The Evergreen State College Institutional Repository

Whisman_MMESthesis2013.pdf

Media

Part of Revegetation of Post-Dam-Removal Riparian Sediments in the Lower Elwha River, Wa

extracted text (extracttext:extracted_text)

REVEGETATION OF POST-DAM-REMOVAL RIPARIAN SEDIMENTS IN THE
LOWER ELWHA RIVER, WA

by
Marisa Whisman

A Thesis
Submitted in partial fulfillment
of the requirement for the degree
Master of Environmental Studies
The Evergreen State College
June 2013

©2012 by Marisa Whisman. All Rights Reserved

This Thesis for the Master of Environmental Study Degree
by
Marisa Whisman

has been approved for
The Evergreen State College
by

Dr. Carri LeRoy
Member of the Faculty

Date

ABSTRACT
Revegetation of Post-Dam-Removal Riparian Sediments in the Lower Elwha River,
WA
Marisa Whisman

With the deconstruction of Glines Canyon Dam nearly complete, Olympic National
Park’s Elwha Revegetation Crew has implemented the second season of a 7-year native
riparian plant restoration effort. A critical component of restoring the lower Elwha River
ecosystem is the establishment of late seral riparian forests to provide stream shading,
woody debris, nutrient inputs and erosion control. As a restoration strategy, installing
woody plants may prevent invasive weed colonization, facilitate native plant recruitment,
and provide seed sources beyond intact forest edge. The greatest challenge facing native
plant establishment in the dewatered Lake Mills reservoir is the survival of vegetation in
post-dam-removal sediments. Fine sediments (alluvial silt and clay) lack porosity, are
subject to inundation and desiccation during wet and dry periods, and create hypoxic
growth conditions. Coarse sediments (gravel, cobbles, sand) are highly porous and prone
to desiccation and wind erosion. Determining which plant species can survive in specific
sediment textures may provide guidelines for successful plant establishment as Elwha
River post-dam-removal plant restoration progresses.
This study focuses on the survival and performance of six native woody plant species in
three post-dam-removal sediment textures on the dewatered Lake Mills reservoir during
the initial 2012 revegetation efforts. Selected species included ocean spray (Holodiscus
discolor), Nootka rose (Rosa nutkana), thimbleberry (Rubus parviflorus), western
redcedar (Thuja plicata), black cottonwood (Populus balsamifera ssp. trichocarpa), and
Douglas-fir (Pseudotsuga menziesii). Total individuals surveyed equaled 860 (n=80-180
samples per species). Sediment moisture, particle size and nutrient content were also
assessed. By late September plant mortality reached a total of 66, with 82% of dead
individuals being Douglas-fir. The lowest mortality (<1%) occurred for black
cottonwood. Proportionally, plant mortality was highest in the coarse sediment. These
and similar data collected over the next 5 years may be useful in determining woody
plant species selection in other post-dam-removal restoration efforts.

Table of Contents
I.

Ecology of Dam Removal
1.1
1.2
1.3
1.4

II.

Elwha River Ecosystem and Dams
2.1
2.2
2.3

III.

Impacts of dams and benefits of dam removal.............................................1
Implications of dam removal .......................................................................3
Decommissioning dams: A new approach to management and…...
restoration of rivers .....................................................................................7
Dam removal case studies in the United States ..........................................8

Geography and vegetation ........................................................................14
History and features of the Glines Canyon and Elwha River dams ..........15
Environmental impacts of the Elwha River dams.....................................18

Projected outcomes of the Elwha River dam removal
3.1
3.2

IV.

Methods
4.1
4.2
4.3
4.4
4.5
4.6
4.7

V.

Restoration: Research needs and site conditions ....................................23
Research question ....................................................................................28

Study sites ................................................................................................29
Selection of woody plants and prescriptions............................................30
Experimental design .................................................................................34
Sediment analysis .....................................................................................38
Vegetation monitoring .............................................................................39
Laboratory analysis ..................................................................................41
Statistical analysis ....................................................................................42

Results
5.1
5.2
5.3
5.4
5.5

Site conditions ..........................................................................................44
Plant performance ....................................................................................44
Sediment field and laboratory analysis ....................................................52
Natural plant regeneration........................................................................56
Herbivory observations ............................................................................58

iv

VI.

Discussion

6.1
6.2
6.3
6.4

VII.

Plant performance ....................................................................................61
Sediment field and laboratory analysis ....................................................65
Natural plant regeneration and invasive plants ........................................69
Future research .........................................................................................70

Significance
7.1
7.2
7.3
7.4
7.5

Broader impacts of study .........................................................................73
Cultural impacts of Elwha River dam removal on the Lower Elwha
Klallam Tribe ...........................................................................................75
Elwha River nearshore habitat .................................................................80
Interdisciplinary Context..........................................................................83
Conclusion ...............................................................................................84

Appendix A. Full species list for 2012 ONP planting prescriptions.................................85
Appendix B. Sediment moisture assessment methods: Options and limitations .............86
Plant growth measurement: plant selection and totals ................................87
Bibliography ......................................................................................................................88

v

List of Figures
Map 1.

Map of study sites around Lake Mills, Olympic National Park ...........30

Map 2.

Map of road and creek weed seed sources surrounding Lake Mills .....57

Figure 1.

Median growth of measured plants by species .....................................50

Figure 2.

Mean tensiometer readings by month...................................................52

Figure 3.

Available sediment moisture versus gravimetric water content, Lake
Mills 2012 ............................................................................................53

Figure 4.

Lake Mills Available Nitrate and Phosphate by Sediment ...................55

Figure 5.

Photos of the grey midget moth (Nycteola cinereana) .........................59

List of Tables
Table 1.

Study sites and planting prescriptions included in 2012 Lake Mills
Assessment ............................................................................................31

Table 2.

Native plants selected and totals for each species and site ...................36

Table 3.

Lake Mills weather data for the 2012 growing season .........................44

Table 4.

Plant mortality by species and site ........................................................45

Table 5.

Plant mortality: Percent mortality of species by site .............................46

Table 6.

Chi-square test table for plant survival by site ......................................46

Table 7.

Chi-square test table for plant survival for combined sites by month...47

Table 8.

Qualitative leaf data by site ...................................................................48

Table 9.

Qualitative leaf data by species and month ...........................................48

Table 10.

One-way ANOVA of leaf color by month, sediment and species ........49

Table 11.

Measured plant growth ..........................................................................51

Table 12.

Kruskal-Wallis table..............................................................................51

Table 13.

Lake Mills 2012 sediment moisture data ..............................................52
vi

Table 14.

Lake Mills 2012 fine sediment mean particle size data ........................54

Table 15.

Lake Mills sediment nutrient and pH data ............................................55

Table 16.

All plants included in ONP 2011-2012 planting prescriptions .............85

Table 17.

Table of total plants measured ..............................................................87

vii

Acknowledgements

I would like to thank the following individuals, whose support and guidance made the
successful completion of this research possible:
Dr. Carri LeRoy, my faculty reader and advisor. Her expertise and encouragement led me
to expand my knowledge base to not only a broader understanding of plant ecology, but
also of soil science and even moth-rearing.
Dr. Richard Bigley, consulting ecologist, who helped guide some of my early study
design
Joshua Chenoweth, Restoration Botanist, Olympic National Park, and Dave Allen, Plant
Propagation Specialists, Olympic National Park. Their insight and input regarding
research needs and methods truly aided me in shaping this study
The Elwha Revegetation Crew – supportive co-workers and friends; Steve Acker, Plant
Ecologist, Olympic National Park; Mike Tetreau, Field Biologist Olympic National Park
and Jack of many trades who assisted me in some of my preliminary field study setup and
provided useful tips
Gay Hunter, cultural resources curator at Olympic National Park, as well as Lars Crabo
and Rich Zack of Washington State University entomology staff, and Dennis and Beverly
Strenge, for their assistance in the identification a moth which fed on my cottonwoods
Darlene Zabowski, Soil Scientist and University of Washington Professor of Forest Soils
and Soil Genesis and Classification, whom I consulted for a number of sediment
assessment questions
Greg Hart, technical support, Soil Moisture Corp, for his enduring patience in answering
endless questions regarding tensiometer operation
My partner James Ditchman, and family and friends for their continual support

viii

I.

Ecology of Dam Removal

1.1 Impacts of Dams and Benefits of Dam Removal
Impacts of Dams
In the United States and worldwide, dams influence the majority of river systems. Since
the founding of the U.S., more than 965,000 of over 5.6 million kilometers (km) of U.S.
rivers have been impounded, with only 2% of U.S. rivers remaining uninfluenced by
dams (Grant and Parks, 2009). Put in perspective, roughly 80,000 dams have been
constructed in the U.S., nearly the equivalent of building one dam per day from the time
of U.S. establishment to the early 21st century (Grant and Parks, 2009). Worldwide,
virtually 80% of the upper one third of all large river discharge has been at least partially
obstructed in order to meet human needs (Bednarek, 2001).
Overwhelmingly, the influences of dams on river systems are negative. Dams
impact riverine ecosystems in a variety of ways: serving as barriers to anadromous fish
passage, impeding seed delivery to downstream riparian reaches, influencing channel
dynamics and natural flood events, and limiting freshwater nutrient inputs to estuaries
and other marine systems (Brown and Chenoweth, 2008; Downs et al., 2009; Nilsson et
al., 2010). In some cases, hydropower benefits of dams which reduce human dependence
on polluting energy sources are thought to outweigh negative influences of dams. But in
many cases, dam construction may produce high ecological, economic and social costs
without the intended benefits coming to fruition. Water shortages, for example, are
becoming more prevalent and widespread despite the potential for reservoirs to store
water resources and avoid such shortages (Stanley and Doyle, 2003). Sedimentation of
dam reservoirs poses eventual public safety hazards in the absence of costly dredging.
1

Additionally, CO2 emissions produced from decay of this sediment and organic matter
trapped behind the dams potentially contribute to climate change (Parekh, 2004). From a
cost-benefit standpoint, negative ecological impacts of dams increasingly outweigh
societal gains when compared in the development of science and policy supporting dam
removals.
Dams alter river flow, which can impact aquatic and riparian biodiversity. A
decrease in periodic and seasonal floods and increased periods of low flows can reduce
aquatic faunal diversity through changes in temperature or dissolved oxygen (DO)
(Bednarek, 2001). As a result, impounded rivers may eventually support only animals
suited to constant flows and reduced flood conditions, while organisms poorly adapted to
increased temperature and low DO may become locally extinct. Riparian floral diversity
may also be negatively influenced by altered flow regimes, as downstream transport of
seeds and the stems of resprouting woody plants is hindered (Vesk et al., 2006; Brown
and Chenoweth, 2008). Riparian plant species composition can then become fragmented,
with greater similarity of riparian plant communities at sites among impoundments than
those of reaches in-between (Jansson et al., 2000). Dams and impoundments can
decrease river velocity, reducing the downstream delivery of floating diaspores (seeds,
fruits or spores), which may instead sink to the river bottom or become beached (Jansson
et al., 2000). In contrast, diaspores in unimpounded rivers can pass freely and rapidly
downstream in conjunction with seasonal floods and high flows, and subsequently
riparian plant communities vary more gradually in species composition downstream
compared to dammed reaches (Jansson et al., 2000). Dammed rivers may also undergo
hourly, daily or weekly flows which are far more severe than those of free-flowing rivers,

2

depending on water energy needs or other water resource consumption patterns
(Bednarek, 2001).
In coastal regions such as the Pacific Northwest, river impoundments have farreaching effects on both riverine and marine productivity. Retention of sediment and
woody debris (WD) behind dams leads to loss of freshwater nutrient delivery to estuaries
and limits marine productivity (Bednarek, 2001). Reciprocally, dams prevent tidal surges
in coastal rivers from travelling significant distances upstream, and limit the degree of
cyclical flooding, both of which are mechanisms for upstream and downstream migration
of anadromous fish and shrimp (Bednarek, 2001). Loss of riverine sediment and WD
delivery to coastal regions depletes resources key to the building and maintaining of
nearshore habitat (Shaffer et al., 2008).
The decline of fish populations common in regulated rivers can have a cascading
effect on riparian corridors. Dams are a primary contributor to the reduction of salmon
populations in the U.S., along with fishing, habitat alteration, fish hatcheries, climate
change and the presence of invasive species (Pess et al., 2008). Dams negatively
influence anadromous fish by blocking passage and altering the sediment and flood
regimes of rivers. Over time, this flow alteration leads to the decline of sustaining food
chains in streams (Bednarek et al., 2001). The decline of anadromous fish populations
can adversely influence terrestrial wildlife directly, as with a loss of protein and high
calorie resources, or indirectly through the absence of fish carcasses, which fertilize
riparian plants (Helfield and Naiman, 2002; Sager-Fradkin, 2008).
1.2 Implications of Dam Removal
A common focal point for the ecological benefits of dam removal, in some regions of the
3

northern hemisphere, is the return of native salmon runs. In Pacific Northwest riverine
ecosystems, restoring anadromous fish populations often means restoring riparian habitat
for a variety of plants and animals, supporting diverse aquatic wildlife, and restoring
salmon-centered indigenous cultures (Valadez, 2002; Doyle et al., 2005). While salmon
recovery indeed provides ecological benefits, the roles of aquatic insects and bottomdwelling aquatic organisms (zoobenthos) in supporting fish and other higher-trophic
aquatic organisms is of equal significance (Morley et al., 2008). A shift in aquatic insect
and substrate-dwelling organisms can mean a shift in the higher-trophic organisms a river
ecosystem can support (Morley et al., 2008). Due to their short generation time,
invertebrate populations can recover from disturbance more rapidly than other life forms
such as riparian plant communities, in which different successional pathways of
undetermined temporal scales may result in variability of species diversity over the first
decade or so (Doyle et al., 2005). Concurrently, the sources of available carbon needed
to support aquatic invertebrates depend on detrital inputs from riparian plant
communities, nutrient delivery from downstream marine systems, and algal communities
within the river. These interdependent relationships illuminate the complexity of postdam-removal recovery of a river ecosystem; a multitude of factors beyond the removal of
fish passage barriers are at play (Casper et al., 2006). Hence, in riverine restoration
planning, it is critical to addresses entire ecosystems rather than a single species such as
anadromous fish (Doyle et al., 2002, 2005).
Degree of flow in a riverine ecosystem can influence food webs and ecosystem
processes such as litter decomposition rates by modifying macroinvertebrate
communities. Casper et al. (2006) studied zoobenthos and changes in trophic food webs,

4

via isotopic signatures, following the Edwards Dam removal on the Kennebec River,
Maine. Overall, at three locations along the river, a greater density increase of zoobenthos
was observed in the site closest to the former dam. Diversity was variable, but few
differences between pre-and post-dam-removal species richness or evenness were
apparent. A more apparent change took place in post-dam-removal insect community
composition, with the addition of 8 genera to the existing population consisting of
mayflies, caddisflies, oligochaetes, and a variety of predatory and non-predatory
chironomid midges. The difference between zoobenthic communities in restored and
free-flowing sites was less distinct over time while stable isotope signatures changed,
indicating that the feeding tactics and quantities of aquatic organic matter were altered
(Casper et al., 2006). Muehlbauer et al. (2009) conducted a study in which leaf litter bags
were strategically placed above and below Fossil Creek dam, near Strawberry, AZ, and
measured decomposition rates and presence of macroinvertebrates and fungal
decomposers prior to and following the decommissioning of the dam. They found that,
while litter decomposition rates did not vary above or below the dam prior to dam
removal, the macroinvertebrate and fungal decomposer communities increased below the
dam post-removal. Restoration of full water flow appeared, in this case, to positively
impact decomposer populations (Muehlbauer et al., 2009).

While dam removal is generally expected to benefit aquatic systems in the longterm, this restoration approach is not devoid of potential ecological detriments,
particularly in the short-term. Water quality may be negatively affected by high
concentrations of total suspended solids (TSS) during initial post-dam-removal high-flow
events (Downs et al., 2009). Fine sediment released and deposited on river bed surfaces
5

may reduce conductivity and water infiltration, and increase the downstream flooding and
burial hazards of salmonid and macroinvertebrate habitat by sediment deposition. In
urban, agricultural or other developed river basins, buildup of toxins in reservoir
sediment may pose a health hazard to humans and wildlife with their release (Stanley and
Doyle, 2003; Wildman and MacBroom, 2005). Additionally, the sediment exposed
following dam removal can serve as a vector for invasive weed colonization, providing
seed sources for other locations along riparian corridors (Orr and Stanley, 2006). A lack
of complete, pre-and post-dam-removal monitoring datasets in conjunction with the small
dam removals which have occurred over the past decades only increases these concerns
when addressing the removal of large dams (Downs et al., 2009; Muehlbauer et al.,
2009). Of the 500 dam removal projects which have taken place in the U.S., few have
received pre-and post-dam-removal monitoring to measure ecosystem changes over time
(Duda et al., 2008; Woodward et al., 2008).

Sediment
A key aspect of dam removal recovery centers around geomorphic impacts, in
terms of sediment storage. Thus, a key component of post-dam-removal river ecosystem
restoration is an understanding of the direction, distance and speed of sediment transport,
to control release rates and minimize harm to aquatic organisms (Doyle et al., 2002).
Over time, dams result in upstream sediment storage, while sediment delivery and
accumulation downstream is diminished (Bednarek, 2001; Shaffer et al., 2008). It is
generally expected that dam removal results in a reversal of this trend, based on the
erosion of upstream sediment and aggradation of downstream reaches from fluvial
transport (Konrad, 2009). The overall geomorphic response of a riverine system to dam
6

removal depends on: 1) the level of sediment which has accumulated in the reservoir; 2)
the flow and ability of the river to transport this sediment; 3) rates of sediment erosion,
driven by high flows during wet seasonal periods; and 4) response of channel
morphology downstream, which may determine the overall geomorphic response of the
system to dam removal (Konrad, 2009). River systems with higher rates of discharge
and/or higher slope generally move sediment loads more rapidly than rivers with lower
discharges and slopes. Additionally, the texture of impounded sediment influences
removal rates, with fine, silt-dominated sediment delivery occurring more rapidly than
that of coarse sediment (Shafroth et al., 2002; Mussman et al., 2008). Based on the
limited dam removal ecosystem recovery data available, geomorphic adjustments tend to
occur within the first 1 to 5 years (Doyle et al., 2005).

1.3 Decommissioning Dams: A New Approach to Management and Restoration of
Rivers
Due to a greater recognition of the ecological impacts of dams, more progressive
regulations for dam maintenance and operation are in effect. Non-federal hydroelectric
projects require licensing under the Federal Power Act (FPA, 16 U.S.C. 791-828c as
amended; Chapter 285, June 10, 1920; 41 Stat. 1063). Testament to a changing public
perception of the role of dams, the Federal Energy Regulatory Commission (FERC)
modified its relicensing process for existing, non-federal dams in the early 1990s. For the
first time, general consensus supported the decommissioning of operating dams for
ecological restoration alone, not simply due to public safety concerns (Gowan et al.,
2006). Greater priority has been assigned to weighing the benefits of the restoration of
impounded rivers in relation to the utility of maintaining existing dams. Rather than

7

basing renewal chiefly on the provision of public service, the environmental impacts of
these dams are now investigated when considering the extension of the 30-to-50-year
licenses issued to private and public power producers. Currently, factors such as
increased minimum flow, added or enhanced fish passages, allowance of periodic high
flows, and adjusted flow regulation to accommodate riparian environments are included
in the criteria for dam relicensing (Bednarek, 2001).

Contemporary public views on river impoundments with the passage of time have
placed aging dams under increasing scrutiny. Since the late 1990s, the licenses of
hundreds of hydroelectric dams have expired; the majority of these dams were
constructed in the 1950s and 1960s, and are subject to relicensing investigations. If dams
fail to meet specific safety and ecological provisions, they are decommissioned (Grant
and Parks, 2009). An estimated 85% of aging dams in the U.S. will reach the end of their
operational lives by the year 2020 (Doyle et al., 2003). As with other aging structures,
public safety concerns play a role in determining whether older dams remain operable. In
2002, the Federal Emergency Management Agency (FEMA) categorized 9,200 U.S.
dams as “high hazard” due to structural instability and risk to downstream human
developments in the event of a dam break (Downs et al., 2009). In such cases, the utility
of river restoration may clearly outweigh the benefits of the prolonged and potentially
hazardous operation of outmoded dams.

1.4 Dam Removal Case Studies in the United States
Previous dam removal operations, if monitored and documented, may provide insight to
the rates of recovery and cost-benefit analysis entailed post-dam-removal river restoration

8

efforts. Critical data include timeframe of sediment release and erosion, altered channel
dynamics, riparian vegetation recovery, water quality improvement, and eventual
improvement of fish populations once passage is restored (Bernhardt et al., 2005).
Generally, the goals of post-dam-removal river restoration include restoration of fish
passage, improvement of water quality, management of riparian zones and enhancement
of aquatic habitat. Timing of drawdown and dam deconstruction in relation to high and
low flows, the degree to which a river is regulated (i.e., number of obstructions), and the
size of dams all play a role in the amount of time required for ecosystem recovery
(Bednarek, 2001).
Marmot Dam, Sandy River, OR
The 2007 Marmot Dam removal on the Sandy River, Oregon provides an example of an
ecologically successful dam decommissioning. The former 15-m-tall dam lay 43 km
above the confluence of the Sandy and Columbia Rivers, and its deconstruction created
the largest post-dam-removal sediment release up to that time (Grant, 2000; Podolak and
Pittman, 2010). A cofferdam, protecting the Marmot Dam removal construction site, was
breached in 2008 (Stillwater Sciences Technical Memorandum, 2009). The 90 km Sandy
River drains a 1300 km2 basin in the southwestern Cascade Range, near the base of Mt.
Hood (Podolak and Pittman., 2010). Since the construction of Marmot Dam in 1913,
sediment had accumulated to its crest height, composed of approximately 750,000 m3 of
cobble, and gravel sediment overlying a finer sand layer (Stewart and Grant, 2005;
Downs et al., 2009). The sand and gravel, originating from volcanic debris on Mt. Hood,
were promptly released into the river following dam removal. Both headward and lateral
erosion removed the sand and gravel which had accumulated for nearly a century,

9

following a swift alteration of the river’s profile, termed a “knickpoint” (Downs et al.,
2009; Major et al., 2008). In less than three days following the breaching, the Sandy
River channel morphology 2 km downstream from the former dam changed from singlethread to multiple-channel, and the channel bed had aggraded nearly 4 m (Major et al.,
2008).
Sandy River sediment removal occurred at a rate more rapid than projected, with
15% of all stored sediment transferred within the first 48 hours (Major et al., 2010). By
late winter of 2008, approximately half of the sediment had been redistributed over the
first 3.2 km downstream of the dam. New braided channels, bars and riffles had formed,
flooding was not extreme (only 3 to 4 times that of average summer low flows), and a
greater variety of salmon habitat was soon apparent (Grant and Parks, 2009). One year
following dam removal, the topographic complexity of the Sandy River was comparable
to pre-dam conditions in at least one site, the primary difference being higher channel bed
elevation in the immediate vicinity of the former dam and a temporary decrease in
channel complexity following the coffer dam breaching (Stillwater Sciences, 2009). Postdam-removal channel bed recovery of the Sandy River was projected to be at least 10
years, but serious concerns of fish impediment were alleviated within days of dam
removal (Downs et al., 2009). Relicensing Marmot Dam would have cost $20 million,
including the enhancement measures which would have been required by FERC, with
additional costs of upgrading the aging dam structures; in contrast, removal of the dam
cost $17 million, an investment which restored 43 km of riverine habitat (Grant and
Parks, 2009; Kober, http://www.water. ca.gov/ fishpassage/docs /dams /dams.pdf).

10

Grangeville and Lewiston Dams, Clearwater River, ID
The removal of the Grangeville and Lewiston Dams on the Clearwater River, Idaho
produced successful river ecosystem recovery prior to downstream redamming. The
dams, 17 and 14 m tall, respectively, were constructed in 1903 and 1927. These
impoundments impeded salmon and steelhead runs, adversely impacting the 1855 fishing
and treaty rights of the Nez Perce tribe. The 1963 and 1973 removals of these dams
initially improved salmon runs, freeing 68 km of river and hundreds of km of tributaries.
The silt released during dam removal had little adverse impact (American Rivers, 1999).
In fact, the removal of the Lewiston Dam was the first occasion in which the Army Corp
of Engineers cooperatively decommissioned a federally licensed dam for the purpose of
river restoration (American Rivers, 1999). Unfortunately the 1970s construction of the
Dworshak Dam on the north fork Clearwater River, along with the installation of 4
federally-operated dams on the lower Snake River, all but negated the benefits of the
earlier dam removals and led to a drastic decline in salmon populations in the Snake
River (American Rivers, 1999).
Other U.S. Dam Removals
As indicated in the previous case studies, dam removal can improve the long-term health
of riverine ecosystems. Some dam removal efforts, however, have produced negative
results, particularly in areas of urban or agricultural development. The two-stage removal
of the Fort Edwards Dam on New York’s Hudson River resulted in the release of toxins
contained in accumulated sediment (Stanley and Doyle, 2003). The initial and partial
removal in 1973 released petroleum wastes and polychlorinated biphenyls (PCBs)
downstream and required expensive cleanup efforts (Stanley and Doyle, 2003). Released

11

sediment restricted the flow of the river, blocking access both for navigation and sewage
waste distribution downstream and created additional health hazards (Stanley and Doyle,
2003). Following the second and final stage of the Fort Edwards dam removal in 1991,
PCB concentrations in striped bass proved to be twice that of previous tests (Stanley and
Doyle, 2003; American Rivers, 1999). A similar problem arose with the removal of the
Anaconda and Union City dams on the Naugatuck River, Connecticut. Polyaromatic
Hydrocarbons (PAHs) were discovered in post-dam-removal sediments, and had to be
removed to prevent further downstream contamination (Wildman and MacBroom, 2005).

Since the 1960s, the state of Wisconsin has undertaken numerous small dam
removals. Initial dam removals were carried out due to failing structural integrity.
Beginning in the 1990s, dam decommissioning occurred increasingly for the purpose of
ecological restoration, with mixed success. Following the 1988 removal of the Woolen
Mills Dam on Wisconsin’s Milwaukee River, river flushing and accompanying fine
sediment removal downstream took place over a 6 month time period, and an overall
improvement in passage was seen for fish and other aquatic organisms (Bednarek, 2001).
Invasive common carp populations diminished due to faster-flowing waters, while native
smallmouth bass populations increased (Bednarek, 2001). In contrast, removal of the
Fulton Dam on Wisconsin’s Yahara River led to the replacement of native cattail and
sedge with non-native wet meadow grasses, and a decline in muskrat and duck
populations resulting from the loss of the reservoir. Displacement of desirable flora and
fauna must be carefully considered in the planning of dam removals (Bednarek, 2001).

Few studies have followed the long-term recovery of drained reservoirs,

12

particularly in relation to riparian vegetation following dam removal. However, limited
data are available from vegetation surveys conducted in thirteen post-dam-removal sites
in Wisconsin (Orr and Stanley, 2006). Vegetation surveys depicted grasses and forbs as
the dominant riparian plants within the first few years following dam removal, coupled
with trees in later years. Aquatic and semiaquatic plants, such as rushes, sedges and reeds
showed negative trends over time. Invasive plants also colonized the exposed sediment,
persisting throughout the study, with the exception of sites dominated by shrubs and a
stinging nettle (Urtica dioica) (Orr and Stanley, 2006). Reed canary grass (Phalaris
arundinacea), a highly invasive rhizomatous grass introduced from Asia, became
established despite an apparent lack of local seed source. Other than a general trend of
increasing tree frequencies as time following dam-removal elapsed, plant species richness
and species composition proved variable during the 2-to-50-year post-dam-removal time
period, illuminating the unpredictability of riparian plant succession following a major
substrate-altering disturbance (Orr and Koenig, 2006).

With little known about post-dam-removal plant succession or success of artificial
plant regeneration on drained reservoir sediments, performance of the Elwha
Revegetation Crew plantings must be documented throughout the project. The intent of
this study is to provide potential expected outcomes when selected woody plant species
are installed in specific dewatered reservoir sediment types. This small body of research
may contribute to the limited knowledge base of revegetated reservoir sediments postdam-removal, particularly where active plant restoration is to take place.

13

II. Elwha River Ecosystem and Dams
2.1 Geography and Vegetation
The Elwha River, 72 km in length, lies 10 km west of the city of Port Angeles on the
northern central Olympic Peninsula of Washington State. The Elwha River basin covers
830 km2 and forms 161 km of tributaries (Acker et al., 2008; Brenkman et al., 2008). The
headwaters of the Elwha River are situated at the base of Mount Barnes in the Bailey
Range of the Olympic Mountains. The mouth of the river enters the Strait of Juan de
Fuca, a marine passage connecting the Pacific Ocean with the waters of Puget Sound and
British Columbia (Duda et al., 2008; Brenkman et al., 2008). Elevations in the watershed
range from sea level to 1372 m at the headwaters (Duda et al., 2008; Brenkman et al.,
2008). The Elwha River basin typically experiences warm, arid summers and cool, wet
winters. Total annual precipitation averages just over 143 cm, with an average total
snowfall of 37 cm and an average annual temperature range of 4° to 14° C (Western
Regional Climate Center, http://www.wrcc.dri. edu/cgi-bin/cliMAIN.pl?waelwha).

Soil types in the Elwha River basin are classified as Haplumbrepts, typical of
temperate to warm regions, of the order inceptisol, with parent materials of postPleistocene origin (Jackson and Kimerling, 2003; Chernicoff and Venkatakrishnan,
1995). Much of the Elwha River basin, Olympic Mountain Range and Strait of Juan de
Fuca geology and topography were formed by seafloor scraping of the Pacific and North
American plates, and glacial incision. The Juan de Fuca shoreline was formed by the
southern advance of the Cordilleran ice sheet approximately 16,000 years ago (Warrick et
al., 2009; Shaffer et al., 2008).

14

The Elwha River basin is comprised of many Olympic Peninsula vegetation zone
classifications. Lower elevations fall within the western hemlock (Tsuga heterophylla)
zone, generally dominated by Douglas-fir (Pseudotsuga menziesii) and co-dominated by
western hemlock, with western redcedar (Thuja plicata) in moister sites (Henderson et
al., 1989; Duda et al., 2008). Drier areas around Lake Mills and the Lillian River tributary
consist of vegetation classes associated with the Douglas-fir zone. Higher elevations lie
in the subalpine fir (Abies lasiocarpa) zone on the drier eastern side, and the mountain
hemlock (Tsuga mertensiana) zone on the wetter western side, within the Cascade
Subalpine Forest Zone Complex. Floodplains, river terraces and valley bottoms consist of
red alder, black cottonwood, grand fir (Abies grandis), and bigleaf maple (Acer
macrophyllum) associations, with canopy dominance of the different species varying
(Duda et al., 2008; Jackson and Kimerling, 2003).

2.2 History and Features of the Glines Canyon and Elwha Dams
The Glines Canyon and Elwha hydroelectric dams were constructed in the early 20th
century. The former Elwha Dam, 33 m in height and located 7.9 km upstream from the
mouth of the Elwha River, was built in 1913, forming the Lake Aldwell reservoir; the
Glines Canyon Dam, 64 m in height and located 21.6 km upstream, was completed in
1927, forming the Lake Mills reservoir. The dams, designed to power a lumber mill and
enhance economic development on the North Olympic Peninsula, were not built to
accommodate salmon passage (Duda et al., 2008). These dams were classified as high
head (>30 m) storage dams, one of the two major structural dam categorizations (Poff
and Hart, 2002; Woodward et al., 2008). Storage dams bear large hydraulic heads, afford
storage volume, long hydraulic residence time (HRT), and control the rate of water
15

release. The other major dam classification, “run of the river” dams, usually pertain to
smaller dams in which hydraulic head and storage volume are minimal, HRT is short, and
control of water release is minor to nonexistent (Poff and Hart, 2002). The Lake Aldwell
reservoir was 1.08 km2 in size, and Lake Mills was 1.68 km2. The two reservoirs together
once flooded more than 9 km of riverine habitat. To-date, the Glines Canyon Dam is the
tallest dam to be removed in U.S history (Duda et al., 2008). Of the post-dam-removal
Lake Mills dewatered reservoir surfaces exposed, 30% were projected to be floodplain,
and 40% were projected to be upland terraces consisting of abandoned residual sediment,
with the remainder forming steep valley walls (Duda et al., 2008).
Approximately 21.7 million m3 (28 million yd3) of sediment were retained by the
Glines Canyon Dam during its operation (ONP, 2013). Upland dewatered reservoir
landform soils were covered by 8 cm to 6 m of sediment; the floodplains as much as 60 m
(2011 ONP soil probe surveys). Post-dam-removal substrates retained on the benches,
upland terraces and valley walls in the former Lake Mills reservoir consist of fine,
alluvial silt and clay sediments, coarse sand and gravel sediments, and fine-coarse
sediment mixtures (Winter and Crain, 2008). Restriction of sediment transport following
the Glines Canyon Dam construction led to the formation of a delta at the head of Lake
Mills up to 924 m in length, 24 m thick, and supplying much of the coarse gravel, sand
and fine sand substrate remaining on the dewatered reservoir in 2013 (Winter and Crain,
2008).
Removal of the Glines Canyon and Elwha dams was initially set in motion by the
Elwha River Ecosystem and Fisheries Restoration Act (PL 102-495), mandated by
Congress in 1992, which required the full restoration of the Elwha River ecosystem and
16

its native anadromous fish (Gowan et al, 2006). Environmental Impact Statements (EIS)
completed in 1995 by the National Park Service, U.S. Fish and Wildlife Service, Bureau
of Reclamation, Bureau of Indian Affairs, and Lower Elwha Klallam Tribe determined
that restoration of this ecosystem could be achieved only through the removal of both
dams (Duda et al., 2008). Removal of only one dam would mean maintaining an
upstream or downstream obstruction to fish passage and sediment retention, thus negating
full restoration of river continuity and overall restoration progression (Bednarek, 2001). It
was determined that removal of both dams would result in ecosystem recovery sufficient
to justify the expense of removal and restoration efforts, based on studies of the efficacy
of different fish passage facilities, prior dam removals, and earlier dam-related Elwha
River fisheries and wildlife conservation efforts (Winter and Crain, 2008).

The 1913 Elwha Dam construction occurred prior to the formulation of the
Federal Power Act (FPA), hence the owners of the dam were not obligated to obtain a
license. The Glines Canyon Dam, completed in 1927, was granted a 50-year operating
license (1926). Following an investigation of unlicensed non-federal hydroelectric
projects by FERC, beginning in the 1960s, a combined license (No. 588) was issued for
both dams in 1979, upon determination that the two dams were “hydraulically,
electrically, and operationally interconnected” (Winter and Crain, 2008). By the 1980s
the matter of relicensing the Elwha River dams, never officially licensed to begin with,
became a source of public debate. That the Lake Mills reservoir lay within Olympic
National Park further fueled this controversy, where the Glines Canyon Dam was
concerned (Duda et al., 2008). Given that dam re-licensing fell under the regulatory
umbrella of the National Environmental Policy Act (NEPA), FERC facilitated an EIS
17

procedure which required applicants to gather new environmental data, to be considered
along with data contributed by various federal agencies, when constructing a cost-benefit
analysis of keeping versus removing dams. This was a precursor to the 1992
congressional Elwha River Ecosystem and Fisheries Restoration Act (EREFRA). Prior to
the passage of the EREFRA, no hydroelectric dam removal in the U.S. had been based
centrally on fish and wildlife benefits (Winter and Crain, 2008). Thus began the legacy of
the largest dam removal restoration attempted in U.S. history prior to 2013 (Woodward et
al., 2008).

2.3 Environmental Impacts of the Elwha River Dams
As is often the case with large dams, the Glines Canyon and Elwha dams negatively
impacted the health of the lower Elwha River, the river basin, and nearshore habitat. For
nearly 100 years the dams obstructed access to over 113 km of high-quality anadromous
fish habitat. Loss of access to spawning grounds dramatically reduced salmon runs,
which in turn altered the foraging patterns of Elwha River terrestrial mammals and birds.
Reduced activity of mammals along the Elwha River corridor has affected the
distribution of nutrients and seeds through zoochory (Duda et al., 2008). Hydrochorous
seed transport has also been hindered, preventing the downstream colonization of
resilient, diverse riparian vegetation and altering plant species composition (Brown and
Chenoweth, 2008; Duda et al., 2008). Additionally, the dams have prevented the delivery
of sediment and woody debris, crucial sources of marine nutrients and beach-building
materials, to the Elwha estuary, delta, and nearshore habitat in the Strait of Juan de Fuca
(Shaffer et al., 2008).

18

The Elwha Estuary, Delta and Nearshore Habitat
A common tendency of dams, particularly storage dams, is the retention of sediment and
woody debris which would otherwise be distributed throughout the river basin. During
their existence, the Elwha River dams and reservoirs retained sediment and large woody
debris (LWD) critical to building shoreline beaches and bluffs. From the years 19392006, shoreline erosion around Port Angeles, WA was measured at approximately 0.6 m
per year, or 24,000 m3 (Warrick et al., 2009). Historically, the Elwha River delivered
roughly 160,000 m3 of fine and coarse sediment per year to the mouth of the river. This
delivery of sediment and wood provided nutrients responsible for estuarine and marine
productivity, as well as contributing beach-building materials to the approximately 21 km
of shoreline between Freshwater Bay and Ediz Hook, bordering the city of Port Angeles.
Installation of shoreline bulkheads during the 1950s compounded the problem of beach
erosion, obstructing sediment input from eroding coastal bluffs. The formerly mixedgrain-size nearshore habitat and beach texture of the Strait of Juan de Fuca was replaced
by cobble-and-boulder-dominated substrate. Estuarine habitat near the Elwha River
mouth declined from 0.4 km2 to 0.12 km2 (Shaffer et al., 2008). Estuarine decline is of
particular concern, as the Elwha River estuary was documented as having the greatest
plant species diversity recorded in Pacific Northwest coastal wetland surveys (Shafroth et
al., 2009). Additionally, riverine fine sediment deposition is critical to the sustenance of
coastal marshes in a future of climate change and sea level rise (Poff and Hart, 2002;
Stoeker Ecological, 2011).

The nearshore habitat near the mouth of the Elwha River historically supported a
variety of marine wildlife. Riverine nutrient input from the Elwha River to the Strait of
19

Juan de Fuca, in conjunction with deep marine water, powerful winds, and strong
currents conducive to mixing, maintain well-mixed, cool and nutrient-rich conditions in
the Strait. Loss of sediment delivery has led to a decline of wetlands and associated
shellfish, clam, eelgrass and kelp populations (Shaffer et al., 2008). Estuarine and
nearshore habitat between Ediz Hook in Port Angeles and Cape Flattery, Neah Bay are
known to be critical early saltwater rearing zones for Chinook salmon (Shaffer et al.,
2008). Kelp forests make up 40% of the Strait of Juan de Fuca shoreline, and consist of
giant kelp (Macrocystis integrifolia), bull kelp (Nereocystis luetkeana), and understory
kelp (Pterygophora californica), while eelgrass (Zostera marina) forms 20% of the
shoreline. Modification of low-tide beaches from mixed-grain to coarse-grained textures
has resulted in unsuitable shellfish habitat. Removal of the Elwha River dams will likely
benefit the estuarine and shoreline habitat near the river mouth and delta, with the return
of riverine nutrients and debris. However, much of this land lies outside the boundary of
Olympic National Park. Full restoration of this marine nearshore habitat must be a
collaborative effort among Tribal, private, city, and state landowners, to fully restore the
severely altered shoreline (Shaffer et al., 2008).

Dam Impacts on Salmon
The presence of the Elwha River dams has had adverse effects on salmon diversity and
overall salmon populations. Since the construction of the dams, native salmon runs have
declined by 90% in the Elwha watershed (Pess et al., 2008). The majority of the Elwha
River lay above the dams, unavailable for use as spawning habitat. Historically, annual
returns of anadromous salmonid populations were estimated at 380,000 to 500,000.
Average numbers between 1990 and 2000 were less than 5000 returning salmon, with
20

ranges of <5000 to 19,800 per year (Pess et al., 2008). Prior to damming, the Elwha River
was home to all anadromous salmonids native to the Pacific Northwest: coho
(Oncorhynchus kisutch), Puget Sound Chinook (Oncorhynchus tshawytscha), sockeye
(Oncorhynchus nerka), pink (Oncorhynchus gorbuscha), Strait of Juan de Fuca/Hood
Canal summer chum (Oncorhynchus keta), steelhead trout (Oncorhynchus mykiss), and
cutthroat trout (Oncorhynchus clarkii). Potamodromous bull trout (Salvelinus
confluentus) and rainbow trout (Oncorhynchus mykiss) have been confined to areas above
the dams, when historically they migrated within the freshwater reaches above, between
and below both dams (Brenkman et al., 2008).
Salmon-Dependent Wildlife
The Elwha River dams, in addition to impacting the life cycles of aquatic organisms,
have impacted migratory patterns of terrestrial organisms, among them the North
American black bear (Ursus americanus) (Sager-Fradkin et al., 2008). The black bear,
prevalent in the wilderness of Olympic National Park, plays an important role in nutrient
distribution between marine and terrestrial environments in the Elwha and other river
basins. Typically, bears feeding on healthy riverine salmon populations provide crucial
nutrients to riparian and inland forests through fish consumption, defecation and
transportation of salmon carcasses inland (Helfield and Naiman, 2002; Sager-Fradkin et
al., 2008). Bears preparing for hibernation require fatty food sources such as salmon, but
lacking significant fish populations will substitute this food for more consistently
abundant calorie sources, such as huckleberries (Vaccinium sp.) or whitebark pine seed
(Pinus albicaulis) where available at higher elevations (Smith et al., 2008). Currently,
resident black bears in the Olympics spend the fall period leading up to hibernation in

21

higher elevations feeding primarily on huckleberry fruits. In particular, female bears
linger at higher elevations throughout most of the spring, summer and fall seasons
leading up to hibernation, which draws male bears up in elevation during the early
summer mating season (Sager-Fradkin et al., 2008). Historical reports indicate that black
bears in western Washington would more typically feed on salmon prior to hibernation. A
decline in salmon runs following the installation of the Elwha and Glines Canyon dams
may have altered the movement and feeding patterns of Olympic black bears (SagerFradkin et al., 2008). As a component of a 5-year pre-dam-removal bear tracking study
conducted by ONP, U.S. Geological Survey, and the University of Idaho Department of
Fish and Wildlife Resources, salmon carcasses were placed in riparian areas of the
Olympics near the time of hibernation, and were readily eaten by black bears (SagerFradkin et al., 2008). The removal of the Elwha River dams may eventually result in the
movement of black bears back to the river if salmon populations are restored (SagerFradkin et al., 2008).

22

III. Projected Outcomes of the Elwha River Dam Removal

Few examples of post-dam-removal ecosystem recovery following the removal of large
dams are available to aid in predicting the long-term outcomes of the Elwha River dam
removals. The best current example would be a sediment transport model, designed for
the Elwha River, which predicted increasing sediment stability over time, with occasional
flooding disturbance disrupting that stability, following dam removal (Konrad, 2009).
Original Lake Mills sediment load estimates amounted to 20.4 million m3 of sediment (or
204 million km3), and were later determined to be nearly 21.4 give or take 3 million m3
(approximately 214 million km3) (USGS sediment team, 2012). While the majority of
U.S. rivers are impounded by dams, most are small dams on small-to-midsize rivers, and
dam removal efforts thus far have been no exception (Doyle et al., 2005). The enormity
of the Elwha River dam removal and restoration has been compared to the recovery of
the North and South Fork Toutle River basins, following the 1980 eruption of Mount
Saint Helens, WA (Bednarek, 2001). This eruption, produced by highly explosive
pyroclastic lava, released nearly 2.3 billion m3 of sediment into the North Fork Toutle
River basin, and over 38.2 million m3 into the South Fork (Bednarek, 2001). These
massive mudslides resulted in the burial of 90% of salmon habitat, and virtually all
riparian vegetation. Despite these unfavorable odds, salmon populations began to recover
a mere 3 months following the volcanic sediment delivery (Bednarek, 2001).

3.1 Restoration: Research Needs and Site Conditions
The removal of the Elwha River dams has resulted in the large-scale exposure of
sediment, much of it well beyond the range of seed rain from the intact forest edge

23

(Michel et al., 2011). Restoration measures are in effect to create river-sustaining riparian
forest habitat, prevent weed colonization, support riparian wildlife and to aid in eventual
soil development (Bradshaw, 1982; Chenoweth et al., 2011). Since the fall of 2011,
Olympic National Park’s Elwha Revegetation Crew has been planting upland terraces
and benches of the dewatered Lake Mills reservoir to meet such needs. All plants
installed are propagated from seed collected within the lower Elwha River basin and
processed at the Matt Albright Native Plant Center near Agnew, WA, 4th Corner Nursery
(Bellingham, WA), and Corvalis Plant Materials Center (Salem, OR); many of the
conifer seedlings are propagated by Silvaseed Nursery in Roy, WA.

Revegetation of the exposed sediments in the Lake Mills and Aldwell reservoirs
may trigger more rapid riparian plant recovery and development of late-seral forests
(Chenoweth et al., 2011). The roots of installed plants may provide a carbon source for
microbes participatory in soil formation, fragment compacted sediment through the
formation of root pathways, and eventually develop macropores which aid in fluid
transport when roots penetrate soil (Angers and Caron, 1998). Enhancement of riparian
forest development can also improve stream ecosystem health. Riparian forests provide
shade critical to stream temperature regulation and large woody debris (LWD) input,
which increases channel complexity beneficial to spawning fish (Collins and
Montogmery, 2002).

Plants installed at Lake Mills consisted of potted and bare-root specimens, planted
according to three different planting prescriptions (Table 1, Section IV). The timely
establishment of woody riparian vegetation, in particular, is a goal of the ONP and the

24

Lower Elwha Klallam Tribe (LEKT) restoration plans. Woody plants provide substantial
shading competition to non-native plants in newly exposed ground, provide a sheltered
growth medium and microclimate for volunteer and planted seedlings, and provide seed
and detritus to areas far from seed sources (Castro et al., 2004; Chenoweth et al., 2011).
A particularly challenging aspect of post-dam-removal revegetation in the
dewatered Lake Mills and Lake Aldwell reservoirs is the survival of plants in residual
sediments. Post-dam-removal sediment in Lake Mills includes fine, silt-and-clay-sized
particles, 50-67% of which is predicted to erode and wash downstream over the severalyear period following dam removal, with the remainder consisting of coarse sand, gravel
and cobble (Mussman et al., 2008; Czuba et al., 2011). Fine sediment, composed of
alluvial silt and clay, is low in nutrients, lacks porosity and hydraulic conductivity, and
creates hypoxic growing conditions for riparian plants. Approximately 20% of the fine
sediment consists of clay, with the remaining ~80% consisting of silt, typical of Lake
Mills sediment (Mussman et al., 2008; Warrick et al., 2009; Czuba et al., 2011). Coarse
sediments, found primarily on the reservoir delta and future valley walls, are
predominantly sand and highly porous, erosive, nutrient-poor gravel possessing little
moisture retention properties. The erosive nature of coarse substrates coupled with windy
site conditions at the former Lake Mills poses the risk of plant root exposure (Chenoweth
et al., 2011).

Over time, with the establishment of vegetation in post-dam-removal sediment,
organic matter can contribute greatly to mineralization of the substrate (Berendse, 1998).
Vegetative litter created by isopod defoliation can harbor litter-decomposing microbes,
which in turn aid in soil development, mineralization and nutrient input (Facelli and
25

Pickett, 1991). However, neighboring forest vegetation close enough to provide a seed
source may not consist of species tolerant of soil saturation, drought, low nutrients, or pH
extremes. This illuminates the potential benefits of introducing a variety of native
riparian plants to determine which species possess tolerance of such extreme soil
conditions (Ash et al., 1994).

While the colonization of Lake Mills post-dam-removal sediment may prove
challenging for many native plants, a multitude of neighboring invasive grasses, forbs
and shrubs have the capability to thrive in such substrates (Orr and Stanley, 2006). In
2008, seed bank analysis of sediment cores collected from Lake Mills revealed a higher
presence of invasive plant species than native species as potential colonizers of exposed
sediment beyond the range of forest edge seed rain (Brown and Chenoweth, 2008).
Invasive weeds such as Himalayan blackberry (Rubus armeniacus) favor substrates
dominated by gravel and other coarse materials (Caplan and Yeakley, 2006), like those
found on the Lake Mills delta. Hence, weed control is vital to the establishment of a
riparian environment conducive to stream shading, nutrient input, and contribution of
woody debris to support a riverine ecosystem. Invasive plants can modify succession of
riparian plant communities through alteration of microbial communities, competition for
natural resources, allelopathy, and habitat alteration (Haubensak and Parker, 2004;
Rudgers and Orr, 2009; Peltzer et al, 2009).

Examples of habitat alteration from non-native plant invasion are plentiful. Local
invasive grasses such as tall fescue (Lolium arundinaceum) host the fungal endophyte
Neotyphodium coenophialum, known to slow the progression of disturbed grasslands to

26

native forests over time through the growth inhibition of non-host plants (Rudgers and
Orr, 2009). In southern Wisconsin, following the removal of the Oak Street Dam in
Baraboo and the Rockdale Dam in Rockdale, invasive reed canary grass colonized
exposed post-dam-removal sediment, even in the absence of an on-site seed source (Orr
and Stanley, 2006; Orr and Koenig, 2006). Attempts to establish primarily graminoid
native vegetation through outplanting showed little success three years following dam
removal (Stanley and Doyle, 2003; Orr and Koenig, 2006). Establishment of native
woody plants may aid more effectively in the establishment of riparian forests, providing
soil formation, terrestrial wildlife forage, and bank stabilization in addition to weed
control (Castro et al., 2004; Chenoweth et al., 2011).
Previous studies have revealed the potential for greater homogeneity in post-damremoval sediment than in underlying topsoils. Such environments may favor colonization
by invasive plants such as reed canary grass, which can thrive in wet substrates with
homogeneous chemical, textural and nutrient properties. This advantage over native
plants, many of which fare better in heterogeneous substrates, can allow this grass to alter
the successional pathways of native vegetation (Wells et al., 2008). Conversely, the
establishment of native plant communities of sufficient biodiversity and species richness
may create a competitive environment more likely to resist invasion by exotic plant
species (Kennedy et al., 2002).
Post-dam-removal sediment is not a comparable substrate to developed soil which
supports most plant communities. Therefore, no existing plant communities or
associations can serve as models on which to base the species composition of the ONP
Lake Mills revegetation efforts (Dave Allen, 2012, personal communication). For this

27

reason, testing a variety of native vegetation types may be equally important to invasive
weed control in the dewatered Lake Mills reservoir.

3.2 Research Question
The post-dam-removal sediment remaining in the dewatered Lake Mills reservoir affords
an environment which is potentially inhospitable to native plant colonization.
Establishment of native riparian vegetation is an important component of the restoration
of a riverine ecosystem (Chenoweth et al., 2011). The research question guiding this
study centers around which woody plants, native to the lower Elwha River, show the
greatest survivability in the post-dam-removal sediments of Lake Mills. Does species
selection influence woody plant survival in specific sediment textures? An additional
research question focuses on the revegetation treatments employed by ONP. Do planting
adjacent to woody debris or adjacent to other vegetation affect plant performance?
Answering these questions may aid in the selection of suitable species for future planting
efforts at the former Lake Mills reservoir site. Additionally, these findings may help set
the precedent for future post-dam-removal plant restoration efforts.

28

IV. Methods

4.1 Study Sites
The purpose of this study was to determine whether certain woody plant species perform
better than others in Lake Mills post-dam-removal sediments, and additionally whether
proximity to WD and other vegetation enhances plant performance. To meet the
objectives of this study, the survival and growth performance of six native woody plant
species were compared in three different Lake Mills sediment textures: 1) fine sediment:
composed of alluvial silt and clay; 2) coarse sediment: composed of cobbles, gravel and
sand; and 3) a mixed sediment composed of silt, clay and sand. The restoration plots
assessed in this study lie on the southern, southwestern, and southeastern shores of the
former Lake Mills reservoir, numbered Site 1, Site 2, and Site 3, respectively (Map 1).
Site 1, the southernmost study site containing coarse sediment, lies on a bench at the
Boulder Creek delta of the former Lake Mills. Here, several planted sites totaling 0.30 ha
were included in the study. Sites 2 and 3 lie on the southwestern and southeastern shores,
totaling 0.61 ha and 0.20 ha, respectively. Depending upon aspect, each site is subject to
varying degrees of wind exposure, with sites 1 and 3 experiencing especially windy
conditions. The interactive effects of heavy wind, low soil moisture, and high
temperature can negatively influence the performance of woody plants (Heiligmann and
Schneider, 1974).

Three planting prescriptions, with different planting densities and lifeforms, were
assessed in this study (Table 1). Site 1 plots in this study include three, 0.10-ha plantings.
Two of these plots were planted at a density of 8,000 plants per acre of shrubs only,

29

totaling 12,000 plants (prescription 1). The third plot, east of and separated from the other
two plots by a small gully, contains trees only at 3,000 plants per acre (prescription 3).
The majority of
plants monitored
lie in Site 2,
south of Stukey
Creek and 0.53
km north of
Boulder Creek.
Map 1. Lake Mills study site locations

In this 0.65hectare site
consisting of
Map1. Lake Mills woody plant restoration 2012 study sites

fine sediment,

six, 0.10-hectare plots were planted with trees, shrubs and forbs, at a density of 9,000
plants per acre (prescription 2); the other three plots were planted according to
prescription 3. Site 3, directly across the drained reservoir from Site 2, is 0.20 ha in size,
with a sediment texture of mixed fine and sandy sediment. At this site, two 0.10-ha plots
were planted according to prescriptions 2 and 3. The plot planted according to
prescription 2 lies downslope from and west/northwest of the plot planted according to
prescription 3, in a terrace formation.

4.2 Selection of Woody Plants and Prescriptions
We selected woody plant species to include in this study by reviewing primary literature,
consulting plant ecologists, and basing our decisions on planting trials conducted by ONP
30

Location
Site

1

Site
Size

Planting Expos.
Density/
lifeforms

South
(Delta)

0.60 ha
8,000/acre
(3, 0.20-ha Shrubs
plots)
only

Southwest

Southeast

Sediment
Texture

Prescriptions
Applied

Coarse
(Gravel,
cobbles,
sand)

1, 3

0.60 ha
9,000/acre NE
(6, 0.10-ha Trees,
plots)
shrubs,
herbs

Fine (Silt,
clay)

2, 3

0.20 ha
3,000/acre NW
(2, 0.10-ha Trees only
plots)

Mixed

2, 3

N

2

3

Table 1. Study sites and planting prescriptions, Lake Mills 2012 survivability study

during the fall of 2010. The following plants were selected for this study (Table 1):
oceanspray (Holodiscus discolor), Nootka rose (Rosa nutkana) (ONP Planting trials,
2011), thimbleberry (Michel et al., 2011), western redcedar (D’Amore et al., 2009), black
cottonwood (Populus balsamifera ssp. trichocarpa) (Naiman and DeCamps, 1997, 2005;
Naiman et al., 2010), and Douglas-fir (Pseudotsuga menziesii) (Joshua Chenoweth and
Dave Allen, personal communications).

Ocean spray and Nootka rose were selected due to their performance in
experimental sediment planter boxes installed at ONP’s Matt Albright Native Plant
Center near Agnew, WA. In this experiment, three substrates were placed in multi-celled
wooden planter boxes with the following substrates: 1) coarse sediment and 2) fine
sediment, both collected from the northern Lake Mills shoreline neighboring the Glines
Canyon dam; and 3) a potting soil control. Oceanspray and Nootka rose plugs were
planted in the cells, tagged and observed over time for survival, mortality and growth
31

performance. The majority of these shrubs survived, with only two oceanspray
mortalities which likely resulted from sudden freezing temperatures at the ONP nursery
site (Joshua Chenoweth, personal communication). Thimbleberry was selected for this
study due to its tolerance of a variety of substrates and moisture regimes, general
hardiness in riparian environments, dense growth pattern, ability to spread via rhizome
and seed, as well as its successful germination and growth in fine sediment in a recent
seed rain study (Michel et al., 2011).
Trees selected in this study included both late and early-successional riparian
species. Western redcedar was selected due to its abundance in the lower Elwha River
and tolerance of relatively inundated growing conditions (D’Amore et al., 2009; Naiman
et al., 2010). Additionally, cedar trees can grow in nitrogen-poor substrates, and
contribute calcium-rich litterfall which may increase N turnover over time by increasing
pH (D’Amore et al., 2009). The major challenge for this tree may be its survival during
drier periods, particularly in fine sediment. Douglas-fir seedlings were abundant in
plantings due to highly successful germination, and this tree is prolific in late-seral forests
along the lower Elwha River (Wendel and Zabowski, 2010). While Douglas-fir is a latersuccessional tree in riparian environments, the dynamic and transitional nature of the
restoration sites warrants its inclusion in order to test a wide range of species.
Black cottonwood was selected due to its survivability in a variety of disturbed
environments and substrates. Cottonwoods and willows tend to establish in flood-prone
exposed sites, but impoundment of the Elwha River has reduced the extent and frequency
of flood disturbance conducive to creating such appropriate environments (Shafroth et al.,
2002). Cottonwood trees can perform well in poorly drained hypoxic soils, largely due to

32

the development of adventitious roots (Van DerKamp and Gokhale, 1979; Hanson, 1997;
Naiman and Decamps, 1997; Haycock et al., 2003; Naiman et al. 2010). Black
cottonwood trees and willows of the family Salicaceae are phreatophytes, possessing
extensive, deep root systems capable of accessing riparian water tables during dry periods
(Naiman, Decamps and McClain, 2005). The roots of these trees grow especially fast as
soil moisture diminishes, in some cases over 38 cm within the first two months (Naiman,
Decamps and McClain, 2005). Establishment of Populus varieties can be enhanced in
environments where the spring season brings high levels of moisture or inundation
followed by dry summer seasons (Naiman, et al., 2005).
In addition to their survivability in extreme conditions, black cottonwoods can
provide many ecological benefits in stream and riparian restoration. Cottonwood trees are
known for their ability to colonize unfruitful, inundated or semi-arid riparian habitats that
may hinder initial establishment of other trees, creating a facilitative canopy cover, and
also possess resprouting capabilities when water movement shears saplings and transports
them downstream (Rood et al., 2003; Gurnell et al., 2005). Nitrogen-fixing bacteria
housed in the stems of cottonwoods allow these trees to produce nitrogen even in
extremely nitrogen-poor soils (Doty et al., 2009; Gang Xin et al., 2009). Additionally,
cottonwood may play an essential role in post-dam-removal floodplain and perched
terrace stabilization, and contribute LWD to the Elwha River. The short, 80-150 year
lifespan and rapid growth of this tree species enables a more rapid production of LWD
available to streams than that of longer-lived native conifers, and larger WD than that of
other short-lived early successional broadleaf trees such as red alder (Collins and
Montgomery, 2002; Chenoweth et al., 2011).

33

Red alder, while included in the 2012 ONP Lake Mills plantings (Table A1,
Appendix A) and currently more common than black cottonwood in floodplain
environments of the lower Elwha River, may not be as well-suited as other earlysuccessional riparian trees to post-dam-removal fluvial geomorphology (Shafroth et al.,
2002; Chenoweth et al., 2011). Therefore, this tree was not included in the 2012 woody
revegetation study. Large quantities of sediment, carried and deposited by the river, may
create hypoxic burial and inundation states which can be inhospitable to red alder
survival, as these trees have shown poor performance in fine, hypoxic sediments
(Shafroth et al., 2002). Additionally, red alders lack the deep rooting systems of
cottonwoods and willow crucial to accessing increasingly lowered water tables. In 2003
seeding trials conducted by ONP, red alder did not survive beyond one growing season
(Chenoweth et al., 2011).
4.3 Experimental Design
Plantings assessed in this study were installed by the ONP Elwha Revegetation Crew and
Washington Conservation Crew (WCC) according to a technique which facilitates
shading, moisture retention, nutrients and protection from herbivory, known as a
facilitation patch strategy (del Moral and Wood, 1993; Chenoweth et al., 2011). Dense
islands of trees and/or shrubs were installed, followed in certain treatments by the
interplanting of less dense corridors of woody plants, forbs and graminoids. Dense
placement of woody plants affords a nurturing microsite which may aid in the survival
and establishment of other plantings and volunteer seeds, and eventually may serve as a
seed source in exposed sediments >50 m from forest edge (Chenoweth et al., 2011). The
shed leaves and litter of the planted shrubs in turn contribute organic matter to the litter-

34

depauperate sediment, providing a more suitable environment for seed establishment.
Eventually, these islands of woody plants may reach a state of plant species competition
leading to natural selection of the most site-suitable species (Chenoweth et al., 2011).
Additional techniques of ONP planting strategies entail plant placement. Plants
were placed among onsite woody debris when possible, and “armored” thorny shrubs,
less palatable to herbivores, were strategically placed around plants more desirable to
herbivores (Padilla and Pugnaire, 2006). Western redcedar and berry-producing shrubs in
particular require these or similar protection measures. Over time, seeds of established
berry-producing plants may be distributed throughout the exposed reservoir through
zoochory (Chenoweth et al., 2011). The three planting prescriptions tracked in this study
are different variations of facilitation patches in three southern Lake Mills locations.
Plantings were positioned at least 50 m from the edge of the intact forest. Within this
distance to forest edge, seed rain from established trees and vegetation aide in natural
plant regeneration, negating a need for artificial regeneration efforts (Chenoweth et al.,
2011; Michel et al., 2011). Predictive models in previous vegetation studies depicted a
drastic decline in seed dispersal beyond 170 m from forest edge (Greene and Johnson,
1996).
Field Study Design
For each of the 6 woody species tracked in this study, 30 replicates per 0.10-ha
restoration plot were tagged, totaling N=860 individual trees across all sites. The
exception was western redcedar, for which only 20 plants were installed in each plot
(Table 2). The field component of this study required two major stages: 1) selection and
marking of replicates, followed by 2) monitoring throughout the growing season. Plants

35

Woody Plants Selected for 2012 Lake Mills Survivability Study
Species
Oceanspray
(Holodiscus discolor)
Nootka rose
(Rosa nutkana)
Thimbleberry
(Rubus parviflorus)
Western redcedar (Thuja plicata)
Douglas fir
(Pseudotsuga menziessii)
Black cottonwood (Populus balsamifera)
Grand Total =

Totals
180
120
180
80
150
150
860

Totals
Site 1
60
0
60
0
30
30
180

Totals
Site 2
90
90
90
60
90
90
510

Totals
Site 3
30
30
30
20
30
30
170

assessments
included
qualitative
vigorMills
(leaf coloration,
size and
shape),
Table
2. Native
plantsurvival,
species assessed
in Lake
study, and totals
number
of and
individuals for each species per site
assessed in this study were selected using random number generators for azimuth and
distance, and marked with numbered aluminum tags. Plant performance assessments
included survival, qualitative vigor (leaf coloration, size and shape), and minimal growth
performance measurements. The temporal scale of this study (i.e., one growing season) is
likely not sufficient for significant measurable plant growth. Growth can be
underestimated because bare root and potted plants, when newly installed into developed
soil, invest much of their initial energy in root growth rather than foliar growth during the
first 1 to 2 growing seasons, and may exhibit lower root growth rates than those of
cuttings (Alpert et al., 1999). The Lake Mills restoration sites lack surficial developed
soil and pose yet a greater challenge for plant growth beyond root establishment. It may
require at least two to three growing seasons for substantial between-site growth
differences to be significant (Woodruff et al., 2002). Nevertheless, the height and
diameter or lateral growth of 155 tagged plants were randomly measured using a meter
stick and dial caliper (Table 17, Appendix B). To establish baseline data, measurements
were initially conducted in April 2012, early in the growing season, and repeated in late

36

September 2012. Diameter measurements were taken from 5 cm aboveground.

In the process of plant selection and tagging, vegetation and woody debris (WD)
within 1m2 of replicates were documented (Eränen and Kozlov, 2007). Vegetation in
close proximity to plantings may result in either beneficial or harmful interactions
(Callaway, 1995). Competition for moisture and space can occur between adjacent plants
both above and belowground, particularly during dry periods or other environmental
conditions that intensify plant stress (DeSteven, 1991; Roberts et al., 2005; Smit et al.,
2006; Sthultz et al., 2007). Conversely, the presence of the neighboring plants may serve
as “nurse” vegetation, aiding in shading, moisture retention or provide a nurturing
microclimate for certain species (Coffman, 1975; del Moral and Wood, 1993; Callaway,
1998; Padilla and Pugnaire, 2006). For newly introduced seedlings which are not adapted
to an open, disturbed environment, shelter of adjacent plants may be especially beneficial
(DeSteven, 1991). WD provides moisture retention, a shaded microclimate and nutrients
to adjacent vegetation (Naiman et al., 2010). The moisture retention neighboring plants
receive from WD is of potentially greater benefit than that provided by adjacent
vegetation, as WD creates no interspecific competition (Coffman, 1975). LWD in
particular can serve as nurse logs, providing the aforementioned benefits as well as
elevating plants which are less tolerant to inundated conditions (Fetherson et al., 1995).
Finer woody debris, particularly litter, can potentially produce a positive or negative
effect on seed and seedling establishment, contributing to competition for light and
moisture or moisture retention and nutrient competition (Goodson et al., 2003). Woody
debris encountered was classified as: coarse (C) = diameter of 10 cm or greater; fine (F)
= diameter of <10 cm; and litter (L) = diameter of 1 cm or less.
37

4.4 Onsite Sediment Analysis
Post-dam-removal sediment may create an inhospitable environment for plant growth in
comparison to developed soil. As a means of measuring environmental stress and testing
the hypothesis that substrate influences plant performance, three abiotic factors were
measured for sediment at each site: 1) moisture using two different methods; 2) particle
size; and 3) nutrient analysis. Soil moisture, coupled with texture, particle size and
nutrient content, can largely impact plant survival and growth (Sthultz et al., 2007),
particularly potted and bare root specimens. The Lake Mills post-dam-removal sediments
are subject to greater seasonal moisture extremes than are developed soils.

Sediment moisture was assessed every 3 to 4 weeks using a Quick Draw 2900F1
portable tensiometer (Soil Moisture Corp, Santa Barbara, CA). Tensiometers measure the
capillary force or suction which roots must exert in order to draw water from soil
particles at different moisture contents and in a wide range of sediment textures. In other
words, these instruments measure the amount of moisture available to be utilized for
plant growth in units of positive or negative pressure (Soil Moisture Equipment Corp,
2012). With measurements in centibars (1 centibar = 1/100th of a bar, 1 kilopascal or 0.14
pounds per square inch (PSI)), they reflect plant root stress within 1.5% accuracy
throughout the growing season while affording complete onsite assessment Stoeckler and
Aamodt, 1940; Robertson et al., 1999). Adhesion of soil particles to water varies with soil
pore size and texture. Soil particles possess an attraction to water contained the soil; the
finer the soil particles, the greater their binding to water molecules, and the lower the
availability of water to plant roots (Soil Moisture Corp, May 2012). Finer sediments,
while inundated during wetter periods of a growing season, provide less available water
38

for plants during drier seasonal periods, since the water contained therein is “bound”
(Stoeckler and Aamodt, 1940; Robertson et al., 1999). To gather heuristic baseline
sediment moisture data, gravimetric water content (GWC) was also assessed with
collection of sediment cores from each moisture sample site, at 20 cm depths, in May and
late September (Black, 1965).

Sediment moisture sampling sites were divided among planted sites, with 10 to 11
sample sites in each 0.10-ha restoration plot (n = 111) (Gotelli and Ellison, 2004). The
tensiometer probe was placed at depths of 20 cm to measure available moisture (Roberts
et al., 2005). High variability between and within plots and planting sites required a large
sample size for moisture measurement, and this number proved to be the maximum
number feasible in relation to time constraints. Moisture sampling sites were randomly
selected using the same technique as for replicate tagging (Appendix A). In addition to
moisture assessment, pH was recorded at each sample site through the use of a portable
Kelway® soil pH and moisture meter (Kel Instruments, Inc., Wyckoff, NJ).

4.5 Vegetation Monitoring
Monitoring, conducted from May through September of the 2012 growing season, was a
vital component of assessing plant survival, performance and changes over time. The
primary aspect of plant performance monitored in this study was the survival or mortality
of tagged plants in fine, coarse and mixed fine-coarse sediment textures. Visual
inspection of live biomass on each plant was the primary method for determination of
survival or mortality (Schaff and Pezeshki, 2003; Greer et al., 2006). Mortality was
indicated by a lack of foliage, all or mostly dead branches and twigs, and scratching

39

trunks or stems to inspect for green tissue if necessary. It was assumed in this study that
plants which did not produce foliage by late summer or fall, beyond the peak of the
growing season, did not survive; however, future monitoring will determine whether this
assumption was correct.

Plant vigor was assessed qualitatively through leaf color, leaf shape (i.e., curled
under, curled up or other deformation) and rough size (“full” or “diminished”). Leaf color
can be linked to nutrient availability in a given substrate, or environmental conditions. In
particular, nutrient limitation may have a greater influence and be more visible in
younger, early successional plants (Chapin et al., 1986). Nitrogen is especially critical to
riparian plant and ecosystem development (Bradshaw, 1982). Yellow leaves may indicate
chlorosis from a lack of nitrogen, while red leaf color may indicate sun stress (Stewart,
1999). Leaf pigmentation results from anthocynanin production, which is associated with
the protection of photosynthetic structures in leaf tissue during periods of excessive light
exposure. Anthocyanin production has also been associated with environmental stresses
such as drought or nutrient limitation (Hoch et al., 2000; Hoch et al., 2003), two key
factors that may affect plant growth in the dewatered Lake Mills restoration sites. The
water-soluble pigments created in anthocyanin production responsible for red leaf
coloration may aid leaves in water retention during periods of drought (Chalker-Scott, L.,
WSU). “Scorched” leaf edges or spots can be indicative of low soil potassium levels
(Stewart, 1999). Additionally, leaf scorch can result from drought, drying winds or poor
root development (Purdue University Plant and Pest Diagnostic Laboratory, 2002).
Herbivory, an additional factor affecting plant performance, was also documented during
monitoring. Signs of plant predation such as chewed or broken off stems or leaves were
40

recorded, including signs of insect herbivory on leaves (i.e., holes or skeletonization).
4.6 Laboratory Analysis
Sediment Moisture and Nutrient Analysis
Sediment cores were collected at all moisture sample sites in May and late September at
depths of 20 cm (Henderson et al., 1989; Roberts et al., 2005). To assess gravimetric
water content (GWC), sediment samples were sieved and dried at 70°C for 72 h; samples
were weighed after 48 h of drying, then at 72 h to ensure samples had completely dried.
Ideally, soil samples are dried at the standard 105°C temperature (Black, 1965).
However, two soil ovens were required to dry a large number of samples with limited
time constraints, and one of these ovens reaches a maximum temperature of only 70°C.
In order to be consistent with methodology, this temperature had to be applied to all
sediment samples.

Using an AQ1 discrete autoanalyzer (Seal Analytical Inc., Mequon, Wisconsin) a
subset (N=34) of the fall 2012 sediment cores collected from moisture sample sites were
tested for nitrate-N and phosphate-P concentrations (mg N/L and mg P/L). Ten g of wet
sediment were extracted into 100 ml of potassium chloride (KCl). Bottles were shaken
for 1 h and allowed to settle overnight. The supernatant was then filtered through KClinfused Grade 42 Whatman filters to create an extract, and frozen until analysis
(Mulvaney, R.L., 1996). Extracts were tested for nitrate-N and phosphate-P using the
AGR 31-B Nitrate-N+Nitrite-N and AGR-204-B Orthophosphate-P in 2 M KCl Extracts
of Soil methods, respectively. Reagents employed included 2% copper sulfate (CuSO4), a
high standard reagent, KCl, and a nitrate buffer.

41

Sediment particle size can shape organic matter-derived nutrient retention
capabilities during early soil development (Naiman et al., 2010). In addition to moisture
and nutrient analysis, sediment particle size was characterized from sixteen randomly
selected subsamples of the Lake Mills sediment. Nine of these sediment samples were
assessed using the hydrometer method (Bouyoucos, 1962; Gee and Bauder, 1986). A
Bouyoucos ASTM No. 152H hydrometer with Bouyoucos scale in g/L was employed to
measure clay, silt and fine sand (H-B Instrument Company, Trappe, PA). Due to a very
low presence of organic matter in all study sites, sediment samples were not pretreated
with a hydrogen peroxide (H2O2) digestion. Seven coarse samples from Site 1 were
assessed through dry sieving to separate gravel (particles >2mm), sand (0.05 - 2 mm), silt
(0.002-0.05 mm) and clay (<0.002 mm) (Caplan et al, 2006).
4.7 Statistical Analysis
The categorical and continuous nature of the data collected in this study and the variables
hypothesized to influence plant performance required the use of both parametric and nonparametric analytical methods. R and JMP software were utilized to conduct statistical
analysis. Categorical variables included survival or mortality, plant species, planting in
woody debris, presence of other vegetation within 1m,2 leaf color, and sediment type.
Continuous data consisted of plant growth, sediment moisture, percent mortality and pH.
Therefore, Pearson’s Chi-square tests for independence were appropriate to test for
interactions between plant survival and treatments. Levene’s tests for equality of variance
and Shapiro-Wilk’s tests of normality for plant growth data revealed non-normal datasets,
inappropriate for analysis of variance (ANOVA) tests (Alpert et al., 1999). Plant height
growth by sediment type, moisture, and pH were analyzed by performing Kruskal-Wallis

42

rank sum tests. Percent leaf color and percent plant mortality were transformed using an
acrsine-square root transformation to perform one-way ANOVAs. Total leaf color by
species and sediment type were also examined using one-way ANOVAs, to detect
relationships between leaf color, plant species and sediment type. Tensiometer and GWC
data were analyzed with repeated-measures ANOVAs, testing the influence of sediment
texture and month on moisture availability and GWC. Percent moisture data from GWC
assessments were transformed through arcsine square root transformations prior to these
tests.

43

V. Results

5.1 Site Conditions
Overall, weather during the Lake Mills 2012 growing season remained cool and wet until
the months of late July through September (Table 3). By October, rainfall resumed and
temperatures dropped by nearly 9° C (http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?
wa2548). Windy conditions were frequent at all sites. The Boulder Creek delta (Site 1)
received especially high gusts during 2012 surveys. Archival 2009 and 2010 wind data
collected from an anemometer formerly installed on the Glines Canyon Dam, reported
average wind speeds of 10 kph, with wind speeds periodically reaching speeds of greater
than 29 kph (USGS data, courtesy of ONP GIS office, 2012).
5.2 Plant performance
Overall, mortalities of tagged plants were low at all sites during the 2012 growing
Season (Table 4); however, variation was detected in plant performance among species
and sites. During the month of May, a total of 30 plants had died (mortality = 23

Month
Apr
May
Jun
Jul
Aug
Sep
Oct

Lake Mills Weather Data Spring through Fall 2012
Average
Average
Max
Average
Mean
Air
Min Air
Air
2012
Average
Temp
Temp
Temp Rainfall
Rainfall
(C)
(C)
(C)
(cm)
(cm)
13.4
16.4
17.1
22.2
23.9
22.2
13.5

3.3
4.8
6.9
9.8
10.7
8.5
5.1

8.4
10.6
12.0
16.0
17.4
15.3
9.3

9.2
3.3
6.3
2.0
0.2
0.3
20.1

8.5
4.5
3.1
1.9
2.8
4.4
13.3

Table 3. Lake Mills 2012 weather data, taken from the Elwha
Ranger Station Weather Station
44

Site 1 – Coarse
Sediment
Species

HOLDIS
RUBPAR
*ROSNUT
*THUPLI
PSEMEN
POPBAL
Total

Survival

57
59
_
_
12
30
158

Site 2 – Fine
Sediment

Mortality Species

3
1
_
_
18
0
22

HOLDIS
RUBPAR
ROSNUT
THUPLI
PSEMEN
POPBAL
Total

Site 3 - Fine/Sandy
Sediment

Survival Mortality Species

89
89
87
56
57
90
470

1
1
3
2
33
0
40

HOLDIS
RUBPAR
ROSNUT
THUPLI
PSEMEN
POPBAL
Total

Survival Mortality

30
30
30
17
27
29
166

0
0
0
0
3
1
4

Table 4. Survival and mortality by species and site. *Nootka rose and western redcedar
were not planted in Site 1. HOLDIS=oceanspray, RUBPAR=thimbleberry, ROSNUT=
Nootka rose, THUPLI=western redcedar, PSEMEN=Douglas-fir, POPBAL=black
cottonwood
Douglas-fir, 4 oceanspray, 3 Nootka rose). During the month of July, mortality had
increased to 43 (additional mortality = 10 Douglas-fir, 1 thimbleberry, 2 western
redcedar). By August, tagged plant mortality reached 59 (additional mortality = 14
Douglas-fir, 1 cottonwood, 1 thimbleberry). In the final September 2012 monitoring,
plant mortality totaled 66 for all sites, with the remaining 7 mortalities occurring among
Douglas-fir seedlings. Of the 6 species selected for this study, Douglas-fir had the highest
mortality rate at 36% of tagged Douglas-fir seedlings and 82% of total tagged plant
mortalities. Proportionally, the highest percentage of plant mortalities occurred in Site 1
(coarse sediment), at 12% mortality of onsite tagged plants, compared to 8% in Site 2
(fine sediment) and 2% in Site 3 (fine/sandy sediment). By individual species, Douglasfir, oceanspray and thimbleberry experienced the highest mortality, relative to individual
site sample size, in the coarse sediment (Table 5). Nootka rose and western redcedar
mortalities occurred only in the fine sediment (these species were not planted in the
coarse prescriptions tracked in this study), while the only black cottonwood mortality
occurred in the fine/sandy sediment, possibly as a result of browsing.

45

Lake Mills 2012: Percent Mortality for Plant Species by Individual Site
HOLDIS

ROSNUT

RUBPAR

POPBAL

THUPLI

PSEMEN

Coarse

5%

N/A

2%

0%

N/A

60%

Fine

1%

3%

1%

0%

3%

36%

Fine/Sandy

0%

0%

0%

3%

0%

10%

Table 5. Plant species percent mortalities, proportional to sample size, by individual sites
According to Pearson’s Chi-Square tests for independence for individual study
sites, sediment type and plant species significantly influenced plant survival for sites 1
and 2, but not for Site 3 in the mixed sediment (Table 6). Adjacent woody debris showed
significant influences only in the fine sediment (Site 2), while woody debris size was not
significant to survival in any of the study sites. Adjacent vegetation significantly
influenced survival in the fine and fine/sandy mixed sites, while browsing
Significantly influenced plant survival in the coarse and fine sediment types. Pearson’s
Chi-Square tests for independence for combined study sites by month revealed significant
Pearson’s Chi-Square Test for Independence, Individual Sites:
Plant Survival by Treatments, Browsing
Sediment

Coarse

Species
X²

= 77.16

df

Fine/sandy

X²

X²

=3

p-value = <0.0001

Fine

WD

df

WD Size

= 1.89
=2

p-value = 0.39

= 135.86 X²

df

= 5.84

X²

=5

p-value = 0.32

df

= 10

df

=4

p-value = 0.34

X² = 14.67

df

= 0.87
=1

Browsed
X²

= 11.61

df

= 2

p-value = 0.35 p-value = 0 .003

= 341.07 X²
= 19.88
df
= 10 df
=6
df = 12
df
= 4 df
= 2
p-value =
p-value = 0.26
p-value = <0.0001
p-value = <0.0001
p-value =<0.0001 <0.0001
X²
= 13.88
X²
= 4.53
X²
= 1.77
X²
= 9.11
X²
= 1.06
p-value = 0.18

= 233.86

X²

Other Veg

df

= 10

p-value = 1.00

X²

df

=2

p-value = 0.01

df

=2

p-value = 0.59

Table 6. Chi-Square table: plant survival compared to treatments and browsing by
sediment

46

Pearson’s Chi-Square Test for Independence, Combined Sites:
Plant Survival by Treatments, Browsing
Species

Month

May

July

August

WD

Vegetation 1 m2

X²

= 12.31

X²

= 6.66 X²

df

= 10

df

=4

df

= 28.43
=4

Sediment type

Browsed

X²

= 13.00

X²

= 5.47

df

=4

df

= 2

p-value = 0.27

p-value = 0.18 p-value = <0.0001 p-value = 0.01

p-value = 0.06

X²

= 38.30

X²

= 4.68 X²

= 31.01

X²

= 23.50

X²

=

7.96

df

= 15

df

=6

=6

df

= 6

df

=

3

df

p-value = 0.001

p-value = 0.67 p-value = <0.0001 p-value = 0.0006 p-value = 0.05

X²

= 41.41

X²

= 3.45

X²

= 862.06

X²

= 26.02

X²

= 10.00

df

= 20

df

=8

df

=8

df

=8

df

= 4

p-value = 0.003 p-value = 0.90 p-value = <0.0001 p-value = 0.001

September X²

= 19.96
df
= 20
p-value= 0.03

p-value = 0.04
= 10.22
X²
= 2.79 X²
= 344.59 X²
= 23.31 X²
df
= 2
df
= 4
df
= 4
df
= 4
p-value
= 0.006
p-value = 0.59 p-value = <0.0001 p-value = 0.0001

September (Table 6), while woody debris did not appear to significantly influence plant
Table 7. Chi-Square table, plant survival compared to treatments for all sites

differences in plant survival by species and adjacent vegetation for the months of July
through September (Table 7), while woody debris did not significantly influence survival
for any month of the growing season. The influences of sediment type and browsing on
plant survival were significant for all or most months of the 2012 growing season.
Leaf color varied among the three Lake Mills study sites, but yellow and red leaf
coloration increased throughout the growing season in all sites. Leaves were primarily
green during the first assessment in May, with red and yellow leaf coloration making up
only 3 and 7% of all plants examined, respectively (Table 8). By August, red leaf
coloration reached its peak at 54% of the total studied plant population and highest in the
coarse sediment, declining slightly by the late September assessment. Overall yellow leaf
coloration rose to 45% by late September, with the highest percent in the fine/sandy
sediment. Scorched leaf edges, leaf spotting, leaf curling and diminished leaf size also
increased throughout the growing season for most plant species in all sites (Table 9).
47

Month

May
July
August
September

% Yellow Leaf Color
By Sediment Texture
Coarse
Fine
Fine/Sandy
0%
3%
2%
18%
33%
46%
32%
34%
50%
41%
38%
55%

% Red Leaf Color
By Sediment Texture
Coarse
Fine
Fine/Sandy
0%
7%
0%
58%
44%
24%
61%
45%
55%
56%
26%
22%

Table 8. Yellow and red leaf coloration (% of total tagged plants) by sediment
Percent leaf color by species and month varied considerably. By the end of the growing
season, Nootka rose and western redcedar displayed the highest percentage of yellow leaf
coloration at 70% and 60% (Table 9). Highest red leaf coloration was observed in
Douglas-fir and thimbleberry, both at 59% by fall 2012. Thimbleberry displayed the
greatest percent scorched leaf edges (71%) and leaf deformation (69%); oceanspray the
2012 Lake Mills Qualitative Leaf Data by Species
Species

Month

% Yellow
Lvs.

% Red
Lvs.

Oceanspray

May
Jul
Aug
Sept
May
Jul
Aug
Sept
May
Jul
Aug
Sept
May
Jul
Aug
Sept
May
Jul
Aug
Sept
May
Jul
Aug
Sept

0
22
30
33
2
19
21
39
5
50
33
39
3
55
63
70
0
24
34
35
0
40
44
60

7
63
52
52
1
22
15
10
8
45
63
59
0
18
14
18
3
57
60
59
10
44
39
29

Black
cottonwood

Douglas-fir

Nootka rose

Thimbleberry

Western
redcedar

%
Scorched
Edges
0
3
25
30
0
0
2
1
0
0
0
1
0
1
13
23
0
8
57
71
0
1
3
4

%
Spotting

% Lvs.
Deformed

% Lvs.
Diminished

0
32
54
66
0
9
24
22
0
1
3
3
0
3
35
39
0
23
31
43
0
0
3
24

0
6
11
27
0
15
23
29
0
23
27
32
0
8
8
15
0
45
62
69
0
8
9
11

0
69
69
30
0
96
94
5
0
41
45
42
0
85
83
14
0
51
57
41
0
79
79
16

Table 9. Qualitative Leaf data by plant species and month

48

One-Way Analysis of Variance: Leaf Color and Mortality by Month, Sediment
Red Leaves
Yellow Leaves
Mortality
F = 7.99
F = 19.30
F = 0.31
Month
DF = 3
DF = 3
DF = 3
P = 0.82
P = 0.009
P = 0.0005
F = 0.13
F = 0.44
F = 32.23
Sediment type
DF = 2
DF = 2
DF = 2
P = 0.88
P = 0.65
P = <0.0001
F = 1.10
F = 0.43
F = 3.08
Species
DF = 5
DF = 5
DF = 5
P = 0.006
P = 0.74
P = 0.01

Table 10. One-way ANOVA, leaf color by month, sediment and species

greatest leaf spotting (66%); and Douglas-fir the highest percentage of diminished
needles (41%). During July and August up to 96% of black cottonwood specimens had
diminished leaves, 73% of which was due to caterpillar defoliation, but late summer
foliar growth reduced this percentage to 5% by late September. Results of One-way
ANOVAs indicated that month during the growing season significantly influenced
percent leaf color, while sediment type and plant species significantly influenced plant
mortality (Table 10).
Plant growth increases generally proved minimal, but varied by site (Table 11;
Figure 1). Measured oceanspray and thimbleberry were browsed in all three sites,
resulting in negative or minimal growth trends in the coarse and mixed sediments. By
individual site, the fine/sandy sediment site experienced the highest plant growth, with
western redcedar, Nootka rose and cottonwood total height growths reaching 285, 127
and 165 mm, respectively (Table 11). Following a Kruskal-Wallis rank sum test for all
sites (Table 12), species, pH and GWC significantly influenced plant growth while
sediment type did not. Testing the relationship between growth of individual plant
species, sediment type and pH, sediment significantly influenced oceanspray lateral

49

Plant Growth by Species
Height (mm)

Lateral or Diameter (mm)

*P = <0.0001

HOLDIS POPBAL PSEMEN ROSNUT RUBPAR THUPLI

*P = <0.0001
HOLDIS POPBAL PSEMEN ROSNUT RUBPAR THUPLI

*P = 0.53
*P = 0.12

Coarse

Fine

Fine/Sandy

Coarse

Fine

Fine/Sandy

Figure 1. Median plant growth by species and sediment type. *Kruskal-Wallis rank
sum tests for plant growth by species and sediment texture

50

Species

Height (mm)
Lat. Grwth. (mm) Diam. (mm)
Comments
Mean
Stdev Mean
Stdev Mean
Stdev
Site 1 - Coarse

Holodiscus discolor
Rubus parviflorus
Pseudotsuga menziessii
Populus balsamifera

-8.00
-27.92
8.00
123.29

94.72 -4.25
65.49 99.41
20.30
---192.66
---Site 2 - Fine

50.01
---113.95
------0.11
---1.04

---Browsed
---Browsed
0.28
0.61

Holodiscus discolor
Rosa nutkana
Rubus parviflorus
Thuja plicata
Pseudotsuga menziessii
Populus balsamifera

26.00
76.35
47.93
111.84
19.57
101.38

124.28 184.31
78.12 204.07
56.97 138.07
119.16
---32.1
---135.98
---Site 3 - Mixed

175.09
---80.17
---57.21
------0.79
---0.40
---1.76

---------1.57
0.76
1.20

Holodiscus discolor
Rosa nutkana
Rubus parviflorus
Thuja plicata
Pseudotsuga menziessi
Populus balsamifera

-66.33
127.00
1.33
285.33
39.75
165.11

51.47 257.67
79.68 252.67
88.10 170.67
21.01
---17.07
---200.16
----

48.21
---79.68
---21.01
------3.22
---0.58
---5.52

---Browsed
------88.10
17.07
200.164

Table 11. Height and lateral (shrub) or diameter (tree) growth of selected studied plants
growth, thimbleberry height, and western redcedar lateral growth. No apparent
significant relationship was found between pH and plant growth by individual species.
However, GWC appeared to significantly influence plant height growth (Table 12).

Plant
Growth

Wilcoxon/Kruskal-Wallis Rank SumTests:
Plant Growth by Site Conditions
Sediment
Species
pH
Moisture*

X2
Df
Height
p-value
X2
Lateral
df
or
Diameter p-value

= 4.96
= 2
= 0.53
= 14.28
=2
= 0.12

X2
= 30.32
Df
= 5
p-value = <0.0001
X2
= 78.99
df
=5
p-value = <0.0001

X2
df
p-value
X2
df
p-value

= 30.29
= 14
= 0.007
= 25.53
= 14
= 0.05

X2
Df
p-value
X2
Df
p-value

= 38.88
= 26
= 0.05
= 34.19
= 26
= 0.13

Table 12. Kruskal-Wallis rank sum test data plant growth. *GWC difference

51

Lake Mills 2012 Sediment Moisture Averages by Site

March

Site 1
Site 2
Site 3
Site 1
Site 2
Site 3

April

6.3
4.7
9.4
----------

May

Tensiometer (centibars)
13.1
19.7
41.8
26.6
49.2
68.6
18.8
20.3
43.1

7.6
5.7
10.6
----------

June/July August September October

Gravimetric Water Content (%)
--------0.13
0.39
--------0.32
---------

Increase

46.1
74.7
50.6

-------------

35
68
41

-------------

Decrease
0.07
6%
0.19
20%
0.23
9%

Table 13. Mean centibars suction and GWC for the 2012 Lake Mills growing season
5.3 Sediment Field and Laboratory Analysis
Tensiometer suction in centibars (cb) increased at all sites throughout the growing
season, indicating increasingly lower water availability in post-dam-removal sediments
throughout the summer months. GWC decreased during the growing season; periods of

Centibars suction (1 cb = 1 kPa)

hotter, drier conditions corresponded with the timeframe in which tensiometer suction

80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0

2012 Lake Mills Mean Tensiometer Readings by Month and
Sediment

Coarse
Fine
Mixed

*F(2, 108) = 67.41
*P = <0.0001

March
1/1

April
1/2

May
1/3

June/July
1/4

August
1/5

Sept
1/6

Figure 2. Mean tensiometer readings (available water) by site and month. *Repeated
Measures ANOVA, monthly readings by sediment texture
52

Centibars Suction (1cb=1kilopascal)

Lake Mills Sediment Moisture 2012
100.0
90.0
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0.00

y = 54.281x + 52.45
R² = 0.0733

*F(2, 108) = 75.81

May-12

*P = <0.0001

Sep-12

y = 32.187x + 10.881
R² = 0.4144
0.10
0.20
0.30
0.40
0.50
Gravimetric Water Content (% Moisture)

0.60

Figure 3. Available moisture versus percent moisture content of sediment in all
study sites. *Repeated-measures ANOVA, May and October GWC by sediment
was high and GWC low (Table 13; Figures 2, 3). It is noteworthy that these figures are
averages, and readings were variable within each plot and site. By the end of the 2012
growing season, individual tensiometer readings reached 85 cb in all sites, particularly in
northern Site 2. Mean seasonal pH for coarse sediments was 6.7, fine sediments 6.0, and
mixed sediments 6.3 (Table 15, p. 55). Repeated-measures ANOVAs for tensiometer
readings and GWC by month showed significant effects of sediment type (Figures 2, 3).
As a general trend, water availability and GWC showed the greatest decrease over time in
fine sediment, while overall GWC was consistently the lowest in coarse sediment.
Following particle size analysis, characterization of sediment cores by site was as
follows (Table 14): 1) predominantly sand, followed by gravel/cobbles and silt/clay in
the coarse sediment site; 2) predominantly silt, followed by sand and clay in the fine and
fine/sandy sediments (in the fine sediment, percent sand content was only marginally
higher than percent clay). Additionally, hydrometer analysis showed highly variable

53

2012 Lake Mills Sediment Particle Size Characterization
Mean Particle Size by Sediment Type
Particle Size Class
Coarse
Fine
Fine/sandy
(mm)
Sieve Hydrometer
Sieve Hydrometer
Sieve Hydrometer
27%
----------------------% Gravel/Cobbles
(>2.0)
61%
38%
-----17%
15%
36%
% Sand
(0.05-2.0)
% Silt
(0.002-0.05)

11%

38%

------

69%

85%

57%

% Clay
(<0.002)

----

24%

------

15%

----

7%

Table 14. Sediment particle size ranges by sediment type
particle size ranges in all sites. Fine sediments ranged from 1% to 32% sand, 18% to
34%clay, and 58% to76% silt, while the mixed sediment samples ranged from 15% to
40% sand, 7% to 8% clay, and 54% to 85% silt. Two samples included in hydrometer
assessments from the predominately coarse sediment site (Site 1) were actually exposed
fine sediment, with particle sizes ranging from 31% to 44% sand, 23% to 26% clay, and
33% to 43% silt. Sieve analysis of coarse sediments revealed gravel/cobble content
ranging from 14% to 31%, sand content from 42% to 79%, and silt content from 3% to
37%.
Sediment Nutrient Analysis
Discrete analyzer nutrient tests revealed low available nitrate and phosphate
concentrations in the 34 sediment cores assessed (Table 15). Nitrate-N/L content was
highest in the fine sediment, and ranged from 1) below levels of detection (-0.006) to
0.118 mg Nitrate-N/L in coarse sediment; 2) 0.005 to 0.197 mg N/L in the fine sediment;
and 3) below levels of detection (-0.001) to 0.551 mg N/L in the fine/sandy mixed
sediment. Orthophosphate content was highest in the coarse sediment, and ranged from 1)
0.01 to 0.156 mg phosphate-P/L in coarse sediment; 2) below levels of detection (-0.004)
54

Sediment

2012 Lake Mills Mean NO3-N , PO4-P
and pH by Sediment Type
NO3PO4mg
mg
N/L
SE
P/L
SE

Mean
pH

Coarse

0.05

± 0.01

0.08

± 0.01

6.7

Fine

0.10

± 0.02

0.03

± 0.007

6.0

Fine/sandy

0.10

± 0.03

0.05

± 0.006

6.3

Table 15. Mean Nitrate-N and Phosphate-P concentrations and pH by
Sediment

to 0.111 mg phosphate-P/L in the fine sediment; and 3) 0.004 to 0.13 mg P/L in the
fine/sandy mixed sediment. In Site 3, an anomalous spike in nitrate-N was detected in
one sample, with an outlier of 0.551 mg Nitrate-N/L. While the majority of samples at
this site read lower than 0.10 mg Nitrate-N/L, this outlierresulted in mean Nitrate-N
concentrations of 0.099 mg/L, just below the mean concentrations in the fine sediment.
One-way Analysis of Variance tests were performed to determine if effects of sediment
type significantly influenced nitrate and phosphate concentrations. While phosphate

Nitrate-N and Phosphate-P mg/L

content was significantly influenced by sediment type, nitrate content was not (Figure 4).
Lake Mills 2012 Available Nitrate and Phosphate by
Sediment
0.14

Nitrate-N

0.12

Phosphate-P

0.1
0.08
0.06
*NO3-N
F(2) = 1.14
P = 0.33

0.04
0.02
0

Coarse

Fine
Fine/Sandy
Sediment Texture

* PO3-P
F(2) = 3.63
P = 0.04

Figure 4. Nitrate-N and phosphate-P (mg/L) by sediment type. *Oneway ANOVAs for nutrient concentrations by sediment
55

5.4 Natural Plant Regeneration
At all restoration sites, native graminoids and horsetails rapidly recolonized moist
areas. Horsetail (Equisetum arvense) showed the highest frequency (41%, ONP 2012
vegetation surveys). Baltic rush (J. balticus), Common rush (Juncus effusus), daggerleaved rush (J. ensifolius), Bolander’s rush (J. bolanderi), toad rush (J. bufonius),
spreading rush (J. supiniformis) and tapered rush (J. acuminatus) were common species
among native, naturally regenerating rushes (Hitchcock and Cronquist, 1973; PojarMackinnon, 1994; ONP 2012 vegetation surveys). Common regenerating native grasses
included slender hairgrass (Deschampsia elongata), spike bentgrass (Agrostis exarata),
blue wildrye (Elymus glaucous), and bromes (Bromus spp.). Native sedges included
thick-headed sedge (Carex pachystachya), saw-bead sedge (C. stipata), and dewey sedge
(C. deweyana), and others not observed flowering (Hitchcock and Cronquist, 1973;
Pojar- Mackinnon, 1994; ONP 2012 vegetation surveys). Over time, as the exposed
sediment changes from mesic to xeric, the rushes and other naturally regenerating plants
prone to moist conditions may not persist (Shafroth et al., 2006).

Naturally regenerating forbs included several varieties of forget-me-not (Myosotis
sp.), fringed willowherb (Epilobium ciliatum), chaparral willowherb (E. minutum), pearly
everlasting (Anaphalis margaratacea), sow thistle (Sonchus spp.), western dock (Rumex
occidentalis), American speedwell (Veronica americana) and goats beard (Aruncus
dioceae) (Hitchcock and Cronquist, 1973; Pojar- Mackinnon, 1994; ONP 2012 vegetation
surveys). Germinants of Douglas-fir, grand fir (Abies grandis), western hemlock (Tsuga
heterophylla), bigleaf maple (Acer macrophyllum) western redcedar, and oceanspray
were also commonly encountered at all sites. An additional naturally regenerating native
56

woody plant, observed in greater abundance than expected throughout the dewatered
Lake Mills reservoir, is resprouting Sitka willow (Salix sitchensis) (Hitchcock and
Cronquist, 1973; Pojar-MacKinnon, 1994; ONP 2012 surveys).
In addition to naturally
regenerating native plants, a

Stukey Creek
Whiskey Bend
Road
West Lake
Mills Trail

Fine

variety of non-native forbs
and graminoids also emerged
in the exposed Lake Mills

Sedge Creek
Fine/sandy

sediments. A trail on the west
lakeshore and an unpaved

Boulder Creek

Coarse

road above the east shore
coupled with several creeks
flowing from these human

Map 2. Creeks and Whiskey Bend Road, bordering
Lake Mills

developments to the
dewatered reservoir likely

serve as weed vectors (Map 2). Three common invasive grasses and several common
forbs included: common velvet grass (Holcus lanatus), creeping bentgrass (Agrostis
stolonifera), orchard grass (Dactylus glomerata), mainly at northern boundaries of the
dewatered reservoir, hairy cat’s ear (Hypochaeris radicata), bull thistle (Cirsium
vulgare), and creeping buttercup (Ranunculus repens) (Hitchcock and Cronquist, 1973;
Pojar- Mackinnon, 1994; ONP 2012 vegetation surveys). Common invasive forbs
observed to less frequently were: Canada thistle (Cirsium arvense) and Robert’s
geranium (Geranium robertianum) (Hitchcock and Cronquist, 1973; Pojar-Mackinnon,

57

1994). Robert’s geranium was generally encountered within forested edges of the
northeastern and northwestern shores of the reservoir, where Whiskey Bend Road and the
West Lake Mills Trail serve as potential weed seed sources from which adjacent creeks
carry the seed to the edges of the exposed sediment (Map 2). The Sedge Creek vicinity
harbors a large infestation of Robert’s geranium originating from Whiskey Bend Road
which requires periodic treatment and monitoring.

5.5 Herbivory Observations
During the July 2012 monitoring, small green caterpillars appeared on the leaves of black
cottonwood at all study sites. The caterpillars skeletonized affected leaves and led to a
decline in qualitative vigor data for many tagged trees. In all, 73% of the cottonwoods in
this study showed varying degrees of caterpillar herbivory. For the purpose of moth
identification, several caterpillar specimens were collected from Sites 2 and 3 and reared
in a jar containing green leaves (Carri LeRoy, 2012, personal communication). One moth
emerged and was submitted to moth specialist Dr. Lars Crabo, co-founder of Western
Washington University’s “Pacific Northwest Moths” website, for identification. The
moth was identified as the grey midget (Nycteola cinereana), Neumögen & Dyar, 1893,
in the family Nolidae (Lars Crabo 2012, personal communication).

Grey midget caterpillars are semi-translucent, green and relatively hairless,
bearing long setae. Caterpillars feed on members of the tree family Salicaceae (Populus
and Salix sp.). Larvae can forage as colonial tent-makers where populations are high, or
as “single defoliators,” as with the Lake Mills population. Moths are light grey-brownand-white mottled, with rectangular wings reaching 2.5 to 3.1 cm in wingspan

58

(www.entomology.museums.ualberta.ca_Searching_species_details.php?s=2776). Life
history traits include a broader “single peak” of emergence in early summer than other
Nycteola species, which are generally bimodal in spring and late summer. Grey midget
moths can overwinter in either larval or caterpillar life stages. A common physical
characteristic of the moth is a dark “mustache” from the basal area to posterior margin.
This moth is generally found from Newfoundland to southern B.C., although sightings in
western Washington State and Oregon are not uncommon. This moth is not known to be
a significant pest in Pacific Northwest forests or croplands (http://pnwmoths.biol.
wwu.edu).

Figure 5. Photos of the grey midget moth as mature moth (upper left,
lower right) and in caterpillar and larval stages (lower left, upper right)

Ungulate herbivory was apparent in all sites, particularly with thimbleberry,
oceanspray and western redcedar. In all sites, potted plants were pulled up by Roosevelt

59

elk (Cervus canadensis roosevelti) or blacktail deer (Odocoileus hemionus columbianus).
Elk and deer tree herbivory and damage had been observed during the 2012 growing
season, and became increasingly apparent by the winter and spring of 2013 following this
study. In addition to grazing on a variety of planted forbs, shrubs and young trees, a
resident elk bull utilized these saplings to rub the velvet from its antlers during molting,
resulting in significant bark shredding. In some cases this led to displacement of black
cottonwood tags from the 2012 survivability study. This bull, initially sited during this
study at on the West Lake Mills Trail during the winter of 2012, was apparently displaced
from a neighboring herd upon reaching maturity and is currently being tracked via radio
collar by ONP wildlife biologists.

60

VI. Discussion
6.1 Plant Performance and Site Conditions

2012 Lake Mills revegetation plant mortality, while lower than expected, showed a
distinct trend in relation to species. It was not entirely unexpected that Douglas-fir
showed the highest morality rate as this tree is generally not an early successional
riparian species and often requires a greater degree of organic matter and nutrients,
lacking in the Lake Mill substrates, in order to establish. The most likely cause of the
poor Douglas-fir seedling performance is believed to be a lack of mycorrhizal fungi
(Dave Allen, Joshua Chenoweth, personal communications). Douglas-fir trees are
dependent on ectomycorrhizal fungi in order to establish and survive. Mycorrhizae allow
host plants to maximize available nutrients in nutrient-poor, low-fertility soils typical in
severely disturbed lands such as the dewatered Lake Mills reservoir. Additional benefits
trees receive from mychorrhizae are disease resistance (Haselwandter and Bowen, 1996),
greater absorption of available water, and increasing tolerance to drought stress and high
temperatures (Simmard, 2009). With the exception of the periodic, naturally reestablishing plants in moister areas, the exposed post-dam-removal sediments were
devoid of live trees, the roots of which serve as carbon sources for mychorrhizal fungi,
prior to replanting efforts (Jones et al., 2003). Nursery inoculation of Douglas-fir
seedlings with locally-derived ectomychorrizal fungi may enhance their survivability and
performance in future ONP restoration efforts (Chenoweth et al., 2011).

Environmental factors coupled with low organic matter and poor soil
development likely contributed to high Douglas-fir mortality during the 2012 Lake Mills

61

growing season. Windy site conditions may have been detrimental to Douglas-fir
performance and that of other installed plants. High winds can increase the transpiration
rates of seedlings, resulting in stomatal closure and lowered water potential (Heiligmann
and Schneider, 1974). Additional research (Philipson, 1988) indicates that Douglas-fir
seedlings are dependent on current photosynthate for new root growth. In other words,
Douglas-fir trees must obtain carbohydrates for new root development from foliage,
rather than from buds, cambium or starch reserves as with other conifer seedlings such as
Sitka spruce (Picea sitchensis) (Philipson, 1988). Additionally, root damage incurred
from rough handling during transport and planting can exacerbate drought stress in
Douglas-fir seedlings, reducing the potential absorption of water lost during transpiration
in periods of drought due to inherently low root growth potential (RGP) (Philipson,
1988).

The high survivability of black cottonwood in this study despite nutrient-poor,
undeveloped substrates supports further examination of its performance in Lake Mills
post-dam-removal sediments. In addition to their deep rooting capabilities as an
adaptation to drying soil conditions (Naiman and Decamps, 2005), recent research
indicates that cottonwoods and other trees of the family Salicaceae share symbiotic
microbial relationships which further enhance their adaptability to nutrient-poor
substrates (Doty et al., 2009). Extraction of cells from the surface-sterilized stems of
cottonwood (Populus trichocarpa) trees revealed the presence of endophytic bacteria
(Burkholderia vietnamiensis, wild poplar strain B) responsible for nitrogen fixation. In a
nursery experiment, inoculation of Kentucky bluegrass (Poa pratensis) with this
endophytic bacterium enhanced its growth in a nitrogen-free substrate (Doty et al., 2009;
62

Xin et al., 2009). Black cottonwood was included in the 2013 ONP plantings of the
dewatered Lake Mills and Lake Aldwell reservoirs, and will likely be included
throughout the remaining planting trials through 2017. Planting black cottonwood,
willow and other native woody and herbaceous species adapted to nutrient-poor soils may
be a key strategy in the survival of the revegetation in Lake Mills, and the eventual
development of late-seral forests critical to river restoration. While black cottonwood
showed the greatest survivability of woody species included in this study, their long-term
survival remains to be seen as water tables drop over time (Scott et al., 1999).

In addition to the highest and poorest-performing woody plant species from this
study, the other species were assessed for their utility in future restoration efforts. The
remaining four woody plants tested during the 2012 ONP planting trials, oceanspray,
Nootka rose, thimbleberry, and western redcedar, all experienced low mortality rates and
were included in the 2013 ONP Elwha restoration plantings. Additionally, woody plants
such as western bittercherry (Prunus emarginata), red alder, and all of the Rubus and
Ribes species in the 2012 Lake Mills planting trials were unofficially observed to perform
well in all sediment types where planted. Low germination success rates and challenges
with rodent seed predation prevented the inclusion of high numbers of western bittercherry in the 2013 plantings, but this species will be propagated over the coming 2013
ONP seed collection season for future plantings.

Pearly everlasting (Anaphalis margaritacea) and fireweed (Chamerion
angustifolium), included in the 2012 Lake Mills plantings and also observed as natural
regeneration in all planted sites, were known to colonize low-nitrogen volcanic

63

depositions following the 1980 Mt. St. Helens pyroclastic volcanic eruption in southern
Washington (del Moral and Wood, 1993). Pearly everlasting was observed to adapt to the
severe substrate in these post-eruption environments with the development of branched
root systems (Chapin, 1995). It remains to be seen how these and other pioneer plants
may or may not persist in post-dam-removal sediments, or whether they will aid in the
native vegetation colonization, but these plants may show greater persistence than the
current, predominantly mesic rush and sedge regeneration as exposed sediments dry out
(Shaffroth et al., 2006). During the 2012 Lake Mills woody plant performance studies, all
plant species in all restoration sites appeared to have completed their flowering and
fruiting cycles, and a resulting seed legacy and new germinants may be detected over the
coming growing 2013 season.

Low plant growth measurements were not unexpected during the relatively short
timeframe of this study. Follow-up monitoring over the next several years may reveal
increasing growth rates of surviving vegetation once plant roots are established and
energy is transferred to stem growth (Woodruff et al., 2002). Growth measurements will
continue along with monitoring survivability and vigor of the replicates included in this
study during the next several growing seasons.

While the fine sediments in Lake Mills were originally considered to be of greater
concern than coarse sediments for native plant establishment, the 2012 planting trials
have changed this perspective. Although this sediment affords poor drainage and hypoxic
growth conditions, its cohesive texture protects plant roots more effectively than the
highly erosive coarse sediment, particularly in the windy site conditions experienced at

64

Lake Mills. These very properties also inhibit herbivores from uprooting plantings, in
comparison to high occurrences of uprooting observed in the planted coarse substrates.
Limited vegetation survival in coarse sediments can potentially delay the development of
this substrate into soil, lacking the structure provided by organic matter, root contact
points and capillary draw (Angers and Caron, 1998).

6.2 Sediment Field and Laboratory Analysis
Nutrient limitations, particularly N, may delay the establishment of native
vegetation in the dewatered Lake Mills reservoir, as was the case with mudflow deposits
out of the range of legacy seed and organic matter following the eruption of Mount St.
Helens (del Moral and Clampitt, 1985). Past analysis of Lake Mills fine sediments
showed total N contents of 9 to 11%, compared to 15 to 22% in climax western hemlock
forest values, and P contents of 1.0 to 2.0 ppm, compared to 7.4 to 9.2 ppm in climax
western hemlock forests (Henderson et al., 1989; Chenoweth et al., 2011). Additionally,
pre-dam-removal assessments of Lake Mills sediments conducted by ONP revealed low
concentrations of potassium, calcium and magnesium content and the micronutrient
boron (Chenoweth et al., 2011). The fall 2013 nitrate and phosphate testing of Lake Mills
sediment, with concentrations ranging from below level of detection to 0.10 mg/L further
supported low nutrient content in the post-dam-removal sediments. Phosphate
concentrations detected during this study were higher in coarse sediment than in fine,
contrary to earlier nutrient assessments of Lake Mills sediments (Cavaliere and Homann,
2012). However, the sediment nutrient assessment conducted in this study reflects only a
single snapshot in time during the fall of 2012. Ideally, numerous nutrient assessments
would have taken place over time from these sample sites to detect changes during the
65

2012 growing season, since detectable levels of available N and P can vary during
different seasonal periods (Carri LeRoy, personal communication).

During early stages of forest development, nutrient-poor sediments are not
uncommon in Pacific Northwest river floodplains (Naiman et al., 2010). Over time (>50100 years of floodplain soil development), sediment erosion and N-fixing plants such as
red alder, cottonwood, willow or lupines (Lupinus sp.) (del Moral and Wood, 1993)
provide rapid nutrient input. Fine silt and clay sediment particles can be efficient in the
adsorption of C and N obtained through organic matter during the initial 100 years of soil
development (Naiman et al., 2010), while coarse sediments may not readily retain
nutrients due to leaching, typical of an “open” system. A “closed” system is attained with
the formation and accumulation of soil and organic matter and the development of plantsoil nutrient recycling more typical in later-seral forests (Naiman et al., 2010). N
retention and availability are especially dependent upon the presence of organic matter
(Chappell et al., 1991). Therefore, nutrient assessments of sediments in the dewatered
Lake Mills reservoir should continue over time as more vegetation becomes established.

The low available NO3 concentrations detected in Lake Mills sediment cores
potentially explains the high yellow leaf coloration of 38 to 55% of tested woody plants
in the 2012 Lake Mills plantings by September (Stewart, 1999). However, further
research and consideration must be given pertaining to the different responses (i.e., vigor)
of individual plant species to environmental stress. Hot, dry weather potentially
contributed to the rising percentages of plants displaying red leaf coloration by late
summer 2012 (Hoch et al., 2001, 2003). The highest red leaf coloration among all plants

66

occurred in the coarse sediment, in which windy site conditions coupled with erosive,
low-moisture substrate may have exacerbated drought and root stress. The introduction of
potted or bare root plants into a new site can result in “transplant shock,” where trees and
shrubs display leaf scorching during the first two to three years of development while
undergoing root establishment (Purdue University Plant and Pest Diagnostic Laboratory,
2002). Increasing leaf scorching as high as 27% by the end of the 2012 Lake Mills
growing season followed a pattern indicative of drought stress and/or root damage, where
dry winds during the late summer drought were likely the greatest contributors.

Mean pH ranges by sediment type measured during this study were more basic
than those of climax western hemlock forest ranges (Henderson et al., 1989), but
consistent with those found in a preliminary ONP Lake Mills sediment assessment (6.06.5) (Chenoweth et al., 2011). Some fine sediment sample sites, however, yielded pH
readings in the range of 5.4 – 5.9, more consistent with baseline climax western hemlock
forest ranges (Henderson et al., 1989).

Overall, the negative GWC trend from the early to late growing season was
consistent with what may be expected in increasingly dry site conditions between May
and late September 2012, coupled with lower water availability (Figure 1). In coarse
sediment, due to large gravel and sand pore sizes providing greater moisture availability,
lower suction in centibars, compared to fines occurred in conjunction with low GWC.
However, average tensiometer readings in coarse sediment were 46 cb by late summer
and fall. In agricultural operations, soils high in coarse sand would receive irrigation
upon reaching 20 to 40 cb (Soil Moisture Corp, 2012). In fine sediment, low porosity and

67

fine sediment particles yielded higher suction in centibars, coupled with GWC
significantly higher than that of the coarse substrate. Tensiometer readings had climbed
as high as 85 cbs suction in the fine sediment by August and September. Soils high in silt
and clay, in agricultural production, would require watering upon reaching 50 to 70 cb,
depending on clay content (Soil Moisture Corp, 2012). The permanent wilting point for
most soils is 150 cb (Tolk, 2003). The fine/sandy sediment moisture availability and
GWC proved more variable within-site, but showed a greater trend of decreasing water
availability during tensiometer measurements than in the coarse sediment, displaying
some of the moisture characteristics. The Heterogeneity encountered in both GWC and
available water assessments in this study within individual sites is not uncharacteristic of
substrates lacking developed vegetation, since lower density of living plants means the
presence of fewer plant roots to uniformly remove soil moisture at any given time
(Bethlahmy, 1962; Adams et al., 1991). Results from 2012 Lake Mills sediment moisture
assessments illuminate the likelihood of increasingly low moisture storage in all postdam-removal sediment types as they dry with continued exposure.

Use of a tensiometer in coarse sediments proved challenging during this study,
particularly due to the fact that this substrate is far coarser than in developed soils given
this characterization. Abrasive, loose rock often served as a barrier to insertion of the
tensiometer core and probe, shortened the life of the ceramic probe tip due to gouging
and surface abrasion, and high porosity coupled with low cohesion provided less surface
area contact for the probe tip than in the fine and fine/sandy sediments. While the use of a
tensiometer suited the needs for onsite moisture assessment in this study due to extensive
time required in the field, this method may not be well-suited for future high-volume use
68

in similar coarse sediments.
Tensiometer readings will be conducted by ONP for the remainder of the Elwha
revegetation project, subsampling 20 of the 111 2012 Lake Mills sediment moisture sites
(Sites 1 and 2), in addition to 30 new sample sites within the coarse and fine 2013 planted
sites on west Lake Mills. Reassessment of the 2012 sediment moisture sites may indicate
the degree to which sediment water availability declines post-dam-removal with
continued exposure, potentially shifting the plant species these sites can support as
conditions change from mesic to xeric (Shafroth et al., 2002).

Particle size characterization in this study appeared fairly consistent with that of
previous efforts. In two previous Lake Mills studies, fine sediments were found to
contain 82% to 85% silt, 18% to 6% clay, and up to 8% sand while coarse sediment
contained <1% to 4% clay, 2% to 11% silt and 85% to 98% sand (Mussman et al., 2008;
Cavaliere and Homann, 2012). Sediment particle size in this study proved variable due to
within-site heterogeneity, to a lesser degree in the fine sediment (see Section 5.3), and
was consistent with the variability detected in sediment moisture between sites.

6.3 Natural Plant Regeneration and Invasive Plants
Several invasive plants encountered in all study sites will be monitored and treated for the
remainder of the Elwha River restoration revegetation efforts. Robert’s geranium is of
particular concern due to its presence in the Lake Mills vicinity and ability to establish
beneath closed forest canopies and well-established dense native shrub communities
(Chenoweth et al., 2011; Washington State Noxious Weed Control Board,
http://www.nwcb.wa.gov/detail. asp?weed=55). This herb has been known to colonize

69

waste sites, spoil heaps, limestone quarries and a multitude of disturbed areas worldwide,
and can eject seeds as far as 4.5 to 6.0 m. Robert’s geranium prefers moist sites, which
are currently abundant on the dewatered Lake Mills reservoir (Tofts, 2004). Common
velvet grass was common in all sites, and is currently treated by park staff through
mechanical removal when detected. This grass was linked to a slowing of litter
decomposition in coastal California prairies, by deterring key detritivore
macroinvertebrates responsible for grass litter loss (Barstow, 2008). While the lower
Elwha River ecosystem may not be impacted by common velvet grass in the same
manner as the California prairie environments, it is worth future monitoring due to its
potential effects on long-term soil formation.

6.4 Future Research
For at least the next 3 to 4 growing seasons, the survivability and vigor of the Lake Mills
2012 tagged woody plants will be monitored. This long-term monitoring will capture
more fully the plant survival and establishment rates which may not have been reflected
during the first year following their installment. It not uncommon in similar plant
restoration efforts for the majority of potted and bareroot seedling mortalities to occur
within the first growing season; however, gradual mortality over the following several
growing seasons may also be expected (DeSteven, 1991). Additionally, treatments such
as planting adjacent to woody debris, woody debris size class, planting density, proximity
of natural regeneration or dense neighbor species may show greater influence as time
passes.
During the winter of 2013, a student of Peninsula College, Port Angeles, WA is
conducting a nursery ectomycorrhizal inoculation study with the seeds and seedlings of
70

Douglas-fir and grand fir at the ONP nursery. During late February and early March
2013, inoculated soil (256 ounces or one, 2-gallon container) was collected from the base
of Douglas-fir stands or trees at several locations between Glines Canyon Dam and the
ONP park boundary along Olympic Hot Springs road. For both Douglas-fir and grand fir,
400 seeds were inoculated in a 3-layered substrate mesocosm of collected soil and
nursery seeding soil (Sunshine peat mix with 40% perlite), with an additional 400 seeds
for each species sown into seeding soil only as a control. In all, 8 seeding flats were
prepared: six flats of Douglas-fir, (three flats inoculated and three control, 200 seeds per
flat); and two flats of grand fir (one inoculated, one control, 200 seeds per flat).
Additionally, 300 1-year-old Douglas-fir and grand-fir seedlings were inoculated in pots,
accompanied by potted control seedlings, 300 for each species. Inoculated and control
seeding flats are currently stored in the ONP greenhouse, while inoculated and control
potted seedlings reside onsite outdoors. The performance of seedlings will be tested in the
2014 ONP post-dam-removal restoration efforts, to determine whether ectomycorrhizal
conifer inoculation improves their survivability and performance in different substrates.
These findings may build on and at least partially explain the high mortality rate of
Douglas-fir in the 2012 Lake Mills woody plant trials.

A University of Washington graduate student from the Master of Environmental
Horticulture program will conduct thesis research similar to the 2012 Lake Mills woody
plant revegetation study. This research will test the performance of 2-year-old Douglasfir seedlings, to determine whether performance may be more favorable with older
specimens (2012 plantings were 1-year-old seedlings). Additionally, grand fir (Abies
grandis), western white pine (Pinus monticola), bigleaf maple (Acer macrophyllum), and
71

Scouler’s willow (Salix scouleriana) will be included to broaden knowledge of tree and
tall shrub survivability in Lake Mills fine and coarse post-dam-removal sediments. The
plant survivability studies of 2012, 2013 and those of future ONP plant restoration efforts
will capture an increasingly larger picture of native woody plant species performance in
post-dam-removal sediments.

72

VII. Significance
7.1 Broader Impacts of study
Findings from the Lake Mills native plant restoration and similar efforts may set the
precedent for future plant restoration following dam removals, particularly in the Pacific
Northwest. Riparian plant establishment is key to maintaining healthy river ecosystems
which support fish and other associated aquatic and terrestrial organisms. While abundant
research has been conducted on effective riparian plant restoration strategies, little to no
research has focused on selection of plant species suited for specific types of residual
reservoir sediment. Findings from this study will aid in the development of planting
strategies for the remaining five of the seven-year plant restoration efforts of Olympic
National Park. The ONP riparian plant restoration efforts will be carried out in stages,
determining which species to propagate and plant based on survival rates in previous
planting efforts. Determining which woody species to plant in specific post-dam-removal
sediment textures will aid in a more efficient use of time and resources for the duration of
the project.

The removal of Glines Canyon Dam is nearly complete, with 20 m or 30% of the
dam remaining. In March 2013, 4.9 million m3 of sediment had passed from the
dewatered Lake Mills reservoir to downstream reaches of the Elwha River. This is
estimated to be 25% of the sediment load accumulated during the presence of the Glines
Canyon Dam (USGS Sediment Team, 2013). An additional 1.1 million m3 of sediment
have passed downstream from Lake Aldwell (USGS Sediment Team, 2013). This
exceeds estimates from previous Lake Mills sediment core studies (Shaffer et al., 2008;

73

Winter and Crain, 2008). Additionally, due to historic topographic mapping errors,
sediment depth was found to be 6.1 m deeper than previously estimated (USGS Sediment
team). Thus, Lake Mills is believed to have contained 21.7 million m3 of accumulated
sediment (USGS Sediment Team, 2012). This quantity of sediment is approximately 2.3
billion m3 less than the total sediment released into the north and south forks of the Toutle
River during the eruption of Mt. St. Helens (Bednarek., 2001). Establishment of native
riparian vegetation is therefore key to the eventual development of organic matter and
eventual topsoil with large quantities of sediment remaining in the former lake.

Due to high levels of suspended sediment in the lower Elwha River, further dam
deconstruction has been delayed until September of 2013 at the earliest, and before the
expiration of the dam removal contract in September 2014 at the latest (Mapes, 2013).
The Elwha Water Treatment Plant (EWTP), constructed in 2010 in order to mitigate
sedimentation-related harm to water consumers following dam removal, requires
improvements to process the recently high levels of coarse debris flowing out of the
dewatered Lake Mills reservoir. The EWTP treats water utilized by the City of Port
Angeles, Nippon Paper Company (a mill located on Ediz Hook), the Washington
Department of Fish and Wildlife’s hatchery rearing channel, and the LEKT fish hatchery
(ONP Press release, February 1, 2013). The need for EWTP amendments became
apparent during the fall of 2012 when fish screens and pumps were overwhelmed with
organic matter and sediment, lessening the feasible amount of water the plant could
process and increasing the required equipment maintenance time. The EWTP was built to
process up to 53 million gallons of water per day, with sediment loads as high as 40,000
parts per million (ppm) Total suspended solids. Sediment loads thus far have only
74

reached 10,000 ppm TSS (Maynes, ONP Press release, February 1, 2013). The
inadequacies of the EWTP have become so chronic that personnel must be continually
onsite just to keep it running, when the treatment facility was originally designed to be
operated remotely (Mapes, 2013). The Nippon Paper plant is currently the only recipient
receiving treated water from the EWTP, and the City of Port Angeles has had to rely on
its well for drinking water (Mapes, 2013).

7.2 Cultural Impacts of Dam Removal on Lower Elwha Klallam Tribe
The Lower Elwha Klallam Tribe (LEKT), inhabiting the Lower Elwha River and
neighboring bluffs since as early as 750 B.C., are acknowledged in the U.S. 1855 Treaty
of Point No Point as having resided in this location since “time immemorial.” Current
Elwha tribal lands make up close to 405 ha of the Elwha River watershed. Ancestral land
of the LEKT included villages on the north and south sides of the Strait of Juan de Fuca.
Areas of residence south of the Strait included the regions of Hoko, Clallam Bay, Pysht,
Deep Creek, Freshwater Bay, the Elwha valley and river mouth, the Port Angeles vicinity
and areas east of modern-day Port Angeles (Valadez, 2002). The current LEKT
reservation was not officially designated a reservation until 1968, when the U.S. Federal
government assessed stipulations the Treaty of Point No Point and determined that it
entitled the tribe to their own historic land (http://www.elwha.org/elwhariver
restoration.html; Valadez, 2002).

When the Treaty of Point No Point was signed in 1855, the LEKT were expected
to relocate to the Skokomish reservation on the Hood Canal of the southeast Olympic
Peninsula. The LEKT, however, resisted this push to be relocated, preferring instead to

75

reside in their homeland where their ancestors were buried (Valadez, 2002). European
settlers began occupation of LEKT land in the 1860s, displacing tribal members from
their traditional villages and hunting and fishing grounds. Because the LEKT were not
U.S. citizens by technical definition, they were unable to purchase their own land until
the signing of the Indian Homestead Act in 1884 (Valadez, 2002). While this act allowed
as many as thirteen Klallam tribal families to own land in their historic residences, the
price of such ownership was the severance of LEKT homesteaders from their tribal
relations. While some individuals were able to adapt to life as independent homesteaders,
many did not wish to break away from their tribal heritage and as a result lost their land.
The 1934 Wheeler-Howard Act passed under the Franklin D. Roosevelt administration,
also named the Indian Reorganization Act, eventually provided federal funding for the
LEKT to acquire private land in their usual and accustomed dwellings until the official
designation of a reservation in 1968 (Valadez, 2002).

The construction of the Elwha River dams has adversely affected the culture and
economy of the LEKT for nearly a century through a drastic decline in salmon runs
(http://discoveringourstory.wisdomoftheelders.org/history-of-the-lower-elwha-klallampeople). Historically, the culture and economy of the Elwha tribe centered around salmon
fishing of all five Pacific salmon species. The tribe has actively proposed the removal of
the dams since the 1980s, and has been increasing the reestablishment of traditional tribal
knowledge since this proposal proved closer to reality (Valadez, 2002). In addition to
serving as barriers to anadromous fish, the dams led to loss of LWD delivery into the
lower Elwha River. Woody debris formerly carried downstream from upstream reaches
of the river was retained by the Glines Canyon Dam and to a lesser degree the Elwha
76

Dam for nearly a century following their construction. This loss of woody debris delivery
to the lower reaches of the river eliminated logjams which had been key structural
elements in formation of pools, side channels and other features needed for salmon
resting and spawning grounds. To address this problem, the Elwha tribe has been
installing engineered log jams (ELJ) since 1999 (Chenoweth et al., 2011). ELJs have
been proven potentially effective for restoration of juvenile Elwha River salmon (Pess et
al., 2012), but natural WD delivery and placement from an undammed river may further
enhance salmon habitat.

In order to enhance fisheries recovery in the lower Elwha River, a 5-year fishing
moratorium on the Elwha River has been proposed by the LEKT, State of Washington
Department of Fish and Wildlife (WDFW), and ONP (Warren, 2010). In accordance to
this moratorium, fishing on the Elwha River ceased in the spring of 2012 and will not
resume until the spring of 2017. It is hoped that this temporary fishing cessation will
allow the reestablishment of native anadromous fish stock in the middle and upper
reaches of the Elwha River (Warren, 2010).

The LEKT manages numerous fish hatchery programs based on the Hatchery and
Genetic Management Plan (HGMP). The HGMP was created cooperatively by the LEKT
and Hatchery Scientific Review Group (HSRG), in addition to a number of federal and
state government agencies such as National Marine Fisheries, U.S. Fish and Wildlife, and
National Park Service. This plan was intended to reduce endangerment risks to threatened
and endangered salmonid species (Case 3:12-cv-05109-BHS, Document 126, 2013).
A controversial fisheries recovery measure proposed by the LEKT entailed the

77

stocking and upstream release of non-native, Chambers Creek steelhead into the Elwha
River. The last release of Chambers Creek steelhead into the lower Elwha River took
place in 2011, before major dam deconstruction had taken place. Tribal members in favor
of the non-native steelhead stocking believed the LEKT had waited long enough to reap
the benefits of their historic fishing rights. Opponents emphasized the risk of nonnative
steelhead competing with native fish for food and resources, and the potential loss of
genetic integrity resulting from nonnative steelhead breeding with Elwha River native
fish.
Upon announcement of the plan to release non-native steelhead into the Elwha
River, immediate legal actions were taken against the LEKT and National Park Service.
Several nonprofit interest groups, including the Wild Fish Conservancy, Wild Steelhead
Coalition, Federation of Fly Fishers Steelhead Committee, the Wild Salmon Rivers and
the Conservation Anglers, filed lawsuits suing the LEKT and other agencies involved
with the steelhead restocking, under the claim that this action violates the Endangered
Species Act (Case 3:12-cv-05109-BHS, Document 126, 2013). In February 2013, U.S.
District Court Judge Benjamin Settle threw out the lawsuit, due to the fact that the tribe
had obtained all necessary federal fisheries permits to carry out their hatchery operations
since the initial complaint. The LEKT, in the meantime, agreed to cease the release of
any additional Chamber Creek steelhead, opting only to catch the remainder of hatchery
stock released in 2011. As an alternative option the tribe will attempt to obtain permits to
fish wild native Elwha River steelhead following the moratorium, even if these fish are
still listed as endangered, so long as this action can be deemed as non-harmful to the fish
populations (Case 3:12-cv-05109-BHS, Document 126, 2013).

78

Just one year following initial dam deconstruction, changes in the lower Elwha
River seemed hopeful for the LEKT and the prospect of restoring historic fish runs. By
the summer and fall of 2012, steelhead and Chinook, coho and pink salmon were
observed upstream of the former Elwha dam (http://www.usgs.gov/ blogs/features/
usgs_top_story/elwha-one-year-later/). During the spring of 2013, however, hatchery
Chinook released into the river experienced high mortality due to high river
sedimentation. Acclimation of hatchery fish to conditions which better represent the
current state of the lower Elwha River (i.e., cloudier water) and adjustments to the timing
of release in relation to periods of high suspended sediment may prevent such fish kills in
the future. Over time, the restoration of sediment delivery to the mouth of the Elwha
River, particularly fine sediment, may restore shellfish beds, allowing an additional
historic form of sustenance to be available once more to the LEKT (Shaffer, 2004).

Elwha tribal members are currently working with ONP, WDFW and the Bureau
of Reclamation to restore the lower Elwha River. In addition to hatchery operations,
fisheries monitoring and research, LEKT crews are actively revegetating and seeding the
dewatered Lake Aldwell reservoir. Additionally, the LEKT crew treats invasive plants
detected in the vicinity of the dewatered Lake Mills reservoir. Nearshore restoration
efforts of the LEKT include the organization of cleanup at the abandoned, contaminated
Rayonier mill site in Port Angeles, and the Ennis Creek Conceptual Restoration Plan,
which entails removal of a pier, jetty and concrete to restore natural stream meanders and
native riparian vegetation (The Watershed Company, 2011). It is hoped that this
cooperative effort will restore not only the livelihood and traditionally sustaining salmon

79

food source to the Elwha tribe, but also provide cultural renewal to a community that has
triumphed in maintaining its homeland despite a legacy of strife (Valadez, 2002).

7.3 Elwha River Nearshore Habitat
As early as one year following the 2011 Elwha River dam deconstructions, changes at
and beyond the mouth of the Elwha River were taking place as freed sediment was
transported downriver. However, the complete restoration of Elwha River nearshore
habitat requires interdisciplinary efforts among government and private entities.
Testament to the large-scale sediment release following the full and partial removal of the
Elwha River impoundments, sandbars were forming at the mouth of the Elwha River by
fall 2012 (http://www.oregonlive.com/environment/ index.ssf/ 2012/12/
sandbars_forming_at_ mouth_of_r.html). While natural recovery processes such as
downstream sediment delivery may improve the health of the lower Elwha River
ecosystem over time, human development may impede other recovery processes. At least
66% of tidelands along the Strait of Juan de Fuca are privately owned, with <12% of
shoreline bluffs or uplands adjacent to public beaches operating under public ownership
(Shaffer, 2006).

Nearshore waters along the Strait are more tidally influenced than open marine
waters, and therefore more subject to stratification and hypoxia, particularly in heavily
populated areas. Deeper marine waters undergo mixing from high winds, currents and
freshwater input from a multitude of rivers (Shaffer, 2006). Loss of sediment delivery
formerly contained by the Elwha River dams had been a major culprit of beach loss in
nearshore habitat, but armoring of privately owned bluffs is an additional contributor.

80

Pollutant runoff from industrial and everyday human activity in neighboring
developments introduces toxins and risk of eutrophication to the Strait. Where
government land regulation is replaced by a mosaic of private and city land ownership,
collaborative changes in land use are critical to restore the beaches surrounding the
mouth of the Elwha River and beyond.

Beyond the agency and tribal post-dam-removal restoration efforts of the lower
Elwha River, nearshore recovery is being monitored by private entities. With
downstream sediment delivery processes resumed, the possibility of rebuilding beaches
and nearshore habitat arises for the first time in nearly a century (Shaffer et al., 2008).
The Elwha Nearshore Consortium (ENC), coordinated by Anne Shaffer of Washington
State Department of Fish and Wildlife, brings together scientsists, land managers, city
planners and other stakeholders to address post-Elwha-River-dam-removal nearshore
habitat recovery. This organization conducts research regarding public and private land
management issues in shorelands between the mouth of the river and the City of Port
Angeles (Shaffer, 2009). The ENC developed the DRFT Shoreline Management Plan in
2011. This plan operates under the priniciples of the Shoreline Management Act of 1971,
which encompasses specific water of Washington State in addition to their shorelands.
Focus in this Act is placed on shorelands associated with streams possessing a mean
annual flow of at least 20 cfs, as well as lakes greater than 8.1 ha (20 acres). The
Shoreline Management Act of 2011 defines shorelands as “lands projecting at least 200
feet in all directions on a horizontal plane from the high water mark,” and includes all
wetlands and floodplains associated with streams, lakes, and tidal waters (Shaffer, 2006).

81

The ENC conducts a variety of studies related to nearshore, post-Elwha Riverdam-removal recovery. Among these studies are the monitoring and mapping of Lower
Elwha River river sedimentation and fish passage, particularly forage fish and their
nursery beds, monitoring eelgrass populations, and monitoring for euchalon (also known
as candlefish) in the Elwha estuaries. Historically, these estuaries served as critical
feeding grounds for Chinook, coho, pink, and cutthroat salmon as well as anchovies,
euchalon and surf perch (Shaffer, 2006, 2009). Of particular interest is whether the bull
kelp beds of Freshwater Bay and throughout the nearshore will be replaced once again
with eel grass as riverine-derived cobble material is replaced with finer sediment
(Shaffer, 2006).

Several key components of nearshore restoration addressed by the ENC include
beach loss, beach modification, and the pollution of marine waters. The ENC samples
the beach profile from Crescent Bay to Discovery Bay to detect changes over time
following dam removal (Shaffer, 2006). Bulkheads and dikes installed to prevent erosion
of private coastal bluffs which actually serve as “feeder bluffs” for beach-building will
continue to deprive beaches of material despite the post-Elwha-River-Dam-removal
sediment delivery. Of equal concern is point and non-point-source pollution from
residential and industrial sources in and around the City of Port Angeles, particularly that
of a leaching decommissioned landfill constructed in the 1940s (Shaffer, 2006).
Other ENC restoration goals include the inventory of critical nearshore habitat,
development of an Ediz Hook master plan, protection and restoration of native
vegetation, and reconnection of feeder bluffs to the Strait of Juan de Fuca’s marine
system (Shaffer, 2006). The ENC is carrying out a 2-year study tracking the movement of
82

fine-grained sediment transported down the Lower Elwha River to the river mouth.
Between November and June of 2012, an offshore sediment surface plume appeared to
have increased in suspended sediment concentration by 60%, although this was later
transported away from the delta by tidal action and a “net northeast travel direction”
(www.coastalwatershed institute.org).

7.4 Interdisciplinary Context
Completion of this study required an expansion of my skill sets beyond that of my
previous scholarly or work experience. Developing a better understanding of soil science
was a necessary part of this study, and truly an asset. Learning the use and concepts of a
tensiometer, planning and implementing the sampling of sediment water availability, and
the collection and processing of 111 sediment cores for GWC and nutrient analysis, much
of this with full-time college course and workloads, were genuine tests of endurance.
Exposure to the use of an AQ1 discrete analyzer and the necessary soil preparations was
also a new experience I am pleased to have gained. Particularly, the knowledge acquired
in research design, data analysis, and professional writing skills will influence my future
work.

In addition to expanding my scientific research skills, a variety of I have gained a
greater understanding of the processes entailed in dam removal and post-dam-removal
restoration. Through literature review, I learned of the legalities and functions which
dictate the operation and licensing of hydroelectric dams. A greater understanding of the
cultural history of the Lower Elwha Klallam Tribe, and the degree to which the LEKT
had had to struggle to remain in their historical land, was gained through this project. I

83

had taken for granted that the LEKT would have, at the very least, been designated
reservation land to call their own long before 1968. Reading about the policies and
legalities of post-dam-removal fisheries and learning more about the work of the ENC
were additional benefits of this study. With much time, commitment and effort behind
me, I leave the MES Program with a greater understanding of where my career path in
the environmental sciences may lead.

7.5 Conclusion
In conclusion, the 2012 woody plant survival study is merely one piece of a much
larger puzzle to the long term post-dam-removal recovery of the lower Elwha River.
Species selection appeared to influence woody plant survival in Lake Mills post-damremoval sediments, while planting treatments showed varying levels of influence by
sediment type. Native riparian plant restoration of Lake Mills during the 2012 growing
season saw high survival rates overall. Plant restoration efforts of ONP and the LEKT
may aid the eventual provision of stream shading, channel complexity and other
ecological contributions of a riparian forest which are key to maintaining a healthy river
ecosystem. Interdisciplinary government and private entities will continue to work
toward the full ecosystem of the lower Elwha River basin and nearshore habitat through
active and passive restoration work and monitoring. Only time will tell to what degree
this modified ecosystem returns to its pre-dam state following the largest U.S. dam
removal operation to-date.

84

Appendix A.
Table A1: Full Species Composition of ONP Planting Prescriptions

2012 ONP Planting Prescriptions, Lake Mills: Species Composition
Prescription 1
Shrubs Only - Site 1

Prescription 2
Prescription 3
Trees, Shrubs and
Trees Only – Sites 1, 2, 3
Forbs - Sites 2, 3
Acer circinatum
Acer macrophyllum
Acer macrophyllum
Alnus viridis
Alnus rubra
Malus fusca
Populus balsamiferassp.
Cornus sericea ssp. Occidentalis Crataegus douglasii
Trichocarpa
Gaultheria shallon
Holodiscus discolor
Holodiscus discolor
Oemleria cerasiformis Prunus emarginata var. mollis
Lonicera involucrata
Pseudotsuga menziesii Pseudotsuga menziesii
Salix scouleriana
Mahonia aquifolium
Rhamnus purshiana
Mahonia nervosa
Rosa nutkana
Oemleria cerasiformis
Rubus parviflorus
Philadelphus lewisii
Sambucus cerulea
Physocarpus capitatus
Symphoricarpos albus
Thuja plicata
Ribes divaricatum
Ribes lacustre
Herbaceous Plants
Ribes sanguineum
Anaphalis margaritacea
Rubus leucodermis
Aquilegia formosa
Rubus parviflorus
Artemisia suksdorfii
Rubus spectabilis
Aruncus dioicus
Rubus ursinus
Chamerion angustifolium
Salix sitchensis
Erigeron philadelphicus
Sambucus cerulea var. cerulea
Fragaria vesca
Sambucus racemosa
Petasites frigidus
Spirea douglasii
Solidago Canadensis
Symphoricarpos albus
Table A1. Plants included in ONP prescriptions: Full species list of prescriptions
followed in this study. Note: Nootka rose and western redcedar not included in Site 1

85

Appendix B: Methods of Sediment Moisture Assessment, and Limiting Factors in
Study Sites
A portable tensiometer was utilized in this study to allow onsite assessment of sediment
moisture. Many leave-in instruments, which are vulnerable to damage from freezing
temperatures, are also very costly. Tensiometers, however, may not be advantageous in
fine sediments during extremely dry periods, when the sediments desiccate and crack,
and tensions of greater than 80 to 100 centibars are required for plant roots to extract
water. During such periods, tensiometers may measure only 74-85% of the suction
required of plant roots in order to draw water from binding soil particles (Stoeckeler and
Aamodt, 1940).

Due to the wide range of sediment texture, particle size, and accompanying range
of moisture retention potential between and within plots, no perfectly-suited instrument
was available to accurately capture moisture levels in the highly variable sediments of the
Lake Mills study sites (Robertson et al., 1999). Fine soils are commonly measured by the
use of gypsum electrical resistance blocks and meters, while tensiometers are more
commonly applied to measure available moisture in sandier soils (Robertson et al., 1999).
post-dam-removal sediments found on the Lake Mills delta, however, do not resemble the
substrates generally assessed for moisture in agricultural or other planting operations.
Instruments such as neutron probes or capacitators were beyond the economic range of
this study. The gravimetric method, in which core sediment samples are collected, preweighed, oven-dried and reweighted to measure gravimetric water content (GWC), is an
empirical, accurate soil moisture assessment technique. However, the large number of
samples required in this study (N=111), the time involved in sieving and processing

86

samples, as well as 24-72 h for oven-drying, and long distances from study sites to the
campus soil laboratory made this method a time-consuming and less realistic alternative.
The onsite sediment moisture assessments afforded by a portable tensiometer proved a
more feasible option, as this process could occur concurrently with vegetation
monitoring. Additionally, tensiometers simulate the action of a plant root, measuring
suction in units of pressure, providing a favorable representation of the stress undergone
by plant roots during periods of drought.
Plant Growth Measurements
The number of measured plants (Table 17) and their random selection for measurement
was based upon their reasonable proximity to sediment moisture sample sites (i.e., the
resulting feasible number of plant measurements which could be carried out while
waiting for tensiometer readings).
Species
Oceanspray
Nootka rose
Thimbleberry
Black cottonwood
Douglas-fir
Western redcedar
Total =

Number
Measured
31
17
30
32
29
16
155

Table 17. Total number of growth measurements by plant species

87

References
Acker, S.A., Beechie, T.J., Shafroth, P.B (2008). Effects of a Natural Dam-Break on
Geomorphology and Vegetation on the Elwha River, Washington, U.S.A. Northwest
Science, 82(Special Issue):210-223
Adams, P.W., Flint, A.L., Fredriksen, R.L. (1991). Long-Term Patterns in Soil Moisture
and Revegetation After a Clearcut of a Douglas-Fir Forest in Oregon. Forest Ecology
and Management, 41(1991):249-263
Allen, D. (2012). Plant Propagation Specialist, Olympic National Park. Port Angeles,
Washington
Alpert, P.A., Griggs, F.T., Petersen, D.R. (1999). Riparian Forest Restoration Along
Large Rivers: Initial Results from the Sacramento River Project. Restoration
Ecology, 7(4):360-368
American Rivers, Friends of the Earth and Trout Unlimited (1999). Dam Removal
Success Stories: Restoring Rivers Through Removal of Dams That Don’t Make
Sense. Final Report
Angers, D.A. Caron, J. (1998). Plant-Induced Changes in Soil Structure: Processes and
Feedbacks. Biogeochemistry, 42:55-72
Ash, H.J., Gemmell, R.P., Bradshaw, A.D. (1994). The Introduction of Native Plant
Species on Industrial Waste Heaps: A Test of Immigration and Other Factors
Affecting Primary Succession. Journal of Applied Ecology, 31:74-84
Barstow, J.L., Preisser, E.L., Strong, D.R. (2008). Holcus lanatus Invasion Slows
Decomposition Through Its Interaction With a Macroinvertebrate Detritivore,
Porcellio scaber. Biol Invasions, 10:191-199
Bednarek, A.T. (2001). Undamming Rivers: A Review of the Ecological Impacts of
Dam Removal. Environmental Management, 27(6):803-814. © 2001 SpringerVerlag New York Inc.
Berendse, F. (1998). Effects of dominant Plant Species on Soils During Succession in
Nutrient-Poor Ecosystems. Biogeochemistry, 42(1/2):73-88
Bernhardt, E.S., Palmer, M.A., Allan, J.D., Alexander, G., Barnas, K., Brooks, S., Carrr,
J., S. Clayton, C., Dahm, C., Follstad-Shah, J., Galat, D., Gloss, S., Goodwin, P.,
Hart, D., Hassett, B., Jenkinson, R., Katz, S., Kondolf, G.M., Lake, P.S., Lave, R.,
Meyer, J.L., O’Donnell, T.K., Pagano, L., Powell, B., Sudduth, E. (2005).
Synthesizing U.S. River Restoration Efforts. Ecology, 308(29 April 2005):636-637
Bethlahmy, N. (1963). Soil-Moisture Sampling Variations as Affected by Vegetation and
88

Depth of Sampling. Soil Science, 95(3):211-213
Black C.A. (1965). Methods of Soil Analysis: Part I Physical and Mineralogical
Properties. American Society of Agronomy, Madison, Wisconsin, USA
Borgmann, K.L., Rodewald, A.D. (2005). Forest Restoration in Urbanizing Landscapes:
Interactions Between Land Uses and Exotic Shrubs. Restoration Ecology, 13(2):334340
Bradshaw, A.D. (1982). The Reconstruction of Ecosystems: Presidential Address to the
British Ecological Society, December 1982. The Journal of Applied Ecology, 20(1):117
Brenkman, S.J., Pess, G.R., Torgersen, C.E., Kloehn, K.K., Duda, J.J., Corbett, S.C.
(2008). Predicting Recolonization Patterns and Interactions Between Potamodromous
and Anadromous Salmonids in Response to Dam Removal in the Elwha River,
Washington State, USA. Northwest Science, 82(Special Issue 2008):91-106
Brown, R.L. and J. Chenoweth (2008). The effect of Glines Canyon Dam on
Hydrochorous Seed dispersal in the Elwha River. Northwest Science, Special Issue.
82: 197-209.
Buoycos, G.J. (1962). Hydrometer Method Improved for Making Particle Size Analysis
of Soils. Agron J, 54:464-465
Callaway, R.M. (1998). Are Positive Interactions Species-Specific? Oikos, 82(1):202207
Callaway, R.M. (1995). Positive Interactions Among Plants. The Botanical Review,
61(4):306-349
Caplan, J.S., Yeakley, J.A. (2006). Rubus armeniacus (Himalayan blackberry)
Occurrence and Growth in Relation to Soil and Light Conditions in Western Oregon.
Northwest Science, 80(1):9-17
Casper, A.F., Thorp, J.H., Davies, S.P., Courtemanch, D.L. (2006). Ecological
Responses of Zoobenthos to Dam Removal on the Dennebec River, Maine, USA.
Large Rivers, Arch. Hudrobiol. Suppl. 158/4, 16(4):541-555. © 2006 E.
Schweizerbart’sche Verlagsbuchhandlung, Stuttgart
Castro, J., Zamora, R., Hodar, J.A., Gomez, J.M., Gomez-Aparicio, L. (2004). Benefits
of Using Shrubs as Nurse Plants for Reforestation in Mediterranean Mountains: A 4Year Study. Restoration Ecology, 12(3):352-358
Cavaliere, E., Homann, P. (2012). Elwha River Sediments: Phosphorous Characterization
and Dynamics Under Diverse Environmental Conditions. Northwest Science,
89

86(2):95-107
Chalker-Scott, L., Ph.D., Extension Horticulturist and Associate Professor,
Puyallup Research and Extension Center. The Myth of Red Leaves “If plants
develop red leaves, it means they are phosphorus deficient.” Washington State
University, Seattle
Chapin, D.M. (1995). Physiological and Morphological Attributes of Two colonizing
Plant Species on Mount St. Helens. American Midland Naturalist, 133(1):76-87
Chapin, F.S., Vitousek, P.M., Van Cleve, K. (1986). The Nature of Nutrient Limitation
in Plant Communities. The American Naturalist, 127(1):48-58
Chappell, H.N., Cole, D.W., Gessel, S.P., Walker, R.B. (1991). Forest Fertilization
Research and Practice in the Pacific Northwest. Fertilizer Research, 27:129-140
Chenoweth, J. (2011). 2011-2012 Planting Plan, Olympic National Park Elwha
Revegetation Project. Olympic National Park and the Lower Elwha Klallam Tribe,
2011 Port Angeles, WA.
Chenoweth, J., Acker, S.A., McHenry, M.L. (2011). Revegetation and Restoration Plan
for Lake Mills and Lake Aldwell. Olympic National Park and the Lower Elwha
Klallam Tribe, 2011,Port Angeles, WA.
Chernicoff, S., Venkatarishman, R. (1995). Geology: An Introduction to Physical
Geology. Worth Publishers, Inc. New York, NY
Coastal Watershed Institute (2012). Elwha Nearshore Update, An Annual Newsletter of
the Elwha Nearshore Consortium, October 2012. www.coastalwatershedinstitute.org
Coffman, S. (1975). Shade from Bush Increases Survival of Planted Douglas-Fir.
Journal of Forestry, November 1975:726-728
Collins, B.D., Montgomery, D.R. (2002). Forest Development, Wood Jams, and
Restoration of Floodplain Rivers in the Puget Lowland, Washington. Restoration
Ecology, 10(2):237-247
Crabo, Lars, M.D., Western Washington University biology professor and co-founder of
Pacific Northwest Moths website: http://pnwmoths.biol.wwu.edu
Czuba, C.R., Randle, T.J., Bountry, J.A., Magirl, C.S., Czuba, J.A., Curran, C.A.,
Konrad, C.P. (2011). Anticipated Sediment Delivery to the lower Elwha River During
and Following Dam Removal. Ch. 2 of Duda, J.J., Warrick, J.A., Magirl, C.S., eds.,
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.
90

D’Amore, D.V., Hennon, P.E., Schaberg, P.G., Hawley, G.J. (2009). Adaptation to
Exploit Nitrate in Surface Soils Predisposes Yellow Cedar to Climate-Induced
Decline While Enhancing the Survival of Western Redcedar: A new Hypothesis.
Forest Ecology and Management, 258(10):2261-2268
del Moral, R., Clampitt, C.A. (1985). Growth of Native Plant Species on Recent Volcanic
Substrates from Mount St. Helens. American Midland Naturalist, 114(2):374-383
del Moral, R., Wood, D.M. (1993). Early Primary Succession on the Volcano Mount St.
Helens. Journal of Vegetation Science, 4:223-234
DeSteven, D. (1991). Experiments on Mechanisms of Tree Establishment in Old-Field
Succession: Seedling Survival and Growth. Ecology, 72(3):1076-1088
Doty, S.L., Oakely, B., Xin, G., Won Kang, J., Singleton, G., Khan, Z., Vajzovic, A.,
Staley, J.T. (2009). Diazotrophic Endophytes of Native Black Cottonwood and
Willow. Symbiosis, (2009)47:23-33
Downs, P.W. et al. (2009). Managing Reservoir Sediment Release in Dam Removal
Projects: An Approach Informed by Physical and Numerical Modelling of NonCohesive Sediment. Intl. J. River Basin Management, 7(4):433-452.
Doyle, M.W., Stanley, E.H., Harbor, J.M. (2002). Geomorphic Analogies for Assessing
Probable Channel Response to Dam Removal. Journal of the American Water
resources Association, 38(6):1567-1579.
Doyle, M.W., Stanley, E.H., Harbor, J.M., Grant, G.S. (2003). Dam Removal in the
United States: Emerging Needs for science and Policy. EOS, 84(4, 28 January
2003):29,32-33.
Doyle, M.W. et al. (2005). Stream Ecosystem Response to Small Dam Removal:
Lessons from the Heartland. Geomorphology, 71(2005):227-244.
Duda, J.J.. Freilich, J.E.,. Schreiner, E.G. (2008). Baseline Studies in the Elwha River
Ecosystem Prior to Dam Removal: Introduction to the Special Issue. Northwest
Science, 82(Special Issue 2008): 1-12.
Elwha Ranger Station Weather Station, Olympic National Park
http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?waelwh
Eränen, J.K., Kozlov, M.V. (2007). Competition and Facilitation in Industrial Barrens:
Variation in Performance of Mountain Birch Seedlings with Distance. Chempsphere,
67:1088-1095
Facelli, J.M., Pickett, S.T. (1991). Plant Litter: Its Dynamics and Effects on Plant
91

Community Structure. The Botanical Review, 57(1):1-32
Fetherson, K.L., Naiman, R.J., Bilby, R.E. (1995). Large Woody Debris, Physical
Processes, and Riparian Forest Development in Montane River Networks of the
Pacific Northwest. Geomorphology, 13(1995):133-144
Gee, G.W., Bauder, J.W. (1986). Particle Size Analysis, 383:411 in Klute. A. (Ed)
Methods of Soil Analysis, Part 1, Physical and Mineralogical Methods, Agronomy
Monographs No 9 (2nd Edition), American Society of Agronomy, Madison.
Goodson, J.M., Gurnell, A.M., Angold, P.G., Morrissey, I.P. (2003). Evidence for
Hydrochory and the Deposition of Viable Seeds Within Winter Flow-Deposited
Sediments: The River Dove, Derbyshire, UK. River Res. Applic., 19:317-334
Gotelli, N.J., Ellison, A.M. (2004). A Primer of Ecological Statistics. ©2004 by Sinnauer
Associates, Inc., Sunderland
Gowan, C., Stephenson, K., Shabman, L. (2006). The Role of ecosystem Valuation in
Environmental Decision Making: Hydropower Relicensing and Dam Removal on the
Elwha River. Ecological Economics, 56 (2006):508-523
Grant, G., Parks, N. (2009). A Ravenous River Reclaims Its Course: The Tale of
Marmot Dam’s Demise. Science Findings, 111(March 2009):1-5. Pacific Northwest
Research Station, Portland, OR.
Greer, E., Pezeshki, S.R., Shields, FD Jr. (2006). Influences of cutting diameter and soil
moisture on growth and survival of black willow, Salix nigra. Journal of Soil and
Water Conservation, Sep/Oct 2006, 61(5):311-323.
Greene, D.F., Johnson, E.A. (1996). Wind Dispersal of Seeds from a Forest Into a
Clearing. Ecology, 77(2):595-609.
Gurnell, A., Tockner, K., Edwards, P., Petts, G.(2005). Effects of Deposited Wood on
Biocomplexity of River Corridors. Frontiers in Ecology and the Environment,
3(7):377-382
Hanson, M.D. (1997). Riparian Forest Revegetation For Water Quality Improvement.
Restoration and Reclamation Review, Student On-Line Journal, 2(1):1-7, Spring
1997. St. Paul, MN: Department of Horticultural Science, University of Minnesota.
Haselwandter, K., Bowen, G.D. (1996). Mycorrhizal Relations in Trees for Agroforestry
and Land Rehabilitation. Forest Ecology and Management, 81(1996):1:17
Haubensak, K.A., Parker, I.M. (2004). Soil Changes Accompanying Invasion of the
Exotic Shrub Cytisus scoparius in Glacial Outwash Prairies of Western
Washington [USA]. Plant Ecology, 175:71-79
92

Haycock, N.E., Pinay, G., Tjaden, Robert L. and Glenda M. Weber (2003). Riparian
Buffer Effectiveness Literature Review. Prepared by Straughan Environmental
Services, Inc. for John Sherwell, Maryland Department of Natural Resources January
2003.
Helfield, J.H., Naiman, R.J. (2002). Pacific Salmon, Nutrients and the Dynamics of
Freshwater and Riparian Ecosystems. Ecosystems, 5:399-417
Henderson, J.A., D.H. Peter, R.D. Lesher, and D.C. Shaw (1989). Forested Plant
Associations of the Olympic National Forest. USDA Forest Service, Pacific
Northwest Region. R6 ECOL Technical Paper 001-88. Portland, OR
Heiligmann, R., Schneider, G. (1974). Effects of Wind and Soil Moisture on Black
Walnut Seedlings. Forest Sci., 20:331-335
Hitchcock, C.L., Cronquist, I.A. (1973). Flora of the Pacific Northwest. University of
Washington Press, Seattle, WA
Hoch, W.A., Singaas, E.L., McCown, B.H. (2003). Resorption Protection. Anthocyanins
Facilitate Nutrient Recovery in Autumn by Shielding Leaves from Potentially
Damaging Light Levels. Plant Physiology, 133(3):1296:1305
Hoch, W.A., Zeldin, E.L., McCown, B.H. (2000). Physiological Significance of
Anthocyanins During Autumnal Leaf Senescence. Tree Physiology, 21(2001):1-8.
Jackson, P. L. & Kimerling, J. A. (2003). Atlas of the Pacific Northwest, Ninth Edition.
Corvalis, OR: Oregon State University Press, Corvalis, OR
Jansson, R., Nilsson, C., Ren, B. (2000). Fragmentation of Riparian Floras In Rivers
With Multiple Dams. Ecology, 81(4):899-903.
Jones, M.D., Durall, D.M., Cairney, J.W.G (2003). Ectomycorrhizal Fungal
Communities in Young Forest Stands Regenerating After Clearcut Logging. New
Phytologist, 157:399-422
Kennedy, T.A., Naeem, S., Howe, K.M., Knops, J.M.H., Tilman, D., Reich, P. (2002).
Biodiversity as a Barrier to Ecological Invasion. Nature, 417:636-638
Kober, A., American Rivers. Restoring Rivers: Major Upcoming Dam Removals in the
Pacific Northwest. http://www.water. ca.gov/ fishpassage/docs /dams /dams.pdf
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, 35:1104-1115

93

Major, J.J., O’Connor, J.E., Podolak, C.J., Mackenzie, K.K., Spicer, K.R., Wallick, J.R.
Bragg,, H.M., Pittman, S., Wilcock, P.R., Rhode, A., Grant, G.E., (2010). Evolving
Fluvial Response of the Sandy River, Oregon, Following Removal of Marmot Dam.
Pre-print, Proceedings, 9th Federal Interagency Sedimentation Conference, Las
Vegas, NV, June 27-July 1, 2010:1-6.
Major, J.J., O’Connor, J.E., Spicer, K.R., Bragg,, H.M., Rhode, A., Tanner, D.Q.,
Anderson, C.W., Wallick, J.R. (2008). Initial Fluvial Response to the Removal of
Oregon’s Marmot Dam. EOS, 89(27):24-252
Mapes, L.V. (2013). Elwha Dam Removal Hostage to Water Plant Repairs. The Seattle
Times, Sunday, April 21, 2013 issue.
Maynes, B., McKenna, R. (2013). Corrections Continue at Elwha Water Facilities: Dam
Removal Progress Temporarily Postponed. Olympic National Park News Release,
March 27, 2013
Michel, J.T., J.M. Helfield, and D.U. Hooper (2011). Seed rain and revegetation of
exposed substrates following dam removal on the Elwha River. Northwest Science.
58(1): 15-29.
Morley, S.A., Duda, J.J., Coe, H.J.,. Kloehn, K.K., McHenry, M.L. (2008). Benthic
Invertebrates and Periphyton in the Elwha River Basin: Current Conditions and
Predicted Response to Dam Removal. Northwest Sciences, 82(2008):179-196
Muehlbauer, J.D., LeRoy, C.J., Lovett, J.M., Flaccus, K.K., Vlieg, J.K., Marks, J.C.
(2009). Short-Term Response of Decomposers to Flow Restoration in Fossil Creek,
Arizona, USA. Hydrobiologia, (2009) 618:35-45.
Mulvaney, R.L. (1996). Methods of Soil Analysis. Part 3. Chemical Methods – SSSA
Book Series no. 5. Soil Science Society of Americal and American Society of
Agronomy, Madison, WI
Mussman, E.K., Zabowski, D., Acker, S.A. (2008). Predicting Secondary Reservoir
Sediment Erosion and Stabilization Following Dam Removal. Northwest Science,
82(2008):236-245.
Naiman, R.J., Bechtold, J.S., Beechie, T.J., Latterell, J.L., Van Pelt, R. (2010). A
Process-Based View of Floodplain Forest Patterns in Coastal River Valleys of the
Pacific Northwest. Ecosystems, (2010)13:1-31
Naiman, R.J., Decamps, H. (1997). The Ecology of Interfaces: Riparian Zones.
Annual Review of Ecology and Systematics, 28 (1997):621-658. Annual Reviews.
Naiman, R.J., H. Decamps and M.E. McClain. 2005. Riparia: Ecology, Conservation and
Management of Streamside Communities. Elsevier Academic Press. Burlington, MA.
94

Nilsson, C., Brown, R.L., Jansson, R., Merritt, D.M. (2010). The Role of Hydrochory in
Structuring Riparian and Wetland Vegetation. Biological Reviews, 2010:1-24
Orr, C.H., Koenig, S. (2006). Planting and Vegetation Recovery on Exposed Mud Flats
Following Two Dam Removal in Wisconsin. Ecological Restoration, 24(2):79-86
Orr, C.H., Stanley, E.H. (2006). Vegetation Development and Restoration Potential of
Drained Reservoirs Following Dam Removal in Wisconsin. River Research and
Applications, 22:281-295 (2006).
Padilla, F. M., Pugnaire, F.I. (2006). The Role of Nurse Plants in the Restoration of
Degraded Environments. Front Ecol Envir, 4(4):196-202
Peltzer, D.A., Bellingham, P.J., Kurokawa, H., Walker, L., Wardle, A., Yeates, G.W.
Punching Above their Weight: Low-Biomass Non-Native Plant Species Alter Soil
Properties During Primary Succession. Oikos, 118:1101-1014
Pess, G.R., HcHenry, M.L., Beechie, T.J. and J. Davies (2008). Biological Impacts of the
Elwha River Dams and Potential Salmonid Responses to Dam Removal. Northwest
Science, 82(Special Issue 2008):72-90
Pess, G.R., Liermann, M.C., McHenry, M.L., Peters, R.J., Bennett, T.R. (2012). Juvenile
Salmon Response to the Placement of Engineered Log Jams (ELJs) in the Elwha
River, Washington State, USA. River Research and Applications, 28(7):872-881
Parekh, P. (2004). A Preliminary Review of the Impact of Dam Reservoirs on Carbon
Cycling. Internation Rivers Network, Department of Earth, Atmospheric, and
Planetary Sciences, Mass Institute of Technology, Cambridge, MA, November 2004.
Phillipson, J.J. (1988). Root Growth in Sitka Spruce and Douglas-Fir Transplants:
Dependence on the Shoot and Stored Carbohydrates. Tree Physiology, 4:101-108
Plant Pest and Diagnostic Laboratory (P&PDL), 2002. Purdue University West Lafayette,
IN. www.extension.purdue.edu/extmedia/BP/BP_25_W.pdfShare
Podolak, C., Johns Hopkins University, Pittman, S., Graham Matthews & Associates
(2010). Marmot Dam Removal Geomorphic Monitoring & Modeling Project. Annual
Report June 2008-May 2009. Prepared for: Sandy River Basin Watershed Council,
Sandy, OR
Poff, N.L., Hart, D.D. (2002). How Dams Vary and Why It Matters for the Emerging
Science of Dam Removal. Bioscience, 52(8):659-668
Pojar, K. and MacKinnon, A. (1994). Plants of the Pacific Northwest Coast: Washington,
Oregon, British Columbia and Alaska. B.C. Ministry of Forests and Lone Pine
95

Publishing, Canada
Roberts, S.D., Harrington, C.A., Terry, T.A. (2005). Harvest Residue and Competing
Vegetation Affect Soil Moisture, Soil Temperature, N Availability, and Douglas-fir
Seedling Growth. Forest Ecology and Management, 205(2005):333-350
Robertson, G.P., Coleman, D.C., Bledsoe, C.S., Sollins, P. (1999). Standard Soil
Methods for Long-Term Ecological Research (Long-Term Ecological Research
Network Series. Oxford University Press, New York
Rood, S.B., Braatne, J.H., Hughes, F.M.R. (2003). Ecophysiology of Riparian
Cottonwoods: Stream Flow Dependency, Water Relations and Restoration. Tree
Physiology, 23:1113-1124
Rudgers, A.. Orr, S. (2009). Non-native Grass Alters Growth of Native Tree Species via
Leaf and Soil Microbes. Journal of Ecology, 97:247-255
Sager-Fradkin, K.A., Jenkins, K.J., Happe, P., Beecham, J.J., Wright, G.R., Hoffman,
R.A. (2008). Space and Habitat Use by Black Bears in the Elwha Valley Prior to
Dam Removal. Northwest Science, 82(Special Issue 2008):164-178.
Schaff, S.D., S.R. Pezeshki, and F.D. Shields, Jr. (2003). Effects of soil conditions on
survival and growth of black willow cuttings. Environmental Management. 31(6):
748-763.
Scott, M.A., Shafroth, P.B., Auble, G.T. (1999). Responses of Riparian Cottonwoods to
Alluvial Water Table Declines. Environmental Management, 23(3):347-358
Shaffer, A. (2009). The Elwha Nearshore: Linking Management, Education, and
Research to Achieve Ecosystem Restoration. Priority Recommendations of the
Elwha Nearshore Consortium, (ENC) 2009. Compiled by: Anne Shaffer, ENC
Coordinator, Washington Department of Fish and Wildlife, Port Angeles,
Washington, and Dwight Barry, Director, Center of Excellence, Peninsula College,
Port Angeles Washington
Shaffer, A.J., Crain, P., Kassler, T., Penttila, D., Barry, D. (2012). Geomorphic Gabitat
Type, Drift Cell Forage, Forage Fish and Juvenile Salmon: Are They Linked?
Journal of Environmental Science and Engineering A. 1(2012):681-703, formerly
part of Journal of Environmental Science and Engineering, ISSN 1934-8932
Shaffer, A., Crain, P., Kassler, T., Schilke, J. (2008). Juvenile Chinook Use of the
Nearshore Central and Western Strait of Juan de Fuca. Washington Department of
Fish and Wildlife Habitat Program Sciences Division and Olympic National Park,
Port Angeles, WA
Shaffer, A., Coastal Watershed Institute, Wray, Jacilee, Olympic National Park (2004).
96

Native American Traditional and Contemporary Knowledge of the Northern Olympic
Peninsula Nearshore. Olympic Peninsula Intertribal Cultural Advisory Committee,
Kingston, Wa and Coastal Watershed Institute, Port Angeles, WA
Shaffer, A., ENC Coordinator, Washington Department of Fish and Wildlife, Barry, D.,
Center for Excellence, Peninsula College (2009). The Elwha Nearshore: Linking
Management, Education, and Research to Achieve Ecosystem Restoration. Priority
Recommendations of the Elwha Nearshore Consortium. Priority Recommendations
of the Elwha Nearshore Consortium (ENC) 2009. Port Angeles, WA
Shaffer, A., Washington Department of Fish and Wildlife, Lear, C., Clallam County,
Morrill, D., Beirne, M., Elwha Tribe, Crain, P., Olympic National Park. Elwha and
Glines Canyon Dam Removals: Nearshore Restoration of the Central Strait of Juan
de Fuca.
Shaffer, A.J, 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 (Special Issue 2008):72-90
Shafroth, P.B., Fuentes, T.L., Pritekel, C., Beirne, M.M., Beauchamp, V.B. (2009).
Vegetation of the Elwha River Estuary, ch. 8 of Duda, J.J., Warrick, J.A., Magirl,
C.S., eds., 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. 225-248.
Shafroth, P.B., J.M. Friedman, G.T. Auble, M.L. Scott and J.H. Brattne (2002). Potential
response of riparian vegetation to dam removal. Bioscience. 52(8): 703-712.
Simard, S.W. (2009). The Foundational Role of Mycorrhizal Networks in SelfOrganization of Interior Douglas-fir Forests. Forest Ecology and Management
(2009):1-13
Smith, Wilson, B.. Rasheed, S., Walker, R.C., Carolin, T., Shepherd, D. (2008).
Whitebark Pine and White Pine Blister Rust in the Rocky Mountains of Canada and
Northern Montana. Can. J. For. Res., 38:982-995
Smit, C., Vandenberghe, C., den Ouden, J., Müller-Schärer, H. (2006). Nurse Plants,
Tree Saplings and Grazing Pressure: Changes in Facilitation Along a Biotic
Environmental Gradient. Oecologia, 152:265-273
Stanley, E.H., Doyle, M.H. (2003). The Ecological Effects of Dam Removal. Frontiers
in Ecology and the Environment, 1(1 Feb., 2003):15-22
Stewart, G., Department of Geosciences Oregon State University, and Dr. Gordon E.
Grant, USDA Forest Service (2005). Potential Geomorphic and Ecological Impacts of
Marmot Dam Removal, Sandy River, OR Final Report. Prepared for Portland General
97

Electric July 2005
Stewart, M. (1999). Nutrient deficiencies and their symptoms in selected crops. Potash
& Phosphate Institute of Canada, Norcross, GA. http://www.nm.nrcs.usda.
gov/technical/handbooks/iwm/NM_IWM_Field_Manual/Se ction16/16dNutrient_Deficiency_Sysmptoms.pdf
Stillwater Sciences (2009). Post-Dam-Removal Channel Complexity Monitoring survey
Data Analysis, Sandy River, Oregon, 1st Year Following Dam Removal. Technical
Memorandum, prepared for Portland General Electric:1-4. Berkley, CA
Stoeckeler, J.H., Aamodt, E. (1940). Use of Tensiometers in Regulating Watering in
Forest Nurseries. Plant Physiology, 15(4):589-607. Contribution from Lake States
Forest Experiment Station, University Farm, St. Paul, Minnesota. Maintained by the
U.S. Department of Agriculture, Forest Service, in cooperation with Division of
Forestry, University of Minnesota
Stoecker, M. (2011). Dam Removal Sediment Management Examples and considerations
for the Matillija Fine Sediment Group. Stoecker Ecological,
www.StoeckerEcological.com
Sthultz, C.M., Gehring, C.A., Whitham, T.G. (2007). Shifts from Competition to
Facilitation Between a Foundation Tree and a Pioneer Shrub Across Spatial and
Temporal Scales in a Semiarid Woodland. New Phytologist, 173:135-145
Toft, R.J. (2004). Biological Flora of the British Isles. List Br. Vasc. Pl. (1958) no. 168,
16. Journal of Ecology, 92:537-555
Tolk, J.A. (2003). Soils, Permanent Wilting Points. Encylopedia of Water Science, DOI:
10.1081/E-EWS 120010337. United States Department of Agriculture (USDA),
Bushland, Tx
United States District Court Western District of Washington At Tacoma (2012). Wild
Fish Conservancy, et al., Plaintiffs, v. National Park Service, et al., Defendents. Case
No. C12-5109 BHS. Order Granting Defendants’ Motion to Dismiss.
Valadez, J. (2002). “Elwha Klallam”, included in, Native Peoples of the Olympic
Peninsula by Jacilee Wray, Ed Norman: University of Oklahoma Press, 2002
Van Der Kamp, B.J., Gokhale, A.A. (1979). Decay Resistance Owing to Near-Anaerobic
Conditions in Black Cottonwood Wetwood. Canadian Journal of Forest Research,
9(1979):39-44
Vesk, P.A. (2006). Plant Size and Resprouting Ability: Trading Tolerance and Avoidance
of Damage? Journal of Ecology, 94:1027-1034

98

Warren, R., State of Washington Department of Fish and Wildlife (2010). Letter to
Robert Elofsen, Lower Elwha Klallam Tribe June 10, 2010.
http://wildfishconservancy.org/resources/publications/wild-fish-runs/wdfw-letter-tolower-elwha-klallam-tribe
Warrick, J.A. et al. (2009). Beach Morphology and Change along the Mixed Grain-Size
Delta of the Dammed Elwha River, Washington. Geomorphology, 111(2009):136148.
Washington State Noxious Weed Control Board, http://www.nwcb.wa.gov/detail.asp?
weed=55
The Watershed Company (2011). Shoreline Restoration Plan for City of Port Angeles’
Shoreline: Strait of Juan de Fuca. Kirkland, WA
Wells, A.J., Balster, N.J., VanWychen, S., Harrington, J. (2008). Differences in Below
Ground Heterogeneity Within a Restoration of a Dewatered Reservoir in
Southwestern Wisconsin. Restoration Ecology, 16(4):678-688
Wendel, R., Zabowski, D. (2010). Fire History Within the Lower Elwha River
Watershed, Olympic National Park, Washington. Northwest Science, 84(1):88-97
Western Regional Climate Center, http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?waelwha
Wildman, L.A.S., MacBroom, J.G. (2005). The Evolution of Gravel Bed Channels After
Dam Removal: Case Study of the Anaconda and Union City Dam Removals.
Geomorphology 71 (2005):245–262
Winter, B.D., Crain, P. (2008). Making the Case for Ecosystem Restoration by Dam
Removal in the Elwha River, Washington. Northwest Science, 82(2008):13-28
Woodruff, D.R., Bond, B.J., Ritchie, G.A., Scott, W. (2002). Effects of Stand Density on
the Growth of Young Douglas-fir Trees. Canadian Journal of Forest Resources,
32:420-427
Woodward, A., Schreiner, E.G., Crain, P., Brenkman, S.J., Happe, P.J., Acker, S.A.,
Hawkins-Hoffman, C. (2008). Conceptual Models for Research and Monitoring of
Elwha Dam Removal - Management Perspective. Northwest Science, 82(2008):59-71
WTU herbarium website, an online native plant guide of the Burke Museum
http://biology.burke.washington.edu/herbarium/imagecollection.php
Xin, G., Zhang, G., Won Kang, J., Staley, J.T., Doty, S.L. (2009). A Diazotrophic,
Indole-3-Acetic Acid-Producing Endophyte from Wild Cottonwood. Biol Fertil Soils
(2009)45:669-674

99

Zabowski, Darlene (2012). Professor of Forest Soils and Soil Genesis and Classification;
Biogeochemical Cycling of Forest soils. University of Washington School of
Environmental and Forest Sciences, Seattle, WA

100