-
Title
-
Eng
Forage Fish Spawning in the Elwha Nearshore: Ecological Form and Function in a Changing Environment
-
Date
-
2014
-
Creator
-
Eng
Wefferling, Leif T
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
FORAGE FISH SPAWNING IN THE ELWHA NEARSHORE:
ECOLOGICAL FORM AND FUNCTION IN A CHANGING ENVIRONMENT
by
Leif T. Wefferling
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
September 2014
©2014 by Leif Wefferling. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Leif T. Wefferling
has been approved for
The Evergreen State College
By
________________________
Carri J. LeRoy, Ph.D.
Member of the Faculty
________________________
Date
ABSTRACT
Forage Fish Spawning in the Elwha Nearshore:
Ecological Form and Function in a Changing Environment
Leif Wefferling
Intertidal beaches within the Elwha nearshore are documented habitat for forage
fish migration and spawning. Sediment processes of the Elwha drift cell, critical
for forage fish spawning habitat, were historically altered by armoring of the
shoreline, lower river alterations, and the in-river Elwha and Glines Canyon dams.
The recent removal of these two dams, and the consequent release and transport
of upwards of 2.5 x 106 m3 of fluvial sediment to the Elwha nearshore, has begun
a partial restoration of sediment processes within the drift cell and is changing the
substrate characteristics of its beaches, potentially restoring nearshore function for
forage fish spawning. We conducted egg surveys for two species of forage fish,
surf smelt (Hypomesus pretiosus) and sand lance (Ammodytes hexapterus), over
four years, including two years before dam removal (2007-8) and two years
during the dam removal process (2012-13). Samples were collected from
geomorphic habitat types (GMHTs) (embayment, bluffs, and spit) within the
impaired Elwha drift cell and from comparative, intact Dungeness and Crescent
Bay drift cells. In order to assess nearshore function, we compared spawning
activity across impaired/intact drift cells, GMHT, and before/during dam removal
time periods. While no sand lance eggs were found during the course of this
study, our surf smelt results show that, overall, the intact Dungeness drift cell
supports more-robust spawning activity than the impaired Elwha drift cell and
that egg productivity did not differ significantly between the two time periods.
We also conclude that egg abundance is highly variable across GMHT, with the
greatest abundance in the intact bluffs site followed by the impaired embayment,
where spawning habitat appears to have expanded during the dam removal period.
Spit sites did not support any spawning activity. Understanding the implications
of dam removal to the ecological functioning of the nearshore is important for full
ecosystem restoration of the Elwha system, where shoreline armoring will remain
an outstanding and long-term restoration issue.
Table of Contents
List of Figures ..................................................................................................... v
List of Tables...................................................................................................... vi
Acknowledgements ......................................................................................... vii
Chapter 1: Introduction and Literature Review .......................................... 1
INTRODUCTION ...................................................................................... 1
LITERATURE REVIEW ........................................................................... 3
The Puget Sound nearshore ........................................................................ 4
Beach formation and sediment processes of the nearshore ........................ 7
Coastal Feeder Bluffs .................................................................................. 8
Forage fish ................................................................................................ 10
Surf smelt ...................................................................................... 12
Pacific sand lance ......................................................................... 13
Spawning habitat ....................................................................................... 15
Shoreline armoring.................................................................................... 20
The Elwha River and nearshore ................................................................ 23
Management issues and nearshore restoration .......................................... 26
Chapter 2: Article Manuscript: Forage Fish Spawning in the Elwha
Nearshore: Restoring Ecological Form and Function in a Changing
Environment ...................................................................................................... 30
ABSTRACT .............................................................................................. 30
INTRODUCTION .................................................................................... 31
METHODS ............................................................................................... 38
Study Sites ..................................................................................... 39
Surf Smelt and Sand Lance Egg Sampling .................................... 40
Statistical Analysis ........................................................................ 42
RESULTS ................................................................................................. 43
Intact vs. Impaired Drift Cells ...................................................... 46
iv
Before vs. During Dam Removal .................................................. 48
Geomorphic Habitat Types ........................................................... 51
DISCUSSION ........................................................................................... 53
Chapter 3: Restoration and Management of the Elwha Nearshore
............................................................................................................................... 62
Introduction ............................................................................................... 62
Nearshore conditions of the central Strait of Juan de Fuca ...................... 65
The Elwha Drift Cell ................................................................................. 67
Dam removal and the potential for nearshore restoration ......................... 71
Port Angeles landfill and other nearshore management issues ................. 73
Elwha nearshore restoration: questions and actions ................................. 76
The Elwha as cautionary tale .................................................................... 80
Conclusions ............................................................................................... 82
Bibliography ...................................................................................................... 84
Appendix .......................................................................................................... 100
v
List of Figures
Figure 1. Changes in Freshwater Bay beach substrate. ........................................ 37
Figure 2. Study sites for surf smelt and sand lance egg surveys........................... 39
Figure 3. Surf smelt egg abundance in impaired and intact drift cells. ................ 47
Figure 4. Surf smelt egg abundance in impaired and intact drift cells a) before and
b) during dam removal. ........................................................................ 47
Figure 5. Surf smelt egg abundance before and during dam removal. ................. 48
Figure 6. Surf smelt egg abundance before and during dam removal in the a)
impaired and b) intact drift cells. ......................................................... 49
Figure 7. Surf smelt egg abundance in Freshwater Bay. ...................................... 50
Figure 8. Freshwater Bay samples containing surf smelt eggs before and during
dam removal. ........................................................................................ 50
Figure 9. Surf smelt egg abundance by geomorphic habitat type (GMHTs). ....... 51
Figure 10. Surf smelt egg abundance by geomorphic habitat type (GMHT) within
a) impaired and b) intact drift cell treatments. ..................................... 52
Figure 11. Looking west along the base of the Elwha bluffs towards the City of
Port Angeles landfill. ........................................................................... 74
Figure 12. City of Port Angeles installing landfill sea wall in 2005. ................... 78
Figure 13. All surf smelt sample locations in the Elwha and Dungeness drift cells.
............................................................................................................ 100
Figure 14. All surf smelt samples containing eggs. ............................................ 100
Figure 15. Surf smelt survey results for all samples collected in the impaired
Elwha drift cell. .................................................................................. 101
Figure 16. Surf smelt survey results for all samples collected in the intact
Dungeness drift cell............................................................................ 101
Figure 17. Surf smelt survey results in the intact Dungeness drift cell before dam
removal (2007-2008).......................................................................... 102
Figure 18. Surf smelt survey results in the intact Dungeness drift cell during dam
removal (20012-2013)........................................................................ 102
Figure 19. Surf smelt survey results in Freshwater Bay (impaired Elwha drift cell)
before dam removal (2007-2008). ..................................................... 103
Figure 20. Surf smelt survey results in Freshwater Bay (impaired Elwha drift cell)
during dam removal (2012-2013). ..................................................... 103
v
List of Tables
Table 1. Sample areas designated by geomorphic habitat type (GMHT) and
location within an impaired or intact drift cell. ...................................... 40
Table 2. Sampling schedule for sand lance eggs. ................................................. 43
Table 3. Sampling schedule for surf smelt eggs. .................................................. 44
Table 4. Surf smelt sampling results by study site and phase of dam removal. ... 45
Table 5. Table of single dead surf smelt eggs not included in analysis. ............... 45
Table 6. Surf smelt data consolidated for statistical analysis. .............................. 46
vi
Acknowledgements
Dr. Carri J. LeRoy, TESC
Anne Shaffer, Coastal Watershed Institute (CWI)
Nicole Harris, CWI
David Parks, WDNR
Dan Penttila, Salish Sea Biological
vii
Chapter 1: Introduction and Literature Review
INTRODUCTION
Nearshore environments are complex and productive ecosystems that provide
refuge, feeding, migratory, nursery, and spawning habitat to a diverse range of
species. Because they occur at the interface between marine and terrestrial
ecosystems, nearshore ecosystems are vulnerable to a wide range of
anthropogenic activities (Fresh et al., 2011; Simenstad et al., 2011). The Elwha
nearshore, located within the Strait of Juan de Fuca and along the north coast of
Washington State’s Olympic Peninsula, provides an example of an environment
with historically impaired ecological processes that is currently undergoing
massive and potentially restorative change. For over one hundred years, two dams
on the Elwha River impounded fluvial sediment that would have otherwise been
transported downstream to replenish nearshore beaches (Czuba et al., 2011a).
Kilometers of the shoreline have also been armored, impairing the process of
coastal bluff erosion that is responsible for most of the sediment input to the
nearshore system (Finlayson, 2006). The result of these two developments has
been severe sediment starvation of the nearshore, causing ongoing erosion of the
coastline (Warrick et al., 2009) as well as coarsening of nearshore beach substrate
to beyond the range necessary for successful forage fish spawning.
In 2011, a project to completely remove both dams on the Elwha River
began. This will be the largest dam removal and sediment release in U.S. history
(Draut & Ritchie, 2013). The project is now nearly complete and sediment has
1
begun to be mobilized out of the former reservoirs and transported downstream
(Kaminsky et al., 2014). This massive pulse of sediment delivered by the mouth
of the Elwha River is transforming the sediment processes governing the creation
and maintenance of nearshore habitat and has the potential to reestablish forage
fish spawning habitat; however, significant impediments to full ecosystem
restoration remain. Armored stretches of the shoreline will remain armored long
after the initial pulse of Elwha River sediment has subsided and the river resumes
its normal, much smaller, annual contribution of fluvial sediment to the nearshore
system. Armoring will continue to impair feeder bluff erosion and sediment
recruitment to nearshore beaches. While little is known about the exact quantity,
timing, location, grain-size, or duration of sediment delivery, questions are raised
as to which restorative actions will best optimize sediment arrival and retain it on
nearshore beaches (MacDonald & Harris, 2013). As the sediment release is
projected to last only 7-10 years, the time to act is now.
This thesis is comprised of three chapters. The chapters are interrelated
and supportive of the others, but each chapter can also stand alone as a separate
treatment with its own focus. The first chapter is a literature review, focusing on
the Puget Sound nearshore environment and the sediment processes that form and
sustain the spawning habitat of surf smelt and sand lance, two species of forage
fish that are crucially important to the marine food web. The life histories and
particular sediment composition and grain-size requirements of these forage fish
are described in this review. The Elwha nearshore is also introduced in the first
chapter, an area with impaired sediment processes that is located in the central
2
Strait of Juan de Fuca and within which the experimental portion of this thesis
was conducted. The second chapter describes this experiment and has been
formatted as a manuscript for publication in a journal of coastal research. It
includes an abstract, introduction, and methods section, reports our results and
conclusions, and includes a discussion of the findings of this study. The third and
final chapter of this thesis examines the interdisciplinary nature of nearshore
management issues, focusing on which actions might be taken to optimize both
short and long-term ecosystem restoration within the Elwha nearshore and its
implications for adjacent nearshore areas.
LITERATURE REVIEW
This literature review is composed of several sections, organized by topic. The
first section, the Puget Sound nearshore, defines the nearshore environment
within the larger context of the Puget Sound Basin, including the Strait of Juan de
Fuca. The next section, beach formation and sediment processes of the nearshore,
examines the sediment processes responsible for supplying, forming, and
maintaining its nearshore beach habitat. The important role of coastal feeder
bluffs is described in the following section because it is the erosion of these bluffs
that contributes the majority of sediment to the nearshore system. After having
described some of the physical and geomorphological conditions and processes of
the nearshore, the next section, forage fish, describes the important biological role
these fish play in the marine food web. The following two sections, surf smelt and
Pacific sand lace, relate some of what is understood about the life histories of
3
these two important forage fish. A lengthy examination of their specific habitat
requirements is described in the subsequent section, spawning habitat. Some of
the various ways in which shoreline armoring can degrade spawning habitat is
reviewed next, followed by a closer examination of the Elwha River and
nearshore, the specifics of its historical impairment, and how forage fish
spawning in this area has been impacted. The final section, management issues
and nearshore restoration, considers the major changes currently underway in the
Elwha nearshore sediment regime and some of the ongoing management actions
that might optimize nearshore restoration.
The Puget Sound nearshore
The Puget Sound is one of the largest estuaries in the United States, encompassing
more than 8,000 square kilometers of marine waters and a watershed of more than
33,000 square kilometers. The physical and ecological complexity found within
the Puget Sound region supports a rich and productive natural environment for
plant, animal, and human communities alike. The nearshore zone is a complex
ecosystem found within the narrow, contiguous ribbon of land and shallow water
that rings the more than 4,000 km shoreline of Puget Sound. The nearshore is
often defined as extending from the upland and backshore areas that directly
influence shoreline conditions to the shallow offshore waters as far as the lower
limit of the photic zone, where sunlight is no longer able to sustain marine
vegetation (EnviroVision et al., 2007; Fresh et al., 2011). The Puget Sound
Nearshore Ecosystem Restoration Project (PSNERP), a collaborative effort
4
between government agencies, universities, tribes, and environmental
organizations, defines the nearshore as an area extending from the top of shoreline
bluffs waterward to the deepest extent of the photic zone (Clancy et al., 2009).
This photosynthetically-defined edge of the nearshore varies in its distance from
shore according to water depth and clarity, but is often delineated around 30
meters below mean lower low water (MLLW) (EnviroVision et al., 2007; Shaffer
et al., 2008). Occupying areas commonly known as the shore, beach, intertidal,
and subtidal zones, the nearshore forms the transitional interface between three
critical edge habitats: the edge between terrestrial and aquatic environments, the
edge between the diverse and productive shallow waters and deeper water, and
the edge between fresh water streams and the marine salt waters (EnviroVision et
al., 2007). Interactions between various physical conditions such as coastal
geomorphology, wave energy, sediment movement, sunlight, and salinity along
these nearshore edges creates a mosaic of different habitats that support a wide
diversity and abundance of life. The nearshore marine habitats of the Pacific
Northwest are therefore critical components of our regional ecosystem. They
provide nursery, migration, and feeding corridors for a number of fish, including
several federally and state-listed salmon, and other wildlife (Fresh, 2006; Penttila,
2007). In fact, the nearshore environments are the most productive waters of
Puget Sound; their condition, therefore, influences the productivity of the entire
Puget Sound basin and many of the ecosystem goods and services important to
human communities (EnviroVision et al., 2007; Fresh et al., 2011; Simenstad et
al., 2006).
5
The Puget Sound nearshore has long attracted human activity. The area
was home to about 50,000 native people when Captain Vancouver first sailed into
Puget Sound more than 200 years ago but now sustains a population of 4.5
million people (Puget Sound Partnership, 2013), about 70% of the people living in
Washington State. Large sectors of the northwest economy are tied to the Puget
Sound and its nearshore environment, including shellfish and commercial fishing
industries, ports, refineries, and trade activities, and a variety of recreational
opportunities. This concentration of human population and intensity of activity
within the nearshore makes it especially vulnerable to human impacts.
In recognition of this vulnerability, the Shoreline Management Act (SMA)
of 1971 provides a framework requiring the evaluation of existing nearshore
conditions and the establishment of policies and regulations that will protect
nearshore ecological functions. As part of a required process to inventory and
characterize shoreline conditions, local governments are directed to identify areas
of critical saltwater habitat and designate them as “critical areas.” In addition to
kelp and eelgrass beds, mudflats, and areas associated with priority species,
critical areas are defined to include, “spawning and holding areas for forage fish,
such as herring, smelt and sand lance” (WAC 173-26-221(2)(iii)(A)). Critical
habitats are also recognized as requiring a higher level of protection in order to
preserve the important ecological functions they provide. The law calls for
regulatory provisions for critical areas that protect existing ecological functions
and ecosystem-wide processes (WAC 173‐26‐221(2)(B)(iv)) by integrating the
management of shorelands and submerged areas (RCW 90.58.090(4) and WAC
6
173‐26‐221(2)(iii)(A)). As beaches are the primary feature that defines the
landward edge of the nearshore zone, they provide important ecological functions
but are often directly impacted by human activities in the nearshore. Beaches
therefore, warrant a closer look at how they are formed, what processes sustain
them, and what role they play in nearshore habitat form and function.
Beach formation and sediment processes of the nearshore
Beaches are formed by the accumulation of mobile material between the upland
environment and an area of deeper water where substrate is not influenced by
wave action and so is not active (EnviroVision et al., 2007). They occur along the
shoreline where there is both an abundant supply of sand and gravel sediment and
a sufficient degree of wave action to rework this material (Shipman, 2008). The
erosion of coastal bluffs provides most of the sediment inputs into the Puget
Sound nearshore system, with large rivers and small streams contributing
additional inputs (Finlayson, 2006; Johannessen & MacLennan, 2007). It is this
interaction between sediment input and sediment transport that controls the
structure of beaches. The steady erosion of these bluffs, therefore, is critical for
maintaining beaches and spits over the long term.
Beach substrate is in constant flux, mobilized by wave action and
redistributed along the coastline in a process called longshore transport
(Simenstad et al., 2011). Sediment redistribution occurs in a net direction
according to prevailing waves, wind, and currents. The configuration and
orientation of the coastline relative to these prevailing forces divides the shoreline
7
into distinct areas with its own sediment source, area of transport, and area of
deposition (Shipman, 2008). These semi-independent segments of the coast are
called littoral cells, or drift cells, and 860 of them have been identified within
Puget Sound (EnviroVision et al., 2007). The sediment processes that occur
within drift cells result in a gradual change in beach size, shape, and structure, and
are responsible for the present configuration of the shoreline. This link between
geomorphological processes and the physical form of the nearshore creates
habitats that are of critical importance to the survival of a multitude of marine
creatures, including forage fish, juvenile salmon, marine birds and mammals, and
aquatic vegetation such as eelgrass and kelp beds. Since bluffs contribute the
majority of sediment sources into the nearshore system, it is important to
understand their role in maintaining intact coastal sediment processes, as well as
what might impact these processes.
Coastal Feeder Bluffs
Puget Sound’s coastal bluffs are geologic features that have formed since the last
glacial period of the region ended 14,000 years ago, leaving behind an extensive
layer of poorly consolidated sediment across the region at elevations above
modern sea level (Shipman, 2004). Channels cut into this sediment layer by
glacial meltwater became deep troughs and slowly filled with sea water until
reaching the current sea level about 5,000 years ago, at which time bluffs
developed and the modern shoreline to begin to evolve (Shipman, 2004).
8
Bluffs have a significant influence on the region’s nearshore environment
because they are found along more than 60% of Puget Sound’s shoreline and are
the primary source of recruitment for the sand and gravel that make up beach
substrate (Johannessen & MacLennan, 2007). It is primarily through the erosion
of these coastal feeder bluffs that sediment is delivered into the nearshore system
to shape and maintain nearshore habitat. Bluff erosion often occurs through a
process in which wave action removes material at the bluff toe, undercutting the
slope and creating an unstable profile that eventually leads to mass-wasting of
new material onto the beach (Shipman, 2004). The mobilization and distribution
of this new material sustains the sediment processes within the drift cell. The rate
of bluff recession depends on the bluff’s exposure to wave action of sufficient
energy to erode and remove sediment from the toe of the bluff, the geologic
makeup of the bluff which determines its susceptibility to erosion and masswasting, and beach characteristics such as beach width and berm height which
control how frequent and with what intensity waves reach the bluff toe (Shipman,
2004). Complex interactions between all of these factors combine to create a
diversity of ecological forms and habitat types along the marine shoreline. Many
nearshore species depend on habitat that is associated with beach substrate of a
particular composition, size, degree of sorting, or other specific characteristic. For
example, Forage fish require substrate within a particular range of grain-sizes for
successful spawning (Moulton & Penttila, 2001; Penttila, 1995; Penttila, 1978).
The habitat suitability of kelp forest and eelgrass beds, both important for
providing productive refuge for a wide range of marine species, is also
9
determined by sediment characteristics, although these factor differently for each
(Mumford, 2007). Because bluffs are the primary source of sediment replenishing
and maintaining beach substrate, they are indirectly responsible for ensuring that
diverse habitats are available for species with a variety of sediment requirements.
Bluffs, therefore, are a feature of the Puget Sound nearshore that are vital to the
health of nearshore populations, and by extension, to the health of the Puget
Sound region as a whole.
Having described some of the physical conditions and geomorphological
processes shaping the nearshore, we now turn to examine one group of biological
creatures inhabiting this unique environment. Forage fish display particular
requirements for sediment size and composition and so are good indicators of the
physical form and the health of ecological function in Puget Sound’s nearshore
environment.
Forage fish
Three species of forage fish, Pacific herring (Clupea pallasi (Valenciennes
1847)), surf smelt (Hypomesus pretiosus (Girard 1955)), and sand lance
(Ammodytes hexapterus (Pallas 1811)) spawn in the nearshore zone of Puget
Sound beaches (Penttila, 2007). Since all forage fish species rely on nearshore
habitats for at least part of their life history and congregate in the nearshore in
large numbers during spawning, the protection of these habitats is critical to the
long-term sustainability of these species as well as to a number of other fish,
mammal, and bird species which rely on forage fish as prey. Pacific herring
10
spawn on marine vegetation in shallow subtidal regions of the nearshore, but surf
smelt and sand lance spawning habitat occupies the upper intertidal region and so
is particularly vulnerable to both changes in nearshore sediment processes and
human activities which modify the shoreline. The following sections of this
chapter will focus on the life histories of surf smelt and sand lance in particular,
their importance to the marine food web of the Puget Sound region, and how their
spawning activity is crucially linked to the sediment supply processes of the
nearshore.
Surf smelt and sand lance are critically important to the structure of Puget
Sound marine food webs (Cross et al., 1980; Penttila, 2007; Simenstad et al.,
1979; Therriault & Schweigert, 2009). For instance, 35% of the diets of juvenile
salmon (and 60% of the diets of juvenile Chinook salmon) in Puget Sound were
found to be comprised of sand lance (Hershberger et al., 2006). Some have
described forage fish as representing the primary energy bottleneck in the
biological community of the nearshore, exerting both top-down control over
primary and secondary tropic levels (phytoplankton and zooplankton) and
bottom-up control over higher order predators such as salmonids, marine
mammals, and sea birds (Rice, 1995; Simenstad et al., 1979). Variability in forage
fish abundance and distribution, therefore, may have major consequences for
determining the fitness of predator populations (Haynes et al., 2008). The
Washington Department of Fish and Wildlife (WDFW) is charged with managing
forage fish stocks in Washington State; yet, in spite of the apparent ecological
significance of forage fish for other commercial or endangered species, the
11
ecological variables that influence their populations remain largely understudied (
(Penttila, 2007; Robards et al., 1999b). While herring stocks are regularly
monitored, no monitoring strategy has been developed for surf smelt or sand lance
and little is known of their life histories or geographic distribution. Nevertheless,
observations of surf smelt and sand lance spawning behavior present in the
literature afford some insight into their respective habitat requirements as well as
ways in which anthropogenic factors can affect population abundance and
demographics.
Surf smelt
Surf smelt occur from northern California to southern Alaska (Hart, 1973), but are
particularly abundant in Puget Sound (Therriault et al., 2009) where they are also
available much of the year, increasing their importance as a food source within
the marine food web. Surf smelt feed primarily on calanoid copepods (Simenstad
et al., 1979), a key herbivorous species of zooplankton that concentrate
autotrophic carbon into particles of high-density plankton. Surf smelt thus serve
as a crucial trophic link between primary consumers and higher order predators.
Maturing and spawning at one year of age, few fish appear to survive beyond four
years of age (Penttila, 1978). Little is known about the life history of surf smelt
apart from their use of nearshore intertidal beaches for spawning.
Many questions remain unanswered as to the number of distinct surf smelt
stocks, as well as their seasonal distribution and movement throughout the waters
of the Puget Sound region (Penttila, 2007; Pierce et al., 2009). Spawning can
12
occur year round, but in general surf smelt stocks are loosely divided into summer
spawners and winter spawners. Summer spawning occurs in the northern Puget
Sound region, including the Strait of Juan de Fuca, from May to October; winter
spawning takes place in the southern Puget Sound region from September to
March (WDFW, 2010). Spawning habitat has been documented along about 10%
of the shoreline of the Puget Sound Basin (Penttila, 2007), including coastal areas
of the Strait of Juan de Fuca (Moriarity et al., 2002a; Penttila, 1978), with a total
of 259 lineal statute miles of Washington State shoreline known to be surf smelt
spawning beach (WDFW, 2010). Spawning takes place in short intervals during
high tides, in the few inches of water covering the uppermost one-third of the
intertidal zone (Penttila, 2007). Surf smelt eggs are adhesive to particles of beach
substrate and are often found with several grains of sand attached to them which
acts to weigh down the egg and help it mix into the substrate below the beach
surface (Penttila, 1978). Eggs can take from two to eight weeks to incubate,
depending on seasonal temperature (Penttila, 2007).
Pacific sand lance
The Pacific sand lance is a common and widespread species of forage fish found
throughout the nearshore marine waters of the Puget Sound Basin. Like surf
smelt, they feed primarily on calanoid copepods (Miller et al., 1980) and are in
turn preyed upon by a broad array of marine mammal, bird, and fish species. As
many as 31 species of birds, 9 marine mammals, and 27 fishes depend on sand
lance for more than 50% of their diet (Robards et al., 1999b). Sand lance thus
occupy an ecologically-important link in marine food webs similar to surf smelt.
13
Unlike surf smelt, however, sand lance often form dense surface schools,
commonly called “bait-balls,” which attract a variety of predators including
numerous alcid seabirds (Penttila, 2007). Sand lance are also unusual in that they
actively burrow into beach substrate as a predator-avoidance mechanism and as
part of diurnal and seasonal cycles of energy conservation (Quinn, 1999). The
spawning season for sand lance appears to be shorter than for surf smelt,
occurring exclusively in the fall and winter months. The greatest spawning
activity has been detected in November and December, but extends to a lesser
degree into January and February (Penttila, 1995b).
Although taxonomically unrelated, sand lance spawning habitat closely
resembles that of surf smelt. Both spawn on upper intertidal beaches consisting of
sand and fine gravel during high tides when the beach is covered with shallow
water. While mapping the spawning activity of sand lance in beaches along the
western Strait of Juan de Fuca, Moriarity et al. (2002b) found eggs deposited in a
fluffy mixture of fine and coarse sands. This “fluffy” nature of the substrate is
important because females will excavate shallow pits in which to deposit spawn
(Penttila, 1995b; Robards et al., 1999a). The preference for uncompacted
substrate, whether fluffy sand or a loose mixture of sand and gravel, is a
characteristic shared by surf smelt. In fact, the same beaches are often used for
spawning by both sand lance and surf smelt and the eggs of both can co-occur in
the same beach substrate when their spawning seasons overlap in winter (Penttila,
1995a; Penttila, 2001). Since the spawning behavior so closely resembles each
14
other, we will now examine the specific substrate requirements and other shared
characteristics that define surf smelt and sand lance spawning habitat.
Spawning habitat
The upper intertidal zone is one of the most rigorous habitats in the marine
environment. Few organisms have managed to adapt to its fluctuations in
temperature, salinity, and submergence time, as well as to the grinding, abrasive
nature of its shifting substrate (Penttila, 1995a; Penttila, 1978). Both the surf
smelt and the sand lance have adapted their spawning activity to this zone and
have been observed to seek out a specific type of substrate for spawn deposition.
The characteristic beach substrate of an egg-bearing sample is often described as
“pea-gravel,” “coarse sand,” or a “sand-gravel mix.” In one of the earliest
measurements of substrate samples containing forage fish eggs, Penttila (1978)
found that samples containing surf smelt spawn were composed mostly (80% by
weight) of material in the size range of 1 to 7 mm in diameter. Another study
found that the top one-inch surface of sand-gravel beaches used by spawning surf
smelt was comprised mostly of material in the 1 to 10 mm size range (Penttila,
2001).
Sand lance utilize this same range of substrate sizes, but also spawn in
finer classes of sand. Penttila (2001) analyzed the grain-sizes of spawn-bearing
samples throughout Puget Sound and found that while 25% of the material
containing sand lance eggs was characterized as “gravel-coarse sand” resembling
the 1 to 7 mm grain-size range used by spawning surf smelt, the majority (67%)
15
of the material was medium sand between 0.2 to 0.4 mm. Others report spawning
areas consisting of coarse sand and gravel, 20% of which consisted of shell
fragments, with a slightly higher median particle diameter of 1.9 mm, concluding
that the choice substrate for sand lance appears to be highly specific,
characterized as well-washed, drained, and unpacked coarse sand with very little
content of mud or silt (Robards et al., 1999a; Robards et al., 1999b). A controlled
study in which sand lance were given a range of substrate sizes to chose from
found that they preferred a coarse sand with grain-size from 0.5-1 mm (Summers
et al., 2013). Since benthic areas and shorelines that lack sediment have been
found to have no sand lance present (Haynes et al., 2007), it is clear that this
species also displays specific substrate requirements in their spawning activity
and use of the nearshore environment.
One of the more curious uses of the nearshore by sand lance is displayed
in their habit of burrowing into beach substrate. The species has developed a
number of adaptations, including the lack of a swim bladder, a slender body, and
the ability to utilize oxygen-poor interstitial water, which allow them to bury
themselves in loose sandy substrate (Quinn, 1999). This behavior appears to occur
for a variety of reasons and at different times of the seasonal and diurnal cycle.
For example, sand lance have been observed burrowing into sediment at night, as
well as during the day while not foraging to escape from predators (Haynes et al.,
2008; Haynes & Robinson, 2011). They may also burrow into sediment during
winter for months at a time in a state of dormancy, and can remain buried above
the waterline during low tide (Ciannelli, 1997; Quinn & Schneider, 1991). Sand
16
lance may remain in the shallow depths of the coastal environment for much of
their life history. Ostrand et al. (2005) found few sand lance located at depths of
40 m and none at 60 m. Their requirement for a highly specific type of spawning
habitat, consisting of well-drained sand and pebbles without silt or mud, and their
use of sediment as a refuge causes sand lance to be associated with shorelines that
are plentiful in suitable substrate (Haynes et al., 2007).
Spatial and temporal factors also relate to substrate type and affect forage
fish spawning. For example, shoreforms such as sandy spits and beaches at the far
end of drift cells often support sand lance spawning habitat because those are the
locations where the appropriate type of finer-grained sediment accumulates
(Penttila, 2007). In their Field Manual for Sampling Forage Fish Spawn in
Intertidal Shore Regions, Moulton & Penttila (2001) report that the upper third of
the beach is the most likely area to contain eggs of both species. Because of wave
energy acting to sort the substrate at this tidal elevation, this area is characterized
by loose and well-mixed sand and small gravel that is devoid of the very-fine size
classes of silt and mud. Forage fish, therefore, frequently spawn at high tide in
order to reach this high position on the beach (Penttila, 1995b). After being
deposited near the high tide line, some eggs are washed down the beach slope by
receding waves and are distributed widely along the beach face for incubation
(Moulton & Penttila, 2001). Eggs, along with smaller particles of beach material,
are sifted into a lower strata of the beach by the sorting and resorting of the
surface substrate by wave action, eventually coming to rest in a microenvironment providing both capillary moisture and sufficient aeration to
17
maximize spawn survival (Middaugh et al., 1987; Penttila, 1978; Thompson &
Associates, 1936). The spawning activity of these forage fish are thus welladapted to the dynamics of the upper intertidal zone.
Spawning habitat is often not uniformly distributed, but found in a mosaic
of substrate types representing various degrees of potential suitability. Depending
on such factors as the source and composition of raw beach material, wave action
regime, orientation of the shoreline, direction and velocity of sediment drift, and
presence or absence of shoreline structures, the preferred sediment can be
spatially patchy, occurring in limited areas (Haynes & Robinson, 2011), or in
broad bands of material meters wide and kilometers long (Penttila, 1978; 2007).
Geographical distribution can vary as well. Penttila (1995b) found that virtually
every sandy-gravel beach in the series of bays of the northeastern sector of the
Olympic Peninsula supported sand lance spawning activity. In contrast, the
relatively rare pocket beaches of the San Juan Islands (Beamer & Fresh, 2012)
and the protected cove beaches on the outer coast of the Olympic peninsula
(Thompson & Associates, 1936) offer only small and discrete patches of suitable
habitat.
Successful spawning habitat for forage fish depends on the presence of an
adequate amount of beach substrate of the correct composition. Areas where the
substrate of the upper beach is composed entirely of fine sand or large gravel and
cobbles are rarely used by forage fish for spawning. When spawning does occur
in such areas, for instance when located adjacent to heavily-used satisfactory
spots, Penttila (1978) observed spawn of poor quality. Whether deposited on
18
large-sized beach material or on fine pure sand, the light-weight eggs are unable
to mix into the substrate below and so are left on the surface, suffering desiccation
and thermal stress from the continuous exposure to sun and wind.
Further evidence that forage fish require substrate of a particular character
comes from the observation that spawning sites are commonly used year after
year. Haynes & Robinson (2011) found that sand lance exhibit site fidelity during
their first year of life, re-using the same nearshore patch of sediment on time
scales from weeks to months, and for a few sites, inter-annually. After
discovering a suitable sediment patch, the fish stay nearby, presumably because of
their high environmental specificity for substrate type and because of the risk of
not finding another suitable sediment patch. Whether perennial use of isolated
patches of habitat is evidence of a homing ability or of active searching behavior
is not known (Penttila, 1995b)
Other investigations into how variations in the beach environment affect
the suitability of spawning habitat have revealed annual spatiotemporal variations
in the distribution of spawning activity in Puget Sound. Quinn et al. (2012)
evaluated beach characteristics hypothesized to affect the suitability of surf smelt
spawning habitat and found that aspect, fetch, solar radiation, and temperature
were predictors of eggs abundance, but not of embryo mortality. Spawning
activity appears highly variable, both spatially and temporally. Factors other than
substrate characteristics, such as population density, behavioral dynamics, and
other environmental conditions also determine whether a beach supports
spawning. It has been estimated that, in any given year, perhaps as little as 30% of
19
the known spawning beaches in Puget Sound actually support surf smelt
spawning (Penttila, 2007; Quinn et al., 2012). The fact that most beaches with the
suitable type of substrate are not used for spawning suggests that impacting those
relatively few beaches that do support spawning could disproportionately affect
surf smelt production.
The particular sediment requirements of forage fish spawning habitat
makes them especially susceptible to alterations of sediment processes supplying
the nearshore environment. Disruptions of these processes can change the
physical characteristics of a beach, from coarsening the composition of substrate
material to altering the beach slope and width (Fresh et al., 2011). Beaches that
lose the continual inputs of sediment that sustain them can suffer loss of shoreline
habitat. Wave action continues to suspend and carry away the fine sediments from
the beach surface, over time leaving it as an area of hardpan mud, bedrock, and
cobble, unsuitable for spawning forage fish (Middaugh et al., 1987; Moulton &
Penttila, 2001). Changes in the distribution of sediment to size ranges outside
those required by forage fish are likely to affect spawning site selection as well as
egg mortality. Continual inputs of sediment are therefore required to sustain beach
structure and forage fish habitat.
Shoreline armoring
The past 150+ years of development since Europeans began settling the region
have profoundly changed the physical form of Puget Sound’s nearshore
ecosystems with implications for the vitality of ecosystem functions, goods, and
20
services. The modification of beaches and bluffs through the construction of
shoreline armoring results in the reduction of sediment supply and the interruption
of sediment transport processes. According to recent estimates, approximately
27% (about 1,070 km) of the shoreline of Puget Sound is armored (Puget Sound
Partnership, 2013). Furthermore, while 27% of barrier beaches and 8% of pocket
beaches have been armored, a full third of bluff-backed beaches have been
armored along at least half of their length (Fresh et al., 2011).
Placing bulkheads or other armoring structures along a shoreline can result
in forage fish habitat degradation in several ways. Armoring that extends low
enough into the intertidal zone can cover over and physically eliminate the finegrained substrates found on the upper beach that are necessary for forage fish
(Penttila, 2007). By reducing bluff erosion and blocking sediment input to the
beach, armoring can convert spawning areas of fine-grained substrate to coarser
gravel and cobble material, unsuitable as spawning habitat (Fresh et al., 2011).
Bulkheads may accelerate this process by reflecting wave energy back onto the
beach, suspending and transporting away the fine grains of substrate and
contributing to further coarsening (Carrasquereo-Verde et al., 2005). Armoring
can also cause an increase in temperature on the upper beach through the removal
of shade trees, thereby negatively impacting the survival of incubating embryos
(Penttila, 2002). In fact, Rice (2006) demonstrated that anthropogenic shoreline
modifications can create a brighter, hotter, and drier shoreline environment in
which the proportion of surf smelt eggs containing live embryos was reduced in
half.
21
Armoring can also have indirect effects on the abundance and distribution
of forage fish spawning substrate. One study in Thurston County found that while
woody debris on beaches provided structural support for the accretion and
stabilization of sand, its presence or absence was the single most distinguishing
factor between unarmored and armored shorelines (Carrasquero-Verde et al.,
2005). They concluded that the loss of woody debris from armored beaches is
likely to contribute to reduced forage fish spawning habitat (also see Clancy et al.,
2009; Rich et al., 2014). Other indirect impacts to spawning habitat occur through
activities that change the size and shape of the beach, and thus the area available
for spawning, or the size and composition of beach substrate. Perhaps the most
important indirect effect of shoreline armoring is to inhibit bluff erosion and thus
reduce sediment inputs into the entire beach system. Changes in sediment supply
directly affect the volume of sediment that is available for longshore transport
within drift cells (Simenstad et al., 2011), and can lead to lower elevations and
coarser sediments of beaches in the upper intertidal zone (Shipman, 2008).
Longshore transport can be further impeded by groins and jetties or fill that
extends onto or across a beach. Finally, the rate of transport may be altered by
structures parallel to the shore, such as seawalls, which modify how waves
interact with the beach (Simenstad et al., 2011). Armored shorelines are one
important mechanism by which beach sediment processes are impaired. To
adequately protect forage fish habitat requires not only protecting the beaches
where spawning occurs, but also protecting the physical processes that form and
maintain those nearshore habitats that support spawning (Schlenger et al., 2011).
22
Second to coastal bluffs, rivers are the other major input of sediment into
Puget Sound and the Strait of Juan de Fuca. Rivers transport an estimated 6.5 ×
106 t yr-1 of sediment to Puget Sound every year (Czuba et al., 2011b). Disrupting
the delivery of sediment to nearshore beaches leads to changes in its structure and
results in degraded ecosystem function (Schlenger et al., 2011; Simenstad et al.,
2006). An example of impaired sediment processes resulting from the dual effects
of shoreline armoring and in-river dams can be found within the Elwha River
system and its nearshore environment.
The Elwha River and nearshore
The Elwha nearshore is located along a segment of coastline found within the
Strait of Juan de Fuca, a body of water connecting the Pacific Ocean to the inland
marine waters of Puget Sound which provides refuge, feeding, and spawning
habitat for forage fish as well as a critical conduit for several migrating species of
salmon (Shaffer et al., 2003). The Elwha nearshore follows a 21 km stretch of
shoreline extending from the west end of Freshwater Bay east to the tip of Ediz
Hook. Two large dams, built in the early 1900s, have disrupted the delivery of
sediment into the Elwha drift cell, impacting the character of the substrate found
on the beaches dependent on this supply. Over the course of their lifetime, the
dams trapped an estimated 21 to 26x106 m3 of sediment within their reservoirs
and reduced the Elwha River’s delivery of fluvial sediment to the coast to about
2% of the pre-dam load (Draut & Ritchie, 2013). With the dams in place, exposed
bluffs along the lower Elwha River remained the only substantial source of
23
sediment in the short section of river downstream of the lower Elwha Dam (Draut
et al., 2011). The sediment supply to Ediz Hook, the spit at the terminal end of the
Elwha drift cell, has been further impacted by the presence of approximately 5 km
of bluff armoring located between the river mouth and the spit (Galster, 1989).
The Elwha drift cell, with the two dams on the Elwha River and stretches of
armored bluffs on the coast, provides an interesting case study on the effects that
an impaired sediment delivery process has on the supply and composition of
nearshore sediment, and the subsequent effects on forage fish spawning habitat.
Warrick et al. (2009) conducted a comprehensive study on the
morphological changes ongoing in the Elwha delta and adjacent beaches and
found that, between 1936 and 2006, the shoreline eroded at a rate consistent with
the reduction in sediment supply from the Elwha River. Prior to dam construction,
the river freely discharged sediment at a rate which maintained a steady shoreline
position during the past ~7000 years. However, after dam construction and from
1939 to 2006, approximately 100,000 m2 of coastal plain was lost to erosion as a
result of the sediment reduction (Warrick et al., 2009). Not only did the shoreline
recede in response to the reduced sediment input, but the intertidal zone
substantially coarsened over the 20th Century, as evidenced by the cobbled lowtide terrace consisting of lag clasts that are stuck in place and rarely move
(Warrick et al., 2009). This observation of change in the character of nearshore
substrate is corroborated by oral histories of the Lower Elwha Klallam Tribe,
predating dam construction, which describe a low-tide beach of soft sediment
ideal for shellfish harvesting (Reavey, 2007). The storage of sediment behind the
24
dams has impacted the river’s delta and the coastal beaches within the Elwha drift
cell through increased erosion of the coastline and a dramatic coarsening of the
beach substrate.
The long-term reduction in sediment inputs into the Elwha nearshore has
caused diminished ecological function in a number of ways. Shaffer et al. (2012)
compared fish abundance, density, and diversity between the nearshore
environments of the sediment-impaired Elwha drift cell and the adjacent, intact
Dungeness drift cell. They found that the degraded habitat in the Elwha drift cell
had lower fish species richness and diversity than did the intact Dungeness drift
cell. Interestingly, although surf smelt presence and diversity varied somewhat by
geomorphic habitat type (embayments, bluffs, spits, and the lower reaches of
rivers), they were consistently good indicators of habitat quality and ecological
function at the overall drift cell scale. The higher surf smelt densities found in the
intact drift cell may be due to the increased availability of substrate found there
that meets their specific grain-size requirement (1-7 mm) for suitable spawning
habitat (Shaffer et al., 2012). Since the researchers also found salmon from as far
away as the Columbia River and Kalamath systems using the Strait of Juan de
Fuca shorelines, they concluded that actions taken to restore and preserve
nearshore ecosystem processes and ecological function are most appropriately
designed at the scale of the drift cell, rather than targeted at single species or
specific locations, and could thereby have cross-regional benefits.
Another study examining the dynamics of sediment supply and forage fish
spawning activity within the Elwha drift cell found that adjacent, intact drift cells
25
had both more sediment of the appropriate grain size for surf smelt spawning and
significantly higher densities of surf smelt spawn than the impaired Elwha drift
cell (Parks et al., 2013). Although sediment characteristics displayed seasonal
variation with pulses of delivery in the spring and fall, all geomorphic habitat
types within the intact drift cell consistently showed higher numbers of samples
with grain sizes preferred by surf smelt (1-7 mm). In contrast, all geomorphic
habitat types within the impaired Elwha displayed coarser size classes of
sediment, with higher numbers of samples with grain sizes larger than 7 mm. The
researchers conclude that disrupting the delivery of sediment into and across a
drift cell causes the distribution of sediment size to be significantly more variable
and significantly lowers the functional habitat required for forage fish spawning
(Parks et al., 2013).
A project to remove both dams, the largest such project and sediment
release in U.S. history, began in September 2011 and is expected to be completed
by September 2014. Dismantling the dams, and the subsequent release of
sediment trapped in the reservoirs, represents a unique and unprecedented
opportunity to restore the sediment-starved and ecologically-degraded Elwha
nearshore. However, significant impediments to full ecosystem restoration are
present and require attention.
Management issues and nearshore restoration
Before the construction of shoreline armoring and in-river dams, feeder bluffs
provided an estimated 70% of the sediment contribution to the entire Elwha drift
26
cell (Parks et al., 2013) and 85% of the sediment that formed and sustained Ediz
Hook (Galster, 1989). However, 68% of the entire length of these feeder bluffs
are now armored (Flores et al., 2013; Kaminsky et al., 2014). Of the portion of
bluffs within the Port Angeles city limits, 91% are armored, including a sea wall
which was constructed at the city’s landfill site in response to bluff erosion that
had caused garbage to fall onto the nearshore beaches below (City of Port
Angeles, 2012b; Neal, 2013). These armoring structures have significantly
impaired the sediment processes within the Elwha drift cell by greatly reducing
bluff erosion and its associated sediment input (Kaminsky et al., 2014), and will
likely remain an issue for the long-term restoration of the Elwha nearshore. Still,
some are optimistic that the sediment processes within the Elwha nearshore
environment could be at least partially restored with the reestablishment of fluvial
sediment sources (Parks et al., 2013; Shaffer et al., 2008; Winter & Crain, 2008).
With ongoing dam removal, sediment from the two reservoirs is moving
downstream (Warrick et al., 2012) and deposited on Elwha nearshore beaches
(Draut & Ritchie, 2013; Kaminsky et al., 2014). As of spring 2013, a total of 6.1 x
106 m3 of sediment had been transported out of the two former reservoirs (Draut
& Ritchie, 2013). Within the first two years, 2.5 x 106 m3 of sediment had reached
the nearshore environment (Kaminsky et al., 2014). Over the next five years, the
natural flow of the Elwha River is expected to mobilize and transport downstream
between one-third to one-half of the total volume of sediment stored within the
two former reservoirs (Konrad, 2009; Randle et al., 1996). The large amount of
sediment already delivered to the Elwha nearshore is changing the nature of its
27
beach substrate and potentially creating habitat of the composition and grain size
required for forage fish spawning.
The large pulse of sediment, however, will be short lived. After 7-10
years, the easily-erodible sediment in the reservoirs will be exhausted and the
river will resume supplying its normal, much-lower amount of naturally-eroded
sediment (Czuba et al., 2011a). Restoration of nearshore function, therefore, could
be enhanced and perhaps prolonged by targeted action and management
strategies. The Elwha Neashore Consortium (ENC), a group of scientists,
managers, and citizens dedicated to understanding and promoting the restoration
of the Elwha nearshore, advocates for an adaptive management approach to
respond to the changing conditions, management needs, and best science as it
becomes available (MacDonald & Harris, 2013). The presence of large woody
debris (LWD) on nearshore beaches has been shown to reduce erosion and help
stabilize beach substrate (Clancy et al., 2009; Rich et al., 2014). Deliberate
placement of LWD, perhaps in combination with beach nourishment using large
cobble, may be a useful strategy for capturing and retaining the new fluvial
sediment as it arrives on sediment-starved beaches (Shaffer, 2013). Restoration
will also depend on the continued preservation of unarmored coastal areas within
the drift cell, such as Freshwater Bay near the mouth of the river, as well as
adjacent areas with intact sediment processes. Ecosystem service valuation has
been proposed as an additional management strategy, helping to justify
investments in environmental restoration by revealing the economic value of
intact areas of the shoreline as compared to impaired areas (Flores et al., 2013).
28
Such an approach could be useful for prioritizing restoration and conservation
goals, and validating the need to ensure the continued protection of intact areas.
The ecological degradation of the sediment-starved, impaired Elwha drift
cell may serve as a cautionary example to stewards of intact coastal regions,
underscoring the detrimental consequences that shoreline armoring and impaired
sediment processes have on nearshore function. Conversely, those nearshore areas
with intact processes and robust ecological function can serve as an inspiration to
guide effective and timely action in the Elwha nearshore. The rich diversity and
biological activity found within intact coastal areas may also provide a baseline of
healthy ecological function, to which we can calibrate our restoration goals and
aspire to achieve in impaired nearshore areas. The dynamic changes now
occurring in the Elwha nearshore offer an unprecedented opportunity to
successfully restore this unique area. It is an opportunity that should not be
squandered.
29
Chapter 2: Article Manuscript:
Formatted for submission to a journal of coastal research.
Forage Fish Spawning in the Elwha Nearshore:
Restoring Ecological Form and Function in a Changing Environment
Abstract
Intertidal beaches within the Elwha nearshore are documented habitat for forage
fish migration and spawning. Sediment processes of the Elwha drift cell, critical
for forage fish spawning habitat, were historically altered by armoring of the
shoreline, lower river alterations, and the in-river Elwha and Glines Canyon dams.
The recent removal of these two dams, and the consequent release and transport
of upwards of 2.5 x 106 m3 of fluvial sediment to the Elwha nearshore, has begun
a partial restoration of sediment processes within the drift cell and is changing the
substrate characteristics of its beaches, potentially restoring nearshore function for
forage fish spawning. We conducted egg surveys for two species of forage fish,
surf smelt (Hypomesus pretiosus) and sand lance (Ammodytes hexapterus), over
four years, including two years before dam removal (2007-8) and two years
during the dam removal process (2012-13). Samples were collected from
geomorphic habitat types (GMHTs) (embayment, bluffs, and spit) within the
impaired Elwha drift cell and from comparative, intact Dungeness and Crescent
Bay drift cells. In order to assess nearshore function, we compared spawning
activity across impaired/intact drift cells, GMHT, and before/during dam removal
time periods. While no sand lance eggs were found during the course of this
study, our surf smelt results show that, overall, the intact Dungeness drift cell
supports more-robust spawning activity than the impaired Elwha drift cell and
that egg productivity did not differ significantly between the two time periods.
We also conclude that egg abundance is highly variable across GMHT, with the
greatest abundance in the intact bluffs site followed by the impaired embayment,
where spawning habitat appears to have expanded during the dam removal period.
Spit sites did not support any spawning activity. Understanding the implications
of dam removal to the ecological functioning of the nearshore is important for full
ecosystem restoration of the Elwha system, where shoreline armoring will remain
an outstanding and long-term restoration issue.
30
INTRODUCTION
The nearshore marine habitats of the Pacific Northwest are critical components of
our regional ecosystem. They provide nursery, migration, and feeding corridors
for shore birds, marine mammals, and a number of fish species, including several
federally and state-listed salmon (Fresh, 2006; Penttila, 2007; Shaffer et al.,
2008). The nearshore environment also provides spawning grounds for small,
schooling fishes known as forage fish. Surf smelt (Hypomesus pretiosus) and sand
lance (Ammodytes hexapterus) are forage fish species that serve a crucial role in
the complex marine food web of Puget Sound and the Strait of Juan de Fuca
(Penttila, 2007; Robards et al., 1999b; Wilson et al., 1999). Forage fish represent
a primary energy bottleneck in the biological community of the nearshore,
exerting both top-down control over primary and secondary trophic levels as
consumers of phytoplankton and zooplankton, and bottom-up control over higher
order predators by serving as a prey species for other fish, birds, and marine
mammals (Bargmann, 1998; Rice, 1995; Robards et al., 1999b). Surf smelt occur
from northern California to southern Alaska (Hart, 1973), but are relatively
abundant in Puget Sound (Simenstad et al., 1979; Therriault et al., 2009) where
they are accessible for much of the year, increasing their importance as a food
source within the local marine food web. Pacific sand lance are also an important
prey species, constituting large portions of the diets of all life-stages of salmon,
especially Coho (Oncorhynchus kisutch) and Chinook (Oncorhynchus
tshawytscha) (Beacham, 1986; Brodeur, 1990; Hart, 1973). For example, sand
lance were found to constitute, on average, 35% of the diet of juvenile salmon and
31
60% of the diet of juvenile Chinook salmon in Puget Sound (Hershberger et al.,
2006). Consequently, variability in forage fish abundance and distribution may
have major consequences for determining the fitness of these and other
economically important predator populations (Wilson et al., 1999). This
realization has led scientists and natural resource managers to become
increasingly interested in forage fish conservation and protection (Penttila, 2007).
The ecological functioning of nearshore habitats, therefore, are of special concern
because forage fish rely on them for their spawning activity (Penttila, 2007).
Both the surf smelt and sand lance have successfully adapted their
spawning activity to the habitat of the upper intertidal beach zone, rigorous for its
fluctuations in temperature, salinity, submergence time, and to the harsh and
grinding regime of shifting substrate (Penttila, 1978). Egg deposition occurs in the
upper elevation of the beach, near the high tide line, in areas where waves and
currents have sorted the substrate into a characteristic loose mixture of sand and
fine-gravel, consisting mostly of material in the size range of 1 to 7 mm in
diameter in the case of surf smelt, to finer size-classes of sand in the case of sand
lance (Penttila, 2001; Quinn, 1999; Robards et al., 1999a). Such substrate allows
wave action to work deposited eggs into the protective interstitial spaces below
the beach surface where they are kept moist and aerated, and thus protected from
desiccation and thermal stress (Middaugh et al., 1987; Penttila, 1978).
Conversely, in areas where the substrate of the upper beach is composed of very
fine sand or large gravel and cobbles, the lightweight eggs are unable to mix into
the subsurface substrate and are consequently left exposed to the drying effects of
32
sun and wind. Since successful spawning habitat for forage fish depends on the
presence of adequate sediment, altering the sediment processes that supply coastal
beaches can degrade the quality of forage fish spawning habitat and impair
nearshore ecological function (Johannessen & MacLennan, 2007; Schlenger et al.,
2011). The links between coastal sediment processes, nearshore function, and
forage fish spawning makes the status of forage fish populations a useful indicator
of the health and productivity of nearshore systems (Parks et al., 2013; Puget
Sound Partnership, 2009; Simenstad et al., 2006).
The Puget Sound nearshore, including the nearshore of the Strait of Juan
de Fuca, has been classified into geomorphic habitat types (GMHTs) that reflect
the close relationship between sediment processes, coastal landforms, and habitat
formation (Finlayson, 2006; Shaffer et al., 2012; Shipman, 2008). Nearshore
ecological function is strongly influenced by the geomorphic processes that erode,
transport, and deposit sediment across the coastal landscape and determine its
physical form (Simenstad et al., 2006). Coastal bluffs and river estuaries are the
two main sources of sediment responsible for the formation and maintenance of
marine beaches within Puget Sound and the Strait of Juan de Fuca (Finlayson,
2006). Sediment processes in the nearshore occur within drift cells—semiindependent segments of the coastline that include both sources and sinks of
sediment and within which net long-term sediment transport occurs (Shipman,
2008). The sediment processes within a drift cell are kept intact by the continual
input of new fluvial and bluff sediment to replenish that which is lost to erosion,
but can become impaired when anthropogenic activities, such as building dams
33
and armoring shorelines, reduce these sediment inputs. Dams trap sediment within
their reservoirs, disrupting fluvial sediment transport downstream (Finlayson,
2006; Johannessen & MacLennan, 2007). Shoreline armoring, designed explicitly
to prevent the erosion of coastal bluffs, disconnects beaches from their major
source of sediment nourishment (Shipman, 2010).
Armoring structures can have numerous additional direct and indirect
detrimental effects on forage fish spawning habitat. Their placement in the upper
intertidal zone covers-over and replaces the upper beach, directly reducing the
area of available spawning habitat (Dugan et al., 2011; Penttila, 2007). Wave
energy reflected from armored shorelines tends to suspend and transport away the
finer-grained substrate, thereby converting the beach to unsuitable coarse gravel
and cobble (Carrasquero-Verde et al., 2005; Fresh et al., 2011). By displacing
shoreline trees, armoring structures can create a brighter, hotter, and drier
shoreline environment that negatively impacts the survival of incubating embryos
(Penttila, 2002; Rice, 2006) and disconnects the nearshore from terrestrial sources
of food, nutrients, and organic matter, including woody debris which provides
important structural support for the accretion and stabilization of beach substrate
(Carrasquereo-Verde et al., 2005; Penttila, 2001; Rich et al., 2014). The
disruption and impairment of nearshore sediment processes caused by the
damming of rivers and construction of shoreline armoring can alter the physical
characteristics of beaches and degrade important nearshore function, including
that of forage fish spawning (Simenstad et al., 2006).
34
The Elwha nearshore provides an example of a drift cell with an
historically-impaired sediment process that is currently undergoing massive
change. Classified as including spit, bluff, and embayment GMHTs, the shoreline
of the Elwha drift cell has been significantly degraded due to the disruption of
habitat-forming processes from the construction of in-river dams and shoreline
armoring (Schlenger et al., 2011; Shaffer et al., 2012). Two dams built on the
Elwha River in the early 1900s trapped an estimated 21 to 26 x 106 m3 of
sediment within their reservoirs and reduced the Elwha River’s delivery of fluvial
sediment to the coast to about 2% of the pre-dam load (Draut & Ritchie, 2013).
Consequently, evidence of sediment starvation has been documented in both the
below-dam river channel and the Elwha nearshore (Draut et al., 2011; Warrick et
al., 2009). Without the replenishing input of Elwha River sediment,
approximately 100,000 m2 of coastal plain within the Elwha delta was lost to
increased erosion of the coastline between 1939 to 2006, and coastal beaches
underwent a dramatic coarsening of their substrate (Warrick et al., 2009). This
observation is corroborated by oral histories of the Lower Elwha Klallam Tribe
which describe a low-tide beach of soft sediment ideal for shellfish harvesting
prior to dam construction (Reavey, 2007). A project to remove both dams, the
largest such project and sediment release in U.S. history, began in September
2011 and is expected to be completed by September 2014. During the ongoing
dam removal, released sediment is washing downstream (Warrick et al., 2012)
and arriving on Elwha nearshore beaches (Draut & Ritchie, 2013). Over the next
five years, the natural flow of the Elwha river is expected to mobilize between
35
one-third to one-half of the total volume of sediment within the two former
reservoirs (Konrad, 2009; Randle et al., 1996). Although 93% of the feeder bluffs
and spit within the impaired Elwha drift cell are currently armored and are likely
to remain so (City of Port Angeles, 2012b), sediment processes within the Elwha
nearshore environment may be at least partially restored with the complete
removal of the dams and the reestablishment of fluvial sediment sources (Parks et
al., 2013; Shaffer et al., 2008; Winter & Crain, 2008). The large pulse of sediment
already delivered to the Elwha nearshore is changing the nature of its beach
substrate (Figure 1) and is potentially creating habitat of the composition and
grain size required for forage fish spawning.
This study examines changes in the amount and distribution of forage fish
spawning activity within the Elwha drift cell since the dam removal process began
in the fall of 2011. Our research quantifies this change and adds to the currently
sparse literature on forage fish response to the rapid alteration of nearshore
conditions due to a large river restoration project. We also report results for forage
fish egg surveys that were conducted concurrently in the adjacent, comparative
Dungeness and Crescent Bay drift cells, both of which retain intact sediment
processes. By comparing spawning activity between before and during dam
removal time periods, between impaired and intact drift cells as a whole, and
between the various GMHTs within the drift cells, we hope to detect patterns in
forage fish spawning behavior and start to determine the role that sediment
processes play in forming favorable habitat conditions. We observe that the
changes brought about by the influx of large quantities of sediment into the Elwha
36
nearshore are changing the physical form of its habitat, thus potentially enhancing
the long-term ecological function of forage fish spawning. Specifically, we pose
the following hypotheses: a) forage fish spawning activity has significantly
increased within the impaired Elwha drift cell during the dam removal phase; b)
no significant changes between before and during dam removal have occurred
Figure 1. Freshwater Bay (impaired Elwha drift cell) has undergone dramatic changes in the
composition of its beach substrate between the time period before dam removal (top) and during
dam removal (bottom). Photo: Coastal Watershed Institute.
37
within the comparative, intact drift cells; c) the intact drift cells continue to
support greater spawning activity than the impaired drift cell during the dam
removal phase; and d) differences between GMHTs are significant, with bluffs
supporting more forage fish spawning than embayments or spits. We believe that
a better understanding of forage fish spawning activity on beaches within intact
and impaired drift cells will highlight the close relationship between habitatforming processes and habitat function, and will help scientists and resource
managers implement successful long-term coastal management and restoration
projects within the nearshore environment.
METHODS
This study describes the 2007-2013 forage fish component of the Coastal
Watershed Institute-led long-term assessment of the Elwha nearshore. The
nearshore study is intended to define nearshore ecological restoration response to
dam removals, utilizing fish as the ecological metric and has three phases: before
dam removal, during dam removal, and (in the future) after dam removal. To
determine the role sediment plays in providing forage fish spawning habitat, we
selected drift cells with impaired sediment processes (Elwha drift cell) and intact
sediment processes (Dungeness and Crescent Bay drift cells). Habitat areas within
these drift cells were categorized into the geomorphic habitat types (GMHTs) of
embayment, bluffs, and spit in order to detect how different habitat types support
forage fish spawning (Parks et al., 2013; Shaffer et al., 2012; Shipman, 2008).
Forage fish data presented in this paper were collected over four years: 2007-2008
(pre-dam removal) and 2012-2013 (during dam removal).
38
Study Sites
The Elwha, Crescent Bay, and Dungeness drift cells lie adjacent to each other
within the central Strait of Juan de Fuca and along the northern coast of
Washington’s Olympic Peninsula (Figure 2). The Elwha drift cell includes three
impaired GMHTs: an embayment (Freshwater Bay), bluffs (Elwha bluffs), and a
spit (Ediz Hook). Matching GMHTs were selected in the Crescent Bay drift cell,
consisting of a single embayment (Crescent Bay), and the Dungeness drift cell,
consisting of bluffs (Dungeness Bluffs) and a spit (Dungeness Spit), to serve as
comparative study sites with intact sediment processes. Samples were collected
from each GMHT within each drift cell, for a total of 6 sample areas (Table 1).
Figure 2. Study sites for surf smelt and sand lance egg surveys.
39
Table 1. Sample areas designated by geomorphic habitat type (GMHT) and location
within an impaired or intact drift cell. Samples were collected from three sample areas
within the impaired Elwha drift cell and from matching GMHTs within the intact
Dungeness (2 sample areas) and Crescent Bay (1 sample area) drift cells.
Sample Areas
Intact Drift Cells
Impaired Drift Cell
GEOMORPHIC
HABITAT TYPE
Crescent Bay
Dungeness
Elwha
EMBAYMENT
Crescent Bay
----------
Freshwater Bay
BLUFFS
----------
Dungeness Bluffs
Elwha Bluffs
SPIT
----------
Dungeness Spit
Ediz Hook
Surf Smelt and Sand Lance Egg Sampling
Sampling for surf smelt and sand lance took place during their respective
spawning seasons. In the Strait of Juan de Fuca, the documented spawning season
for surf smelt is summer and for sand lance is winter (Moriarity et al., 2002a;
Penttila, 2007). Surf smelt samples for this study were collected in the months of
July, August and September, and sand lance samples were collected in the months
October through January. Sampling for all forage fish eggs was conducted using a
modified Moulton and Penttila (2001) technique. Bulk samples of beach substrate
were collected using a hand scoop to skim from the top 2-3 cm of the beach
surface at each sampling location. As forage fish spawning and incubation areas
are normally in the +7 to +9 foot mean lower low water (MLLW) tide zone
(Moulton & Penttila, 2000), samples were collected from the upper third of the
beach, near the high tide mark, or 1 to 2 vertical feet below the driftwood log line.
Between 5 and 8 scoops were used to collect about 15 kg of substrate from each
40
sampling location and placed in a plastic bag, constituting one sample. A
modified systematic random design was used to select 2 to 21 sampling locations
spaced roughly equally across the length of sample area beaches. For each
sample, a variety of metadata were collected, including the date, sample number,
GPS coordinates of the sampling location, and geomorphological unit of the drift
cell.
Once collected into plastic bags, the bulk samples were transported to
Peninsula College in Port Angeles, WA for processing. Each sample was washed
through a series of screens in order to sort the sediment grain sizes and collect the
light fraction, thereby condensing the sample to a manageable size and
concentrating the portion most likely to contain fish eggs. This was accomplished
by placing a rack of Nalgene sediment screens, sizes 4, 2, and 0.5 mm, graded
from the largest mesh size on top to the smallest on bottom, over a 5-gallon
plastic bucket and thoroughly washing the sample through the screen set using
water from a garden hose. Once washed, the sediment remaining on the top two
screens was discarded while the material collected on the bottom (0.5 mm) screen
was placed into a plastic dishpan and covered with 2-5 cm of water. The sample
was then elutriated by hand in order to allow the relatively light eggs to migrate
upward through the sediment towards the surface. After elutriation for 1-2
minutes, the lighter fraction was skimmed from the surface using a 235 ml plastic
collecting jar. This winnowing process was repeated twice more on the remainder
of the sample and added to the same jar, to which Stockard’s solution (50 ml
formalin (37% formaldehyde), 40 ml glacial acetic acid, 60 ml glycerin, and 850
41
ml distilled water) was add to preserve the eggs. All processed samples were sent
to Dan Penttila, of Salish Sea Biological, for examination under a dissecting
microscope and to determine the presence or absence of eggs. All eggs were
identified, counted, and their life-history stage was recorded.
Statistical Analysis
Treatments for analysis were defined as before and during the dam removal
process, by each GMHT (embayment, bluff, and spit), and by drift cell as either
intact or impaired. Because of the nature of the geographic location of our study
sites and sampling schedule, our data are not independent but linked both
physically and temporally. To control for unequal numbers of samples among
different treatments, we calculated the average number of eggs found within each
sampled beach on each date of collection. This average number of eggs served as
a normalized metric of the egg productivity for each beach that could be
combined with other beaches and dates of collection in order to compare
spawning activity among different treatments and combinations of treatments.
Egg count data were analyzed to determine egg abundance within each
drift cell and GMHT, both before and during the dam removal process. Because
our data did not meet the assumptions of normality or equality of variance
required by parametric ANOVA, Monte Carlo resampling methods were used to
generate null distributions (10,000 random iterations) with which non-parametric
analyses could be conducted. When comparing two treatments (i.e. impaired vs.
intact drift cell), the absolute differences (DIF) between the two means were
resampled. When comparing more than two variables (i.e. embayments, bluffs,
42
and spits), the among treatment sums of squares (SSamong) were resampled.
Bonferroni error corrections were used to control the familywise error rate arising
from multiple hypothesis tests on all subsequent pairwise analyses. Statistical
analyses were performed using the Resampling Stats 4.0 add-in for Excel. All
figures show back-transformed means and standard error bars (±1 standard error).
RESULTS
We collected a total of 568 samples over the course of this study. Due to
fluctuations in the amount of available volunteer hours, funding resources, and
site conditions, the number of samples collected varied between years and sites.
Sand lance sampling in the winter months was complicated by the difficulty posed
by short daylight hours and evening low tides, especially for the bluff-backed
beaches of the Dungeness Bluffs site, where the danger of being caught between
the high bluffs and a rising tide at night prevented a more-extensive sampling
regime. Sand lance surveys resulted in a total of 156 samples collected from
across the study site over the course of this study (Table 2). Of this total, 30
Table 2. Sampling schedule for sand lance eggs in impaired and intact drift cells, before
and during dam removal. Numbers in the table refer to the number of individual samples
collected within the given treatment.
Sand Lance Sampling Schedule
Intact
Drift Cells
Impaired
Drift Cell
TOTAL
Before Dam
Removal
During Dam
Removal
Total
Samples
Samples
containing
eggs
17
28
45
0
13
98
111
0
30
126
156
0
43
Table 3. Sampling schedule for surf smelt eggs by month and year. Numbers in the table
refer to the number of individual samples collected within the given treatment.
Surf Smelt Sampling Schedule
Before
During
2007 2008 2012 2013
July
August
September
TOTAL
11
20
19
50
33
39
35
107
0
0
9
9
70
71
72
213
Total
Samples
Samples containing
eggs
114
130
135
379
31
35
20
86
samples were collected before dam removal and 126 samples were collected
during dam removal (Table 2). All 28 samples collected from the intact treatment
during dam removal were collected exclusively from the Crescent Bay drift cell.
No sand lance eggs were found in any of the samples during the course of this
study. Similarly, no eggs of either species were found in any of the 65 Crescent
Bay samples (32 sand lance and 33 surf smelt). Accordingly, all of the sand lance
data and all of the Crescent Bay samples have been excluded from the statistical
analysis below, since conclusions about the relative strength of ecological
function between treatments cannot be determined without any spawning activity
with which to make comparisons. Our analysis, therefore, only includes surf smelt
spawning data from the Elwha and Dungeness drift cells.
Summer sampling for surf smelt in the Elwha and Dungeness drift cells
resulted in a total of 379 collected samples, 86 of which contained one or more
surf smelt eggs (Table 3). Of the157 samples collected before dam removal, 26
were egg-bearing and yielded 457 eggs; of the 222 samples collected during dam
removal, 60 samples were egg-bearing and yielded 617 eggs, resulting in a total
44
Table 4. Surf smelt sampling results by study site and phase of dam removal. Sites in
italic are within the impaired Elwha drift cell.
Surf Smelt Survey Results
Samples
Collected
Site
Samples
Containing
Eggs
Eggs
Found
Samples
Collected
Before Dam Removal
Embayments
Freshwater
Bay
Bluffs
Elwha
Bluffs
Dungeness
Bluffs
Spits
Ediz
Hook
Dungeness
Spit
Total
Samples
Containing
Eggs
Eggs
Found
Total
Eggs
Found
During Dam Removal
27
8
64
33
5
21
85
25
0
0
65
0
0
0
53
18
393
67
55
596
989
26
0
0
27
0
0
0
26
157
0
26
0
457
30
222
0
60
0
617
0
1074
of 1,074 surf smelt eggs found during the course of this study (Table 4). Of the
five beaches sampled, only two beaches were found to support surf smelt
spawning activity: Freshwater Bay (impaired; 85 surf smelt eggs found) and
Table 5. Table of single dead surf smelt eggs not included in analysis because of their
empty state and low number suggest that they may have drifted-in from another area.
Single Dead Surf Smelt Eggs Not Included in Analysis
Date
7/24/2013
Location
Elwha
Bluffs
Dungeness
Spit
Dungeness
Spit
Dungeness
Spit
Coordinates
48.13457, 123.52144
48.16743, 123.16096
48.16288, 123.16733
48.15796, 123.17378
8/18/2013
Dungeness
Spit
48.1762, 123.1362
9/25/2012
7/22/2013
7/23/2013
Number of Eggs Notes
Empty shell with
1
sand grain attached
1
Empty shell
1
Empty shell
1
Empty shell
Empty shell with
attached sand
grains
1
45
Dungeness Bluffs (intact; 989 surf smelt eggs found) (Table 4). A small number
of single dead surf smelt eggs were collected in the Dungeness Spit and Elwha
bluffs locations that were not included in our analysis because their empty state
and low number suggests they may have drifted-in from a different area (Table 5).
For the purposes of statistical analyses below, the consolidation of all surf smelt
samples from each given beach by each sampling date reduced the total number
of surf smelt samples (n=379) to 41 data points (Table 6).
Table 6. Surf smelt data consolidated for statistical analysis. Numbers refer to the number
of data points of the given treatment available for statistical analysis.
Consolidated Surf Smelt Samples for Analysis
Sample Site
Embayments
Freshwater Bay
Bluffs
Elwha Bluffs
Dungeness Bluffs
Spits
Ediz Hook
Dungeness Spit
Total
Before Dam
Removal
During Dam
Removal
Total # of data points for
analysis
5
3
8
5
7
4
3
9
10
4
4
3
3
7
7
25
16
41
Intact vs. Impaired Drift Cells
Overall, the intact Dungeness drift cell supported a significantly greater
abundance of surf smelt eggs than the impaired Elwha drift cell (DIFF=4.267,
p=0.018) (Figure 3). Surf smelt egg abundance in the intact treatment was almost
10 times greater than in the impaired treatment. Even though a greater number of
samples were collected in the impaired Elwha treatment (n=203) than in the intact
46
Egg Abundance in Impaired and Intact Drift Cells
14
12
DIFF=4.267
p=0.018
Average # eggs
10
8
6
4
2
0
Impaired Drift Cell
Intact Drift Cell
Figure 3. Surf smelt egg abundance in impaired and intact drift cells.
Dungeness treatment (n=176), only 13 (6%) samples from the impaired drift cell
were egg-bearing, while 73 (41%) samples from the intact treatment were eggbearing. The difference in the number of eggs is also striking, with a total of 85
surf smelt eggs found in the impaired drift cell compared to a total of 993 surf
smelt eggs found in the intact drift cell. All eggs found in the impaired Elwha drift
14
12
Egg Abundance During Dam
Removal
Egg Abundance Before Dam
Removal
14
DIFF=4.142
p=0.160
12
10
Average # eggs
10
Average # eggs
DIFF=4.414
p=0.035
8
6
4
2
8
6
4
2
0
0
Impaired Drift Cell Intact Drift Cell
Impaired Drift Cell Intact Drift Cell
Figure 4. Surf smelt egg abundance in impaired and intact drift cells a) before and b)
during dam removal.
47
cell were collected in the Freshwater Bay site, and all eggs found in the intact
Dungeness drift cell were collected in the Dungeness Bluffs site. The Dungeness
Bluffs site is the most productive beach within our study area and is the site
responsible for causing the intact drift cell to consistently yield a greater
abundance of surf smelt eggs than the impaired drift cell, both before and during
the dam removal process (Figure 4).
Before vs. During Dam Removal
We compared all surf smelt samples taken before dam removal (n=157) to all
samples collected during dam removal (n=222) and found that surf smelt egg
abundance did not differ significantly between the two time periods (DIF=0.034,
p=0.915) (Figure 5). Surf smelt egg counts seem to track sampling effort, with a
total of 457 eggs found before dam removal and 617 eggs found during the dam
removal time period. We also compared egg abundance between the two time
Egg Abundance Before and During Dam Removal
DIFF=0.630
p=0.744
7
Average # eggs
6
5
4
3
2
1
0
Before Dam Removal
During Dam Removal
Figure 5. Surf smelt egg abundance across the combined impaired and intact drift cells
before and during dam removal.
48
Egg Abundance in the Impaired
Drift Cell
14
12
Average # eggs
10
8
6
4
2
DIFF=0.191
p=0.959
14
DIFF=0.463
p=0.923
12
Average # eggs
Egg Abundance in the Intact
Drift Cell
10
8
6
4
2
0
0
Before Dam
Removal
During Dam
Removal
Before Dam
Removal
During Dam
Removal
Figure 6. Surf smelt egg abundance before and during dam removal in the a) impaired
and b) intact drift cells.
periods in both the impaired and intact drift cells (Figure 6). The change in surf
smelt egg abundance between before and during dam removal is not significant
for either drift cell treatment. This might be expected for the intact drift cell since
the nearshore sediment processes, by definition, remained intact between the two
time periods. However, the dam removal project has apparently not yet had the
expected boosting effect on surf smelt spawning activity in the impaired Elwha
drift cell despite the changes this process has brought to the nearshore sediment
supply and beach composition. Instead, surf smelt egg abundance within the
impaired drift cell appears to have decreased (~30% less), although not
significantly, during the dam removal process (Figure 6a).
Freshwater Bay remained the only beach within the impaired drift cell to
support surf smelt spawning activity throughout the duration of this study (Figure
7). Although egg abundance decreased by about 38% in the during dam removal
49
Egg Abundance in Freshwater Bay
7
DIFF=1.187
p=0.963
Average # eggs
6
5
4
3
2
1
0
Before Dam Removal
During Dam Removal
Figure 7. Surf smelt egg abundance in the impaired Freshwater Bay before and during
dam removal.
phase, spawning not only continues to occur in those areas we documented as surf
smelt spawning habitat prior to the beginning of dam removals, but appears to
have expanded during the dam removal time period (Figure 8). Surf smelt are now
using areas further to the east, close to the Elwha River mouth, where there was
Figure 8. Freshwater Bay samples containing surf smelt eggs before and during dam
removal.
50
no suitable habitat prior to dam removal (see also Figure 19 and 20 in the
Appendix).
Geomorphic Habitat Type
The differences in surf smelt egg productivity by geomorphic habitat type
(GMHT) were not significantly different between embayment, bluff, and spit sites
(SSamong=151.814, p=0.132) (Figure 9). Bluff sites, as a whole, supported the
greatest surf smelt egg abundance, almost three times that of embayments. Spit
sites did not appear to support surf smelt spawning activity.
Surf smelt spawning activity occurred within different GMHTs between
the impaired and intact drift cell treatments (Figure 10). Within the intact drift
cell, only the bluff GMHT (Dungeness Bluffs) supported surf smelt spawning
activity; intact embayment and spit sites did not support any spawning activity. In
contrast, within the impaired Elwha drift cell, only the embayment (Freshwater
Egg Abundance by Geomorphic Habitat Type
14
SSAmong=151.814
p=0.132
12
Average # eggs
10
8
6
4
2
0
Embayment
Bluff
Spit
Figure 9. Surf smelt egg abundance by geomorphic habitat type (GMHTs).
51
Egg Abundance in Impaired
Geomorphic Habitat Types
Egg Abundance in Intact
Geomorphic Habitat Types
25
25
SSAmong=16.276
p=0.016
15
10
5
DIFF=8.063
p=0.053
20
Average # eggs
Average # eggs
20
15
10
5
0
0
Embay.
Bluffs
Spit
Bluffs
Spit
Figure 10. Surf smelt egg abundance by geomorphic habitat type (GMHT) within a)
impaired and b) intact drift cell treatments.
Bay) supported surf smelt spawning activity; none of the samples taken in the
Elwha’s bluff or spit GMHTs contained any eggs. Egg abundance between these
two GMHTs differed greatly, with intact bluffs yielding more than 5.6 times
greater egg abundance than impaired embayment. Given the productivity of the
intact Dungeness Bluffs site, it is striking that no surf smelt eggs were found in
any of the 90 samples collected at the Elwha Bluffs site (see Figure 15 in the
Appendix). Surf smelt spawning activity also differed in its magnitude among the
two productive GMHTs. Of the 60 samples collected in the impaired Freshwater
Bay site, only 13 (22%) samples contained one or more surf smelt eggs, while in
the in intact Dungeness Bluffs site, 73 (61%) of the 120 samples were eggbearing.
52
DISCUSSION
Despite the documented presence of abundant juvenile sand lance in the Elwha
and Dungeness drift cells (Shaffer et al., 2012), we were unable to find any sand
lance eggs at any of our sites during the course of this study. Other surveys have
succeeded in finding sand lance spawning activity within our study area, as well
as along adjacent portions of the Strait of Juan de Fuca shoreline (Moriarity et al.,
2002b; Penttila, 1995b; WDFW, 2014). However, unlike our sites, these
spawning areas were located within sheltered embayments, such as along the
inner margins of Dungeness Spit and Ediz Hook, and within Sequim Bay,
Discovery Bay, and a protected embayment near the Pysht River. It may be that
sand lance prefer a low-energy beach spawning habitat and that the beach sites
within our survey area are too exposed to tidal and wave energy, thus
discouraging their spawning behavior. It may also be simply too early in the
nearshore restoration process as beaches rapidly change in response to the influx
of Elwha River sediment, and that sand lance will begin spawning in upcoming
years. Previous work documenting forage fish spawning habitat has shown spatial
and temporal variability in habitat conditions and unpredictable fluctuations in
spawning behavior, revealing the necessity to conduct multiyear surveys to
accurately define spawning areas (Moriarity et al., 2002b; Parks et al., 2013;
Penttila, 2007; Quinn et al., 2012). The Coastal Watershed Institute will continue
to survey these beaches during the post dam removal time period.
The results of our surf smelt surveys show that areas of the central Strait
of Juan de Fuca are actively used by this species as spawning habitat. Spatial
53
patterns of surf smelt spawning activity reveal preferences for habitat type and
illustrate the important role that sediment processes play in forming favorable
habitat conditions. The greatest abundance of surf smelt eggs was consistently
found at the Dungeness bluffs site. Importantly, these bluffs remain unarmored.
Placing armoring at the base of coastal feeder bluffs in this region has been shown
to reduce bluff recession rates by 50-80% (Kaminsky et al., 2014), significantly
decreasing the amount of new sediment delivered to nearshore beaches of the size
and composition required by surf smelt as spawning habitat. It is significant that
the Elwha bluffs site, which is mostly armored, appears not to provide any surf
smelt spawning habitat as none of our samples collected at the site yielded eggs.
The Elwha bluffs only produce half as much sediment per alongshore distance as
the Dungeness bluffs (Kaminsky et al., 2014), resulting in beaches which are
relatively starved of sediment and unfavorable as spawning habitat. We conclude
that the greater supply of sediment provided to coastal beaches by unarmored
bluffs is connected to our observations of favorable surf smelt spawning habitat
along unarmored stretches of bluffs.
Comparing surf smelt egg abundance between impaired and intact drift
cells as a whole demonstrates the importance of sediment processes operating at a
larger scale. We found that, overall, the drift cell with intact sediment processes
(the Dungeness) clearly supports a significantly greater abundance of surf smelt
eggs than the impaired drift cell (the Elwha). While this result is obviously
connected to the relative egg abundances found in the Dungeness bluffs and
Freshwater Bay sites as mentioned above, it has important implication in its own
54
right. Because sediment is continually entering and moving through a drift cell
system, efforts to restore specific beach locations may be thwarted if they cannot
be linked to larger-scale, intact sediment processes. Having intact sources of
sediment, and shorelines free of impediments to its movement, is important for
maintain longshore sediment drift and crucial for sustaining nearshore habitat
along the entire length of the drift cell. As Chinook and coho salmon from as far
away as the Columbia and Kalamath River systems have been found utilizing
Strait of Juan de Fuca shorelines (Shaffer et al., 2012), nearshore habitat within
intact drift cell systems can have both regional and cross-regional benefits.
Defining priorities for the preservation and restoration of nearshore processes and
ecological function is therefore most appropriately accomplished at the drift cell
scale.
The differences between spawning activity within different GMHTs were
not surprising since GMHTs can function differently for different species and in
different sites (Shaffer et al., 2012). Accretionary shore forms such as sandy spits
that form at the distal ends of drifts cells are known to support sand lance
spawning habitat in other locations within Puget Sound, but may limit surf smelt
spawning by their overly fine, sandy character (Penttila, 2007). The spit GMHTs
encompassed within this study appeared to be used very sparsely for spawning, or
possibly not at all since the origin of the few eggs found on Dungeness Spit are
uncertain and no eggs were found on Ediz Hook. The four dead eggs that were
each found in separate samples from Dungeness Spit were all empty surf smelt
egg shells that were likely spawned on the Dungeness Bluff beach to the west and
55
transported to the spit by currents. This drift of eggs may indicate that beaches
with prolific spawning activity could provide a source of “seed eggs” to newlyformed beaches with suitable habitat where they could incubate and hatch to form
new surf smelt populations.
Surf smelt have continued to use the impaired Freshwater Bay as
spawning habitat during the dam removal phase. Not only are we are seeing
continued use in the areas we documented as surf smelt spawning habitat prior to
the beginning of the dam removals, but, interestingly, it appears that the spawning
area in Freshwater Bay may be expanding to the east, adjacent to the growing
river delta. Surf smelt are now using areas where there was no suitable spawning
habitat prior to dam removal; however, overall egg abundance within this
embayment has not yet changed significantly since the beginning of dam removal.
The event of two dam removals on the Elwha River and the subsequent,
ongoing delivery of fluvial sediment into the Elwha nearshore system have not yet
had a strong effect on spawning behavior on the beaches of the impaired Elwha
nearshore. However, the dam removal process is still ongoing and the post-dam
removal response and restoration of the nearshore has not yet begun. It is
therefore too early to determine the response in surf smelt (and sand lance)
spawning to this dam removal event. The effects of the dam removal process on
the Elwha nearshore environment will clearly be a long-term process and will
require continued monitoring to detect trends and outcomes in the ongoing
ecological response.
56
Surf smelt spawning activity in both the impaired (Elwha) and
comparative intact (Dungeness) drift cell varied considerably between the two
time periods. The nearshore is inherently a variable system and these findings are
consistent with other observations that the abundance of forage fish in a localized
region can fluctuate from year to year in response to factors such as inter-annual
variations in beach substrate composition (Parks et al., 2013) as well as factors
such as ocean conditions, recruitment success, pressure from predators and
fisheries, and habitat quantity and quality (Liedtke et al., 2013). Variability in
year to year forage fish use of the same beaches has also been observed within the
Puget Sound region where only a small fraction of beaches with appropriatelysized sand and gravel substrate are used for spawning in any given year; in fact,
the majority of Puget Sound beaches that appear to have the suitable substrate and
habitat structure to support spawning are not documented surf smelt spawning
sites (EnviroVision et al., 2007; Moulton & Penttila, 2000; Quinn et al., 2012).
Further research into habitat selection and inter-annual and longer-term cycles of
forage fish usage could expand our knowledge of this poorly-understood
phenomenon.
It is still very early in what will surely be a long-term nearshore
restoration of the Elwha system. The composition and timing of sediment
mobilization depends on the rate and stage of dam removal, local morphology, the
driving riverine and marine hydrology during and in the years following the
removal, and the amount and grain size of the sediment, particularly in the
reservoirs (Czuba et al., 2011a; Draut & Ritchie, 2013; Randle et al., 1996). As of
57
spring 2013, a total of 6.1 x 106 m3 of sediment had been mobilized downstream
from the deposits in both reservoirs (Draut & Ritchie, 2013), representing only
about 20% of the total 13 to 20 x 106 m3 projected sediment load to be released
into the Elwha system over the next several years (Randle & Bountry, 2012).
Much of this early-stage sediment release has been very fine-grained material
which has formed ubiquitous mud deposits along the Elwha River channel
margins and floodplain instead of being exported to the coast as was expected.
This was largely due to an unusual lack of winter flood flows in the winters of
2011 and 2012 resulting in unusually low fluvial transport capacity (Draut &
Ritchie, 2013). As dam removal progresses, the coarser sand and gravel sediment
fractions are expected to be increasingly mobilized and released downstream over
the next 7-10 years (Czuba et al., 2011a). The potential volume of sand and gravel
is substantial; 50% of the total sediment (21.6 ± 3.0 x 106 m3) in the upper Lake
Mills reservoir and 32% of the lower Lake Aldwell reservoir total sediment (4.6 ±
1.5 x 106 m3) is estimated to be sand and gravel (Czuba et al., 2011a; Draut &
Ritchie, 2013). The delivery of this sand and gravel sediment to the Elwha
nearshore will change the abundance and distribution of suitable surf smelt
spawning habitat. These changes present an opportunity to investigate and better
understand shifts in habitat form and function and associated spatial and temporal
patterns of surf smelt usage in the Elwha nearshore.
The bulkheads at Elwha bluffs represent an outstanding and long-term
restoration issue. Historically, feeder bluffs provided an estimated 70% of the
sediment contribution to the entire Elwha drift cell (Parks et al., 2013) and 85% of
58
the sediment that formed and sustained Ediz Hook (Galster, 1989). Currently,
68% of the entire length of these feeder bluffs are now armored with bulkheads
(Flores et al., 2013; Kaminsky et al., 2014). The prevention of the sustained
erosional input of sediment from these bluffs has significantly impaired the
sediment processes of the Elwha nearshore in the past; their persistence will likely
continue to impair long-term sediment processes as long as they remain in place.
After the initial pulse of fluvial sediment into the Elwha nearshore system
resulting from dam removal, the delivery of sediment from the Elwha River is
expected to reduce and equilibrate around the pre-dam rate of 120,000 to 290,000
m3 of sediment annually (BOR, 1996; Czuba et al., 2011a). While this sediment
input will benefit the unarmored beaches near the river mouth, including the
unarmored Freshwater Bay, the feeder bluffs of the Elwha drift cell will still
remain armored. It is unclear whether this sediment will accumulate along the
armored portion of the Elwha Bluffs beach. The Army Corps of Engineers, the
City of Port Angeles, and The Coastal Watershed Institute are currently working
to determine what may occur in this regard. Sediment accumulation along
armored areas has been shown to be limited to the shorter beach face below the
base of the armoring structure and is less likely to accumulate and persist in the
high-energy beach environment associated with armored shorelines (Johannessen
et al., 2014; Johannessen & MacLennan, 2007; Rice, 2006). In order to take
advantage of the restorative pulse of Elwha River sediment arriving on nearshore
beaches, an adaptive management approach is needed to respond to changing
conditions, management needs, and best scientific information as it becomes
59
available. Specific actions to enhance nearshore restoration may include the
placement of large woody debris and beach nourishment with cobble in order to
capture and retain the Elwha sediment as it arrives (Clancy et al., 2009; Rich et
al., 2014; Shaffer, 2013). The preservation of Freshwater Bay, which remains
unarmored, as well as intact adjacent bluff areas will be important restoration and
conservation actions for the entire coastal region.
Healthy nearshore ecosystems support sustainable economic activity and
can provide a wide variety of valuable public benefits. The value of nearshore
ecosystem services along the Strait of Juan de Fuca, such as carbon storage and
sequestration as well as habitat creation for fish and wildlife, including for forage
fish, has been estimated to contribute more than $15 million annually to the local
and regional economies (Flores et al., 2013). The sediment transfer value of
feeder bluffs within the Dungeness and Elwha drift cells contribute between
$99,000 to $506,000 every year, with intact, unarmored bluffs providing more
value than armored sections of the shoreline (Flores et al., 2013). However,
nearshore ecological function depends on maintaining those processes that shape
its physical form. Intact sediment processes are crucial for sustaining beaches that
provide forage fish spawning habitat. Documented sites of forage fish spawning
habitat are currently protected from net loss through Washington State’s
Hydraulic Code (WAC 220-110) and by shoreline master programs and critical
area ordinances, but widespread privatization of tidelands throughout Puget
Sound may necessitate further regulations in order to ensure effective stewardship
of the public’s forage fish resources. Interest is also growing in armor removal
60
projects in the region (Johannessen et al., 2014). However, the amount of new
armoring constructed in Puget Sound continues to outpace the amount of
armoring that is removed every year (Puget Sound Partnership, 2013). The
persistence of armored shorelines within the Elwha drift cell represents a
continued and long-term impairment of nearshore sediment processes and thus, of
ecological function. The massive influx of sediment to the Elwha nearshore
resulting from the dam removal project represents an unprecedented opportunity
to promote ecosystem restoration at a drift cell scale. However, this opportunity
will be short lived, so the time to take advantage of it is right now.
61
Chapter 3: Restoration and Management of the Elwha Nearshore
Introduction
The Strait of Juan de Fuca is a critical migratory corridor for federally threatened
Chinook and Hood Canal summer chum salmon, as well as a number of other
culturally and economically important marine fish and wildlife species that
migrate to and from the Pacific Ocean. This region’s nearshore environment,
often defined as extending from the upland coastal bluffs and riparian forests to
the shallow offshore waters of about 30 meters depth (Shaffer et al., 2008), is an
important zone that provides spawning, rearing, and forage habitat for a number
of bird, fish, and marine mammal species. The susceptibility of the nearshore
environment to anthropogenic impacts, combined with its great ecological value,
make it an important area for heightened measures of protection as well as for
efforts at restoration and stewardship. The nearshore of the central Strait of Juan
de Fuca offers a unique location for learning about the relationship between
coastal geomorphic processes, physical habitat form and ecological function. The
sediment recruitment and transport processes of the Dungeness drift cell remain
largely intact and support the creation and maintenance of functioning nearshore
habitat, including that which supports forage fish spawning. In contrast, the
sediment processes of the Elwha drift cell have been impaired for over a century
by ongoing industrial and urban development, including extensive shoreline
modifications by the armoring of bluffs and spit, and the construction of two inriver dams and lower river dikes on the Elwha River. These alterations, and the
62
consequent sediment starvation of the Elwha nearshore, have caused the erosion
and coarsening of beaches and the degradation of nearshore habitat.
The stark contrasts in ecological form and function evident between these
two adjacent segments of the central Strait of Juan de Fuca coastline offer lessons
for coastal management practices as well as for restoration actions. The ecological
health and function of those coastal areas with still-intact ecological function can
serve as an example of potential function achievable by the proper restoration of
degraded areas. In turn, areas of impaired ecological processes and degraded
function can serve as a cautionary tale, an example to be avoided by proper
management practices. In order to pursue both ecological restoration in the Elwha
nearshore and management recommendations in areas of intact ecological
processes requires an interdisciplinary approach. Sound science must inform
policy decisions and the wide range of stakeholders must be considered and
incorporated into the decision process. Historical and present conditions of
nearshore areas must be considered with the best available science and used to
inform our calculations of the consequences of proposed actions. Collaborations
between scientists and local governments are enhanced with input from the local
residents and property owners that would be affected by management decisions.
The present conditions in the nearshore of the central Strait of Juan de Fuca vary
from degraded urban shorelines and armored feeder bluffs to drift cells with intact
sediment processes supplying and maintaining beach and spit habitats. This range
in ecological function spans both a spatial scale as well as a temporal one since
conditions are changing fast and action must be taken immediately. The potential
63
to optimize the sudden influx of 100 years of Elwha River sediment to the
nearshore offers an unprecedented restoration opportunity that requires the
unusual blend of careful consideration and speedy action from the perspective of a
wide range of disciplines. Making the best ecological and community decisions
necessitates weighing how management decisions will affect not only the ecology
of the nearshore, but also the impacts to the region’s economic activity and its
affect on people’s lives in both the short-term as well as the long-term. Good
management decisions occur at the intersection of science and policy. Coastal
geomorphology, ecology, and technology can intersect with such disciplines as
history, economics, law, and anthropology to inform and equip policy makers
with a vision of management that can be effective and respond adaptively to the
needs of the environment and the community.
This chapter is about nearshore management in the Elwha and Dungeness
drift cells. The release of 100 years worth of fluvial sediment into the Elwha
nearshore as a result of the dam removal project represents a unique and
unprecedented opportunity to restore the sediment-starved and ecologicallydegraded Elwha nearshore. However, a number of challenges exist. Lessons
learned from unwise management practices of the past, as well as the attempts at
ecological restoration in the present and near future, can offer a cautionary tale
and insight into how to approach questions of management in other areas of the
nearshore. This chapter begins with an assessment of nearshore conditions in the
central Strait of Juan de Fuca. Our attention then focuses on the degraded
ecological conditions of the Elwha nearshore and the history of how its feeder
64
bluffs and Elwha River sediment input processes became so impaired. We then
turn our attention towards the removal of the two Elwha River dams and what the
release of long-trapped reservoir sediment could mean for Elwha nearshore
restoration. Shoreline armoring in long stretches of the Elwha drift cell pose a
number of challenges to full restoration of its nearshore. We illustrate some of
these challenges with an examination of the unusual situation at the City of Port
Angeles Landfill site and its associated seawall at the base of the Elwha feeder
bluffs. We then examine some of the restoration questions and actions that have
been proposed to address these challenges, such as the role that large woody
debris might play in recruiting and stabilizing the influx of sediment. Ecosystem
service valuation can also be a management tool that is useful for prioritizing
restoration and conservation goals, and help justify investments in environmental
restoration. We look at the findings of a report on the value of ecosystem services
of the Elwha nearshore, including its feeder bluffs, before comparing these values
to nearshore function in the intact Dungeness drift cell. Such a comparison reveals
the great ecological (and economic) value of intact areas of the shoreline and
illustrates the need to ensure the continued protection of these areas.
Nearshore conditions of the central Strait of Juan de Fuca
The Strait of Juan de Fuca’s nearshore environment appears to be in a generally
healthy and unaltered state, although a few important exceptions require attention.
Clallam County’s Inventory and Characterization Report (ESA et al., 2012))
states that the processes shaping and maintaining the nearshore ecosystem along
65
its shoreline are some of the least altered in the entire Puget Sound basin (see also
City of Port Angeles, 2012a). An extensive assessment by the Puget Sound
Nearshore Ecosystem Restoration Project (PSNERP) supports this assertion when
ranking the level of degradation for a number of ecosystem processes within the
various sub-basins of Puget Sound. In their assessment of sediment processes,
they found generally low levels of degradation for sediment input and transport,
and for the erosion and accretion of sediments along the Strait of Juan de Fuca
shoreline (Schlenger et al., 2011). For instance, while shoreline armoring
cumulatively occurs along 27 percent of the entire Puget Sound Basin (and as
high as 63% of the south central Puget Sound sub-basin), the Strait of Juan de
Fuca sub-basin was among those with the least shoreline armoring (16%), and had
one of the longest average length of shoreline reach with no shoreline armoring
(17.2 km; an average of the 10 longest reaches within the sub-basin) (Schlenger et
al., 2011). The Strait of Juan de Fuca also had long portions of shoreline
characterized as “Less Degraded” or “Least Degraded” by environmental
“stressors”; a suite of 12 quantifiable anthropogenic modifications known to
impair nearshore processes. The average length of the 10 longest shoreline
reaches in the Strait of Juan de Fuca sub-basin with no stressor was 12.8 km
(compared to an average of 2.9 km found in the south central sub-basin), and the
longest reach with no stressor in the entire Puget Sound study area was a 38.2 km
reach also found in the Strait of Juan de Fuca sub-basin (Schlenger et al., 2011).
While much of the Strait of Juan de Fuca shoreline remains in an un-degraded
66
state, anthropogenic alterations to the shoreline have occurred in a few areas and
have impaired the ecological function of its nearshore.
The Elwha Drift Cell
The Elwha drift cell is a glaring exception to the relatively unaltered shoreline
environment of the north Olympic Peninsula. Located in the central Strait of Juan
de Fuca, the Elwha drift cell spans approximately 21 km of shoreline from the
western extent of Freshwater Bay to the eastern tip of Ediz Hook and
encompasses a mosaic of shoreline habitats including 6 km of embayment
(Freshwater Bay), the Elwha River estuary, 4.9 km of feeder bluffs (Elwha
Bluffs), and a 5.5 km spit (Ediz Hook) (Figure 2 in previous chapter). However,
years of urban and industrial development in the area have impacted the shoreline
of the central Strait, severely impairing the sediment processes of the Elwha drift
cell and degrading its ecological function. For instance, the beach habitats of the
Elwha nearshore have been documented to be less suitable for forage fish
spawning than comparative beaches in the adjacent Dungeness drift cell (Parks et
al., 2013). In fact, the only process units in the entire 329 km Strait of Juan de
Fuca sub-basin classified as “Most Degraded” encompass the City of Port
Angeles and Ediz Hook, areas with highly modified shorelines that are almost
completely armored (Schlenger et al., 2011; USCOE, 1971). Intact sediment
processes are crucial components of a healthy nearshore. However, disruption to
the sediment processes in the Elwha drift cell has occurred primarily from
extensive armoring of the Elwha feeder bluffs. In addition, the construction of two
67
in-river dams on the Elwha River has contributed further to the sediment
starvation observed on the beaches of the Elwha nearshore.
The Elwha feeder bluffs are critically important to the overall sediment
budget of the Elwha drift cell, but their contribution has been impacted by a series
of shoreline armoring projects over the past 80 years or so. Large volumes of
continually-eroded sediment from the Elwha bluffs are largely responsible for the
formation and maintenance of Ediz Hook, a long spit lying to the east of the bluffs
at the distal, depositional end of the Elwha drift cell. When the sea level
essentially stabilized about 5,000 years ago (Downing, 1983), the Elwha bluffs
lay 900 to 1,500 meters to the north of their present position (Galster, 1989). The
steady erosion of these bluffs supplied an estimated 70% of the sediment
contribution to the Elwha littoral system (Kaminsky et al., 2014; Parks et al.,
2013) and 85% of the sediment that formed and sustained Ediz Hook, even
allowing a progressive extension of the spit by about 1.5 m/yr (Galster, 1989).
The remainder of sediment influx to the Elwha drift cell was furnished by the
fluvial sediment of the Elwha River until the construction of two dams in the early
1900s largely curtailed its transport and delivery to the Elwha nearshore. In 1930,
an industrial waterline was buried along 5.3 km of the toe of the Elwha bluffs, and
a series of armoring projects were completed in 1961 to protect over 2 kilometers
of the pipeline. By 1961, the cumulative effect of dam construction and bluff
armoring had reduced the sediment budget sustaining Ediz Hook by 89% (Galster,
1989; USCOE, 1971). The dramatic reduction in the supply of littoral sediment to
Ediz Hook caused an alarming rate of erosion of the spit itself, prompting the
68
Army Corps of Engineers to design a massive revetment and beach nourishment
project to protect the spit (and the Port Angeles harbor sheltered behind it), which
was completed in 1977-78 (Galster, 1989). Almost the entire length of the spit
remains armored today but continues to erode and requires periodic nourishment
with sand and gravel (USCOE, 2002). At present, 68% of the entire length of the
Elwha feeder bluffs are armored (Flores et al., 2013; Kaminsky et al., 2014). Of
the approximately 3 km of bluffs within the Port Angeles City limits, 91% are
now armored with a rock revetment (City of Port Angeles, 2012b). Armoring has
dramatically slowed the processes of bluff erosion (Kaminsky et al., 2014) and
severely starved the Elwha nearshore of its replenishing sediment supply.
Adding to the Elwha nearshore sediment starvation was the dramatic
reduction in the fluvial sediment contribution of the Elwha River. Two in-river
dams on the Elwha River impounded an estimated 21 to 26 × 106 m3 of sediment
in their reservoirs and reduced fluvial sediment transport to the coastal waters to
about 2% of the pre-dam load (Draut & Ritchie, 2013). Without the replenishing
input of Elwha River sediment, approximately 100,000 m2 of coastal plain within
the Elwha delta was lost to increased erosion of the coastline between 1939 to
2006, and coastal beaches underwent a dramatic coarsening of their substrate
(Warrick et al., 2009). While shoreline armoring has had the most significant
impact on Elwha nearshore sediment processes, the construction of the two inriver dams contributed to the dramatic sediment starvation observed throughout
all the beaches of the Elwha drift cell. The combined effect of these two impacts
has been to significantly impair the sediment delivery processes to the Elwha
69
nearshore and highlights the role that intact coastal geomorphic processes play in
creating and sustaining nearshore ecosystem structure and function.
Disrupting the continual input of sediment into a drift cell can change the
physical characteristics of downdrift beaches, from changing the composition of
substrate material to altering the beach slope and width (Fresh et al., 2011). A
drift cell is a segment of the shoreline along which sediment moves at a
measureable rate and direction depending on wave energy and currents and
includes sources of sediment (such as bluffs and river mouths), a zone of
transportation, and an area of deposition. Modifying the shoreline to interfere with
sediment input or its transport can affect the structure of downdrift beaches.
Observed effects of reduced sediment supply in the Elwha drift cell include the
steepening of the beach profile of Ediz Hook (City of Port Angeles, 2012a) as
well as the coarsening and higher variability in grain-size of beach substrate
throughout the Elwha nearshore beaches (Parks et al., 2013). Such dramatic
changes to the physical structure of beaches reduces or degrades habitats for a
wide variety of marine plants and animals that require the presence of fine
sediment, including forage fish, shellfish, eelgrass, and birds (Penttila, 2007;
Schlenger et al., 2011). Indeed, the Elwha drift cell, with its degraded habitatforming sediment processes, was found to have lower fish species diversity and
richness than comparative areas with intact processes (Shaffer et al., 2012). The
Elwha nearshore has also been starved of deposits of large woody debris (LWD),
an important component of the nearshore ecosystems which provides structure
and stability to beaches and spits by helping to trap and retain sediment, buffer
70
wave energy, and prevent erosion of nearshore habitat (Clancy et al., 2009; Rich
et al., 2014). In addition to impounding sediment, the two dams on the Elwha
River prevented LWD delivery to the nearshore, and shoreline armoring has
prevented the recruitment and retention of large woody debris entering the Elwha
drift cell from riparian forests or from other bodies of water (Rich et al., 2014).
While shoreline armoring has been, and will continue to remain, a significant
impediment to restoration efforts, the dam removal project, and its associated
mobilization and delivery of sediment and LWD, represents an unprecedented
opportunity to restore ecosystem function to the Elwha nearshore.
Dam removal and the potential for nearshore restoration
A project to remove both dams on the Elwha River commenced in September of
2011 and is expected to be completed by the end of 2014. The two dams on the
Elwha River impounded an estimated 21 to 26 × 106 m3 of sediment in their
reservoirs since their construction (Draut & Ritchie, 2013). As of spring 2013, a
total of 6.1 x 106 m3 of sediment had been mobilized downstream from the
deposits in both reservoirs (Draut & Ritchie, 2013), representing only about 20%
of the total 13 to 20 x 106 m3 projected sediment load to be released into the
Elwha system over the next several years (Randle & Bountry, 2012; Ritchie,
2013). Much of the early-stage sediment release has been very fine-grained
material but as dam removal progresses, coarser sand and gravel sediment
fractions are expected to be increasingly mobilized and released downstream over
the next 7-10 years (Czuba et al., 2011a). The potential volume of sand and gravel
71
is substantial; 50% of the total sediment (21.6 ± 3.0 x 106 m3) in the upper Lake
Mills reservoir and 32% of the lower Lake Aldwell reservoir total sediment (4.6 ±
1.5 x 106 m3) is estimated to be sand and gravel (Czuba et al., 2011a; Draut &
Ritchie, 2013). Large amounts of sand have already made their way down to the
river’s mouth and are moving into the Elwha drift cell (Ritchie, 2013; Warrick &
Gelfenbaum, 2013). The physical structure of beaches adjacent to the river mouth,
such as Freshwater Bay, is changing dramatically. Sandy substrate now lies where
coarse cobble made up the beach. As sediment continues to be delivered to the
Elwha nearshore, the volume and trajectory of its distribution throughout the drift
cell is unknown; however, its arrival to nearshore beaches downdrift could
potentially, even if partially, restore nearshore form and function.
The restoration of fluvial sediment inputs will occur in two phases. The
first phase is the delivery of large quantities of sediment released by the dam
removal project. As the Elwha River carves through and mobilizes the abundant
supply of unvegetated and unstable sediment in the former reservoirs, it will
deliver a multiple-year-long pulse of sediment to the nearshore that constitutes the
largest sediment release from a dam removal project in history (Draut & Ritchie,
2013).
The second phase will come after the supply of easily-mobilized reservoir
sediment is exhausted and the river resumes equilibrium with its supply of
normal, naturally-eroded sediment. Estimates of the pre-dam sediment load are
160,000 m3yr-1 of fine and coarse sediment ( Randle et al., 1996), or ~217 000–
513 000 t/year (Czuba et al., 2011a). The restoration of river sediment inputs to
72
the Elwha nearshore is an opportunity to promote restoration of nearshore
ecological function, but represents a temporary and only partial restoration of
sediment processes. The pulse of sediment associated with the dam removals will
be short-lived (7-10 years); the subsequent, normal annual fluvial sediment inputs
represent only a small fraction (~15%) of the total (bluff and fluvial) volume of
sediment that was historically delivered to the Elwha drift cell each year. The
Elwha bluffs, representing the bulk of sediment historically supplied to the
nearshore, will remain armored after the dam removals and therefore will
continue to deliver only a minor fraction of their pre-armoring volume (Kaminsky
et al., 2014). In the long run, with much of the bluffs (68%) and almost the entire
spit armored, ecological function in these areas of the Elwha nearshore will likely
remain impaired. Full restoration of the Elwha nearshore will be challenged by
these and other ongoing management issues.
Port Angeles landfill and other nearshore management issues
Even while the Elwha dams come down and restoration of the watershed begins,
large portions of the Elwha nearshore remains heavily managed. The Nippon
paper mill, located at the base of Ediz Hook, armors their shoreline regularly. The
Army Corps of Engineers performs routine maintenance work on an erosion
control project for Ediz Hook which consists of nourishing the spit’s beach with
gravel and cobble and re-keying revetment rocks that have fallen onto the beach
(USCOE, 2002). Lower Elwha River alterations, such as estuarine diking, will
also limit restoration of portions of the Elwha nearshore (Shaffer et al., 2008).
73
Another major management issue involves the future management of the Port
Angeles landfill site which is located atop the Elwha bluffs west of the city center
and managed by the City of Port Angeles.
The landfill was originally privately owned, predating the City, and was
purchased in 1947, becoming operational as a publicly owned city dump in 1979
(Figure 11). A number of pits, (referred to as East 304 cell, valley cell, and West
304 cell) were constructed near the edge of the bluff and filled with
approximately 575,000 cubic meters of raw garbage in the East and West 304
cells alone (Neal, 2013; Puntenney et al., 2013). The thin bluff wall acting as the
Land fill
Figure 11. Looking west along the base of the Elwha bluffs towards the City of Port
Angeles landfill in the distance, 1947. The photo shows a portal (center) for the industrial
water line buried along the toe of the Elwha bluffs and protected by shoreline armoring
(right). Photo courtesy of Coastal Watershed Institute.
74
sole barrier between the Strait of Juan de Fuca and large quantities of garbage
(~18 m deep) is highly unstable and is thinning at a rate of 0.6 to 1.8 m per year
due to wave action undermining the toe of the bluff and causing mass wasting
events (Neal, 2013). A 140 m long seawall was installed, without federal permits,
in 2006 to protect the West 304 and Valley cells but the installation increased
bluff erosion immediately down drift. Waste from East 304 cell became exposed
at the edge of the bluff in June of 2011, triggering concern that bluff failure could
result in landfill waste once again collapsing onto the Strait of Juan de Fuca
shoreline (Parks et al., 2014; Shaffer, 2013). This event sparked a number of
proposed design alternatives to address the problem. The city eventually decided
on a $21.2 million plan to dig up and transfer 202,600 m3 of waste in stages from
the East cell to another cell within the landfill located further inland, as well as
taking action to augment the ends of the existing seawall at the base of the bluff
with transitional energy-defusing scour protection in order to reduce erosion to
the adjacent unarmored shoreline (City of Port Angeles, 2013; Neal, 2013;
Schwartz, 2014). With the waste removed from East 304 cell, the bluff at that
location would be allowed to erode naturally onto the shoreline while city
managers continue to adaptively manage the site with continued monitoring of
bluff erosion rates and shoreline processes over the next 25 to 100 years. Future
management actions could include additional waste removal to allow continued
bluff erosion as well as seawall removal and replacement with softer shoreline
stabilization material if the wave energy and environment permit (Neal, 2013).
Future management decisions will depend, in part, on those actions taken today at
75
the landfill site. One of the important decisions facing managers today regards
what actions to take in order to best optimize the arrival of Elwha River sediment
to nearshore beaches. Actions that enhance the capture and retention of substrate
on the sediment-starved beaches of the Elwha drift cell, including the beach
below the landfill, will help to stabilize them as well as take a step towards
restoring ecological function.
Elwha nearshore restoration: questions and actions
The delivery of Elwha River sediment to the beaches of the Elwha drift cell could
potentially restore nearshore ecological function to its impaired bluffs and spit,
but little is known about the exact quantity, timing, location, grain-size, or
duration of sediment delivery. Additional questions remain as to whether, and for
how long, sediment would remain on beaches with armored shorelines, and
whether management actions could be taken to assist its capture and retention on
these beaches. Another unknown is whether the arrival of sediment to the beaches
below the city landfill would help ameliorate the ongoing issue of erosion at that
site. Answering these questions could help the city of Port Angeles define specific
restoration actions they could take in order to optimize the arrival of Elwha River
sediment and restore ecological habitat function to their hardened shoreline. The
Elwha Nearshore Consortium (ENC), a group of scientists, managers, and citizens
dedicated to understanding and promoting the restoration of the Elwha nearshore,
has pledged to help the city answer these questions and assist them in making the
best management decisions for the environment and the community. The ENC
76
advocates for an adaptive management approach that responds to the changing
conditions, management needs, and best science as it becomes available. Specific
actions may include the placement of large woody debris and beach nourishment
with cobble in order to capture and retain the Elwha sediment as it arrives
(Shaffer, 2013). The ENC also advocates for the preservation of Freshwater Bay,
which remains unarmored, and the restoration of portions of the lower Elwha
River which has undergone channelization and diking, resulting in restricted fish
use of parts of the tidal influenced estuary (Shaffer et al., 2008). Long-term feeder
bluff erosion rate studies should also be undertaken and incorporated into bluff
management decisions.
In addition to releasing large amounts of sediment, the removal of the
dams is also releasing LWD which could potentially help stabilize eroding Elwha
beaches and trap the new inputs of sediment as it arrives. However, if LWD is
prevented to recruit to beaches by the presence of riprap, the structural habitat
improvements may not be realized (Figure 12). To augment Elwha River LWD
inputs, active protection of intact riparian areas within the Elwha drift cell, such
as Freshwater Bay, and adjacent areas could help optimize restoration efforts and
create a more resilient and natural nearshore habitat for forage fish and other
wildlife. Increasing the amount of LWD on nearshore beaches, especially along
Elwha Bluffs and Ediz Hook, with root wads and branches, and limiting and
reducing shoreline armoring would be first steps of active restoration practices in
the nearshore (Rich et al., 2014). Adding LWD and nourishing the beach with
appropriate-sized gravel has been successful elsewhere as a “soft shore” approach
77
to replace hard armoring while still offering protection from wave energy (Rich et
al., 2014).
The Elwha River Ecosystem and Fisheries Restoration Act (Elwha Act,
Public Law 102-495) of 1992 calls for “full restoration of the Elwha River
ecosystem and the native anadromous fisheries” (Section 3(c)). The restoration of
the Elwha nearshore is a crucial component to achieving the successful realization
of this goal because the nearshore is a bottleneck for salmon recovery. Salmon
depend on nearshore habitat as an important migration and forage corridor, and as
a crucial transition point between the freshwater of the river and saltwater of the
marine environment. However, Elwha nearshore restoration presents unique
Figure 12. City of Port Angeles installing landfill sea wall in 2005. Note large wood
unable to recruit to the beach due to newly installed riprap at base of Elwha bluffs. Photo
by Darlene Shanfeld.
78
management complexities as compared to the restoration of the Elwha River
watershed. While over 80% of the Elwha watershed lies within the Olympic
National Park (ONP), the Elwha nearshore lies entirely outside the ONP and is
owned instead by a complex mosaic of private, city, county, state, tribal, and
federal landowners. This matrix of various stakeholders makes decision-making
and coordination of restoration activities much more complex than that within the
watershed. Since 68% of the feeder bluffs within the Elwha drift cell will remain
armored long after the initial pulse of Elwha river sediment enters the nearshore
system, the mosaic of shoreline property owners, natural resource managers,
research scientists, and other stakeholders will have to consider which actions to
take in order to optimize the temporary pulse of sediment delivery, as well as to
minimize future continued degradation of the Elwha nearshore as the amount of
fluvial sediment drops to normal annual levels. Education of the local citizenry as
to the importance and benefits of intact nearshore processes will undoubtedly be
an essential component of building awareness and support for nearshore
restoration as a crucial part of full ecosystem recovery of the entire Elwha system.
Education could take place through workshops, presentations, newspaper
announcements and articles, and community college classes.
Another tool that can be used to both help promote and guide restoration
and management decisions is a consideration of the economic value of ecosystem
services that are provided by nearshore processes and features. Earth Economics,
a non-profit organization providing science-based, economic analysis of
ecosystem services released a report on the value of natural capital in Clallam
79
County focusing on nearshore processes, including feeder bluffs. They found that
the nearshore ecosystem services of carbon storage and sequestration, creation of
habitat, and forage fish supportive value contribute over $15 million annually to
the local and regional economies; commercial and recreational fishing provides a
minimum of $20 million annually; and feeder bluffs contribute on average
between $99,000 to $506,000 annually, the range depending on the health of the
shoreline processes and the presence or absence of shoreline armoring (Flores et
al., 2013). The difference in the value of sediment inputs between armored and
unarmored sections of feeder bluffs is striking; within the Elwha drift cell alone,
armored portions of bluffs had an estimated value of $2.97 to $5.94/foot/year,
while unarmored portions had an estimated value of $9.45 to $18.90/foot/year.
Ecosystem services valuation can be used to help managers prioritize restoration
and conservation goals, better understand the connections between the
environment and the economy, and help justify investing in environmental
outcomes within the context of pressure for economic development.
The Elwha as cautionary tale
The extent to which the Elwha drift cell’s sediment processes have been impaired
becomes apparent when compared to adjacent drift cells with intact sediment
processes. The Dungeness drift cell serves as an appropriate comparison because
it shares many of the same GMHTs and geomorphic processes, but is not
influenced by the presence of armoring along its feeder bluffs or by in-river dams.
Differences in nearshore processes and measures of ecological function make
80
powerful arguments not only for the restoration of impaired areas, but also for the
preservation of those areas that remain intact. For example, our finding of
increased surf smelt spawning in the intact drift cell can serve both as an example
of what might be achievable with restoration of the Elwha drift cell, as well as a
reason to ensure the continued preservation of the functioning Dungeness
nearshore. The degradation of nearshore function in the Elwha drift cell can serve
as a cautionary tale, illustrating to coastal managers a scenario of what to avoid
replicating in intact stretches of the nearshore environment.
Other measures of nearshore function, such as bluff retreat rates and
volumes of sediment inputs, can also support preservation efforts. The Dungeness
bluffs have been found to erode faster and contribute greater volumes of sediment
to the Dungeness drift cell than do the Elwha bluffs to the Elwha drift cell.
Kaminsky et al. (2014) estimate that the unarmored Dungeness bluffs produce
twice as much sediment per alongshore distance as the mostly-armored Elwha
bluffs (avg. 7.5 m3/m/yr vs. 4.1 m3/m/yr, respectively). The broad, flat, selfmaintained beaches of the Dungeness drift cell, supportive of surf smelt
spawning, are testament to the intact habitat forming processes of this portion of
coastline, and should be recognized as such when considering management
decisions. Coastal managers can use bluff recession rates in planning future land
use zoning and growth rates, and regulating setback distances from bluff edges for
new construction.
81
Conclusions
The project to remove both dams on the Elwha River, the largest project of its
kind in U.S. history, presents a unique opportunity for the restoration of the
impaired Elwha nearshore. The pulse of sediment released from the former
reservoirs is currently making its way down the river and entering the nearshore
environment, changing the character of its beaches and restoring the sediment
processes that shape and maintain nearshore habitat. However, this pulse of
Elwha River sediment is projected to be short-lived. After the un-consolidated,
easily-erodible reservoir sediment has been washed out of the system within 5 to
10 years, the river will likely resume its natural, but much lower, rate of sediment
contribution to the nearshore (Czuba et al., 2011a). The opportunity to take action
and optimize this event, therefore, is time sensitive.
A major obstruction to restoration, however, will persist in the Elwha
nearshore. Much of the Elwha Bluffs, which historically contributed the majority
of sediment to the Elwha drift cell, will remain armored with bulkheads and a sea
wall, thereby greatly reducing their rate of recession and sediment contribution,
and potentially interfering with the capture and retention of fluvial sediment as it
arrives on nearshore beaches. Nearshore restoration associated from dam
removals may therefore be temporary, and only partial. Coordination between
scientists, natural resource managers, and the various private, tribal, and
government stakeholders is required to address this and other problems, as well as
to plan and implement the best possible stewardship of this valuable resource.
82
Understanding the links between sediment processes and the impairment of
nearshore function should also be applied towards the preservation of those areas
of the nearshore environment that remain intact.
83
Bibliography
Bargmann, G. (1998). Forage Fish Management Plan - A plan for managing the
forage fish resources and fisheries of Washington. Washington Fish and
Wildlife Commission report. Washington Department of Fish and
Wildlife, 77 pp.
Beacham, T. D. (1986). Type, quantity, and size of food of Pacific salmon
(Oncorhynchus) in the Strait of Juan de Fuca, British Columbia. Fishery
Bulletin, 84(1), 77–90.
Beamer, E., & Fresh, K. (2012). Juvenile salmon and forage fish presence and
abundance in shoreline habitats of the San Juan Islands, 2008-2009: Map
applications for selected fish species. Skagit River System Cooperative,
LaConner, WA.
Brodeur, R. D. (1990). A synthesis of the food habits and feeding ecology of
salmonids in marine waters of the North Pacific. International North
Pacific Fisheries Commission document FRI-UW-9016. Fisheries
Research Institute, University of Washington, Seattle. 39 pp.
Bureau of Reclamation (BOR). (1996). Sediment analysis and modeling of the
river erosion alternative. Elwha Technical Series, PN-95-9. U.S.
Department of the Interior, Bureau of Reclamation, Pacific Northwest
Region, Boise, ID.
Carrasquereo-Verde, J., Abbe, T., & Morrison, S. (2005). Bulkheading in
Thurston County: Impacts on forage fish spawning habitat. Proceedings of
84
the 2005 Puget Sound Georgia Basin Research Conference. Herrera
Environmental Consultants.
Ciannelli, L. (1997). Winter dormancy in the Pacific sand lance (Ammodytes
hexapterus) in relation to gut evacuation time. In: Baxter B. (ed) Forage
Fishes in Marine Ecosystems. Proceedings of the international symposium
on the role of forage fishes in marine ecosystems. Report No. 97–01.
University of Alaska Sea Grant College Program, Fairbanks, p 816.
City of Port Angeles. (2012a). Shoreline inventory, characterization and analysis
report for the City of Port Angeles’ shoreline: Strait of Juan de Fuca. June
2012 Prepared for the City of Port Angeles, WA. The Watershed
Company, Makers Architecture + urban design, and Landau Associates.
City of Port Angeles. (2012b). Shoreline restoration plan for the City of Port
Angeles’ shoreline: Strait of Juan de Fuca. Prepared for the City of Port
Angeles, WA. The Watershed Company and Makers Architecture + Urban
Design. June 2012.
City of Port Angeles. (2013). Landfill bluff cell stabilization project, SW02-2012.
Comments and Questions from March 5, 2013 City Council meeting.
Clancy, M., Logan, I., Lowe, J., Johannessen, J., MacLennan, A., Van Cleve, F.
B., Dillon, J., Lyons, B., Carman, R., & Cereghino, P. (2009).
Management measures for protecting the Puget Sound nearshore. Puget
Sound Nearshore Ecosystem Restoration Project Report No. 2009-01.
Published by Washington Department of Fish and Wildlife, Olympia,
Washington.
85
Czuba, C. R., Randle, T. J., Bountry, J. A., Magirl, C. S., Czuba, J. A., Curran, C.
A., & Konrad, C. P. (2011a). Anticipated sediment delivery to the lower
Elwha River during and following dam removal. Chap. 2 of Duda, J.J.,
Warrick, J.A., and Magirl, C.S., eds. (2011), Coastal habitats of the Elwha
River, Washington—Biological and physical patterns and processes prior
to dam removal: U.S. Geological Survey Scientific Investigations Report
2011‑5120, p. 27-46.
Czuba, J. A., Magirl, C. S., Czuba, C. R., Grossman, E. E., Curran, C. A.,
Gendaszek, A. S., & Dinicola, R. S. (2011b). Sediment load from major
rivers into Puget Sound and its adjacent waters. U.S. Geological Survey
Fact Sheet 2011-3083, 4 p.
Downing, J. (1983). The coast of Puget Sound: Its processes and development.
University of Washington Press, Seattle.
Draut, A. E., Logan, J. B., & Mastin, M. C. (2011). Channel evolution on the
dammed Elwha River, Washington, USA. Geomorphology, 127(1), 71–87.
Draut, A., & Ritchie, A. C. (2013). Sedimentology of new fluvial deposits on the
Elwha River, Washington, USA, formed during large-scale dam removal.
River Research and Applications, Published online in Wiley Online
Library.
Dugan, J. E., Airoldi, L., Chapman, M. G., Walker, S. J., & Schlacher, T. (2011).
8.02 - Estuarine and Coastal Structures: Environmental Effects, A Focus
on Shore and Nearshore Structures. In E. Wolanski & D. McLusky (Eds.),
86
Treatise on Estuarine and Coastal Science (pp. 17–41). Waltham:
Academic Press.
Environmental Science Associates (ESA) in cooperation with Kramer Consulting,
Coastal Geologic Services and Ann Seiter. (2012). Clallam County
Shoreline Master Program Update: Shoreline inventory and
characterization report for portions of Clallam County draining to the
Strait of Juan de Fuca. Final report.
EnviroVision, Herrera Environmental, and Aquatic Habitat Guidelines Working
Group. (2007, revised 2010). Protecting nearshore habitat and functions in
Puget Sound. Washington Department of Fish and Wildlife Publication,
122 pp.
Finlayson, D. (2006). The geomorphology of Puget Sound beaches. Puget Sound
Nearshore Partnership Report No. 2006-02. Published by Washington Sea
Grant Program, University of Washington, Seattle, Washington.
Flores, L., Harrison-Cox, J., Wilson, S., & Batker, D. (2013). Nature’s Value in
Clallam County: The Economic Benefits of Feeder Bluffs and 12 Other
Ecosystems. Earth Economics: Tacoma, Washington.
Fresh, K. L. (2006). Juvenile Pacific Salmon in Puget Sound. Puget Sound
Nearshore Partnership Report No. 2006-06. Published by Seattle District,
U.S. Army Corps of Engineers, Seattle, Washington.
Fresh, K., Dethier M., Simenstad, C., Logsdon, M., Shipman, H., Tanner, C.,
Leschine, T., Mumford, T., Gelfenbaum, G., Shuman, R., & Newton, J.
(2011). Implications of observed anthropogenic changes to the nearshore
87
ecosystems in Puget Sound. Prepared for the Puget Sound Nearshore
Ecosystem Restoration Project (PSNERP). Technical Report 2011-03.
Galster, R. W. (1989). Ediz Hook—A Case History of Coastal Erosion and
Mitigation. Engineering Geology in Washington, Volume II. Washington
Division of Geology and Earth Resources Bulletin 78. Olympia,
Washington: Washington Department of Natural Resources, pp. 1177–
1186.
Hart, J. L. (1973). Pacific fishes of Canada. Fisheries Resource Board of Canada.
Bulletin 180, 740p.
Haynes, T. B., & Robinson, C. L. (2011). Re-use of shallow sediment patches by
Pacific sand lance (Ammodytes hexapterus) in Barkley Sound, British
Columbia, Canada. Environ Biol Fish, 92, 1–12.
Haynes, T. B., Robinson, C. L. K., & Dearden, P. (2008). Modeling nearshore
intertidal habitat use of young-of-the-year Pacific sand lance (Ammodytes
hexapterus) in Barkley Sound, British Columbia, Canada. Environmental
Biology of Fishes, 83(4), 473–484.
Haynes, T. B., Ronconi, R. A., & Burger, A. E. (2007). Habitat use and behavior
of the Pacific sand lance (Ammodytes hexapterus) in the shallow subtidal
region of southwestern Vancouver Island. Northwestern Naturalist, 88(3),
155–167.
Hershberger, P., Fagergren, D., Frazier, P., Stick, K., O’Toole, M., Bargman, G.,
Kirby, G., Dorn, P., Thorsteinson, L., Piatt, J., & Reisenbechler, R. (2006).
88
Puget Sound Forage Fish Plan. Technical Report. Puget Sound Action
Team. 18 pp.
Johannessen, J., & MacLennan, A. (2007). Beaches and Bluffs of Puget Sound
and the Northern Straits. Puget Sound Nearshore Partnership Report No.
2007-04. Published by Seattle District, U.S. Army Corps of Engineers,
Seattle, Washington.
Johannessen, J., MacLennan, A., Blue, A., Waggoner, J., Williams, S., Gerstel,
W., Barnard, R., Carmen, R., & Shipman, H. (2014). Marine Shoreline
Design Guidelines. Washington Department of Fish and Wildlife,
Olympia, Washington.
Kaminsky, G. M., Baron, H. M., Hacking, A., McCandless, D., & Parks, D. S.
(2014). Mapping and monitoring bluff erosion with boat-based LIDAR
and the development of a sediment budget and erosion model for the
Elwha and Dungeness littoral cells, Clallam County, Washington. Final
report to United States Environmental Protection Agency under grant
number PC-00J29801-0 awarded to the Washington Department of Fish
and Wildlife and managed by the Coastal Watershed Institute. 44 pp.
Konrad, C. P. (2009). Simulating the recovery of suspended sediment transport
and river-bed stability in response to dam removal on the Elwha River,
Washington. Ecological Engineering, 35(7), 1104–1115.
Liedtke, T., Gibson, C., Lowry, D., & Fagergren, D. (2013). Conservation and
Ecology of Marine Forage Fishes—Proceedings of a Research
89
Symposium, September 2012. U.S. Geological Survey Open-File Report
2013-1035, 24 p.
MacDonald, J. M., & Harris, N., eds. (2013). Proceedings of the 8th Annual
Elwha Nearshore Consortium Workshop, 27 February 2013, Port Angeles,
WA, 44 pp.
Middaugh, D. P., Hemmer, M. J., & Penttila, D. E. (1987). Embryo ecology of the
Pacific surf smelt, Hypomesus pretiosus (Pisces: Osmeridae). Pacific
Science, 41(nos. 1-4), 44–53.
Miller, B. S., Simenstad, C. A., Cross, J. N., Fresh, K. L., & Steinfort, S. N.
(1980). Nearshore fish and macroinvertebrate assemblages along the Strait
of Juan de Fuca including food habits of the common nearshore fish. FRIUW-8001. Fisheries Research Institute, University of Washington. 222 pp.
Moriarity, R. M., Shaffer, J. A., & Penttila, D. (2002a). Nearshore habitat
mapping of the central and western Strait of Juan de Fuca I: Surf smelt
spawning habitat. A final report to the Clallam County Marine Resources
Committee, Northwest Straits Commission, and Washington Department
of Fish and Wildlife.
Moriarity, R., Shaffer, J. A., & Penttila, D. (2002b). Nearshore mapping of the
Strait of Juan de Fuca IV: Pacific sand lance spawning habitat. A final
report to the Clallam County Marine Resources Committee, Northwest
Straits Commission, and Washington Department of Fish and Wildlife.
Moulton, L. L., & Penttila, D. E. (2000). Forage fish spawning distribution in San
Juan County and protocols for sampling intertidal and nearshore regions.
90
Report to Northwest Straits Commission. Mount Vernon, Washington, 36
pp.
Moulton, L. L., & Penttila, D. E. (2001, revised 2006). Field manual for sampling
forage fish spawn in intertidal shore regions. San Juan County Forage Fish
Assessment Project, 23 pp.
Mumford, T. F. (2007). Kelp and Eelgrass in Puget Sound. Puget Sound
Nearshore Partnership Report No. 2007-05. Published by Seattle District,
U.S. Army Corps of Engineers, Seattle, Washington.
Neal, K. (2013). Port Angeles landfill update and status. In Proceedings of the 8th
Annual Elwha Nearshore Consortium Workshop, 27 February 2013, Port
Angeles, WA, MacDonald and Harris, eds, 44 pp.
Ostrand, W. D., Gotthardt, T. A., Howlin, S., Robards, M. D., & Orr, J. W.
(2005). Habitat selection models for Pacific sand lance (Ammodytes
hexapterus) in Prince William Sound, Alaska. Northwestern Naturalist,
86(3), 131–143.
Parks, D., Shaffer, A., & Barry, D. (2013). Nearshore drift-cell sediment
processes and ecological function for forage fish: Implications for
ecological restoration of impaired Pacific Northwest marine ecosystems.
Journal of Coastal Research, 29(4), 984–997.
Parks, D., Shaffer, A., & Harris, N. (2014). Protecting the Strait of Juan de Fuca
nearshore through Shoreline Master Program improvements, bluff
development buffers and building setbacks, ecosystem services valuation,
and community stewardship: Field metric final report. A Coastal
91
Watershed Institute final report submitted to the Washington Department
of Natural Resources on 16 January 2014 in fulfillment of contract #121119. 56 pp.
Penttila, D. (1995a). The WDFW’s Puget Sound intertidal baitfish spawning
beach survey project. In Puget Sound Research ’95 Conference
Proceedings, Vol. 1, pp. 235-241. Puget Sound Water Quality Authority,
Olympia, Washington.
Penttila, D. (1978). Studies of the surf smelt (Hypomesus pretiosus) in Puget
Sound. Washington Department of Fisheries Technical Report 42, 45 pp.
Penttila, D. (2001). Grain size analysis of spawning substrates of the surf smelt
(Hypomesus) and Pacific sand lance (Ammodytes) on Puget Sound
spawning beaches. Washington Department of Fish and Wildlife, Marine
Resources Division, Manuscript Report.
Penttila, D. (2007). Marine forage fishes in Puget Sound. Puget Sound Nearshore
Partnership Report No. 2007-03. Published by Seattle District, U.S. Army
Corps of Engineers, Seattle, Washington, 27 pp.
Penttila, D. E. (1995b). Investigations of the spawning habitat of the Pacific sand
lance, Ammodytes hexapterus, in Puget Sound. In Puget Sound Research
’95 Conference Proceedings, Vol. 2, pp. 855-859. Puget Sound Water
Quality Authority, Olympia, Washington.
Penttila, D. E. (2002). Effects of shading upland vegetation on egg survival for
summer-spawning surf smelt on upper intertidal beaches in Puget Sound.
92
In Puget Sound Research-2001 Conference Proceedings. Puget Sound
Water Quality Action Team, Olympia, WA, 9 pp.
Pierce, K., Penttila, D., Benson, B., Krueger, K., Quinn, T., & Price, D. (2009).
Spatiotemporal detection of forage fish eggs derived from long-term
spawning surveys. Presentation poster from the 2009 Puget Sound Georgia
Basin Conference.
Puget Sound Partnership. (2009). Ecosystem Status & Trends. A supplement to
the 2009 State of the Sound Report, Puget Sound Partnership, Seattle,
WA, 131 pp.
Puget Sound Partnership. (2013). 2013 State of the Sound: A biennial report on
the recovery of Puget Sound. Tacoma, Washington, 177 pp.
Puntenney, M., Spillane, M., Bourque, T., & Parsons, J. (2013). Port Angeles
Landfill (PALF) Cell Stabilization and Protection. Presentation by city
engineer Mike Puntenney to the Port Angeles City Council during a work
session on January 29, 2013.
Quinn, T. (1999). Habitat characteristics of an intertidal aggregation of Pacific
sandlance (Ammodytes hexapterus) at a north Puget Sound beach in
Washington. Northwest Science, 73(1), 44–49.
Quinn, T., Krueger, K., Pierce, K., Penttila, D., Perry, K., Hicks, T., & Lowry, D.
(2012). Patterns of surf smelt, Hypomesus pretiosus, intertidal spawning
habitat use in Puget Sound, Washington State. Estuaries & Coasts, 35(5),
1214–1228.
93
Quinn, T., & Schneider, D. E. (1991). Respiration of the teleost fish Ammodytes
hexapterus in relation to its burrowing behavior. Comparative
Biochemistry and Physiology Part A: Physiology, 98(1), 71–75.
Randle, T. J., & Bountry, J. A. (2012). Elwha River Restoration: Sediment
Management. In Innovative Dam and Levee Design and Construction for
Sustainable Water Management, 32nd Annual United States Society on
Dams Conference, April 23-27, 2012, 871–886.
Randle, T. J., Young, C. A., Melena, J. T., & Ouellette, E. M. (1996). Sediment
analysis and modeling of the river erosion alternative. U.S. Bureau of
Reclamation, Pacific Northwest Region. Elwha Technical Series PN-95-9.
138 pp.
Reavey, K. (2007). The Elwha: A river and its people. Olympic Park Institute,
Port Angeles, WA. 8 pp.
Rice, C. A. (2006). Effects of shoreline modification on a northern Puget Sound
beach: Microclimate and embryo mortality in surf smelt (Hypomesus
pretiosus). Estuaries and Coasts, 29(1), 63–71.
Rice, J. (1995). Food web theory, marine food webs and what climate change may
do to northern marine fish populations, pp. 561-568. In R. J. Beamish [ed.]
Climate change and northern fish populations. Canadian Special
Publication of Fisheries and Aquatic Sciences 121.
Rich, S. L., Shaffer, J. A., Fix, M. J., & Dawson, J. O. (2014). Restoration
considerations of large woody debris in the Elwha River nearshore,
Olympic Peninsula, Washington. Ecological Restoration, 32(3), 306–313.
94
Ritchie, A. (2013). Elwha River sediment status. In Proceedings of the 8th Annual
Elwha Nearshore Consortium Workshop, 27 February 2013, Port Angeles,
WA, MacDonald and Harris, eds, 44 pp.
Robards, M. D., Piatt, J. F., & Rose, G. A. (1999a). Maturation, fecundity, and
intertidal spawning of Pacific sand lance in the northern Gulf of Alaska.
Journal of Fish Biology, 54(5), 1050–1068.
Robards, M. D., Willson, M. F., Armstrong, R. H., & Piatt, J. F. (1999b). Sand
lance: A review of biology and predator relations and annotated
bibliography. Exxon Valdez Oil Spill Restoration Project 99346 Final
Report. U.S. Department of Agriculture, Forest Service, Pacific Northwest
Research Station, Portland, OR. Research Paper PNW-RP-521, 327 pp.
Schlenger, P., MacLennan, A., Iverson, E., Fresh, K., Tanner, C., Lyons, B., …
Wick, A. (2011). Strategic needs assessment: analysis of nearshore
ecosystem process degradation in Puget Sound. Prepared for the Puget
Sound Nearshore Ecosystem Restoration Project. Technical Report 201102.
Schwartz, J. (2014, June 29). Work to begin in weeks on project to pull refuse
back from bluff in Port Angeles landfill. Peninsula Daily News. Retrieved
from
http://www.peninsuladailynews.com/article/20140630/NEWS/306309993
Shaffer, A. (2013). Management discussion Elwha Nearshore Consortium.
Synopsis in Proceedings of the 8th annual Elwha Nearshore Consortium
Workshop, 27 February 2013, Port Angeles, WA, 44 pp.
95
Shaffer, J. A., Crain, P., Kassler, T., Penttila, D., & Barry, D. (2012). Geomorphic
habitat type, drift cell, forage fish, and juvenile salmon: Are they linked?
Journal of Environmental Science and Engineering A, 1, 688–703.
Shaffer, J. A., Crain, P., Winter, B., McHenry, M. L., Lear, C., & Randle, T. J.
(2008). Nearshore restoration of the Elwha River through removal of the
Elwha and Glines Canyon Dams: An overview. Northwest Science,
82(sp1), 48–58.
Shaffer, J. A., Moriarity, R., Sikes, J., & Penttila, D. (2003). Nearshore habitat
mapping of the Strait of Juan de Fuca: Phase 2. Surf smelt spawning
habitat- May-August 2003. Final report to the Northwest Straits
Commission, Clallam Marine Resources Committee, and Washington
Department of Fish and Wildlife. WDFW, Olympia, WA.
Shipman, H. (2004). Coastal bluffs and sea cliffs on Puget Sound, Washington. In
M. A. Hampton and G. B. Griggs [eds.]. Formation, evolution, and
stability of coastal cliffs— Status and trends, professional paper 1693, pp.
81-94. U.S. Department of the Interior, U.S. Geological Survey, Denver,
CO., 123 pp.
Shipman, H. (2008). A geomorphic classification of Puget Sound nearshore
landforms. Puget Sound Nearshore Partnership Report No. 2008-01.
Published by Seattle District, U.S. Army Corps of Engineers, Seattle,
Washington.
Shipman, H. (2010). The geomorphic setting of Puget Sound: implications for
shoreline erosion and the impacts of erosion control structures, in
96
Shipman, H., Dethier, M.N., Gelfenbaum, G., Fresh, K.L., and Dinicola,
R.S., eds., 2010, Puget Sound Shorelines and the Impacts of Armoring—
Proceedings of a State of the Science Workshop, May 2009: U.S.
Geological Survey Scientific Investigations Report 2010‑5254, p. 19-34.
Simenstad, C. A., Miller, B. S., Nyblade, C. F., Thornburgh, K., & Bledsoe, L. J.
(1979). Food web relationships of northern Puget Sound and the Strait of
Juan de Fuca. Fisheries Research Institute, University of Washington,
Seattle, Washington, under contract to Environmental Protection Agency,
Washington, DC.
Simenstad, C. A., Ramirez, M., Burke, J., Logsdon, M., Shipman, H., Tanner, C.,
Toft, J., & Craig, B. (2011). Historical change of Puget Sound shorelines:
Puget Sound nearshore ecosystem project change analysis. Puget Sound
Nearshore Report No. 2011-01. Published by Washington Department of
Fish and Wildlife, Olympia, Washington, and U.S. Army Corps of
Engineers, Seattle, Washington.
Simenstad, C., Logsdon, M., Fresh, K., Shipman, H., Dethier, M., & Newton, J.
(2006). Conceptual model for assessing restoration of Puget Sound
nearshore ecosystems. Puget Sound Nearshore Partnership Report No.
2006-03. Published by Washington Sea Grant Program, University of
Washington, Seattle, Washington.
Summers, A. P., Balaban, J., Gidmark, N., & Bizzarro, J. J. (2013). Pacific sand
lance burrowing/morphology. In Liedtke, T., C. Gibson, D. Lowry, and D.
Fagergren [eds.], 2013, Conservation and ecology of marine forage
97
fishes—Proceedings of a research symposium, September 2012. U.S.
Geological Survey Open-File Report 2013-1035, 24 pp.
Therriault, T. W., Hay, D. E., & Schweigert, J. F. (2009). Biological overview
and trends in pelagic forage fish abundance in the Salish Sea (Strait of
Georgia, British Columbia). Marine Ornithology, 37, 3–8.
Thompson, W. F., & Associates. (1936). The spawning of the silver smelt,
Hypomesus pretiosus. Ecology, 17(1), 158–168.
U.S. Army Corps of Engineers (USCOE). (1971). Report on survey of Ediz Hook
for beach erosion and related purposes. Part II. Seattle District Army
Corps of Engineers. Seattle, WA.
U.S. Army Corps of Engineers (USCOE). (2002). Ediz Hook Beach Nourishment
and Revetment Maintenance: Biological Evaluation.
Warrick, J. A., Duda, J. J., Magirl, C. S., & Curran, C. A. (2012). River turbidity
and sediment loads during dam removal. Eos, Transactions American
Geophysical Union, 93(43), 425.
Warrick, J. A., George, D. A., Gelfenbaum, G., Ruggiero, P., Kaminsky, G. M., &
Beirne, M. (2009). Beach morphology and change along the mixed grainsize delta of the dammed Elwha River, Washington. Geomorphology,
111(3), 136–148.
Warrick, J., & Gelfenbaum, G. (2013). Freshwater Bay and Elwha River mouth
sediment and linkages of river and nearshore. In Proceedings of the 8th
Annual Elwha Nearshore Consortium Workshop, 27 February 2013, Port
Angeles, WA, MacDonald and Harris, eds, 44 pp.
98
Washington Department of Fish and Wildlife (WDFW). (2010). Washington State
surf smelt fact sheet. Revised 2013. Washington Department of Fish and
Wildlife publication 1219, Forage Fish division, La Connor, WA, 11 pp.
Washington Department of Fish and Wildlife (WDFW). (2014). Forage fish
spawning map- Washington State. WDFW mapping application.
Spawning data for forage fish along Puget Sound and Strait of Juan de
Fuca shorelines.
http://wdfw.maps.arcgis.com/home/webmap/viewer.html?useExisting=1.
Wilson, M. F., Armstrong, R. H., Robards, M. D., & Piatt, J. F. (1999). Sand
lance as cornerstone prey for predator populations. In: Robards, M.D.,
Willson, M.F., Armstrong, R.H. & Piatt, J.F. (Eds). Sand lance: a review
of biology and predator relations and annotated bibliography. Research
Paper PNWRP- 521. Portland, OR: US Department of Agriculture, Forest
Service, Pacific Northwest Research Station. pp. 17–44.
Winter, B. D., & Crain, P. (2008). Making the case for ecosystem restoration by
dam removal in the Elwha River, Washington. Northwest Science,
82(sp1), 13–28.
99
Appendix
Figure 13. All surf smelt sample locations in the Elwha and Dungeness drift cells.
Figure 14. All surf smelt samples containing eggs.
100
Figure 15. Surf smelt survey results for all samples collected in the impaired Elwha drift
cell.
Figure 16. Surf smelt survey results for all samples collected in the intact Dungeness drift
cell.
101
Figure 17. Surf smelt survey results in the intact Dungeness drift cell before dam removal
(2007-2008).
Figure 18. Surf smelt survey results in the intact Dungeness drift cell during dam removal
(20012-2013).
102
Figure 19. Surf smelt survey results in Freshwater Bay (impaired Elwha drift cell) before
dam removal (2007-2008).
Figure 20. Surf smelt survey results in Freshwater Bay (impaired Elwha drift cell) during
dam removal (2012-2013).
103
