio. Journal of Great Lakes Research, 33(2), 127–141.
Pizzuto, J. (2002). Effects of dam removal on river form and process. BioScience, 52(8),
683–691.
Poff, N. L., & Allan, J. D. (1997). The natural flow regime. BioScience, 47(11), 769–
784.
Poff, N. L., & Hart, D. D. (2002). How dams vary and why it matters for the emerging
science of dam removal. BioScience, 52(8), 659–668.
Pohl, M. M. (2002). Bringing down our dams: Trends in america dam removal
rationales. Journal of the American Water Resources Association, 38, 15111519.
75
Roberts, S. J., Gottgens, J. F., Spongberg, A. L., Evans, J. E., & Levine, N. S. (2007).
Assessing potential removal of low-head dams in urban settings: An example
from the Ottawa River, NW Ohio. Environmental Management, 39(1), 113–124.
Sawaske, S. R., & Freyberg, D. L. (2012). A comparison of past small dam removals in
highly sediment-impacted systems in the U.S. Geomorphology, 151–152, 50–58.
Seattle Times Staff (2013, November 17). Elwha River sees largest run of Chinook in
decades Seattle Times Retrieved from http://www.seattletimes.com
Schmitz, D., Blank, M., Ammondt, S., & Patten, D. T. (2009). Using historic aerial
photography and paleohydrologic techniques to assess long-term ecological
response to two Montana dam removals. Journal Of Environmental Management,
(90), 237-248.
Shuman, J. R. (1995). Environmental considerations for assessing dam removal
alternatives for river restoration. Regulated Rivers: Research & Management,
11(34), 249–261.
Stanley, E. H., & Doyle, M. W. (2002). A geomorphic perspective on nutrient retention
following dam removal. BioScience, 52(8), 693–701.
Stanley, E. H., & Doyle, M. W. (2003). Trading off: the ecological effects of dam
removal. Frontiers In Ecology & The Environment, 1(1), 15-22.
Winter, D. D., & Cain, P. (2008) Making the case for ecosystem restoration by dam
removal in the Elwha River, Washington. Northwest Science, 82, 13-28.
76
