-
extracted text
-
HARBOR PORPOISE RETURN TO THE SOUTH PUGET
SOUND: USING BIOACOUSTIC METHODS TO MONITOR A
RECOVERING POPULATION
by
David Anderson
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2014
©2014 by David Anderson. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
David Anderson
has been approved for
The Evergreen State College
by
________________________
Erin Martin, Ph.D.
Member of the Faculty
________________________
Date
ABSTRACT
Harbor porpoise return to the South Puget Sound: using bioacoustic
methods to monitor a Recovering population
David Anderson
Harbor porpoise began returning to the Puget Sound during the
early 2000s after several decades of absence, becoming established in
the South Puget Sound by 2008. This study deployed a C-POD
ultrasonic click detector from March through May 2013 off Steilacoom,
Washington. Porpoise detections were compared to a variety of
environmental and temporal factors, including rate of tidal change,
wind speed, hour of the day and month of the year. A limited set of
visual observations were used to check the accuracy of the acoustic
data. Harbor porpoise were detected all hours of the day, with acoustic
detections peaking in the morning. The rate of detection was highest in
the month of May, and lowest during April. Acoustic detections were
highest during slack water and slow incoming tides, compared to faster
tides and slower outgoing tides. The acoustic data showed that harbor
porpoise did not leave the area when vessels operating echo sounders
transited the study site, though it is not possible to determine with
acoustic data if diving was used as an avoidance response. Given the
proximity of the site to a ferry terminal, there is a possibility that
harbor porpoise have become habituated to the ferries, which
accounted for greater than 80% of the traffic using echo sounders. The
C-POD has proven to be a useful tool in monitoring the harbor porpoise
population, with an ability to monitor day and night in all weather
conditions.
Table of Contents
Table of Contents ..................................................................................... iv
List of Figures .......................................................................................... v
List of Tables ........................................................................................... vi
1 Introduction ........................................................................................... 1
2 Literature Review ................................................................................. 7
2.1 The Harbor Porpoise ............................................................ 9
2.1.1 Species Description ........................................................ 9
2.1.2 Taxonomy and evolution.............................................. 11
2.1.3 Distribution and Abundance ....................................... 14
2.1.4 Life History .................................................................. 17
2.1.5 Social Structure ........................................................... 18
2.1.6 Behavior ....................................................................... 18
2.1.7 Feeding ......................................................................... 20
2.1.8 Sound production, hearing, communication and
echolocation...................................................................................... 24
2.2 Threats ............................................................................... 28
2.2.1 Natural predators ........................................................ 29
2.2.2 Fisheries interactions .................................................. 30
iv
2.2.3 Noise pollution ............................................................. 32
2.2.4 Hunting ........................................................................ 34
2.2.5 Pollution ....................................................................... 35
2.2.6 Climate change ............................................................ 38
2.2.7 Threats conclusion ....................................................... 39
2.3 Visual monitoring............................................................... 40
2.3.1 Vessel surveys .............................................................. 43
2.3.2 Shore surveys ............................................................... 45
2.4 Passive acoustic monitoring .............................................. 46
2.4.1 Hydrophone .................................................................. 48
2.4.2 The POD, T-POD and C-POD ultrasonic sound
detectors ........................................................................................... 52
2.5 Environmental analysis ..................................................... 62
2.5.1 Seasonal patterns ........................................................ 63
2.5.2 Diel patterns ................................................................ 65
2.5.3 Tidal patterns............................................................... 68
2.6 Vessel avoidance ................................................................ 69
3 Methods ............................................................................................... 73
3.1 Study area .......................................................................... 73
v
3.2 Acoustic Survey Methods ................................................... 74
3.2.1 C-POD Click Detectors ................................................ 74
3.2.2 Deployment and Mooring ............................................ 75
3.2.3 Processing C-POD Data ............................................... 76
3.3 Visual Survey Methods ...................................................... 78
3.4 Environmental Data .......................................................... 80
3.4.1 Anthropogenic Impact Data ........................................ 82
4 Results ................................................................................................. 84
4.1 Acoustic results .................................................................. 84
4.2 Click train analysis ............................................................ 85
4.3 Environmental results ....................................................... 87
4.4 Seasonality ......................................................................... 91
4.5 Interaction with vessel echo sounders .............................. 93
4.6 Visual results ..................................................................... 95
5 Discussion ............................................................................................ 97
5.1 Visual observation.............................................................. 97
5.2 Environmental analysis ..................................................... 99
5.3 Anthropogenic disturbance .............................................. 103
5.4 Passive acoustic monitoring ............................................ 105
vi
5.5 Conclusion ........................................................................ 107
6 An interdisciplinary study with broad impact ................................. 110
7 Acknowledgements............................................................................ 114
8 Bibliography ...................................................................................... 115
vii
List of Figures
Figure 2-1: Harbor porpoise adult and calf. ............................... 10
Figure 2-2: Worldwide harbor porpoise distribution. ................ 14
Figure 2-3: Odontocete head ....................................................... 25
Figure 2-4: Dall's porpoise .......................................................... 43
Figure 3-1: Study site. ................................................................. 73
Figure 3-2: Harbor porpoise click trains.. .................................. 76
Figure 3-3: Echo sounder sonar. ................................................. 77
Figure 4-1: Detection positive minutes (DPM) per hour. .......... 84
Figure 4-2: Detection positive minutes (DPM) per day. ............ 85
Figure 4-3: Hurdle model of environmental variables............... 87
Figure 4-4: Variation in the DPM throughout the day. ............. 89
Figure 4-5: Wind speed DPM per hour. ...................................... 90
Figure 4-6: DPM per day throughout deployment. .................... 91
Figure 4-7: Differences between months. ................................... 92
v
List of Tables
Table 4-1: Hurdle model results ................................................. 88
Table 4-2: Results of Kruskalmc post hoc pairwise test. ........... 92
Table 4-3: Truth table of presence/absence before and echo
sounder events recorded by the C-POD. ............................................... 94
vi
1 Introduction
After World War II, harbor porpoise (Phocoena phocoena) were
one of the most common cetaceans in the Puget Sound (Scheffer and
Slipp 1948), yet by the time the Marine Mammal Protection Act
(MMPA) was passed in 1972, harbor porpoise had been extirpated from
the waters south of Admiralty Inlet. It is unknown for certain what led
to the loss of habitat use by the harbor porpoise, though it was likely
due to a combination of habitat degradation, fisheries bycatch, and
recreational human takes which resulted in unsustainable losses to the
population. High levels of pollution were present in the Puget Sound
during this time period and may have contributed to the decline in the
population, as toxins bioaccumulate in higher trophic level animals
like the harbor porpoise (Calambokidis et al. 1985). The bycatch from
gillnet fisheries, a common commercial and tribal fishing method in
the Puget Sound, is also considered to have contributed to the
extirpation from the South Puget Sound waters. Diving seabirds saw a
decline over the same time period; the birds have a similar diet to the
harbor porpoise and are also subject to by catch issues (Bower 2009),
suggesting a probable systemic problem.
Occasional reports of harbor porpoise sightings were received
between the 1970s and early-2000s, but these were always isolated
incidents with no more than a few animals reported, with no animal
1
sighted during surveys in the 1980s and 1990s (Flaherty and Stark
1982, Calambokidis et al. 1985, 1993). During this time period, the
most common cetacean in the Puget Sound was the Dall’s porpoise
(Phocenoides dalli). Dall’s porpoise are noticeably larger than harbor
porpoise and have different coloration and diving behavior, factors
which make species identification straightforward in the field.
It is unknown precisely when harbor porpoise started returning
to the Puget Sound south of Admiralty Inlet. As the harbor porpoise
population in the San Juan Islands and the Strait of Juan DeFuca
increased through the 1990s and early-2000s, there were increased
sightings of animals within the Puget Sound. By 2007 harbor porpoise
had gradually expanded through much of the North and Central Puget
Sound, with a small group of porpoise regularly sighted in the South
Puget Sound by 2008 (Calambokidis, personal communication). Four
porpoise were spotted off the west side of Anderson Island in the South
Puget Sound in April 2008 by the author during a kayak trip, with
many additional sightings later in that year. The population has
continued to grow throughout the Puget Sound and harbor porpoise
are frequently seen by boaters. While the harbor porpoise population
has been increasing in Washington’s estuarine waters, Dall’s porpoise
sightings have dropped drastically, and it is unknown if the decline is
related to the increase in harbor porpoise or other factors. It is also
2
unknown what factors contributed to the return of the harbor porpoise
to the Puget Sound waters, though reduction in levels of some
pollutants or changes in fisheries many have played a part. Some
animals may have expanded their range into southern Puget Sound as
a result of habitat changes or having exceeded the carrying capacity in
areas they previously occupied.
Although the National Marine Fisheries Service (NMFS) is
responsible for conducting periodic surveys to determine population
size and distribution of marine mammals, no aerial or boat based
surveys have been conducted of small cetaceans in the Puget Sound
since harbor porpoise have returned. Congressional budget cuts have
led to the reduction or elimination of many of the research and
management activities related to these and other protected species.
Without these surveys we lack the ability to monitor the recovery of
harbor porpoise populations, or determine if any management
decisions are necessary to protect them.
Conservation of harbor porpoises in the South Puget Sound
requires abundance estimates as well as knowledge about how
porpoises move throughout their range and make use of the habitat. A
variety of environmental factors, such as tidal flow and time of day
have been shown to influence porpoise presence and behavior
(Johnston et al. 2005, Todd et al. 2009). Tidal flows cause eddies or
3
convergent fronts in the water that can concentrate food species at
different times in the tidal cycle and attract porpoises to the area.
Differences in harbor porpoise presence and behavior vary from site to
site depending upon how the tidal current interacts with the local
bathymetry to concentrate prey. Diel variations in behavior can often
be attributed to the daily migration of zooplankton from the depths to
the photic zone and back. Studies in different regions have shown
consistent diel feeding patterns in harbor porpoise on a site by site
basis, but patterns vary between studies (Todd et al. 2009, Haarr et al.
2009). Determining the distribution and timing of harbor porpoise
behaviors within the South Puget Sound is important inform
management decisions and to minimize anthropogenic disturbance in
critical habitat areas.
Traditionally, harbor porpoise have been studied using visual
observations from land, sea, and air. Their small size, dark coloration,
and brief surface time are significant challenges to the study of this
animal through visual observations. Unlike some of the dolphins and
the Dall’s porpoise that are often attracted to vessels or human
activity, harbor porpoise are considered to be a rather shy species,
often seen fleeing vessels and avoiding areas of high activity (Embling
et al. 2010). Even though harbor porpoise are known to react to human
activity, it is unclear from the literature whether they leave the area
4
for an extended length of time, or if it is a short-term response. In
recent years, there has been increased interest in finding ways that
remote sensing technology, such as passive acoustic monitoring (PAM)
could be used to monitor small cetaceans. Acoustic monitoring can be
used to augment visual observation, capture echolocation clicks, and
collect data about underwater activity of porpoises during the day or
night, in all weather conditions, without introducing human induced
disturbance.
This study considered how PAM equipment, combined with
traditional visual methods, could be used to monitor harbor porpoise
presence and behavior in the South Puget Sound. A location in the
South Puget Sound off Steilacoom, Washington was chosen to deploy a
C-POD (Cetacean POrpoise Detector, Chelonia Ltd., UK) ultrasonic
monitor, from March through June 2013, to detect harbor porpoise
echolocation clicks. This location was chosen due to known harbor
porpoise presence, favorable bathymetry and observation points that
provide a view of the deployment site as well as approximately 30 km2
of the surrounding basin. Visual observations were recorded whenever
possible to corroborate acoustic data, as well as recording cetacean
presence and behavior in the greater basin.
In this study, passive acoustic monitoring was shown to provide
detailed information about harbor porpoise presence, distribution and
5
behavior that will be useful for informing management decisions to
protect the Puget Sound harbor porpoise population in the future,
while limiting impacts on human activities. Acoustic monitoring is not
a replacement for visual monitoring, it provides a low cost method of
long-term monitoring of trends in abundance, distribution and
behavior that is not restricted by the time of day or weather conditions,
but it also has many limitations, such as limited range and the
inability to detect silent animals. With proper monitoring and
management, harbor porpoise can successfully inhabit their traditional
habitat in South Puget Sound.
6
2 Literature Review
This literature review will address issues related to the return of
harbor porpoise to the South Puget Sound and the use of acoustic
monitoring methods to complement traditional visual methods to gain
a better understanding of harbor porpoise behavior. Harbor porpoise
are an important high trophic level predator throughout the coastal
waters and estuaries of the temperate and boreal waters of the
northern hemisphere. As with many high level predators, harbor
porpoise and other marine mammals are considered a sentinel species
for the ecosystems they inhabit (Bossart 2011). Harbor porpoise are
one of the most numerous species of cetaceans in the world, and they
inhabit coastal waters that are often near population centers of the
temperate northern hemisphere, making them an easily accessible
study subject. This proximity to human population also leads to
conflict with anthropogenic usage of the coastal ecosystem, which has
led to a great deal of research into human activities and their impact
on the porpoise populations.
While, as a species, harbor porpoise are not endangered, many
populations and subspecies are in decline due to human activity, with
the Baltic Sea population and the Black Sea subspecies considered
endangered or critically endangered (Hammond et al. 2008b). Harbor
porpoise have also returned to Dutch waters in 1990 after three
7
decades of absence (Boonstra et al. 2013), and they have recently
returned to the San Francisco Bay after 65 years of absence (Keener et
al. 2011). The reestablishment of populations in these areas provide an
opportunity to study the factors that might influence the return of
porpoise to these and other regions, and to develop management plans
to protect these animals in a way that will allow their populations to
recover.
Even with their numbers and accessibility, harbor porpoise
research is subject to the same difficulties encountered in any cetacean
research. Cetaceans are only visible at the surface for a small fraction
of their lives with underwater behaviors and movement patterns
hidden from view. Before the last few decades, studies of harbor
porpoise were restricted to visual observations of behavior at the
surface and examination of dead animals that were stranded, caught
in fisheries or killed in a lethal sampling program (Scheffer and Slipp
1948). With recent developments in passive and active acoustic
technology, as well as the development of wildlife tracking tags, many
new opportunities have emerged for advancing our knowledge of
marine mammals. Additionally, the use of remote sensing data by
physical and biological oceanographers has informed research into
marine mammals and other top ocean predators by providing
8
information about bathymetry and ocean productivity that was
unavailable until recent decades.
This literature review will provide a general overview of the
harbor porpoise, including information about its biology, range,
taxonomy and echolocation; cover a variety of threats to the harbor
porpoise including both natural predators and anthropogenic factors;
examine methods used for monitoring harbor porpoise and other small
cetaceans, both visually and acoustically; and the analysis of the effect
of environmental factors and anthropogenic disturbance on harbor
porpoise distribution and behavior.
2.1 The Harbor Porpoise
2.1.1 Species Description
Harbor porpoise are the smallest cetacean found in the waters of
the United States, measuring 1.5-1.7 m and 60-75 kg at maturity. The
small triangular dorsal fin, without any white on the trailing edge,
along with the characteristic rolling surface motion aids in
differentiating harbor porpoise from other small cetaceans within the
harbor porpoise’s range (Bjorge and Tolley 2009). Coloration is a dark
gray on dorsal surfaces, becoming light gray to white ventrally,
without a distinct demarcation line between colors (Figure 2-1). A
9
study of pigmentation patterns within and between different
populations showed that there are variations between individuals in a
population, and that there are trends within populations, but that
pigmentation cannot be used as a morphological way to differentiate
between populations (Koopman and Gaskin 1994). Though they are
extremely rare, white porpoise with only small areas of pigmentation
have been sighted in populations in both the Pacific and Atlantic
Oceans (Keener et al. 2011).
Figure 2-1: Harbor porpoise adult and calf. (Copyright Uko Gorter)
10
2.1.2 Taxonomy and evolution
Harbor porpoise, along with
all other whales, dolphins and
porpoises of the order Cetacea, are
descended from land dwelling
Taxonomic Classification
Order:
Suborder:
Superfamily
Family:
Genus:
Species:
Cetacea
Odontoceti
Delphinoidea
Phocoenidae
Phocoena
Phocoena phocoena
ungulates, with the hippopotamus being their closest extant relative,
splitting off around 53-54 million years ago (Ma) (Berta et al. 2006). All
cetaceans retain the four-chamber stomach of their graminivore
ancestors (Mead 2009). Mesonychians, an extinct taxa believed to be
closely related to early cetaceans, had dentition suggesting that a
change to a carnivorous diet could have occurred before protocetaceans
returned to a semi-aquatic life (Berta et al. 2006).
Cetaceans are divided into two suborders, Mysticeti or the
baleen whales, which filter feed using baleen plates made of keratin
instead of teeth, and Odontoceti, or toothed whales, which includes
dolphins and porpoises. It is believed that the split between mysticetes
and odontocetes happened around 35 Ma, though there is some
disagreement about the time frame (Berta et al. 2006, Fordyce 2009).
The ability for odontocetes to echolocate is believed to have evolved
within a few million years of the split between mysticetes and
odontocetes. Fossils of an extinct branch that diverged approximately
11
32 Ma showed adaptations indicative of the ability to echolocate
(Geisler et al. 2014).
The increasing availability of low cost genetic testing has
enabled geneticists to reexamine and propose changes to the branches
of the phylogentic tree of the porpoise and dolphin species. The
Delphinoidea clade that includes ocean dolphins (Delphinidae) and
porpoises (Phocoenidae) diverged approximately 19 Ma (Chen et al.
2011). Several lines of genetic study have shown that porpoises are
most closely related to belugas and narwhals (Monodontidae), having
diverged from ocean dolphins approximately 16 Ma, with the porpoises
diverging from the narwhals approximately 11 Ma (Waddell et al.
2000, Chen et al. 2011). Riverine dolphins are called “dolphins” due to
their morphological similarities to ocean dolphins, yet they are not part
of Delphinoidea, nor are they a monophyletic group. They are
members of ancient odontocete lineages that were protected from the
competition for resources by true delphids due to their selection of a
freshwater habitat (Cassens et al. 2000).
There are six extant species of porpoise, four of which are in the
genus Phocoena, including the harbor porpoise, Phocoena phocoena.
The relationships between the porpoise species is the subject of
continuing genetic research, some of which suggests that harbor
porpoise are most closely related to Dall’s porpoise, which is currently
12
classed in a different subfamily and genus (Rosel et al. 1995b). Mating
between harbor porpoise males and Dall’s females have been known to
produce viable, hybridized offspring (Willis et al. 2004). This raises
interesting questions, not yet covered in current literature, about
genetics, morphology, genus differentiation and the role of
hybridization between these two species.
There are four recognized subspecies, P. p. relicta in the Black,
Marama and Adriatic Seas; P. p. phocoena in the North Atlantic; P. p.
vomerina in the eastern North Pacific; and an unnamed subspecies in
the western North Pacific (Rosel et al. 1995a, 2003, Rice 1998, Frantzis
et al. 2001). The division into subspecies was conducted using
morphological characteristics and analysis of mitochondrial DNA.
Additional DNA analysis has been conducted to identify separate
breeding populations, with 14 identified groups in the North Atlantic
and Black Sea (Andersen 2003). There has been limited work on
genetically defining harbor porpoise population structure within the
North Pacific Basin, though mitochondrial DNA studies in the 1990s
suggest that there is some gene flow between populations along the
west coast of North America (Rosel et al. 1995a).
As recently as the 1970s the terms “porpoise” and “dolphin” were
used interchangeably in the scientific literature for any small
odontocete (Renaud and Popper 1975), and the lack of distinction
13
continues in common usage. Porpoise are distinguished from dolphins
by the lack of a prominent rostrum, smaller mouths, stouter bodies,
triangular dorsal fins and spade shaped instead of conical teeth (Read
2009). Porpoise make a much more limited range of sounds, most of
which are above the frequency of human hearing, unlike the lower
frequency whistles and clicks of dolphins that are audible to humans
(Frankel 2009).
2.1.3 Distribution and Abundance
Harbor porpoise are distributed throughout the temperate and
boreal coastal waters of the Northern Hemisphere (Figure 2-2),
preferring the shallow waters of bays and estuaries, as well as the
near-shore region along the coast (Bjorge and Tolley 2009). A
Figure 2-2: Worldwide harbor porpoise distribution. (Copyright Wikipedia user BloodIce
(CC BY-SA 3.0)
http://commons.wikimedia.org/wiki/File:Cetacea_range_map_Harbour_Porpoise.PNG)
14
geographically and genetically distinct population is located in the
Black Sea, separated from the porpoise in the Atlantic by the
Mediterranean Sea and the Bosphorus Strait, where porpoise are
rarely sighted (Frantzis et al. 2001, Reeves and Notarbartolo di Sciara
2006). It is thought that harbor porpoise have avoided the
Mediterranean Sea for several thousand years due to the increase in
temperatures since the last ice age, separating the Black Sea
population from those in the Atlantic Ocean. Tests were run on 5
bycaught animals in the Agean Sea showing that they were part of the
Black Sea population (Rosel et al. 2003). The only harbor porpoise
population known to venture into tropical waters are those along the
west coast of Africa, which can be found as far south as Mauritania
(19° N) (Smeenk et al. 1992, Boisseau et al. 2007),
Globally, the harbor porpoise population is thought to be greater
than 700,000 animals, and the species is listed as Least Concern (LC)
on the International Union for Conservation of Nature (IUCN) Red
List of Threatened Species (Hammond et al. 2008b). The North Sea
supports almost half the world population, with an estimated
population of 335,000 animals (Hammond et al. 2002). While the global
population is not considered to be under threat, in many parts of the
world the population is in decline, and some of the distinct populations
are threatened or endangered. In the Baltic Sea, where porpoise
15
density once rivaled the neighboring North Sea, it is thought that the
population is down to a few hundred individuals (Hammond et al.
2008a). The Black Sea is seeing a large decline in abundance, though
no current estimates exist (Birkun Jr. and Frantzis 2008).
In Washington State, harbor porpoise are found on the Pacific
coast as well as in the Strait of Juan DeFuca. Once common in the
waters of the Puget Sound (Scheffer and Slipp 1948), they were
extirpated south of Admiralty Inlet and east of the San Juan Islands
sometime before surveys began in the 1970s (Calambokidis et al.
1992). There is no additional information available on harbor porpoise
abundance in the Puget Sound before extirpation other than the
Scheffer and Slipp (1948) reference to harbor porpoise as being
“common” in the 1940s. Harbor porpoise began returning to the Puget
Sound in the early 2000s (Calabokidis, personal communication), but
no surveys have been conducted since their return to determine
abundance. Aerial surveys of the coastal waters of Washington State
and the Strait of Juan DeFuca were conducted in the 1990s, producing
an abundance estimate of 15,000 harbor porpoise in these waters
(Calambokidis et al. 1993). High concentrations of harbor porpoise
were found in the central and eastern portions of the Strait of Juan
DeFuca and the northern San Juan Islands during aerial surveys of
those areas in 2003 (Chandler and Calambokidis 2003). These high
16
concentrations might have led some of those animals to expand their
range into the Puget Sound.
2.1.4 Life History
Timing of harbor porpoise birth varies by population, with the
females in most areas giving birth from May through August after a
10.5 month gestation period. Females are capable of giving birth
annually, though not all females produce a calf every year, in some
populations the annual birth rate in reproductive age females can run
lower than 0.75 (Read 1990). One study suggests that the females in
the population off Central California reproduce every second year,
instead of annually as do the populations studied in the Atlantic Ocean
(Read and Hohn 1995). Size at birth is approximately 65-80 cm and
around 5 kg, though this can vary by population group and food
availability. Weaning is thought to occur between 8 and 12 months,
which means that females spend much of their time both weaning and
pregnant. Sexual maturity occurs at 3-4 years, with first parturition
at 4-5 year. Harbor porpoise have been shown to live for more than 20
years, though fewer than 5% of adults are believed to live longer than
12 years (Lockyer 2003).
17
2.1.5 Social Structure
Little is known about the social structure within and between
harbor porpoise populations. Harbor porpoise are usually sighted in
foraging groups of one to three individuals, though groups of up to 12
are not uncommon (Flaherty and Stark 1982). When combined with
their coastal habitat and cryptic behavior, the small group sizes
favored by the harbor porpoise might play a role in predator avoidance
(Gygax 2002). Other than cow-calf pairs, it is believed that groups are
rather fluid in their members and structure (Flaherty and Stark 1982,
Saana 2006). Larger groups have been reported at times in the
literature, most of which are thought to be loosely associated feeding
aggregations (Hoek 1992, Saana 2006). Tracking studies, using
position reporting satellite tags and follow-up acoustic surveys, have
shown that there are areas of seasonal high density, but some animals
are found outside these aggregations (Sveegaard et al. 2011b, 2011a).
2.1.6 Behavior
Harbor porpoise spend most of their time foraging in groups of
1-3 individuals though it is not uncommon to have multiple small
groups foraging in the same area. Normal foraging behavior consists of
18
a surface series of 3-5 breaths followed by an extended foraging dive.
Surface events are generally quick consisting of a simple forward roll
that lasts about 2 seconds, with an audible puff as they breathe out,
though in rough water they rise higher out of the water without their
normal rolling motion (Scheffer and Slipp 1948). Aerial and splashing
displays such as breaching and porpoising are rare in harbor porpoise,
and mostly occur in larger groups, suggesting a possible social
significance to those behaviors (Flaherty and Stark 1982, Hall 2011)
Few studies have been conducted regarding harbor porpoise dive
depth and duration. A study in the Bay of Fundy, New Brunswick,
Canada was conducted using time and depth recording tags showing
that while foraging harbor porpoise made an average of 30 dives per
hour, with a mean dive depth of 25 ± 30 m, and a mean dive time of 65
± 53s. The maximum dive time recorded was 321 s, and the maximum
depth by the same animal was 226 m. The maximum depth of the
study area was 230 m, which may have limited the ability to determine
their full diving capabilities (Westgate et al. 1995). A study currently
underway off the west coast of Greenland is using time and depth
recording satellite tags to monitor travel and dive profiles of harbor
porpoise. Seven of the tagged animals recorded dives of 200 m, with
two porpoise recording dives to a depth of 400 m (Nynne Hjort Nielsen,
personal communication). Puget Sound has a maximum depth of 284
19
m; the results of the Greenland study suggest that harbor porpoise are
able to utilize the entire Puget Sound for benthic foraging.
Harbor porpoise are active during all hours, though individual
animals do have periods where they rest at the surface (called
“logging”). Unlike terrestrial animals, cetaceans have developed a
unique sleep system where one side of their brain sleeps at a time
while the other side remains awake, but in a lower-energy state that
allows them to monitor for potential danger (Lyamin et al. 2008).
While dolphins and the Dall’s porpoise are known for riding the
bow wave of boats and participating in other interactions with people,
harbor porpoise are considered quite shy. Several studies have noted
avoidance behavior in relation to survey vessels (Flaherty and Stark
1982, Barlow 1988, Polacheck and Thorpe 1990, Palka 1995), though
some studies have noted a lack of reaction to some boats and even an
apparent attraction at times (Evans et al. 1994, Raum-Suryan 1995).
2.1.7 Feeding
As with most other marine mammals, harbor porpoise feeding
events are rarely observed directly. Most of what is known about
harbor porpoise diets comes from studying the stomach contents of
stranded or bycaught animals. Recent studies have used fatty acid
(FA) analysis or stable isotope analysis to determine important
20
information about the diet of harbor porpoise (Jansen et al. 2012).
Each of these methods enables the investigation of dietary trends over
different timelines, ranging from hours (stomach contents), to weeks or
months (fatty acids), or even years (stable isotopes). Examination of
stomach contents can identify taxa and estimate the size of recently
ingested prey items by teasing out otoliths (fish ear bones), skeletal
bones, and the beaks and eyes of squid and octopus (Pierce et al. 1991,
Santos and Pierce 2003). Fatty acid analysis compares levels of
different fatty acids that are passed from prey species and stored in the
tissues of a predator (Budge et al. 2006). By knowing FA signatures of
the important forage fish, it is possible to analyze blubber samples to
determine the predominant prey of the harbor porpoise over the past
few weeks or months. Analyzing stable isotopes can reveal the trophic
level of an animal based on the concentration of the stable isotopes of
carbon and nitrogen (Jansen et al. 2012). Body tissues have differing
rates of turnover providing the opportunity to analyze diet over
different time scales. The liver turns over very quickly with isotopes
representing evidence of recent diet, while bones and other hard tissue
turn over very slowly providing a longer term nutritional record
(Phillips and Eldridge 2006).
Harbor porpoise are a high trophic level predator, with an
estimated average tropic level of 4.1 based on stomach samples
21
collected from bycaught animals (Pauly et al. 1998). In the North Sea
and Dutch coastal waters, stable carbon and nitrogen isotopes (δ13C
and δ15N) from the muscle and bone of stranded harbor porpoise have
been used to assess differences in tropic level between animals of
different ages, genders, and feeding areas (Jansen et al. 2012). The 15N
becomes enriched with each additional trophic level so animals with
higher δ15N are consuming higher trophic level prey. Neonates were
found to be eating at the highest level, because they are feeding
exclusively on their mother’s milk, which is one level higher than the
mother is eating. Adult females ate at a slightly higher trophic level
than males, and smaller adults ate at a higher level than larger adults.
It was also found that animals in shallower coastal waters ate at a
higher trophic level than those living in deeper water.
Unlike some of the larger whales with thick blubber, that can
have annual prolonged fasts during their migrations, the harbor
porpoise’s small body size and thinner blubber layer requires them to
eat regularly, with starvation a constant threat when food is scarce
(Koopman et al. 2002). Harbor porpoise are opportunistic feeders,
eating a variety of small fish and cephalopods (squid and octopus)
(Gaskin et al. 1974, Pauly et al. 1998). Additionally, the neonates and
juveniles are known to eat euphasiids (krill). Krill shells have been
found in the stomachs of adults, but it is unclear whether the adults
22
consumed the krill directly, or if the krill had been consumed by prey
eaten by the porpoise. Harbor porpoise are generalists when it comes
to their food, able to eat a wide variety of species, though it is unknown
whether their diet is completely opportunistic or if they selectively eat
high quality prey when it is available (Santos and Pierce 2003).
Studies in the Gulf of Maine show seasonal difference in the harbor
porpoise diet and suggest some selectivity of prey. Porpoise fed
primarily on Atlantic herring in the summer months, while in the
autumn, the diet was more varied. Herring still made up a majority of
calories consumed, but porpoises also consumed pearlsides and red and
white hake. Lactating females consumed more hake and less herring
when compared to the diet of non-lactating females. (Gannon et al.
1998).
The literature on the diets of harbor porpoise in the Eastern
Pacific Ocean is fairly limited, and only a few published studies have
covered the inland waters of British Columbia and Washington State.
A study examining the stomach contents of 26 stranded harbor
porpoise collected along the eastern side of Vancouver Island, British
Columbia and the San Juan Islands in Washington State found prey
numbers were divided almost equally between fishes (52.2%) and
cephalopods (46.5%). Findings indicated the fishes included species
from ten families and cephalopod species from three families. Juvenile
23
blackbelly eelpout (Lycodopsis pacifica) represented the greatest
number of prey (49.6%), though this result is likely biased by 61.5% of
the harbor porpoise in this study having been collected in the spring,
which coincides with the seasonal availability of the juvenile eelpout.
(Walker et al. 1998). A more recent study analyzed 36 stranded harbor
porpoise from the Salish Sea (the Strait of Juan DeFuca, Puget Sound
and the Strait of Georgia), and found evidence of 7 families of fishes
along with cephalopods and polychaetes (which were not identified to
taxa) in the stomach contents (Nichol et al. 2013). Both studies found
Pacific herring (Clupea pallasi) to be an important food source, along
with a mix of benthic and midwater species.
2.1.8 Sound production, hearing, communication and
echolocation
Unlike the terrestrial environment, much of the marine
environment is light limited, so animals often adapt to rely more on
senses other than sight to monitor their environment. For marine
mammals, sound and vibration play an important role in foraging,
predator avoidance, mapping their environment and communicating
(Frankel 2009).
As odontocetes, harbor porpoise produce sounds by moving air
through a structure known as the phonic lips, located in the nasal
passage near the blowhole. Dorsal bursae are lipid filled structures
24
attached to the phonic lips that act as a resonator, amplifying the
sounds. The dorsal bursae lay at the back of the melon, a special
structure unique to odontocetes, which sits on top of the skull and
forms the distinctive forehead of toothed whales. The melon contains
special fats that act as acoustic lenses, focusing most of the sound
energy in a cone directly in front of the animal (Au et al. 1999)(Figure
2-3).
Figure 2-3: Odontocete head, showing sound path from phonic lips, reflecting off
bone and being focused in the melon. Incoming sounds, including echolocation
returns, are received through channels of fatty tissue in the lower mandible, and
transmitted to the auditory bullae. (Copyright Wikipedia user Emoscopes (CC BY-SA
2.5) http://commons.wikimedia.org/wiki/File:Toothed_whale_sound_production.png)
Over 50% of the energy produced in a harbor porpoise
echolocation click is directed within a 16° cone directly in front of, and
slightly above the centerline of the head (Au et al. 1999, 2006,
Kastelein et al. 2005, Madsen et al. 2010). Porpoise produce high
frequency sounds in the form of clicks in the frequency range of 100160 kHz (Madsen et al. 2010) at sound pressure levels (SPL) that can
25
exceed 205 dB re 1 µPa pp 1 as was measured in foraging wild porpoise
(Villadsgaard et al. 2007). It is believed that porpoise evolved to use
such high frequency sonar to avoid detection by killer whales (Orcinus
orca) who only hear up to 100 kHz (Andersen and Amundin 1976).
While high frequency clicks provide protection from predation, they do
not travel as far as lower frequency sounds thus limiting the range of
echolocation and communication.
Odontocete hearing follows a different sound pathway than that
found in terrestrial mammals. Instead of having an outer ear that
channels sound through the ear canal, porpoise and other odontocetes
receive sound using their lower jaw. Thin C-shaped mandibles hold a
special acoustical fat, which receives sounds and directs them to the
inner ear (Nummela et al. 2007). Harbor porpoise are thought to have
a hearing range from 250 Hz through 180 kHz with maximum
sensitivity in their echolocation click range of 100-140 kHz (Kastelein
et al. 2002).
Harbor porpoise primarily produce high frequency clicks and are
unable to make the wide range of lower frequency vocalizations that
dolphins are known to produce (Frankel 2009). Research on harbor
Standard shorthand for acousticians used to determine exactly how the dB
level was calculated. “pp” means “Peak to Peak”, which measures maximum energy,
“RMS” is the other common reporting unit, which means “Root Mean Squared” and is
more concerned with the average energy. “re 1 µPa” is “reference of 1 micro-Pascal”,
the standard in water, where dB in air are measured “re 20 µPa”.
1
26
porpoise communication is limited, but preliminary investigations
suggest that porpoise communicate using variations in the inter-click
interval and frequency of their click trains. A study of four captive
harbor porpoise, including a cow and her calf, demonstrated that
certain click patterns were not used for echolocation, but represented
communication between individuals. The high frequency of harbor
porpoise clicks limits their effective communication range to around
1000 m (Clausen et al. 2012).
Harbor porpoise use echolocation while foraging to find and
track their prey, to navigate and orient themselves spatially within
their environment (Verfusß et al. 2005, Au 2009). To echolocate,
animals first produce sound waves, which are transmitted into the
environment. The waves then encounter objects (such as the sea floor,
prey or other porpoise) and bounce back, forming an echo which
returns to the animal. The strength of the echo depends on the
difference in density between the object and the water, with a greater
difference leading to a stronger echo. The timing of the echo lets the
porpoise know the distance to the target, while the amplitude,
frequency and the shape of the returning sound envelope reveal
information about the prey species and its angle relative to the
porpoise (Au et al. 2009). As animals close in on their targets, the
return time of the echoes decrease, allowing a decreasing inter-click
27
interval (ICI), eventually becoming what is referred to as a “buzz”
with over 300 clicks per second (Deruiter et al. 2009, Nuuttila et al.
2013).
The anatomy of the prey species will influence the strength and
envelope shape of the echo that is produced. Ray-finned fishes that
have swim bladders produce a stronger acoustic signature than other
fish, due to the large difference in density between the gas in the swim
bladder and the flesh of the fish (Foote 1980). While squid lack a swim
bladder or skeleton to produce a strong echo, experiments have shown
that their bodies do produce a sufficient echo which can be detected by
odontocete clicks (Madsen et al. 2007).
The development of echolocation has enabled odontocetes, such
as the harbor porpoise, to thrive in their marine environment. By
depending on sound rather than vision, they are able to efficiently
locate prey, navigate in complete darkness, and avoid predators (Au
2009).
2.2 Threats
Harbor porpoise face a variety of threats throughout their range.
Marine mammal eating killer whales have accounted for most known
predation events on harbor porpoise, though other species have also
caused documented porpoise mortality. Anthropogenic factors, such as
28
fisheries bycatch, can have a direct impact on harbor porpoise
populations. Other factors, such as pollution or vessel noise, can cause
indirect impacts such as susceptibility to disease or reduced fertility.
2.2.1 Natural predators
Due to their small size and large numbers, harbor porpoise are
an important prey item for larger predators. The greatest natural
predatory threat to harbor porpoise are the mammal eating ecotypes of
the killer whale, such as the Bigg’s or transient killer whales (Orcinus
orca) of the eastern North Pacific Ocean (Jefferson et al. 1991, Baird
and Dill 1996). Large lamnid sharks, such as the great white
(Carcharodon carcharias) and Greenland sharks (Somniosus
microcephalus), are also major predators of harbor porpoise (Long and
Jones 1996, Heithaus 2001). Recent accounts from Belgium and France
have documented gray seals (Halichoerus grypus) killing and feeding
on harbor porpoise, possibly due to declines in the lesser sandeel,
which are a major source of high-energy forage for the seals (Haelters
et al. 2012, Bouveroux et al. 2014).
Some fatal interactions with other species are not easily
explained by the predator-prey dynamic. Bottlenose dolphins (Tursiops
truncatus) have been known to attack and kill harbor porpoise in the
United Kingdom (Ross and Wilson 1996) and Central California
29
(Cotter et al. 2012). It is assumed that dolphins consider porpoise to be
competitors for a limited food supply, though an alternative hypothesis
is that bottlenose populations with a tendency towards infanticide
confuse harbor porpoise with juvenile dolphins (Patterson et al. 1998).
Similarly, fish eating southern resident killer whales have been known
to occasionally chase harbor porpoise with at least one case that led to
the death of the porpoise. The killer whales left the body without
consuming it, allowing its recovery by researchers (R. W. Baird,
personal communication, May 7, 2014).
2.2.2 Fisheries interactions
Human fisheries are thought to be the greatest anthropogenic
threat to harbor porpoise populations worldwide due to entanglement
with fishing gear (bycatch). In the Eastern North Pacific Ocean,
documentation of harbor porpoise caught in fisheries include takes in
halibut setnets in California, salmon gillnets in the Pacific Northwest
and a variety of fisheries in Alaska (Jefferson and Curry 1994). Tribal
and commercial salmon fisheries are the most common net fisheries in
the Puget Sound. No published information on bycatch of harbor
porpoise is available due to the recent return of this species to this
area. A study of bycatch in salmon fisheries in nearby southern British
Columbia indicated an average annual mortality of 80 harbor porpoise
30
from this fishery alone (Hall et al. 2002). Gillnets represent the
greatest threat to harbor porpoise throughout their range and are
indicated in their decline in many areas (Jefferson and Curry 1994,
Stenson 2002).
It was originally thought that porpoise could not acoustically
detect gillnets and that most porpoise swimming in close proximity to
the nets were in danger of entanglement. Recent studies have shown
that porpoise are able to detect the presence and location of nets from
greater than ten meters, with the harbor porpoise successfully
avoiding contact with the net (Nielsen et al. 2012). The results suggest
that bycatch is a problem of attention shifts on the part of the harbor
porpoise or acoustic masking of the echolocation returns from the
gillnets. New materials and designs are being studied to increase the
acoustic signature of gillnets so that marine mammals will find it
easier to avoid entanglement. Gillnets made out of denser and stiffer
material are being tested to reduce entanglement risk for non-target
animals that make contact with the net (Larsen et al. 2002, Mooney et
al. 2004, 2007). Research has been conducted into the use of acoustic
deterrents, commonly called “pingers”, to alert harbor porpoise to the
presence of the nets, or to alternatively scare them away (Culik et al.
2001, Johnston 2002, Carlström et al. 2009). While they show some
31
initial success in reducing porpoise bycatch, habituation appears to be
a problem requiring cycling of the sounds (Cox et al. 2001).
Lost or abandoned fishing gear, including gillnets, have been
shown to continually kill wildlife for years or decades (Good et al. 2009,
Gilardi et al. 2010). As with many other fishery areas around the
world, the Puget Sound is subject to this issue of “ghost nets”.
Cooperation from numerous government organizations have resulted
in the removal of over 1,200 nets from the Puget Sound and Northwest
Straits (Good et al. 2009). According to the Northwest Straits
Initiative, between 2002 and 2013, 4,605 derelict fishing nets have
been removed from Puget Sound, with the entangled remains of at
least 270 species including 5 harbor porpoise (Northwest Straits
Initiative 2013).
Fisheries can also impact the availability of food through
competition for some of the common forage fish including herring,
sardines and anchovies. In the North Pacific, few of these fisheries
have been brought close to the point of collapse, so this is not likely to
be a factor affecting local porpoise populations (Trites et al. 1997).
2.2.3 Noise pollution
There are many anthropogenic sources of noise pollution in the
marine environment that have been shown to affect harbor porpoise.
32
Extremely high energy sound sources, such as airguns that are used
for seismic explorations, can cause avoidance behavior in harbor
porpoise at distances >70 km from the source (Bain and Williams 2006)
and pile driving can cause reactions >20 km from the source (Tougaard
et al. 2009). Lower energy sources, such as acoustic harassment
devices (AHD) intended to keep seals away from salmon farms, also
exclude harbor porpoise from the vicinity. A study in the Bay of Fundy
found that porpoise maintained a distance of at least 645 m from the
AHD equipped salmon pens (Johnston 2002). Noise from vessel traffic
has also been shown to cause harbor porpoise to leave the area, though
their behavior can vary with the vessel type and the habituation of the
local population (Evans et al. 1994).
The construction and operation of offshore tidal and wind energy
installations are of concern in areas of high harbor porpoise density.
Currently, offshore wind farms need to be located in shallow coastal
waters in order to secure the tower structures by sinking support
pilings into the seabed. Tidal energy projects will also be located in
shallow coastal waters, in locations with high rates of tidal flow such
as bays and inlets. The areas for both of these types of energy projects
coincide with the preferred habitat of harbor porpoise.
There are concerns about the impact of loud, repetitive, long
term sound impacts on marine mammals during both the construction
33
and operational phases of these projects. Pile-driving during
construction produces high energy broadband sounds that are loud
enough to cause porpoise to leave the area, with behavioral responses
detected at distances greater than 20 km from the source (Tougaard et
al. 2009). During the energy-producing operational phase, repetitive
lower frequency sounds can influence behavior. A study of acoustic
masking using simulated low-frequency wind turbine sounds on
captive harbor porpoise suggests that a localized effect on echolocation
ability does occur (Lucke et al. 2007). Additional research is necessary
to determine the impact of energy infrastructure projects upon wild
porpoise populations.
There are no current plans for marine based energy production
in the South Puget Sound. There is a pilot tidal flow generation project
that will be installed in Admiralty Inlet in the North Puget Sound
(Snohomish County PUD 2014). This equipment has been designed to
be installed without pile driving or any other percussive sound and is
intended to assess the viability of tidal driven electrical generation
from an environmental and technological standpoint.
2.2.4 Hunting
Hunting of harbor porpoise has historically occurred in many
areas throughout their range, including the Puget Sound. Most
34
porpoise fisheries are now closed, with the notable exception of
Greenland, where Inuit porpoise hunting continues, with current
catches in excess of 1000 animals per year (Hammond et al. 2008b).
Hunting of harbor porpoise is thought to have contributed to the
population decline in the Black Sea. Though it was outlawed in 1983,
illegal hunting in the Black Sea continued through 1991 (Birkun Jr.
and Frantzis 2008). Historically, some Washington tribes hunted
harbor porpoise for food. Fisherman and hunters have been known to
shoot porpoise for sport or because they were seen as pests in fish nets
and traps (Scheffer and Slipp 1948). Harbor porpoise are now protected
by the MMPA, which has made hunting of harbor porpoise in the
United States illegal.
2.2.5 Pollution
Marine mammals are exposed to many anthropogenic sourced
organic and inorganic pollutants, with coastal species, such as the
harbor porpoise, having even higher exposure rates. Many of these
pollutants are lipophilic, bioaccumulating in the fat through each
trophic level. Harbor porpoise and other cetaceans are predators near
the top of the food chain and have extensive fat stores in the form of
blubber, where high concentrations of many pollutants have been
detected (Bossart 2011).
35
Heavy metal pollution is often highest in coastal areas with
runoff from regions that conduct industrial or mining activities
(Ruilian et al. 2008). Cetaceans seem to be able to manage exposure to
many of these metal pollutants, such as copper (Cu) and zinc (Zn)
without apparent problem. All age classes of stranded animals had
similar concentrations of these metals, suggesting that they are able to
clear excess quantities from their systems. Porpoise tend to
accumulate mercury (Hg) in their livers over time, often reaching
much higher Hg levels than commonly found in terrestrial mammals
at a similar trophic level (Law et al. 1991). A study of stranded harbor
porpoise in Britain and Wales examined the livers of stranded animals.
The livers of animals that died from infectious agents had higher levels
of Hg than those killed by physical trauma, suggesting that the Hg
may have weakened the immune systems (Bennett et al. 2001).
Organochlorides are a class of chemical compounds that includes
some pesticides such as dichlorodiphenyltrichloroethane (DDT) and the
electrical insulator polychlorinated biphenyl (PCB), which are known
to cause reproductive issues in a variety of species. Production and use
of DDT and PCB have been banned since the 1970s and 1980s in most
industrialized countries. These persistent chemicals have remained in
the environment and continue to be detected, though at reduced levels.
PCBs in particular have been implicated in affecting both harbor
36
porpoise immunity and reproductive success (Calambokidis et al. 1985,
Berggrena et al. 1999, Hall et al. 2006). Polybrominated diphenyl
ethers (PBDE) are a group of organobromide compounds that have
been used since the 1970s as a fire retardant in a wide variety of
materials including building materials and household goods. Several
populations of harbor porpoise have been found to have high levels of
PBDE in their blubber. This chemical compound has a similar
structure to PCB, and is thought to have comparable impacts on
reproduction, the immune system and the endocrine system
(Ikonomou et al. 2002, Law et al. 2002, Weijs et al. 2009)
Polycyclic aromatic hydrocarbons (PAH) are released into the
environment through petroleum spills or seeps, as well as the burning
of wood, coal and petroleum products. High levels of PAH have been
associated with a variety of toxic responses in aquatic wildlife,
including harbor porpoise. PAH is known to damage the immune
system and have possible carcinogenic effects (Law and Whinnett
1992, Fair et al. 2010).
Many of these pollutants have been detected in marine life
within the Puget Sound, including the heavy metals lead and mercury;
PCB; DDT and its metabolite DDE; and PAH (West 1997). It has been
suggested that high pollution levels, specifically PCBs, might have
caused lower reproduction rates in harbor porpoise in the South Puget
37
Sound, contributing to their extirpation (Calambokidis et al. 1985).
Studies of harbor seals (Phoca vitulina), which are at a similar trophic
level to harbor porpoise, revealed much higher concentrations of PCBs
in the blubber of seals within the South Puget Sound as compared to
animals in the North Puget Sound or nearby Hood Canal
(Calambokidis et al. 1984).
2.2.6 Climate change
Climate change is expected to have a major impact on many
marine mammal species, including the harbor porpoise, though the
extent of the impact remains unknown. Several papers address the
potential issues, but only a few studies have been conducted to date.
Increased water temperature is a predicted impact of climate change
and is likely to affect harbor porpoise prey distribution (Learmonth et
al. 2006). The west coast of Greenland is now experiencing longer icefree periods. A recent study examined the improvement of harbor
porpoise body condition as they are able to remain in prime foraging
areas for a longer period of time (Heide-Jørgensen et al. 2011). Climate
change has also been cited as a probable cause of increased starvation
among harbor porpoise in Scottish waters. It has been suggested that
climate change has reduced sandeel populations, which are a major
component of the harbor porpoise diet (MacLeod et al. 2007b). Some
38
controversy exists in regards to the small sample size used in this
study and a potential bias in the data, however it is one of the few
studies on this topic and suggests areas for future research (MacLeod
et al. 2007a, Thompson et al. 2007).
Ocean acidification is another aspect of climate change that
could affect harbor porpoise. It has been suggested that squid, a
common prey item for many harbor porpoise populations, will be
especially vulnerable to acidification, because lower blood pH levels
will reduce the ability to transport oxygen to their tissues.
Acidification will also impact the carbonate shells of some species of
phytoplankton, which can have effects up through the food chain
(Simmonds and Isaac 2007).
There has been a great deal of study regarding the impacts of
climate change on the lower trophic levels of marine ecosystems.
However, little research has been conducted on the potential impacts
on marine mammals.
2.2.7 Threats conclusion
Numerous natural and anthropogenic threats have been
implicated in the decline of harbor porpoise populations in various
parts of the world. While a few populations have been increasing in
recent years, such as those in the inland waters of Washington State,
39
many populations are in decline and the status of others remains
unknown. Monitoring of populations, through abundance estimates,
fisheries observer programs and necropsies on stranded animals,
provide important information on the impacts of these threats to local
populations.
2.3 Visual monitoring
Visual monitoring of harbor porpoise can be conducted using a
variety of methods and platforms. This technique is essential to the
estimation of population abundance, as acoustic methods have not
advanced to the point where they can match the precision of a visual
survey. Boat or ship based surveys can be used to estimate distribution
and abundance through either line transect methods or using photo-ID
for mark-recapture studies. Aerial surveys are conducted using line
transect methods to estimate abundance and determine distribution.
Shore based survey methods use point transect methods to estimate
abundance, though they are more commonly used to conduct focal
follows, which track a group of animals to study their behavior over
time and to monitor animal presence. Visual methods excel at
determining group size, detecting the presence of calves, and can cover
a far greater area than can be achieved with acoustic methods (Evans
and Chappell 1994).
40
While visual methods have a long history and remain the best
method for many aspects of research, they are fraught with many
difficulties when studying cetaceans due to the large percentage of
time that the animals spend beneath the surface. These complications
are more pronounced when studying a small, shy animal such as the
harbor porpoise with their dark coloration and quick surface action.
During their normal foraging behavior, a harbor porpoise surface
series consists of a few surface breaths when their rolling back is only
visible for about two seconds each time, followed by a foraging dive
that can last for several minutes (Scheffer and Slipp 1948). Even
during ideal conditions, visual methods compare poorly to acoustic
methods for detecting animal presence, detecting animals before they
are detected acoustically about 15% of the time, and they completely
miss animals that are detected by acoustics about half the time (Evans
and Chappell 1994). Sightings from vessels, shore or aerial surveys are
also limited to daytime hours and times with low winds and clear
weather conditions, which limits their usefulness during large portions
of the year and (Evans and Chappell 1994, Palka 1996).
Untrained observers can confuse harbor porpoise and Dall’s
porpoise, which is the only other common black small cetacean in the
Puget Sound. The Dall’s porpoise has a larger dorsal fin that is
slightly falcate (sickle shaped) with a more vertical trailing edge that
41
often has some white pigmentation, and they also have a bump on
their back near their tail that leads to a different surface motion than
that of the harbor porpoise (Figure 2-4). Dall’s porpoise also have a
distinct separation of the white and black coloration on their flank,
instead of the gradation seen on the harbor porpoise (Gaskin et al.
1974, Jefferson 1988). Dall’s are much more acrobatic, demonstrate
different surfacing behavior and are more likely to be drawn towards
human activity than harbor porpoise (Jefferson 2009). Though rare,
several dolphin species have been sighted in the Puget Sound.
Dolphins are easily differentiated from harbor porpoise by their larger
falcate dorsal fins, different coloration and much showier surface
behavior.
42
2.3.1 Vessel surveys
Ship and small boat line or strip transect surveys are often used
to produce abundance estimates for a number of cetacean species,
including harbor porpoise (Barlow 1988). During line transect surveys,
a vessel travels along a predetermined route called a transect line.
Observers count animals that are spotted along that line and within
narrow strips off to each side. A probability function is then used to
determine species density in the survey area (Buckland et al. 2001).
Many surveys are conducted for multiple species and will continue
even during higher sea states, as a result, sightings of small cetaceans
can be missed during inclement weather (Barlow and Forney 2007). A
Figure 2-4: Dall's porpoise have distinctly different body shape and coloration,
allowing for easy differentiation by observers. (Copyright Uko Gorter)
43
study of observation teams on a ship transect survey showed that
researchers were found to miss porpoise on the track line
approximately 22% of the time (Barlow 1988). The presence of the
survey ship may cause harbor porpoise to leave the vicinity skewing
the population density estimates by a factor of 1.4 to 2.7 (Palka and
Hammond 2001). Even with these limitations, it is believed that a
reasonable abundance estimate can be obtained through the use of
vessel surveys.
Photo-identification (photo-ID) is another method used to
produce abundance estimates and learn about the social structure of a
population using a mark-recapture framework. Small boats are used to
collect photographs of animals and record information about their
location and behavior. The photo-ID process involves the comparison of
these photographs to image catalogs of known individuals. Images of
specific parts of animals that are likely to have distinctive markings,
such as dorsal fins or flukes, are used for comparison. The first time an
animal is sighted, it’s considered a “mark” and the individual is
entered into the catalog and assigned an identification number. If the
individual has been previously sighted, it’s considered a “recapture”
and the sighting information is added to the animal’s history enabling
long term study of population abundance and social structure
(Hammond et al. 1990). Due to the difficulty in consistently obtaining
44
high quality, high definition photographs of harbor porpoise (Flaherty
and Stark 1982, Gaskin and Watson 1985), few attempts have been
made to produce photo-ID catalogs of the species.
2.3.2 Shore surveys
2.3.2.1 Theodolite tracking
Theodolites are surveying instruments that have been
repurposed for ecological studies. Researchers often use theodolites as
a non-invasive method for tracking marine mammals, boat traffic and
other anthropogenic activity (Würsig et al. 1991, Emery et al. 1993,
Harzen 2002). They are used to obtain accurate position data, or
“fixes”, on animals and other objects moving through a study area. An
observer looks through a telescope, usually 30X power, and places the
target animal in the crosshairs of the scope. High-precision
measurements of the horizontal and vertical angles are recorded either
on paper or through a connection directly to a computer.
Given the latitude, longitude and elevation of the theodolite, the
horizontal and vertical angles can then be used to calculate the
position of the target animal to within a few meters by using basic
trigonometry (Gailey and Ortega-Ortiz 2002) (Frankel et al. 2009).
Theodolites have a range of several kilometers depending on weather
45
conditions and target size. In favorable weather conditions, it is
possible to track larger marine mammals, such as baleen whales, out
to distances of 10 km (Würsig et al. 1985). For smaller animals, such
as dolphins and porpoises, the useful range is reduced to about 5 km
(Würsig et al. 1991). Refraction can skew the determination of the fix
position near the horizon, however, this should not be an issue in the
present study (Kinzey and Gerrodette 2003).
Electronic digital theodolites connect to a computer via a data
cable so that angular measurements can be automatically entered into
tracking software. This enables researchers to quickly record data
points with no transcription errors. Software packages use input from
a digital theodolite to record and calculate the fix position as well as
information about species, group size, behavior, weather and other
factors (Gailey and Ortega-Ortiz 2001).
2.4 Passive acoustic monitoring
Passive acoustic monitoring (PAM) involves using underwater
equipment to detect sounds in the marine environment. In the
biological sciences, it is primarily used by researchers to study the
ecology of marine animals, though it also reveals information about
anthropogenic activity. PAM devices can be deployed for long periods of
time, enabling researchers to monitor animal sounds continuously
46
throughout the day and in all weather conditions. These devices
capture acoustic evidence of the presence and behavior of marine
mammals, with minimal impact on animal behavior (Zimmer 2011).
Certain animal behaviors that are difficult to detect using surface
observation, such as communication or echolocation, can be monitored
through the use of acoustic techniques (Villadsgaard et al. 2007).
While there are many advantages to the use of PAM, it also has
its limitations. In order to be detected, animals must be producing
sounds of sufficient strength to be discernible from the background
noise. Depending on the frequency and directionality of the sound, the
range of detection could be hundreds of kilometers in the case of blue
whales (Širović et al. 2007), or just a few hundred meters for species
such as porpoises which focus their high frequency sounds in a narrow
beam (Kyhn et al. 2012). Using current technology, it is difficult to
produce abundance estimates using PAM. Unless individual animals of
a species make signature sounds, it is not possible to identify specific
animals; therefore it is difficult to estimate group size or study
population structures (Marques et al. 2013).
There are a variety of PAM devices ranging in size from large
permanently deployed arrays down to small portable units that weight
less than one kilogram and are easily carried to remote sites (SousaLima et al. 2013). They can be simple hydrophones which allow
47
realtime monitoring of sounds, acoustic recorders which store the data
for later analysis, or they can be devices that conduct some onboard
processing before the data is recorded. These versatile devices allow for
a broad range of deployment options, including monitoring from shore
(Thomas and Fisher 1986) or ship (Nielsen and Møhl 2006, Borchers
and Burt 2007); mooring devices within the water column (Sousa-Lima
et al. 2013); or attaching equipment to autonomous robot gliders
(Wiggins et al. 2010, Klinck et al. 2012) or including small devices in
animal tags (Sousa-Lima et al. 2013).
2.4.1 Hydrophone
Hydrophones are sealed underwater microphones and are the
most basic form of acoustic monitoring device (Gordon and Tyack
2002). All acoustic monitoring devices include some kind of hydrophone
to capture sound from the marine environment. The output from the
hydrophone can be used for real time monitoring, recorded for later
analysis or used as a source of input for more advanced PAM devices.
Shore based stations can be connected to fixed hydrophones
through cable or radio links providing real time monitoring of the
audio stream (Thomas and Fisher 1986, Marques et al. 2013). These
sites are often permanent installations that utilize undersea cables to
power the equipment and provide a data link back to shore. By using
48
multiple hydrophones deployed in an array, many sites are able to
track individual animals as they move thorough the study area
(Marques et al. 2013). Shore based studies may also include the
deployment of smaller temporary hydrophones which require less
infrastructure or portable units used by researchers for short term
sampling (Gordon and Tyack 2002).
Ships and small boats are versatile platforms for the use of
hydrophones. Keel mounted hydrophones are permanently secured to
the ship, which allows for acoustic sampling without having to deploy
additional equipment and can be used even in rough sea states
(Nielsen and Møhl 2006, Borchers and Burt 2007). Hydrophones towed
behind ships, often in arrays, position the equipment away from the
noise of the ship. This enables research teams to monitor for acoustic
evidence of marine mammals while running transect lines (Thomas
and Fisher 1986, Borchers and Burt 2007, Sveegaard et al. 2011a,
Marques et al. 2013). Small dipping hydrophones can be deployed from
vessels of any size to quickly sample for the presence of animals
(Thomas and Fisher 1986, Gordon and Tyack 2002). Ship based
transect surveys are key tools in determining areas of high porpoise
density, however, the presence of a vessel may cause changes to
porpoise behavior which could influence the results (Sveegaard et al.
2011a).
49
Unlike ship and shore based systems, autonomous hydrophones
are self-contained and are deployed for extended periods, recording
their data for later analysis. Other than during times of deployment,
these devices substantially reduce concerns regarding the impact of the
research vessel on animal behavior. Autonomous hydrophones are
ideal for deployment where there is limited ability to service the
devices, such as in remote locations or for projects that have limited
funding for ship time and equipment. Some devices are designed to be
deployed at the surface, in the form of freely floating buoys, and others
can be moored throughout the water column using a variety of ground
tackle options (Sousa-Lima et al. 2013).
The extended deployment times of autonomous hydrophones
lead to large demands on data storage. In an attempt to conserve
storage capacity, researchers adopted the idea of duty-cycling in which
the device records only a few minutes out of each hour (Sousa-Lima et
al. 2013). Early versions of autonomous hydrophones were often
assembled by the researchers themselves and consisted of tape
recorders triggered by mechanical timers that activated the device
according to the duty-cycle (Thomas and Fisher 1986). Innovations in
batteries, digital recording methods and storage have greatly increased
the possible deployment periods, though duty-cycling is still commonly
50
used as a method to increase deployment time (Sousa-Lima et al.
2013).
Other autonomous devices can include hydrophones in their
design to record acoustic activity. Short-term recording tags can be
attached to marine mammals to track their movement and dive
patterns for periods ranging for a few hours to a few days (Sousa-Lima
et al. 2013). Tags such as the D-tag and Acousonde include a tiny
hydrophone to provide researchers with audio that accompanies the
movement data. In additional to vocalizations, it has been found that
the movements of the fluke while swimming can be identified (SousaLima et al. 2013).
Seagliders and Wave Gliders are small, low power, autonomous
robot vehicles that can be equipped with recording hydrophones.
These vehicles are often deployed for several months on low-cost, low
impact transect surveys. Unlike other autonomous devices, gliders are
able to transmit data through satellite links back to the researchers in
near real time. These mobile platforms allow for economical
monitoring of remote sites without the expenses related to using a ship
while reducing disturbance to the study population (Wiggins et al.
2010, Klinck et al. 2012, Marques et al. 2013).
51
An issue common to all these platforms is that the high
frequency sounds produced by harbor porpoise requires sampling at
greater than 400 kHz, which will generate more than 30 TB of acoustic
data per year (Tregenza 2013). Due to these high volumes of data,
recording hydrophones are better suited to short-term experiments
rather than long-term monitoring of harbor porpoise. They are most
commonly used to study aspects of sound production, such as SPL, ICI,
and directionality of their sonar (Au et al. 1999, Kastelein et al. 2005).
Ultrasonic click detectors, such as the C-POD, provide a solution to
data storage issues during long-term deployments.
2.4.2 The POD, T-POD and C-POD ultrasonic sound detectors
The desire to economically monitor for the presence of dolphins
and porpoises on long-term deployments, led to the development of
ultrasonic click detectors. The audio input from the hydrophone is preprocessed before storage; only metadata about certain tonal sounds is
recorded. Since the entire waveform is not recorded, this method
allows for much longer deployment times with moderate levels of
onboard storage.
The first iteration of the POD (POrpoise Detector) was
developed in the in the late 1990s and was simply a data logger that
recorded the number of clicks detected each minute (Tregenza 1998,
52
Baines et al. 1999). The POD worked by comparing sound levels in the
band used by porpoise (120-150 kHz) to sound levels in three other
bands below 100 kHz. By comparing these bands, it was possible to
determine whether it was a broadband sound or one that was likely to
have been produced by a porpoise. Porpoise clicks were only counted
when the energy in the upper band was greater than in the lower
bands, thus limiting false positives (false detections) (Chelonia Ltd.
2013a). This design was based on previous work that used automatic
click detectors attached to a towed hydrophone array (Chappell et al.
1996). Only aggregate counts of clicks detected per second were
recorded by this early device, no data about individual clicks or click
trains were recorded. In addition to porpoise data, noise level, water
temperature and dolphin click counts, were also logged. The peak
energy within the lower “dolphin” band was used to discriminate
between dolphin and porpoise clicks, which allowed the device to be
used for dolphin research as well. This first generation POD provided
aggregate counts of detected clicks, however, without additional
information on individual clicks and click trains the ability to further
filter data to reduce false detections was limited (Chelonia Ltd. 2013a).
The T-POD (Timing POrpoise Detector, Chelonia Ltd., U.K.) was
the next generation porpoise detector, which changed from recording
the clicks detected per second to recording the time and duration of
53
each click within the filter range. Clicks were still identified by
comparing the energy level within the click band to the energy level in
the lower-bands, and only recording metadata about the target sounds.
Each stored click received a timestamp which allowed post-processing
software on a computer to be used to identify characteristic click
trains. Clicks that did not match the duration of an echolocation click
or spurious clicks that were not part of a train were discarded. The
T-POD advanced through five versions before the design was retired in
2008 due to difficulty and expense of obtaining some of the older
electronics as the design neared a decade in service (Chelonia Ltd.
2013a, Dähne et al. 2013).
The most recent design is the C-POD (Cetacean POD, Chelonia
Ltd., U.K.), which was used in the current study. Unlike the earlier
PODs, which performed much of the detection through the use of
analog filters, the C-POD detection logic is an all-digital design. While
the digital design allows for more complex analysis of the detected
sounds, it retains the core concept of using filters to detect tones in the
NBHF range that are at a higher energy level than tones in the lower
bands and background noise. The C-POD differs by its ability to
digitally separate the different tone frequencies, rather than being
limited to the predefined frequency bands in the analog filters. In
addition to the timestamp and duration of the clicks recorded by the
54
T-POD, the digital design enabled the C-POD to record much more
information about each click, including the dominant frequency, the
bandwidth of the sound, how the sound changes over time and limited
information about the sound envelope. This additional metadata
improves the ability of the post-processing software to identify the
sound source (Chelonia Ltd. 2013a). The broad ultrasonic bandwidth of
the C-POD, from 20 kHz through 160 kHz is able to detect echolocation
clicks produced by all odontocetes other than sperm whales, which
have peak echolocation frequencies of 400 Hz to 2 kHz in the males,
and 1.2 kHz to 3 kHz in the females, which is below the 20kHz
minimum frequency of the C-POD (Goold et al. 2000).
Each generation of POD has improved significantly upon the
previous generation in the ability to detect and filter data. The current
C-POD has been used successfully in many studies because of its
ability to record data through long-term deployments in a wide variety
of conditions. The limitations of recording only metadata restricts
C-POD usefulness in animal behavior studies intended to identify
specific animals in a group, such as work with dolphin signature
whistles, or for other in-depth acoustic analysis. Recording
hydrophones are still the preferred method for gathering data for these
applications.
55
2.4.2.1 Calibration
POD devices can vary in sensitivity, therefore when multiple
click detectors are used, it is important to calibrate the devices so that
data can be compared accurately. The differences can be attributed to
variability of piezoelectric transducers used to pick up the sound, the
plastics used in the end caps and final assembly. When different
version of detectors are used in the same study, these differences in
sensitivity are even greater as many parts are replaced and upgraded
with each revision (Simon et al. 2010, Chelonia Ltd. 2013a).
Techniques that are commonly used to calibrate click detectors
include tank tests and field calibration. Tank tests are conducted by
placing the POD in small tanks containing sound sources and several
hydrophones to monitor actual sound levels (Kyhn et al. 2008). The
tanks are designed to reduce the impact of echoes on the device being
tested (Dähne et al. 2013). Field calibration tests are conducted by
deploying several devices in close proximity to each other so they are in
the same sound environment. The SPL received by the different
devices are then compared to determine their relative sensitivity
(Kyhn et al. 2008).
56
As part of their quality assurance procedures, the manufacturer
tank tests every C-POD, rotating the device through 360 degrees, and
recording the SPL received. The minimum threshold for recording the
click is set to a median value, where 50% of the positions during the
rotation could detect the test tone. A calibration file is created for each
C-POD and this data is used by the CPOD.exe software to account for
differing levels of sensitivity (Chelonia Ltd. 2013b).
2.4.2.2 Range of detection
Determining the range of detection of an acoustic monitoring
device is an important factor for many aspects of acoustic research
such as attempts to use acoustic methods to estimate animal
abundance. The high frequencies used by harbor porpoise are quickly
attenuated in water, limiting their range of detection when compared
to species that vocalize at lower frequencies. Clausen et al. (2012)
modeled the maximum range of harbor porpoise communication based
on known parameters of their sound production and hearing. Due to
the highly directional nature of their click production and hearing, it
was determined that their maximum communication range is 1200 m
when facing each other though it drops to 200 m when facing in
opposite directions. The reception distance is also affected by natural
and anthropogenically produced ambient noise, bathymetry,
57
temperature gradients and obstacles in the sound path. The sensitivity
of the C-POD is much lower than the auditory system of the harbor
porpoise, especially when the porpoise are on-axis to the sound source
(Clausen et al. 2012, Dähne et al. 2013). Therefore, the range of
detection of the C-POD will be considerably reduced as compared to
the above estimates of harbor porpoise communication range.
There are a limited number of studies that have attempted to
determine the range of detection of various POD devices. One study
used nine T-PODs in an attempt to develop a method of estimating
abundance and found considerable variation in the sensitivity between
devices. It was found that harbor porpoise had to be within 22 to 104 m
to ensure detection (Kyhn et al. 2012). It is thought that the maximum
on-axis range for the C-POD is 300-400 m (Hardy et al. 2012).
Studies of this nature require visual observers to use a
theodolite to map porpoise transits as they pass the POD. This
requires a considerable amount of additional effort in the field and few
study sites provide appropriate, accessible shore based locations
necessary for this work.
2.4.2.3 Detection error rate
As with all studies, the error rate is of concern when using any
sort of automated method to process data. The click train classifier
58
used by the CPOD.EXE software is designed to be very conservative,
producing few false-positive (false detection) results, which leads to
having a high rate of false-negatives (missed detections). The rate of
false detections is dependent on the background noise level at each
individual site. Settings can be changed in the software to limit
detections based on the quality of the click trains. Researchers can
visually check several hundred click trains to determine the false
detection rate caused by background noise and adjust the quality
settings in the software (Tregenza 2013).
Missed detections are common when using the click train
classifier, both for NBHF species and dolphins. Entire encounters can
be missed due to a lack of any click trains meeting the conservative
requirements of the classifier, especially at the higher quality settings,
though it is unknown what percentage are actually missed. At present,
the only way to reduce the number of missed detections beyond setting
the output to accept all qualities of detections, is to visually review
potential click trains, marking those that appear to be valid. An
analysis of both false detections and missed detections were conducted
in a study of dolphins at Camp Lejeune (Read et al. 2012), by deploying
C-PODs along with DMON autonomous digital recorders (Woods Hole
Oceanographic Institution, Woods Hole, MA, USA). The acoustic
recordings were analyzed by a trained technician and the results were
59
compared to the results produced by the C-POD and the CPOD.exe
software. When used to detect dolphins, the C-POD was found to have
a between 0.53% and 3.67% false detection rate, and a 50.49% to
82.74% rate of missed detections.
2.4.2.4 Abundance estimates
Due to the expense of conducting traditional aerial or ship based
line transect surveys and mark-recapture studies, there is a great deal
of interest in using acoustic methods to produce abundance estimates.
This is a rapidly advancing field within cetacean ecology, but many
difficulties remain (Marques et al. 2009, 2013). A study using T-PODs,
spaced in an array, proved successful in making abundance estimates
for the study area, with the caveats that the estimates required
theodolite tracking from shore to develop the probability of detection
function for each T-POD as the porpoise swam through the site, and to
also determine the average group size (Kyhn et al. 2012). Kyhn et al.
made it clear that their estimation was only valid within the detection
area of the array of the T-PODs, and estimates were only considered
accurate to an order of magnitude (i.e.10(log10 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒±0.5) ). For example,
a density estimate of 67 animals using acoustic data could represent
between 21 and 212 individuals. While research will continue to
60
advance the use of acoustic monitoring to estimate absolute
abundance, the lack of precision limits its usefulness at this time.
A more established method of analysis involves the monitoring
of changes of relative abundance over time in a population. This
method is being used as part of Mexico’s vaquita (Phocoena sinus)
recovery plan. Approximately 50 C-PODs have been deployed within
the Vaquita Refuge, an area of 1271 km2 within their small home
range at the northern end of the Sea of Cortez (Rojas-Bracho et al.
2010). Vaquita are a close relative of the harbor porpoise, and are
considered the most endangered extant species of marine mammal
with fewer than 200 individuals. Any decline in the population needs
to be quickly detected and addressed. Though the Mexican government
has implemented many measures to protect the vaquita, including
buying back licenses from fishermen in the area, and banning gillnet
fishing within the reserve, the acoustic data suggests that the
population is continuing to decline.
An attempt to monitor the critically endangered Baltic Sea
harbor porpoise population is using similar methods to those used in
the vaquita protect. The SAMBAH (Static Acoustic Monitoring of the
Baltic Sea Harbour porpoise, http://www.sambah.org/) project has
deployed 300 C-PODs throughout the shallower areas of the Baltic Sea
over a two year period, from May 2011 to May 2013, with reports due
61
at the end of 2014. The goals of the study are to estimate porpoise
population density, abundance and distribution throughout the Baltic
Sea; determine habitat preferences; locate frequently used areas; and
identify areas at high risk of adverse human-porpoise interactions.
2.5 Environmental analysis
Implementing conservation management strategies requires an
understanding of how animals use their environment, and how
environmental factors affect their behaviors. It is believed that most
harbor porpoise movements are based on their constant search for food,
therefore most studies examine factors that are known to influence
harbor porpoise prey species. Seasonal migrations of prey such as
herring or squid, or concentrations of other prey can lead to seasonal
changes in distribution of harbor porpoise (Gaskin and Watson 1985,
Gannon et al. 1998). While harbor porpoise don’t feed directly on
zooplankton, the diel vertical migration of zooplankton can affect the
availability of prey species that feed on the zooplankton (Alldredge and
King 1985, Ohman 1990). Similarly, tidal changes can cause fronts to
develop downstream of obstructions, which will concentrate plankton
and draw harbor porpoise prey species (Johnston et al. 2005). There
are other environmental factors that could influence harbor porpoise
behavior, yet many of these require measurements that are outside the
realm of most porpoise research studies.
62
2.5.1 Seasonal patterns
Understanding seasonal patterns of harbor porpoise abundance
and distribution is important, not only to inform biological research,
but also from conservation and management perspectives. Harbor
porpoise are a species of concern throughout much of their range, with
many populations in decline. The listing of the harbor porpoise as a
species of conservation importance under the EU Habitats Directive
has led to numerous studies that considered seasonal density as an
important factor when assessing sites for marine protected areas
(Embling et al. 2010).
Knowledge of harbor porpoise seasonality helps to inform
management decisions regarding the regulation of human activities,
and to strike a balance that allows for the greatest level of protection
with the least economic impact. Gilles et al. (2011) produced a
predictive model of the German Bight as part of a program to assess
proposed wind farm locations. It was noted that seasonal population
shifts can occur and if decisions were based solely on summer surveys,
other seasonal hotspots of high porpoise density would be missed.
Higher seasonal abundance can also lead to either closures in certain
fisheries, or the requirement that nets use acoustic pingers to alert
porpoise to their presence (Trippel et al. 1999).
63
As a species, harbor porpoise are not known for conducting long
seasonal migrations, though many populations do display seasonal
movements and there are a few know instances of longer migrations to
follow prey. These movement patterns can alter the distribution of the
porpoise population, with animals favoring different locations
throughout the course of a year. Seasonality of populations in northern
waters is often related to sea ice. When sea ice melts in the spring,
porpoise will move into formerly frozen areas in search of nutrient-rich
food, such as herring in the Baltic Sea (Koschinski 2001) or cod off the
west coast of Greenland (Heide-Jørgensen et al. 2011). When the ice
returns during fall and winter months, porpoise leave the area for open
waters.
When ice is not a driver of porpoise movement, it is assumed
that seasonal changes are due primarily to food availability. Harbor
porpoise along the northeast coast of the United States and Canada
move into areas such as the Gulf of St. Lawrence and the Bay of Fundy
during the summer months to feed on abundant supplies of herring
(Palka et al. 1996, Gannon et al. 1998). During the fall and winter, the
majority of these populations are believed to move onto the continental
shelf in search of other prey, though some animals are occasionally
sighted on their summer feeding grounds year-round (Gaskin and
Watson 1985).
64
Populations such as those in the inland waters of Washington
State are known to maintain a year-round presence throughout their
range, with areas of higher density varying through the seasons
(Flaherty and Stark 1982). Much of this variability can be explained by
changes in food supply, such as herring or squid moving into a region
to spawn, or juvenile eelpouts growing to a size where they become
worthwhile prey for the porpoise to pursue (Flaherty and Stark 1982,
Hall 2004, Nichol et al. 2013). Some changes in density can also be
attributed to females showing a preference to calve in certain area
(Lockyer and Kinze 1995, Gilles et al. 2011). It is unknown what the
criteria are for the selection of calving grounds, though availability of
abundant nutrient dense food is thought to play a role (Gilles et al.
2009).
2.5.2 Diel patterns
Patterns of diel behavior in harbor porpoise have been
addressed by several studies that used visual and acoustic methods, as
well as telemetry from tagged animals. Diel patterns have often been
noted; however, there is little consistency to porpoise presence and
behavior patterns between study locations.
PODs have been used in a number of studies that have noted
diel patterns in harbor porpoise detections. North Sea gas platforms
65
show more echolocation activity at night (Todd et al. 2009), as did
aquaculture cage sites in the Bay of Fundy (Haarr et al. 2009). Two CPODs were deployed in Minas Passage in the Bay of Fundy to
determine if a tidal turbine was impacting the presence of porpoise.
One C-POD was deployed at the turbine and the other in a control
location away from the turbine. At both sites, the times of greatest
porpoise presence were in the middle of the night and lowest at midday
(Tollit et al. 2011). T-PODS were deployed at two sites in the Blasket
Islands, Ireland as part of a project to assess locations for a marine
protected area. Unlike most other studies, these results indicated that
porpoise were more active at night at one site and more active during
the day at the other (Berrow et al. 2009). At a proposed wind farm site
in the Netherlands little variation was found for much of the year,
though detections during the winter months were the lowest in the
afternoon and the highest at night. (Brasseur et al. 2004).
Data retrieved from tags can also provide information on
porpoise activity levels. A study using radio tracking tags in the Bay of
Fundy found that porpoise activity was lowest between midnight and
0600 (Read and Gaskin 1985). This is an interesting contrast to the
findings of both Haarr et al. (2009) and Tollit et al. (2011) which
indicated a higher level of porpoise echolocation activity during the
night in the Bay of Fundy. The work by Read and Gaskin (1985) used
66
the best technology available at that the time. It will be interesting to
see the results of studies that are currently underway which use more
advanced tag and satellite technology.
Danish researches used towed hydrophone arrays to assess
possible locations for a marine protected area in the Danish Straits
and the Kattegat, between Denmark and Sweden. There was no
significant difference between daytime and nighttime harbor porpoise
echolocation activity. It was suggested that porpoise may have been
reacting to the survey ship which could influence the results
(Sveegaard et al. 2011a). This was the only study found that used
towed hydrophone arrays and addressed diel activity.
Visual techniques are limited by their nature to daytime hours,
yet some visual studies were still able to determine diel patterns of
abundance. Harbor porpoise were observed most frequently in a nearshore feeding area in Monterey Bay, California, between 0700 and
1000 (Sekiguchi 1995). During strip transects of southwestern Ireland,
active feeding on pelagic fish was observed in the morning and evening
hours (Leopold et al. 1992). Line-transect boat surveys conducted off
the Northern San Juan Islands, Washington, observed more porpoise
in the mornings and evenings, with fewer sightings at midday (RaumSuryan and Harvey 1998). A study in southwest Britain using focalfollows showed no significant difference in porpoise presence
67
throughout the day, though one area showed a diel influence on group
size and distance from shore (Goodwin 2008). Boat and shore based
observations off Vancouver Island, British Columbia revealed no
significant difference in diurnal activity (Hall 2011).
Many diel patterns of marine behavior are thought to be driven
by the vertical migration of zooplankton. This migration is believed to
be a predator avoidance mechanism which consists of an ascent of
zooplankton into the photic zone in the evening, followed by a decent
back to the aphotic zone in the morning (Lampert 1989, Ohman 1990).
This diel migration draws planktivores, which in turn draw higher
level predators, including harbor porpoise (Robertson and Howard
1978, Alldredge and King 1985). During the morning downward
migration of zooplankton, benthic predators have been found to swim
several meters up into the water column to feed (Genin et al. 1988).
Harbor porpoise are likely to seek out areas where their benthic prey
becomes easier to forage in the early morning hours.
2.5.3 Tidal patterns
Currents generated by the changing tides also cause fronts and
eddies, which concentrate plankton and attract animals up through
the food chain (Owen 1981). Fronts are a line of lateral convergent flow
between two bodies of water and are generated by currents moving
68
across underwater obstructions. Eddies are the rotational motions of
water that occur as currents flow past underwater obstructions, or
they can be caused by the Coriolis effect on waters moving in a
horizontal system. Both fronts and eddies have vertical components
that concentrate biology, making it a productive area for predators to
forage.
A study of seasonal variability of harbor porpoise in the German
Bight found that porpoise preferred areas with strong fronts produced
by currents or upwelling (Gilles et al. 2011). A study off the northern
end of Grand Manan Island in the Bay of Fundy found greater
localized concentrations of harbor porpoise during the flood than ebb
tides (Johnston et al. 2005). This study used satellite imagery to
determine the length of the island’s wake during incoming tides, and
then used active acoustic sonar to map the distribution of prey species.
Areas that developed fronts during flood tides were found to have the
greatest concentration of prey species as well as a high density of
harbor porpoise.
2.6 Vessel avoidance
Harbor porpoise are generally perceived to be relatively shy
animals that avoid moving vessels. Their reactions to survey vessels
have been shown to introduce a negative bias which needs to be
69
accounted for when estimating abundance (Carretta et al. 2009). A
variety of methods have been used to determine whether porpoise were
reacting to survey vessels, including observations from the survey
vessel itself (Raum-Suryan 1995), shore based survey observers
(Flaherty and Stark 1982), and even the use of a helicopter to scout
ahead and monitor the behavior of the porpoise as the survey vessel
approached (Barlow 1995). In most cases, the harbor porpoise
exhibited avoidance behavior (Flaherty and Stark 1982, Barlow 1988,
Polacheck and Thorpe 1990, Palka 1995), but a survey using a slower
boat (Raum-Suryan 1995) and one using a kayak (Gaskin and Watson
1985) noticed no avoidance behavior as they approached the porpoise.
When studies have observed porpoise behavior as they relate to
a variety of vessels in an area, the results become much more
interesting. A shore based observation site in the Shetland Islands was
used to monitor harbor porpoise reactions to a variety of vessel types,
revealing differing reactions to different types of vessels (Evans et al.
1994). Encounters with yachts caused porpoise to move towards the
vessel 2/3 of the time, while encounters with speed boats always
caused an avoidance reaction. Reactions to fishing vessels appeared to
be a relatively even mix of all three conditions: no response, positive
and negative reactions. A large ferry that rarely entered the study area
caused avoidance behavior two out of four times, while the small ferry
70
that made up to 8 crossings each day, had only 22% negative reactions,
46% no response, and 32% positive response. Such a low level of
negative reactions to the ferry suggests that the porpoise may have
habituated to the regular ferry traffic.
An extensive survey with shore based and boat based
components was conducted in the San Juan Islands of Washington
State in the early 1980s (Flaherty and Stark 1982). One aspect of the
study examined the reaction of porpoise to vessels, including the
survey boat. A negative correlation was found between the number of
boats in the area and the number of porpoise observed, with no
porpoise observed during times of heavy boat traffic. This study found
that animals changed their behavior during 11 out of 13 encounters
with boats. Unlike Evans et al. (1994), only once did porpoise approach
a vessel, all other encounters involved some sort of avoidance behavior.
The boat that was approached was a slowly drifting or trolling fishing
boat which would have made minimal noise. The researchers
concluded that porpoise behavior is affected by vessels, perhaps
because of the heavy use of small boats and the noise produced by
outboard engines.
Flaherty and Stark note that in addition to fleeing the area,
porpoise may use diving as an avoidance behavior. Observations
included incidents in which porpoise dove for up to 7 minutes before
71
resurfacing. Porpoise sometimes resurfaced quite a distance from the
initial point where the vessel passed, and at other times resurfaced
very close to their original position. During times when the survey boat
was floating quietly among a widely spread group of porpoise, some
would occasionally surface within 10 m of the boat.
While it is clear that harbor porpoise often react to vessels in
their vicinity, their reaction is not always to leave the area. Their
response seems to be related to the boat noise, size, speed and
behavior. Evans et al. (1994) noted possible porpoise habituation to
ferry traffic, which is an important consideration in the Puget Sound
due to the extensive ferry system in Washington State.
72
3 Methods
3.1 Study area
The primary study area is located at the north end of Cormorant
Passage, in the South Puget Sound of Washington State (Figure 3-1).
Separated from the main channel by Ketron Island, the bathymetry of
Cormorant Passage offered near-shore deployment depths of 45 m,
without the steep slope leading down into the deep basins that
characterize this area of the Puget Sound. Visits to the area at the
south end of Steilacoom, Washington revealed several potential
observations locations, down near the waterline as well as on high
Figure 3-1: The study site, located off Steilacoom in the South Puget Sound. The CPOD was located approximately 200 m off Saltar Point, at the north end of
Cormorant Passage in water approximately 45 m deep.
73
banks with unobstructed views. Harbor porpoise were observed on
these visits, and discussions with residents suggested that porpoise are
regularly sighted in the area.
The selected deployment location was approximately 200 m off
the beach at Saltar’s Point Park in Steilacoom, Washington (47.1698°
N, 122.6147° W). The C-POD was deployed from March 10 through
May 31, 2013. A utility lot (47.1697° N, 122.6017° W, 15 m elevation)
with public access to the east of the deployment site was used for
visual surveys from July, 2012 through June, 2013. The utility lot was
closed to public access at the end of June, 2013 so a new observation
site was located in a vacant lot (47.1681° N, 122.6127° W) to the
southeast of the C-POD.
Both high-bank locations were within 300 m of the C-POD
deployment site and allowed tracking and behavioral observations of
harbor porpoise throughout the detection range of the C-POD. The
high-banks allowed for monitoring the greater basin to a visual
detection range of 5 km, covering an area of approximately 30 km2.
3.2 Acoustic Survey Methods
3.2.1 C-POD Click Detectors
Passive acoustic monitoring for harbor porpoise was conducted
using C-PODs, manufactured by Chelonia, Ltd.
74
(http://www.chelonia.co.uk), which are autonomous ultrasonic tonal
click detectors. Unlike regular recording hydrophones, which record
the sounds throughout their range, C-PODs record metadata about
each tonal sound between 20 kHz and 160 kHz, allowing for much
more efficient use of memory, greatly increasing the deployment time.
A sound is considered to be “tonal” when a narrow frequency band of
sound contains more energy than the rest of the broadband range of
sounds. The C-POD is powered by 10 D-cell batteries and click data is
stored on a 4GB secure digital (SD) flash memory card, allowing for
quick data recovery and redeployment in the field (Tregenza 2013).
The C-POD (POD2164) was successfully deployed on March 9,
2013 and recovered and redeployed April 14, 2013. The second
deployment ended when the equipment got tangled in fishing gear on
May 31, 2013, and was returned by the fisherman with data intact.
3.2.2 Deployment and Mooring
The C-POD was deployed and retrieved by hand from small
boats. A location with a depth of approximately 45 m was selected, and
marked on the GPS, before dropping the ground tackle and the C-POD.
The ground tackle consisted of a 6 kg Danforth anchor, 4.5 m of anchor
chain, a 6.8 kg pyramid anchor and 60 m of anchor line. A large crab
float was attached at the surface, with a 20 m tag line attached to two
75
smaller floats The C-POD was attached to the anchor line 25 m off the
seabed.
3.2.3 Processing C-POD Data
SD cards containing data from the C-PODs are read onto the PC
and processed using the Windows based CPOD.exe program, which is
supplied with the C-POD, for processing click train data and providing
graphical analysis tools. The CPOD.exe program uses a click train
classification algorithm, the KERNO classifier, to identify probable
click trains. Click trains are categorized as ‘high’, ‘mod’, ‘low’ or ‘?’
Figure 3-2: Harbor porpoise click trains. The lower panel shows the raw tonal sound
data, including individual clicks. The upper panel shows click trains identified by
the KERNO classifier.
76
according to how well they fit the expected parameters of a click train:
frequency and duration of the clicks, number of clicks, shape of the
sound and the inter-click intervals (Figure 3-2).
Sound sources of interest categorized by the KERNO algorithm
include porpoise, other cetaceans, or SONAR from the echo sounders of
passing vessels. Harbor porpoise clicks are high-frequency narrowband (HFNB) sounds over 100kHz of short duration that occur in quick
succession. Other cetaceans, such as dolphins, have click trains that
cover a broader band of frequencies and include a component under
100kHz. Echo sounders from passing vessels are easily distinguished
Figure 3-3: Echo sounder sonar as detected by the KERNO classifier. Regularly
timed pulses at a consistent frequency are easily identified.
77
due to their consistent frequency and pulse pattern over extended
periods (Figure 3-3).
Inspections of the graphical displays were used to determine the
background noise levels at the site. Estimates of the rate of false
detections were made by comparing frequency, inter-click intervals and
possible noise sources to the expected parameters for harbor porpoise
click trains. The visual displays were also used to manually review
porpoise presence for 10 minutes before and 10 minutes after the
passage of a vessel using an echo sounder, and checking for missed
detections during these periods when the KERNO classifier failed to
identify a click train.
Upon completion of classification and analysis in CPOD.exe,
data were exported to a .txt file that could be read into Microsoft Excel
and R. Output was created in both an hourly and daily format to allow
for analysis with environmental data of differing time scales.
3.3 Visual Survey Methods
Visual survey methods were used to validate presence and
behavior of porpoise in close proximity to the C-POD, as well as to log
data about harbor porpoise over a much larger geographic area than is
covered by the detection range of acoustic equipment. Visual survey
methods are limited to daylight hours during times of clear visibility,
78
with a Beaufort Sea State of 2 or less, which corresponds to wind
speeds of less than 8 knots (4 m/s). Visibility was also affected in the
evening by sun glare to the west. Attempts were made to conduct at
least one observation session each week when possible.
Two methods were used to collect the visual observation data on
harbor porpoise, scanning with compass binoculars and tracking with a
theodolite. Due to the wet maritime weather in the spring, most
observations were conducted using compass binoculars and data was
recorded on paper data collection forms. Information included group
size, behavior and approximate position of harbor porpoise, as well as
the start and end times of each sighting and any changes in weather
conditions. The theodolite was used to more precisely track harbor
porpoise in relation to the C-POD when there was no risk of
precipitation and there were at least two observers available to act as
spotters and operate the equipment. Due to these limitations, the
theodolite was only used three times during the study period.
Opportunistic sighting reports were also collected from residents
of the area indicating dates and times when they noticed harbor
porpoise near C-POD mooring float. Several times each week, one of
the volunteers would stop for 10 or 15 minutes on her way to work in
the morning in order to observe porpoise activity and take notes.
79
3.4 Environmental Data
Environmental variables recorded on an hourly basis were used
as independent variables for analysis with detection positive minutes
(DPM) per hour as the dependent variable. Analyses were conducted
using hurdle regression analysis (from the pscl package) using R v.
3.02. A common problem with environmental count data, such as DPM
in this study, is a high percentage of zero-count data, which is known
as “zero-inflated”. The hurdle model was developed to deal with zeroinflated data by splitting the analysis into two separate models. The
zero counts are analyzed in the zero hurdle model which determines
whether the independent variable influences the likelihood of there
being a non-zero count. In the zero-hurdle model counts only have two
states, zero and non-zero. If there is a non-zero count it is considered to
have hurdled into the positive count state. The second part of the
model is the zero truncated model, where only the non-zero counts are
analyzed against a negative binomial distribution. The negative
binomial distribution was selected due to the over-dispersion of the
data. Results from these two tests are reported separately and are not
dependent on each other.
Hourly weather data was retrieved from the Tacoma Narrows
Airport weather station (KTIW, 47.2675°N, 122.57611°W Elev: 315ft)
located approximately 10 km north of the observation site in
80
Steilacoom. The Tacoma Narrows Airport station was selected over the
slightly closer station at Joint Base Lewis McChord due to its close
proximity to the water. Exploratory analyses were conducted on air
temperature, wind speed and precipitation. High levels of correlation
were found between the air temperature and both the time of day and
the season, therefore air temperature was excluded from analysis.
Precipitation was not independent of wind speed and did not achieve
significance, so it was excluded from the hurdle model. The only
weather variable to be included in the hurdle model was wind speed.
Water temperature was recorded every minute by the C-POD, with the
water temperature also acting as a proxy for the season and was
excluded from analysis.
There are no tidal gauges in the South Puget Sound, with the
nearest water level monitoring site in Commencement Bay, Tacoma,
WA, where it will be influenced by the Puyallup River. Tidal data was
calculated for the site using the WXTtide32 program
(http://www.wxtide32.com/), which uses station data provided by
NOAA for Cormorant Passage where the C-POD is located. Change in
tide height is used as a proxy for tidally driven currents, and were
calculated from the beginning to the end of each hour. The hourly
relative change in tide height was used as one of the independent
variables in the hurdle model.
81
To determine if there were diel trends in presence at the study
site the hour of the day was analyzed as an independent variable in
the hurdle model. Seasonal variability was analyzed using DPM per
day as the dependent variable and the month as the independent
variable using a non-parametric Kruskal-Wallis one-way analysis of
variance by ranks. Pairwise comparisons between months was
conducted using the Kruskalmc (R package pgirmess). All
environmental analysis was conducted using R v. 3.02.
3.4.1 Anthropogenic Impact Data
Vessel echo sounders are the most readily identified
anthropogenic data available in the acoustic record. Analyses were
conducted to determine if echo sounders had any impact on harbor
porpoise detections at the site. Echo sounder detections were analyzed
for harbor porpoise presence during the 10 minutes immediately prior
to the initial sonar reception by the C-POD, and for the 10 minutes
after echo sounder activity ceased to be detected.
Echo sounder detections were identified by the KERNO
classifier in CPOD.exe. The classifier only detects the strongest portion
of the echo sounder activity during the middle of the train; it does not
classify the beginning and ends of the train properly due to weak
signal strength. Every echo sounder detection was verified via the
82
graphical interface to extend the start and end points of the echo
sounder train and remove any echo sounder false detections. Periods of
10 minutes before the start of the echo sounder signal and 10 minutes
after the end of the signal were checked for porpoise click trains
detected by the classifier. If no porpoise click trains were detected, a
manual check for missed detections was conducted. A McNemar's chisquared test was used to assess whether vessels with active SONAR
were affecting harbor porpoise presence as they transit the study site.
83
4 Results
4.1 Acoustic results
Acoustic data at the Steilacoom site from 10-Mar-2013 through
31-May-2013 had a total of 59,955 detected click-trains. There were
10,104 DPM out of 119,640 minutes of recording, with porpoise
presence detected 8.5% of the time. During the deployment, NBHF
click trains were detected during 906 out of 1994 hours, with 1088
zero-DPM hours (Figure 4-1). Hourly DPM ranged from 0 to 60 with a
mean of 3.94 and a median of 0. The high proportion of zero counts
necessitated the use of a hurdle model when analyzing hourly
Figure 4-1: Histogram of detection positive minutes (DPM) per hour recorded by the
Steilacoom C-POD illustrates the excess of zeros in the count data, requiring the use
of an appropriate zero-inflated model.
84
environmental data, and the variance to mean ratio (VMR) of the
nonzero counts (VMR = 12.11) required the use of a negative binomial
distribution during analysis. Daily DPM ranged from 7 to 471, with a
mean of 121.7 DPM (SD = 104), which corresponds to 8.5% of the
minutes in a day (1440), and a median of 96 DPM (Figure 4-2).
Figure 4-2: Histogram of the detection positive minutes (DPM) per day. The daily
DPM ranged from a low of 7 to a high of 471, with a mean of 121.7 and a standard
deviation of 104.
4.2 Click train analysis
Background noise levels at the Steilacoom site proved to be low
enough that any NBHF click trains that were classified as “low
quality” or better could be used for analysis. A selection of 900 NBHF
click trains were checked for possible false detections, with only two
questionable results that occurred during broadband ship noise,
85
yielding a false-positive rate of <0.23%. While checking for false
detections, it became apparent that there were far more missed
detections than false detections, though no effort was taken to quantify
the missed detections due to the difficulty of identifying each
individual click train. The level of missed detections appeared to be
consistent throughout the results with most encounters generating at
least one detection.
All high, moderate and low quality results for other cetaceans,
such as dolphins, were checked and all results were considered to be
false detections or questionable as to whether they represent actual
detections due to their occurrence during periods of high
environmental or vessel traffic noise. On days when Risso’s dolphins
were known to be in the area, some potential low quality click trains
were detected, though the results are brought into question due to the
high levels of false detections in the other cetacean category. Many of
the false detections in the dolphin range were produce by echo
sounders or other vessel noise. There were a few cases of false echo
sounder detections, which were caused by harbor porpoise click trains,
though these were easily identified and corrected.
86
4.3 Environmental results
The factors included in the hurdle analysis were rate of tidal
change, wind speed and hour of the day (Figure 4-3). Tidal change per
hour had a range of -1.09 to 1.23 m hour-1 (M = 0.00 m hour-1, SD =
0.55 m hour-1). Wind speeds were in the range of 0 to 9.8 m s-1 (M = 2.6
m sec-1, SD = 2.1 m sec-1), with 449 out of 1994 hours having calm wind.
Figure 4-3: Hurdle model results comparing environmental variables to detection
positive minutes per hour. The top plots show the zero truncated portion of the
hurdle model, and the bottom plots show the zero hurdle portion, where zero counts
are compared to non-zero counts (results are jittered for clarity). Regression lines
are in blue, with LOESS local regression lines in red. Hour of the day and wind
speed were found to be significant in the zero truncated portion of the model, and
tide change was significant in the zero hurdle portion of the model.
87
Hourly DPM data were analyzed against the hour of the day, wind
speed and the tidal change over the course of the hour using a negative
binomial hurdle model due to the zero-inflation of the dataset (
). Tidal change was the only independent variable tested to have
a significant impact on the zero hurdle section of the model (p=0.004).
The zero-truncated count section of the model had significant results
from both the hour of the day (p=0.025) and the wind speed (p < 0.001).
The Steilacoom C-POD recorded a significant difference in the
mean DPM per hour count when compared to the hour of the day (p =
0.025). Detections peaked in the morning between 0700-0759, with a
Table 4-1: Hurdle model results from Steilacoom, with hour of the day and wind
speed having a significant influence on the zero-truncated count of detection
positive minutes (DPM) per hour, and tide change having a significant impact on
the zero hurdle binomial model.
Count model
Coefficient
Std. Err.
z value
p value
Hour
-0.0167
0.0074
-2.24
0.025 *
Wind speed
-0.0957
0.0230
-4.17
< 0.001 *
Tide change
-0.0500
0.0898
-0.56
0.577
Zero hurdle model
Hour
0.0015
0.0066
0.226
0.821
Wind Speed
0.0174
0.0221
0.79
0.429
Tide Change
0.2374
0.0830
2.86
0.004 *
Log-likelihood: -4116 on 9 Df
88
mean of 7.1 DPM hour-1 (SD = 12.4) , with the lowest detection level
occurring in the afternoon between 1700-1759, with a mean of 2.1 DPM
hour-1 (SD = 3.9) (Figure 4-4).
Figure 4-4: There is significant variation in the mean of the DPM per hour
throughout the day, with the highest detection levels in the morning hours. The
bars represent the 95% confidence interval (CI).
The wind speed was found to have a significant impact on the
zero-truncated hourly DPM count data (p < 0.001), with decreasing
detections as wind speed increased (Figure 4-5). The mean zerotruncated DPM hour-1 during calm winds (n=198) was 11.2 (SD=12.9),
while wind speeds greater than 6 m sec-1 (n=83) had a mean of 6.6
(SD=6.3). Even though wind speed appeared to have an inverse
relationship with the mean DPM hour-1, out of 161 hours with wind
89
speed greater than 6 m sec-1, there were 83 hours where click trains
were detected, and 11 of those hours had more than 15 DPM.
Figure 4-5: There is significant variation (p < 0.001) in the zero-truncated means of
the DPM per hour at different wind speeds, with decreasing acoustic activity levels
detected at higher wind speeds. The bars represent the 95% confidence interval (CI).
The magnitude and direction of the change in tide height over
the course of an hour was found to cause a significant difference in the
zero hurdle portion of the model (p = 0.004), though it had little effect
on the mean DPM counts in the zero-truncated portion of the model.
Positive DPM hours were most likely to occur during slowly incoming
tides, with a tidal change rate of approximately +0.4 m hour-1. This
suggests that the current produced by the change in tide plays a role in
porpoise choosing to use this area, but it plays little role in how long
they stay within the detection range.
90
4.4 Seasonality
Daily detection positive minutes varied on a monthly time scale
(Figure 4-6) with the highest level of porpoise presence during the
month of May (n=31) with a range of 35 to 471 DPM day-1 (m=201.5,
SD=123.7). A moderate porpoise presence was recorded in March
(n=22) with 28 to 209 DPM day-1 (m=108.5, SD=49.4), and decreased
throughout the month. The lowest levels were in April (n=30) with 7 to
114 DPM day-1 (m=49.0, SD=25.4). A Kruskal-Wallis analysis of
variance (ANOVA) was conducted to compare the effect of the month of
the year on detection positive minutes (DPM) per day from March
Figure 4-6: Detection positive minutes (DPM) per day throughout the Steilacoom CPOD deployment. Lowest levels are in late April, and highest in Early May. March
and April experience much smaller variance from the mean, with most counts
within or close to the 95% CI.
91
Figure 4-7: Significant differences were detected between months (P < 0.001), with
very few detections during April, followed by a spike in detections during May.
through May, 2013. There was a significant effect of the month on the
DPM day-1 at the p<.05 level for the three months (χ2(2) = 38.96, p <
0.0001).
Pairwise comparisons of the mean ranks between months were
conducted using Kruskalmc post hoc test, with the results shown in
Table 4-2. Significant differences were found in the March-April and
Table 4-2: Results of Kruskalmc post hoc pairwise test. Significant differences were
found in the March-April and April-May tests, but not in the March-May test.
Observed
Critical diff.
Significant
difference
March-April
24.268
16.197
Yes
March-May
13.932
16.086
No
April-May
38.200
14.779
Yes
92
April-May pairwise tests, with no difference detected March-May.
These results show that harbor porpoise density in a portion of their
range within the South Puget Sound can vary on a temporal scale of
weeks to months.
4.5 Interaction with vessel echo sounders
Review of the Steilacoom C-POD data revealed 155 echo sounder
events logged during the March-May 2013 deployments. The majority
of these events (82%) matched the schedule for the Steilacoom-Ketron
Island ferry and presented a similar 4-6 minute 50 kHz echo sounder
signature, so those were all assumed to be the ferry. Other vessels had
echo sounders operating in the 50 kHz, 80 kHz and 120 kHz bands,
with the 50kHz band being the most common. The research boat used
to deploy the C-POD was equipped with a dual-band echo sounder, but
upon review only the lower band was apparent in the record. This is a
probable artifact of the attenuation of higher frequency signals,
combined with the detection algorithm of the C-POD, which only
records metadata about the highest energy tonal sound that it detects.
It is probable that other vessels, especially fishing boats, also had dualband systems with only the lower band showing up.
93
Table 4-3: Truth table of porpoise presence/absence during the 10 minutes before
and 10 minutes after 155 echo sounder events recorded by the Steilacoom C-POD.
Porpoise presence before and after SONAR events
Absent after
Present After
Sum
Absent before
79
21
100
Present before
8
47
55
87
68
155
Sum
All echo sounder events identified by the KERNO classifier were
checked for NBHF click trains that were consistent with harbor
porpoise presence for 10 minutes before the start of the echo sounder,
and for 10 minutes after the echo sounder ended, with the results
summarized in Table 4-3. A McNemar's chi-squared test determined
that there was a significant difference between periods when porpoise
were not detected before the echo sounder event then they were
detected afterwards, and periods when their presence was detected
before the echo sounder and not detected after (χ2= 4.97, df = 1, p =
0.026). Contrary to their reputation for shying away from vessels, 85%
of the time when porpoise were detected before the period of echo
sounder activity, they were also detected after as well, with only 15%
of the periods showing a lack of detections after the echo sounder
activity. When no porpoise were detected before the echo sounder
activity, they remained absent afterwards 79% of the time, with
94
detections showing up 21% of the time after an occurrence of echo
sounder activity.
One notable echo sounder event occurred on May 2, 2013
starting at 1728 and continuing until 2036, with few gaps in the 80
kHz signal. During that time there were also some instances of vessels
with 50 kHz systems also operating in the area. Detections of porpoise
remained at moderate to high levels throughout this period, extending
about half an hour after the vessels with the sonar departed.
4.6 Visual results
Poor viewing conditions interfered with most attempts to
conduct visual observations during March and April, which only had
two short sessions each. A total of 18 visual observation sessions were
conducted during the C-POD deployment, for a total time of 911
minutes. Harbor porpoise were sighted within the detection radius of
the C-POD during 12 sessions, these sightings included 22 animals in
14 groups (m=1.57). On one occasion, two groups were foraging in the
area for overlapping periods, which will be treated as one acoustic
encounter. Porpoise were detected by the KERNO classifier during 12
out of the 13 encounters. The one missed encounter was a fast
traveling porpoise, while all other encounters were of foraging animals.
Visual review of the acoustic data during the period of the missed
95
encounter revealed that many clicks were recorded, but few well
ordered click trains were occurring.
During six of the encounters porpoise were already present at
the start of the survey period. The remaining six encounters were
evenly divided between those that were detected first by the observers,
before porpoise entered the acoustic detection zone, and those which
were detected by the C-POD before the animals were sighted. There
were two periods when the C-POD detected click trains of porpoise, yet
only animals outside the standard 400 m range of detection were
observed. During only one instance, no porpoise were sighted in the
area when the C-POD detected several click trains.
96
5 Discussion
The use of a C-POD acoustic porpoise detector has proven to be
an effective complement to the use of visual observation in the study of
harbor porpoise behavior and distribution in the South Puget Sound.
While the number of visual observations was limited by the weather,
the C-POD was able to detect all the visually confirmed foraging
encounters within the detection radius, only missing a fast traveling
porpoise. Acoustic data revealed associations between porpoise
presence and seasonal, diel, wind speed and tidal variables that would
have been impossible to gauge with visual methods alone. Data from
the C-POD also produced surprising results when porpoise presence
was compared to echo sounder activity from fishing boats and the
passing Steilacoom-Ketron Island ferry.
5.1 Visual observation
Even though the poor weather conditions made it difficult to
collect sufficient visual data to conduct any meaningful quantitative
analysis, the C-POD detected all foraging groups of animals within 200
m. The one encounter that was not acoustically detected was of a single
animal that was traveling through the site, while all detected
encounters involved foraging animals. It is likely that traveling
porpoise only click often enough to ensure that they do not swim into
97
anything, while foraging porpoise will ensonify their surroundings in
all directions in an attempt to locate food. This idea is supported by a
study which found that when porpoise are not actively foraging, they
produce click trains far less often. Porpoise send out a high energy,
long distance click train to assist in navigation, followed by a longer
interval between click trains than occurs during foraging (Akamatsu et
al. 2007). Foraging behavior was studied in finless porpoise
(Neophocaena phocaenoides) and unlike traveling dives, foraging dives
were found to involve rolling and scanning the environment during
31% of the dive time, and echolocating over 4 times more frequently
than during dives without searching behavior (Akamatsu et al. 2010)
There were three acoustic detections that occurred when no
porpoise were sighted within 200 m of the C-POD. These simply could
have been missed by the observers. It is also possible that the porpoise
were outside the 200 m range, but if they were on-axis with the C-POD
it would have extended the range of detection (Clausen et al. 2012)com.
A much larger sample of matched visual and acoustic observations
need to be collected before it is possible to draw any meaningful
conclusions.
98
5.2 Environmental analysis
Understanding seasonal distribution and abundance is
important to inform future research and management decisions.
Seasonality is of particular importance when considering measures to
protect harbor porpoise populations, such as the efforts being
undertaken in the EU to define marine protected areas (Sveegaard and
Teilmann 2008, Berrow et al. 2009, Sveegaard et al. 2011b). If it is
determined that restrictions on human activities need to be
implemented to protect the harbor porpoise in the South Puget Sound,
such as limitations on gillnet fisheries, knowledge of the times and
locations of the highest porpoise density will allow for the most
effective protection while limiting impact on the restricted activities.
Even though harbor porpoise were acoustically detected during
every day of the deployment, there was significant temporal variability
in the use of the site. On a monthly timescale, the difference between
the moderate and low levels of detections during March and April
respectively, were followed by the much higher level of acoustic activity
during May, with the greatest peak happening during the first two
weeks of the month. This suggests that there are some longer-term
factors, such as an increase in the availability of prey at the site, which
influenced the porpoise to spend more time foraging in the area.
99
On a diel timescale, it was found that harbor porpoise were most
acoustically active at the site in the morning, between 0400 and 1000,
with a peak during the 0700 hour. The acoustic results showed high
levels of morning activity in the area which was supported by
volunteer observations with near-shore harbor porpoise activity
observed most mornings within a short 15 minute window, often
within the detection range of the C-POD (Laurie Shuster, personal
communication). Much lower levels of detection occurred in the late
afternoon, with the lowest levels during the 1700 hour. The higher
rates of detection during the morning hours roughly corresponded to
sunrise, which occurs on March 10 at around 0630 Pacific Standard
Time (PST) and 0420 PST on May 31. Peak activity closely associated
with sunrise suggests that harbor porpoise presence at the site is
related to the morning descending vertical migration of zooplankton,
which happens during the hours after sunrise. The zooplankton attract
benthic planktivorous fish that are common prey species of the harbor
porpoise (Alldredge and King 1985, Genin et al. 1988). The relatively
shallow shelf area at the north end of Cormorant Passage, where the
C-POD was deployed, would allow for easier predation by the porpoise
on these benthic species than in the deeper waters of the main
channel. While PAM is able to reveal these trends in diel behavior,
other techniques such as the use of multi-frequency sonar to map
100
biomass migration (Genin et al. 1988, Johnston et al. 2005) would
provide a more complete picture of what is likely to be driving harbor
porpoise presence during the morning hours.
Harbor porpoise are known to congregate and feed along fronts
and in eddies that form on the leeward side of islands and headlands,
suggesting that swiftly moving tidally driven currents might play a
role in porpoise’s site selection (Johnston et al. 2005). The rate of
detection at the C-POD site showed that harbor porpoise were more
likely to be present during the relatively slack waters of slowly
incoming tides, and the lowest chance of detection occurred during the
times of greatest tidal change. This suggests that harbor porpoise are
not drawn to this site by strong currents. A possible explanation for
the higher probability of porpoise detections during slack water lies in
the local bathymetry of the region. Strong fronts develop off the north
side of Ketron Island and at the mouth of Balch Passage, between
Anderson and McNeil Islands, on outgoing tides, where porpoise are
often observed feeding for extended periods (personal observation).
During incoming tides, the strongest fronts develop mid-channel, with
only weak fronts developing in the near shore location of the C-POD
(personal observation). Porpoise may be drawn away from the study
site, to the areas where fronts are developing during times of stronger
101
tidal flow, spending more of their time foraging in the study site during
times of slack water.
Wind speed could be influencing porpoise detections in a
number of different ways. As wind speed increased, there was a decline
in the mean number of DPM hour-1. Wind waves at the water surface
are a major source of broadband noise in the marine environment,
which can mask the porpoise echolocation clicks, making detection
more difficult (Clark et al. 2009, Hildebrand 2009). At a deployed
depth of greater than 20 meters, the surface noise should be somewhat
attenuated, though it would still affect the detection radius of the
C-POD. While the counts were lower during periods of high wind, the
percentage of hours with at least one detection was higher than the
mean for the deployment period and many hours had greater than 15
DPM. It can be concluded noise from wind waves had no more than a
moderate effect on the rate of detections.
Wind also affects water movement, mixing surface water layers,
driving currents and causing upwelling in some areas, which could
affect the distribution of prey species (Koseffl et al. 1993, McManus et
al. 2005). Increases in wind waves could also have a direct effect on
harbor porpoise, requiring more energetic surface behavior instead of
their normal low rolling surface activity (Scheffer and Slipp 1948).
Harbor porpoise might choose to forage in more protected areas during
102
high winds to minimize their energy expenditures. Even though the
winds are having an effect on porpoise detections by the C-POD, the
detection rate is much better than would be expected from visual
techniques under the same conditions.
5.3 Anthropogenic disturbance
The examination of the time periods before and after echo
sounder activity produced a surprising result; vessels transiting this
portion of the Puget Sound do not necessarily cause harbor porpoise to
flee the area. Harbor porpoise were found to remain in the area 85% of
the time when vessels passed close enough to the study site for their
echo sounder to be detected. During most of these events, porpoise
echolocation continued throughout the vessel passage, suggesting that
they stay in the area and continue foraging, rather than leaving and
returning when the vessel has passed. When porpoise were in the area
and a vessel with an echo sounder passed, they left the area 15% of the
time; 21% of the time, porpoise were not detected in an area until after
a vessel with an echo sounder passed through the study area. The
difference between animals leaving the area and those arriving is
sufficient to suggest that porpoise might be attracted to some aspect of
vessel passage. Evans et al. (1994) showed that harbor porpoise
reacted to vessels in different ways depending on the vessel type, speed
and behavior. While the expected avoidance response did occur, there
103
were many instances where no reaction was recorded and some
occasions where the porpoise moved towards the vessel.
Visual observations of porpoise feeding when vessels transited
the area suggest that diving for several minutes is a common
avoidance behavior. Instead of fleeing the area, they often reappear in
the same general vicinity after the vessel has passed. Porpoise that are
traveling through the area appeared to be more likely to leave the area
and not be seen again when a vessel encounter occurs (personal
observation). Similar behavior was noted in Flaherty and Stark (1982)
where many porpoise were witnessed using diving as a preferred
method of vessel avoidance. While acoustic methods excel at
determining porpoise presence, they do not reveal details about
porpoise surface behavior or dive length, which would be necessary for
assessing whether the porpoise are using dives as an avoidance
behavior.
The sample size of echo sounder activity is relatively small, and
collected from a single site in close proximity to regular ferry lanes.
With greater than 80% of the detected echo sounder events
attributable to the Steilacoom-Ketron Island ferry, and additional ferry
runs from Steilacoom that service Anderson and McNeil Islands, it is
reasonable to infer that harbor porpoise in the waters off Steilacoom
may have become habituated to the regular presence of these vessels.
104
A similar situation was discussed in Evans et al. (1994), where one of
the vessels was a small ferry that transited the site as many as 8 times
per day. In that study, harbor porpoise were found to move closer to
the ferry 32% of the time, away 22% of the time and showed no
response 46% of the time. More research needs to be conducted at a
variety of sites, using both visual and acoustic methods in order to
determine the impact of different forms of vessel traffic on porpoise
behavior and the possibility of habituation occurring.
5.4 Passive acoustic monitoring
The C-POD has proven to be a useful tool for monitoring harbor
porpoise presence in conditions that make other forms of observation
unworkable. With the ability to deploy the C-PODs for several months
at a time, then quickly recover the data and redeploy the equipment
within a matter of minutes, they provide an excellent long-term
monitoring solution.
The customized software developed to detect and classify click
trains in the C-POD data are can save a significant amount of time
processing the acoustic data. It provides a graphical interface for easy
review of data, classifies the source of click trains and outputs data in
a format that can be used by Microsoft Excel or statistical packages.
The detection algorithms are quite conservative, providing very low
105
rates of false detections, which unfortunately leads to a high rate of
missed detections. A cursory examination of missed detections found
that well over half of the porpoise click trains remained undetected,
even during encounters with large numbers of detected click trains. In
many cases the high rate of missed detections is not an issue because it
is rare that an entire encounter will go undetected even if every click
train is not accounted for. During analyses when the high rate of
missed detections is unacceptable, such as during the echo sounder
analysis, the identification and marking of the missing click trains is a
fairly simple, yet time consuming, process.
While the C-POD has many advantages, it is not without its
limitations. Instead of capturing full recordings of sounds, the device
only stores metadata describing the sound. The single omnidirectional
hydrophone also limits the ability to determine the number of animals
in a group, making it difficult to come up with reasonable abundance
estimates. Furthermore, its range of detection is limited to only a few
hundred meters which limits the size of the study area to a smaller
range than covered by visual observation techniques.
While this study has addressed the usefulness of C-PODs in
monitoring harbor porpoise in the South Puget Sound, and has
produced some baseline data, a proper monitoring program will require
multiple C-PODs to gain an understanding of the changes in harbor
106
porpoise habitat selection, range, seasonal distribution and behavior.
To understand whether apparent changes in relative abundance are
actually occurring or if the changes have more to do with shifts in
yearly or seasonal distribution, visual and acoustic monitoring is
needed at additional long term study sites throughout the South Puget
Sound region.
5.5 Conclusion
The Puget Sound is a remarkable ecosystem that has seen
considerable abuse over the years. We have overfished and polluted the
waters and dammed the rivers that make the Sound such a productive
estuarine environment. Marine mammals in the region were killed for
their furs, their oil and because they were seen as competition for
resources, with several species brought to the point of near extinction.
When one of these species returns to an ecosystem where they haven’t
been seen for several decades, it is cause for a certain amount of
optimism that conditions are improving. We don’t know for certain why
porpoise disappeared from the South Sound, nor do we know what led
to their return, yet we continue to pollute our waters and we shortchange environmental education and conservation.
Passive acoustic methods used in this study have proven to be
useful for monitoring several key behavioral aspects of the harbor
107
porpoise population that has reestablished in the South Puget Sound.
The timing of this study, during the normally stormy late winter and
spring months, demonstrated the ability to collect acoustic data during
times when visual data collection is often not possible. Even with
limited ability to conduct visual observations during much of the
deployment period, the C-POD continued to collect data about the use
of the site by harbor porpoise during all hours of the day and night, in
all weather conditions. The ability to collect data at night enabled
analysis of diel activity levels around the clock, not just during
daylight hours. With the high winds and rain that is very common
during April and May, it is unlikely that the limited visual
observations would have revealed the extent of the seasonal variation
that was evident in the acoustic record. The apparent lack of response
by harbor porpoise within the acoustic record to the passage of vessels
operating an echo sounder was one of the more surprising results of
this study and is worthy of future research.
The use of passive acoustic monitoring devices, like the C-POD,
is providing many new opportunities to improve our understanding of
the marine environment that we all depend on. These tools are being
used throughout the world to monitor a wide variety of marine species
and represent a large step forward in our ability to collect new forms of
data that were previously unavailable. Researchers don’t need to be
108
physically present in remote locations during data collection and the
equipment gathers information throughout the day and night in all
weather conditions. Passive acoustic monitoring provides a cost
effective method of gathering information about species of concern,
especially as human pressure on marine resources continues to
expand.
Harbor porpoise are a high trophic level sentinel species whose
abundance and success reflects the health of the ecosystem. By
monitoring the continuing viability of the porpoise in South Puget
Sound, it expands our understanding of human impacts on our marine
ecosystem. Harbor porpoise, along with salmon, killer whales, and
harbor seals, can serve as living reminders of our past mistakes and
the need to protect the marine environment.
109
6 An interdisciplinary study with broad impact
The return of the harbor porpoise into the waters of the Puget
Sound is an important ecological event that suggests that the
environment of Washington’s inland waters have improved to a level
that can once again support a harbor porpoise population. While we
will never know precisely what led to their extirpation, by studying
their return, we can monitor porpoise health and growth rates as a
benchmark for the health of the marine environment. The results of
this and future studies will provide valuable information to support
conservation and resource management decisions by policy makers.
The deployment of a single C-POD has shown that it can be a
useful tool for monitoring behavior, but a device in a single location
does not reveal much about the spatial and temporal use of the South
Puget Sound by the porpoise, nor can it tell us about the health of the
population. Deploying an array of C-PODs on permanent moorings
throughout the area would provide important information about
changes in porpoise abundance over time, as well as their movements
throughout the South Puget Sound. The data from multiple C-PODs
would be useful for informing policy and management decisions as well
as providing an opportunity for scientists to expand on behavioral
studies about the species.
110
As a protected species under the Marine Mammal Protection
Act, policy decisions can have a wide ranging impact on a variety of
industries. Fisheries policies might need to take porpoise presence into
account if bycatch or other fisheries interactions increase to a level
that the population cannot support. Pollution from runoff, heavy
industry, the proposed coastal coal train route, and other
anthropogenic sources may need to be regulated if pollutants are found
at dangerous levels in samples of marine mammal tissues.
Resident marine mammal populations act as excellent sentinel
species for monitoring anthropogenic impacts to our marine waters
(Bossart 2011). Porpoise are long-lived, high trophic level predators,
therefore toxins bioaccumulate in their fatty tissues to levels which are
easily detectable. The heavy subdermal blubber layer accumulates
many of the fat soluble toxins which the animal has ingested. These
tissues are easily sampled using nonlethal biopsy darts, which enables
researchers to monitor toxin levels without relying solely on stranded
animals. This can serve as another measure of the presence of toxins in
the marine environment, some of which may have accumulated to the
point of becoming a concern.
Marine mammals are considered to be charismatic megafauna
and are much loved by the people of the Puget Sound region. On many
occasions, members of the public stopped by the research site and
111
expressed interest in the harbor porpoise story. Most people were
unaware of the existence of harbor porpoise in these waters and
wanted to learn more about these animals and their return to the
Puget Sound. This provides an excellent opportunity to educate people
about issues related to the health of our local waters. Recent articles
covering the return of the harbor porpoise have appeared in several
local newspapers spurring additional public interest in the topic.
Citizen scientist groups are forming throughout the world
around a variety of environmental issues. With increasing public
interest in the harbor porpoise, Puget Sound provides an excellent
opportunity for citizen scientists to contribute observations about
porpoise location and behavior and to help educate the general public.
Some citizen scientists volunteer their time to assist professional
researchers with data collection in the field or office based work with
analysis projects. Professional mariners spend a considerable amount
of time on the water and frequently contribute invaluable information
regarding location and behavioral patterns of many marine mammal
species.
Harbor porpoise have returned to the South Puget Sound after
an absence of several decades. Careful monitoring of the range and
abundance of these animals is essential to conservation efforts.
Acoustic monitoring, when combined with visual surveys, can provide
112
essential information to properly monitor and manage this population
within the South Puget Sound. Further study and published research
is needed in order to document the recovery of this population and to
contribute to the global knowledge base and conversation regarding
this widely distributed species.
113
7 Acknowledgements
I would like to thank Cascadia Research Collective for
sponsoring this project. John Calambokidis provided the opportunity
and impetuous for this project and was extremely generous with his
time, support, and willingness to share his extensive knowledge of
marine mammals. The interest, expertise, and encouragement of the
staff and interns at Cascadia, especially that of Dr. Robin Baird and
Dr. Glenn Gailey, was greatly appreciated.
When I was interested in attempting to monitor porpoise in
other portions of the South Puget Sound, Dr. Brad Hanson of the
Northwest Fisheries Science Center and Dr. Jason Wood of SMRU,
loaned me C-PODs to deploy and I appreciate their generosity.
My reader, Dr. Erin Martin, provided me with excellent
guidance and feedback, while graciously tolerating my independent
streak. I feel extremely fortunate to have had her as my reader.
Special thanks to goes out to Uko Gorter for allowing me to use
his excellent illustrations of Dall’s and harbor porpoises in this thesis.
Thank you to Tom & Barb of Steilacoom, Kim & Scott of
Steamboat Island, John in Gig Harbor, and the neighbors at the
research sites for your enthusiasm and interest.
Most importantly, I need to thank the incredible woman who
has put up with me all these years, Laurie Shuster, for all her love and
support throughout this process. Not only did she live with the time
demand that school put on me, she was an enthusiastic field assistant
and a first rate thesis editor.
114
8 Bibliography
Akamatsu, T., J. Teilmann, L. a. Miller, J. Tougaard, R. Dietz, D.
Wang, K. Wang, U. Siebert, and Y. Naito. 2007. Comparison of
echolocation behaviour between coastal and riverine porpoises.
Deep Sea Research Part II: Topical Studies in Oceanography
54:290–297.
Akamatsu, T., D. Wang, K. Wang, S. Li, and S. Dong. 2010. Scanning
sonar of rolling porpoises during prey capture dives. The Journal
of Experimental Biology 213:146–52.
Alldredge, A. L., and J. M. King. 1985. The distance demersal
zooplankton migrate above the benthos: implications for predation.
Marine Biology 84:253–260.
Andersen, L. W. 2003. Harbour porpoises (Phocoena phocoena) in the
North Atlantic: Distribution and genetic population structure.
NAMMCO Sci. Pub. 5:11–29.
Andersen, S. H., and M. Amundin. 1976. Possible predator-related
adaptation of sound production and hearing in the harbour
propoise (Phocoena phocoena). Aquatic Mammals 4:56–57.
Au, W. W. L. 2009. Echolocation. Pages 348–357 in W. F. Perrin, B.
Wűrsig, and J. G. M. Thewissen, editors. Encyclopedia of Marine
Mammals. Second. Academic Press, Burlington, MA.
Au, W. W. L., B. K. Branstetter, K. J. Benoit-Bird, and R. a Kastelein.
2009. Acoustic basis for fish prey discrimination by echolocating
dolphins and porpoises. The Journal of the Acoustical Society of
America 126:460–467.
Au, W. W. L., R. a Kastelein, K. J. Benoit-Bird, T. W. Cranford, and M.
F. McKenna. 2006. Acoustic radiation from the head of
echolocating harbor porpoises (Phocoena phocoena). The Journal of
experimental biology 209:2726–33.
Au, W. W. L., R. A. Kastelein, T. Rippe, and N. M. Schooneman. 1999.
Transmission beam pattern and echolocation signals of a harbor
porpoise (Phocoena phocoena). The Journal of the Acoustical
Society of America 106:3699–705.
115
Bain, D., and R. Williams. 2006. Long-range effects of airgun noise on
marine mammals: responses as a function of received sound level
and distance. Pages 1–13 Int. Whal. Comm. Working Pap. SC/58
E.
Baines, M. E., N. J. C. Tregenza, and C. J. L. Pierpoint. 1999. Field
Trials of the POD - A Self-Contained, Submersible Acoustic DataLogger. Pages 455–456 Proceedings of the Thirteenth Annual
Conference of the European Cetacean Society. Valencia, Spain.
Baird, R. W., and L. M. Dill. 1996. Ecological and social determinants
of group size in transient killer whales. Behavioral Ecology 7:408–
416.
Barlow, J. 1988. Harbor Porpoise, Phocoena phocoena, Abundance
Estimation for California, Oregon, and Washington: I. Ship
Surveys. Fisheries Bulletin 86:417–432.
Barlow, J. 1995. The abundance of cetaceans in California waters. Part
1: Ship surveys in summer and fall of 1991. Fishery Bulletin.
Barlow, J., and K. A. Forney. 2007. Abundance and population density
of cetaceans in the California Current ecosystem. Fisheries
Bulletin 105:509–526.
Bennett, P. M., P. D. Jepson, R. J. Law, B. R. Jones, T. Kuiken, J. R.
Baker, E. Rogan, and J. K. Kirkwood. 2001. Exposure to heavy
metals and infectious disease mortality in harbour porpoises from
England and Wales. Environmental pollution 112:33–40.
Berggrena, P., R. Ishaq, Y. Zebühr, C. Näf, C. Bandh, and D. Broman.
1999. Patterns and Levels of Organochlorines (DDTs, PCBs, nonortho PCBs and PCDD/Fs) in Male Harbour Porpoises (Phocoena
phocoena) from the Baltic Sea , the Kattegat-Skagerrak Seas and
the West Coast of Norway. Marine Pollution Bulletin 38:1070–
1084.
Berrow, S., J. O’Brien, I. O’Connor, and D. McGrath. 2009. Abundance
estimate and acoustic monitoring of harbour porpoises (Phocoena
phocoena (L.)) in the Blasket Islands’ candidate special area of
conservation. Biology & Environment: Proceedings of the Royal
Irish Academy 109:35–46.
116
Berta, A., J. L. Sumich, and K. M. Kovacs. 2006. Marine Mammals:
Evolutionary Biology. Page 547. Second. Academic Press,
Burlington, MA.
Birkun Jr., A. ., and A. Frantzis. 2008. Phocoena phocoena ssp. relicta.
IUCN 2013. IUCN Red List of Threatened Species. Version 2013.2.
Bjorge, A., and K. A. Tolley. 2009. Harbor porpoise: Phocoena
phocoena. Pages 530–533 in W. F. Perrin, B. Würsig, and J. G. M.
Thewissen, editors. Encyclopedia of Marine Mammals. Second.
Academic Press, Burlington, MA.
Boisseau, O., J. Matthews, D. Gillespie, C. Lacey, A. Moscrop, and N.
El Ouamari. 2007. A visual and acoustic survey for harbour
porpoises off North-West Africa: further evidence of a discrete
population. African Journal of Marine Science 29:403–410.
Boonstra, M., Y. Radstake, K. Rebel, G. Aarts, and K. C. J.
Camphuysen. 2013. Harbour porpoises (Phocoena phocoena) in the
Marsdiep area , the Netherlands : new investigations in a
historical study area. Lutra 56:59–71.
Borchers, D. L., and M. L. Burt. 2007. Investigation of towed
hydrophone monitoring power for harbour porpoise on the SCANS
II survey . CREEM Technical report 2007-4. Pages 1–8. St.
Andrews, Scotland.
Bossart, G. D. 2011. Marine mammals as sentinel species for oceans
and human health. Veterinary pathology 48:676–90.
Bouveroux, T., J. J. Kiszka, M. R. Heithaus, T. Jauniaux, and S.
Pezeril. 2014. Direct evidence for gray seal (Halichoerus grypus)
predation and scavenging on harbor porpoises (Phocoena
phocoena). Marine Mammal Science.
Bower, J. L. 2009. Changes in Marine Bird Abundance in the Salish
Sea: 1975 to 2007. Marine Ornithology 37:9–17.
Brasseur, S., P. Reijnders, O. D. Henriksen, J. Carstensen, J.
Tougaard, J. Teilmann, M. Leopold, K. Camphuysen, and J.
Gordon. 2004. Baseline data on the harbour porpoise, Phocoena
phocoena, in relation to the intended wind farm site NSW, in the
Netherlands.
117
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L.
Borchers, and L. Thomas. 2001. Introduction to Distance
Sampling: Estimating abundance of biological populations. Page
432. First. Oxford University Press, New York.
Budge, S. M., S. J. Iverson, and H. N. Koopman. 2006. Studying
Trophic Ecology in Marine Ecosystems Using Fatty Acids: a
Primer on Analysis and Interpretation. Marine Mammal Science
22:759–801.
Calambokidis, J., J. C. Cubbage, J. R. Evenson, S. D. Osmek, J. L.
Laake, P. J. Gearin, B. J. Turnock, S. J. Jefferies, and R. F. Brown.
1993. Abundance estimates of harbor porpoise in Washington and
Oregon waters. Page 56. Olympia, WA.
Calambokidis, J., J. R. Evenson, J. C. Cubbage, P. J. Gearin, and S. D.
Osmek. 1992. Harbor porpoise distribution and abundance off
Oregon and Washington from Aerial surveys in 1991. Seattle, WA.
Calambokidis, J., J. Peard, G. H. Steiger, J. C. Cubbage, and R. L.
DeLong. 1984. Chemical contaminats in marine mammals from
Washington State. NOAA Technical Memorandum NOS OMS 6.
Page xiv + 167. Rockville, MD.
Calambokidis, J., S. M. Speich, J. Peard, G. H. Steiger, J. C. Cubbage,
D. M. Fry, and L. J. Lowenstine. 1985. Biology of Puget Sound
marine mammals and marine birds: population health and
evidence of pollution effects. Page 169. Rockville, Maryland.
Carlström, J., P. Berggren, and N. J. C. Tregenza. 2009. Spatial and
temporal impact of pingers on porpoises. Canadian Journal of
Fisheries and Aquatic Sciences 66:72–82.
Carretta, J. V., K. A. Forney, M. S. Lowry, J. Barlow, J. Baker, D.
Johnston, B. Hanson, R. L. Brownell, J. Robbins, D. K. Mattila, K.
Ralls, M. M. Muto, D. Lynch, and L. Carswell. 2009. U . S . Pacific
Marine Mammal Stock Assessments: 2009. Page 343.
Cassens, I., S. Vicario, V. G. Waddell, H. Balchowsky, D. Van Belle, W.
Ding, C. Fan, R. S. Mohan, P. C. Simões-Lopes, R. Bastida, a
Meyer, M. J. Stanhope, and M. C. Milinkovitch. 2000. Independent
adaptation to riverine habitats allowed survival of ancient
cetacean lineages. Proceedings of the National Academy of
Sciences of the United States of America 97:11343–7.
118
Chandler, T., and J. Calambokidis. 2003. 2003 aerial surveys for
harbor porpoise and other marine mammals off Oregon,
Washington and British Columbia. Page 25. Olympia, WA.
Chappell, O. P., R. Leaper, and J. Gordon. 1996. Development and
Performance of an Automated Harbour Porpoise Click Detector.
Rep. Int. Whal. Commn 46:587–594.
Chelonia Ltd. 2013a. Design history.
http://www.chelonia.co.uk/design_history.htm.
Chelonia Ltd. 2013b. C-POD standarisation and calibration.
http://www.chelonia.co.uk/c-pod_standardisation.htm.
Chen, Z., S. Xu, K. Zhou, and G. Yang. 2011. Whale phylogeny and
rapid radiation events revealed using novel retroposed elements
and their flanking sequences. BMC evolutionary biology 11:314.
Clark, C., W. Ellison, B. Southall, L. Hatch, S. Van Parijs, a Frankel,
and D. Ponirakis. 2009. Acoustic masking in marine ecosystems:
intuitions, analysis, and implication. Marine Ecology Progress
Series 395:201–222.
Clausen, K. T., M. Wahlberg, K. Beedholm, S. Deruiter, and P. T.
Madsen. 2012. Click communication in harbour porpoises
Phocoena phocoena. Bioacoustics 20:1–28.
Cotter, M. P., D. Maldini, and T. a. Jefferson. 2012. “Porpicide” in
California: Killing of harbor porpoises (Phocoena phocoena) by
coastal bottlenose dolphins (Tursiops truncatus). Marine Mammal
Science 28:E1–E15.
Cox, T. M., A. J. Read, A. Solow, and N. Tregenza. 2001. Will harbour
porpoises (Phocoena phocoena) habituate to pingers? Journal of
Cetacean Research and Management 3:81–86.
Culik, B., S. Koschinski, N. Tregenza, and G. Ellis. 2001. Reactions of
harbor porpoises Phocoena phocoena and herring Clupea harengus
to acoustic alarms. Marine Ecology Progress Series 211:255–260.
Dähne, M., U. Katharina Verfuß, A. Brandecker, U. Siebert, and H.
Benke. 2013. Methodology and results of calibration of tonal click
detectors for small odontocetes (C-PODs). The Journal of the
Acoustical Society of America 134:2514–22.
119
Deruiter, S. L., A. Bahr, M.-A. Blanchet, S. F. Hansen, J. H.
Kristensen, P. T. Madsen, P. L. Tyack, and M. Wahlberg. 2009.
Acoustic Behaviour of Echolocating Porpoises During Prey
Capture. The Journal of Experimental Biology 212:3100–3107.
Embling, C. B., P. a. Gillibrand, J. Gordon, J. Shrimpton, P. T. Stevick,
and P. S. Hammond. 2010. Using habitat models to identify
suitable sites for marine protected areas for harbour porpoises
(Phocoena phocoena). Biological Conservation 143:267–279.
Emery, M., A. D. Goodson, and S. Datta. 1993. Integrating Theodolite
and Video Tracking Data. Pages 256–259 Proceedings of the
Seventh Annual Conference of the European Cetacean Society.
Inverness, Scotland.
Evans, P. G. H., Q. Carson, P. Fisher, and W. Jordan. 1994. A study of
the reactions of harbour porpoises to various boats in the coastal
waters of southeast Shetland. European Research on Cetaceans
8:60–64.
Evans, P. G. H., and O. Chappell. 1994. A Comparison of Visual and
Acoustic Techinques for surveying Harbour Porpoises. Pages 172–
175 Proceedings of the Eighth Annual Conference of the European
Cetacean Society. Montpellier, France.
Fair, P. A., J. Adams, G. Mitchum, T. C. Hulsey, J. S. Reif, M. Houde,
D. Muir, E. Wirth, D. Wetzel, E. Zolman, W. McFee, and G. D.
Bossart. 2010. Contaminant blubber burdens in Atlantic
bottlenose dolphins (Tursiops truncatus) from two southeastern
US estuarine areas: concentrations and patterns of PCBs,
pesticides, PBDEs, PFCs, and PAHs. The Science of the total
environment 408:1577–97.
Flaherty, C., and S. Stark. 1982. Harbor Porpoise (Phocoena phocoena)
Assessment in “Washington Sound”. Page 98. Seattle, WA.
Foote, G. 1980. Importance of the Swimbladder in Acoustic Scattering
by Fish: A Comparison of Gadoid and Mackerel Target Strengths.
Journal of the Acoustical Society of America 67:2084–2089.
Fordyce, R. E. 2009. Cetacean Evolution. Pages 201–207 in W. F.
Perrin, B. Würsig, and J. G. M. Thewissen, editors. Encyclopedia
of Marine Mammals. Second. Academic Press, Burlington, MA.
120
Frankel, A. S. 2009. Sound Production. Pages 1056–1071 in W. F.
Perrin, B. Wűrsig, and J. G. M. Thewissen, editors. Encyclopedia
of Marine Mammals. Second. Academic Press, Burlington, MA.
Frankel, A. S., S. Yin, and M. a. Hoffhines. 2009. Alternative methods
for determining the altitude of theodolite observation stations.
Marine Mammal Science 25:214–220.
Frantzis, A., J. Gordon, G. Hassidis, and A. Komnenoy. 2001. The
Enigma of Harbor Porpoise Presence in the Mediterranean Sea.
Marine Mammal Science 17:937–944.
Gailey, G., and J. G. Ortega-Ortiz. 2001. Pythagoras: A computerBased System for Theodolite Tracking. Page 354 Proceedings of
the Fifteenth Annual Conference of the European Cetacean
Society. Rome, Italy.
Gailey, G., and J. G. Ortega-Ortiz. 2002. A Note on a Computer-Based
System for Theodolite Tracking of Cetaceans. Journal of Cetacean
Research and Management 4:213–218.
Gannon, D. P., J. E. Craddock, and A. J. Read. 1998. Autumn Food
Habits of Harbor Porpoises, Phocoena phocoena, in the Gulf of
Maine. Fisheries Bulletin 96:428–437.
Gaskin, D. E., P. W. Arnold, and B. A. Blair. 1974. Phocoena phocoena.
Mammalian Species 42:1–8.
Gaskin, D. E., and A. P. Watson. 1985. The harbor porpoise, Phocoena
phocoena, in Fish Harbour, New, Brunswick, Canada: Occupancy,
distribution, and movements. Fisheries Bulletin 83:427–442.
Geisler, J. H., M. W. Colbert, and J. L. Carew. 2014. A new fossil
species supports an early origin for toothed whale echolocation.
Nature advance on.
Genin, A., L. Haury, and P. Greenblatt. 1988. Interactions of migrating
zooplankton with shallow topology: predation by rockfishes and
intensification of patchiness. Deep Sea Research Part A.
Oceanographic Research Papers 35:151–175.
Gilardi, K. V. K., D. Carlson-Bremer, J. a June, K. Antonelis, G.
Broadhurst, and T. Cowan. 2010. Marine species mortality in
derelict fishing nets in Puget Sound, WA and the cost/benefits of
derelict net removal. Marine pollution bulletin 60:376–82.
121
Gilles, A., S. Adler, K. Kaschner, M. Scheidat, and U. Siebert. 2011.
Modelling harbour porpoise seasonal density as a function of the
German Bight environment: implications for management.
Endangered Species Research 14:157–169.
Gilles, A., M. Scheidat, and U. Siebert. 2009. Seasonal distribution of
harbour porpoises and possible interference of offshore wind farms
in the German North Sea. Marine Ecology Progress Series
383:295–307.
Good, T. P., J. A. June, M. A. Etnier, and G. Broadhurst. 2009. Ghosts
of the Salish Sea: Threats to Marine Birds in Puget Sound and the
Northwest Straits from Derelict Fishing Gear. Marine Ornithology
37:67–76.
Goodwin, L. 2008. Diurnal and Tidal Variations in Habitat Use of the
Harbour Porpoise (Phocoena phocoena) in Southwest Britain.
Aquatic Mammals 34:44–53.
Goold, J. C., S. E. Jones, and M. Bridge. 2000. Time and Frequency
Domain Characteristics of Sperm Whale Clicks. journal of the
Acoustical Society of America 98:1279–1291.
Gordon, J., and P. L. Tyack. 2002. Acoustic Techniques for Studying
Cetaceans. Pages 293–324 in P. G. H. Evans and J. A. Raga,
editors. Marine Mammals: Biology and Conservation. Kluwer
Academic/Plenum Publishers.
Gygax, L. 2002. Evolution of group size in the dolphins and porpoises:
interspecific consistency of intraspecific patterns. Behavioral
Ecology 13:583–590.
Haarr, M. L., L. D. Charlton, J. M. Terhune, and E. a. Trippel. 2009.
Harbour Porpoise (Phocoena phocoena) Presence Patterns at an
Aquaculture Cage Site in the Bay of Fundy, Canada. Aquatic
Mammals 35:203–211.
Haelters, J., F. Kerckhof, T. Jauniaux, and S. Degraer. 2012. The Grey
Seal (Halichoerus grypus) as a Predator of Harbour Porpoises
(Phocoena phocoena)? Aquatic Mammals 38:343–353.
Hall, A. 2011. Foraging Behaviour and Reproductive Season Habitat
Selection of Northeast Pacific Porpoises. The University of British
Columbia (Vancouver).
122
Hall, A., G. Ellis, and A. W. Trites. 2002. Harbour Porpoise
Interactions with the 2001 Selective Salmon Fisheries in Southern
British Columbia and License Hoder Reported Small Cetacean ByCatch. Page 51.
Hall, A. J., K. Hugunin, R. Deaville, R. J. Law, C. R. Allchin, and P. D.
Jepson. 2006. The Risk of Infection from Polychlorinated Biphenyl
Exposure in the Harbor Porpoise (Phocoena phocoena): A Case–
Control Approach. Environmental Health Perspectives 114:704–
711.
Hall, A. M. 2004. Seasonal Abundance , Distribution and Prey Species
of Harbour Porpoise (Phocoena phocoena) in Southern Vancouver
Island Waters. University of British Columbia.
Hammond, P. S., G. Bearzi, A. Bjørge, K. A. Forney, L. Karczmarski, T.
Kasuya, W. F. Perrin, M. D. Scott, J. Y. Wang, R. S. Wells, and B.
Wilson. 2008a. Phocoena phocoena (Baltic Sea subpopulation).
IUCN 2013. IUCN Red List of Threatened Species. Version 2013.2.
Hammond, P. S., G. Bearzi, A. Bjørge, K. Forney, L. Karczmarski, T.
Kasuya, W. F. Perrin, M. D. Scott, J. Y. Wang, R. S. Wells, and B.
Wilson. 2008b. Phocoena phocoena. IUCN 2013. IUCN Red List of
Threatened Species. Version 2013.2.
Hammond, P. S., P. Berggren, H. Benke, D. L. Borchers, a. Collet, M.
P. Heide-Jørgensen, S. Heimlich, a. R. Hiby, M. F. Leopold, and N.
Øien. 2002. Abundance of harbour porpoise and other cetaceans in
the North Sea and adjacent waters. Journal of Applied Ecology
39:361–376.
Hammond, P. S., S. A. Mizroch, and G. P. Donovan. 1990. Individual
Recognition of Cetaceans: Use of Photo-Identification and Other
Techniques to Estimate Population Parameters. Page 448.
Hardy, T., R. Williams, R. Caslake, and N. Tregenza. 2012. An
investigation of acoustic deterrent devices to reduce cetacean
bycatch in an inshore set net fishery. Journal of Cetacean
Research and Management 12:85–90.
Harzen, S. E. 2002. Use of an Electronic Theodolite in the Study of
Movements of the Bottlenose Dolphin (Tursiops truncatus) in the
Sado Estuary, Portugal. Aquatic Mammals 28:251–260.
123
Heide-Jørgensen, M. P., M. Iversen, N. H. Nielsen, C. Lockyer, H.
Stern, and M. H. Ribergaard. 2011. Harbour porpoises respond to
climate change. Ecology and evolution 1:579–85.
Heithaus, M. R. 2001. Predator–prey and competitive interactions
between sharks (order Selachii) and dolphins (suborder
Odontoceti): a review. Journal of Zoology 253:53–68.
Hildebrand, J. 2009. Anthropogenic and natural sources of ambient
noise in the ocean. Marine Ecology Progress Series 395:5–20.
Hoek, W. 1992. An Unusual Aggregation of Harbor Porpoises
(Phocoena phocoena). Marine Mammal Science 8:152–155.
Ikonomou, M. G., S. Rayne, M. Fischer, M. P. Fernandez, and W.
Cretney. 2002. Occurrence and congener profiles of
polybrominated diphenyl ethers (PBDEs) in environmental
samples from coastal British Columbia, Canada. Chemosphere
46:649–63.
Jansen, O. E., G. M. Aarts, K. Das, G. Lepoint, L. Michel, and P. J. H.
Reijnders. 2012. Feeding ecology of harbour porpoises: stable
isotope analysis of carbon and nitrogen in muscle and bone.
Marine Biology Research 8:829–841.
Jefferson, T. A. 1988. Phocoenoides dalli. Mammalian Species:1–7.
Jefferson, T. A. 2009. Dall’s Porpoise: Phocenoides dalli. Pages 296–
298 in W. F. Perrin, B. Wűrsig, and J. G. M. Thewissen, editors.
Encyclopedia of Marine Mammals. Second. Academic Press.
Jefferson, T. A., and B. E. Curry. 1994. A Global Review of Porpoise
(Cetacea: Phocoenidae) Mortality in Gillnets. Biological
Conservation 67:167–183.
Jefferson, T. A., P. J. Stacey, and R. W. Baird. 1991. A review of Killer
Whale interactions with other marine mammals: predation to coexistence. Mammal Review 21:151–180.
Johnston, D. W. 2002. The effect of acoustic harassment devices on
harbour porpoises (Phocoena phocoena) in the Bay of Fundy,
Canada. Biological Conservation 108:113–118.
Johnston, D. W., A. J. Westgate, and A. J. Read. 2005. Effects of finescale oceanographic features on the distribution and movements of
124
harbour porpoises Phocoena phocoena in the Bay of Fundy. Marine
Ecology Progress Series 295:279–293.
Kastelein, R. A., W. W. L. Au, and D. de Haan. 2002. Audiogram of a
Harbor Porpoise (Phocoena phocoena) Measured with NarrowBand Frequency-Modulated Signals. journal of the Acoustical
Society of America 112:334–344.
Kastelein, R. a., M. Janssen, W. C. Verboom, and D. de Haan. 2005.
Receiving beam patterns in the horizontal plane of a harbor
porpoise (Phocoena phocoena). The Journal of the Acoustical
Society of America 118:1172.
Keener, W., I. Szczepaniak, Ü. Adam, M. Webber, and J. Stern. 2011.
First Records of Anomalously White Harbor Porpoises (Phocoena
phocoena) from the Pacific Ocean. Journal of Marine Animals and
Their Ecology 4:19–24.
Kinzey, D., and T. Gerrodette. 2003. Distance measurements using
binoculars from ships at sea: accuracy, precision and effects of
refraction. Journal of Cetacean Research and Management 5:159–
171.
Klinck, H., D. K. Mellinger, K. Klinck, N. M. Bogue, J. C. Luby, W. a
Jump, G. B. Shilling, T. Litchendorf, A. S. Wood, G. S. Schorr, and
R. W. Baird. 2012. Near-real-time acoustic monitoring of beaked
whales and other cetaceans using a SeagliderTM. PloS one
7:e36128.
Koopman, H. N., and D. E. Gaskin. 1994. Individual and geographical
variation in pigmentation patterns of the harbour porpoise,
Phocoena phocoena (L.). Canadian Journal of Zoology 72:135–143.
Koopman, H. N., D. A. Pabst, W. A. McLellan, R. M. Dillaman, and A.
J. Read. 2002. Changes in Blubber Distribution and Morphology
Associated with Starvation in the Harbor Porpoise (Phocoena
phocoena): Evidence for Regional Differences in Blubber Structure
and Function. Physiological and Biochemical Zoology 75:498–512.
Koschinski, S. 2001. Current knowledge on harbour porpoises
(Phocoena phocoena) in the Baltic Sea. Ophelia 55:167–197.
Koseffl, J. R., J. K. Holenl, S. G. Monismith, and J. E. L. Cloern. 1993.
Coupled effects of vertical mixing and benthic grazing on
125
phytoplankton populations in shallow , turbid estuaries. Journal of
Marine Research 51:843–868.
Kyhn, L. A., J. Tougaard, J. Stenback, D.- Kerteminde, and J.
Teilmann. 2012. From echolocation clicks to animal density —
Acoustic sampling of harbor porpoises with static dataloggers.
journal of the Acoustical Society of America 131.
Kyhn, L. A., J. Tougaard, J. Teilmann, M. Wahlberg, P. B. Jørgensen,
and N. I. Bech. 2008. Harbour porpoise (Phocoena phocoena) static
acoustic monitoring : laboratory detection thresholds of T-PODs
are reflected in field sensitivity. Journal of the Marine Biological
Association of the United Kingdom 88:1085–1091.
Lampert, W. 1989. The adaptive significance of diel vertical migration
of zooplankton. Functional Ecology 3:21–27.
Larsen, F., O. R. Eigaard, and J. Tougaard. 2002. Reduction of harbour
porpoise by-catch in the North Sea by high-density gillnets by.
Pages 1–12.
Law, R. J., C. R. Allchin, M. E. Bennett, S. Morris, and E. Rogan. 2002.
Polybrominated diphenyl ethers in two species of marine top
predators from England and Wales. Chemosphere 46:673–81.
Law, R. J., C. F. Fileman, A. D. Hopkins, J. R. Baker, J. Harwood, D.
B. Jackson, S. Kenedy, A. R. Martin, and R. J. Morris. 1991.
Concentrations of Trace Metals in the Livers of Marine Mammals
(Seals , Porpoises and Dolphins) from Waters Around the British
Isles. Marine Pollution Bulletin 22:183–191.
Law, R. J., and J. a. Whinnett. 1992. Polycyclic aromatic hydrocarbons
in muscle tissue of harbour porpoises (Phocoena phocoena) from
UK waters. Marine Pollution Bulletin 24:550–553.
Learmonth, J. A., C. D. MacLeod, M. B. Santos, G. J. Pierce, and R. A.
Robinson. 2006. Potential effects of climate change on marine
mammals. Oceanography and Marine Biology: An Annual Review
44:431–464.
Leopold, M. F., P. A. Wolf, and J. Van Der Meer. 1992. The Elusive
Harbour Porpoise Exposed: Strip Transect Counts off
Southwestern Ireland. Netherlands Journal of Sea Research
29:395–402.
126
Lockyer, C. 2003. Harbour porpoises (Phocoena phocoena) in the North
Atlantic: Biological parameters. NAMMCO Sci. Pub. 5:71–89.
Lockyer, C., and C. Kinze. 1995. Status , ecology and life history of
harbour porpoise (Phocoena phocoena), in Danish waters.
NAMMCO Sci. Pub. 5:143–175.
Long, D. J., and R. E. Jones. 1996. White Shark Predation and
Scavenging on Cetaceans in the Eastern North Pacific Ocean.
Pages 293–307 in A. P. Klimley and D. G. Ainley, editors. Great
White Sharks: The Biology of Carcharodon carcharias. Elsevier
Ltd.
Lucke, K., P. a. Lepper, B. Hoeve, E. Everaarts, N. van Elk, and U.
Siebert. 2007. Perception of Low-Frequency Acoustic Signals by a
Harbour Porpoise (Phocoena phocoena) in the Presence of
Simulated Offshore Wind Turbine Noise. Aquatic Mammals
33:55–68.
Lyamin, O. I., P. R. Manger, S. H. Ridgway, L. M. Mukhametov, and J.
M. Siegel. 2008. Cetacean sleep: an unusual form of mammalian
sleep. Neuroscience and biobehavioral reviews 32:1451–84.
MacLeod, C. D., G. J. Pierce, and M. Begoña Santos. 2007a. Starvation
and sandeel consumption in harbour porpoises in the Scottish
North Sea. Biology Letters 3:535–536.
MacLeod, C. D., M. B. Santos, R. J. Reid, B. E. Scott, and G. J. Pierce.
2007b. Linking sandeel consumption and the likelihood of
starvation in harbour porpoises in the Scottish North Sea: could
climate change mean more starving porpoises? Biology letters
3:185–8.
Madsen, P. T., D. Wisniewska, and K. Beedholm. 2010. Single source
sound production and dynamic beam formation in echolocating
harbour porpoises (Phocoena phocoena). The Journal of
experimental biology 213:3105–10.
Madsen, P., M. Wilson, M. Johnson, R. Hanlon, a Bocconcelli, N.
Aguilar de Soto, and P. Tyack. 2007. Clicking for calamari: toothed
whales can echolocate squid Loligo pealeii. Aquatic Biology 1:141–
150.
Marques, T. a, L. Thomas, S. W. Martin, D. K. Mellinger, J. a Ward, D.
J. Moretti, D. Harris, and P. L. Tyack. 2013. Estimating animal
127
population density using passive acoustics. Biological reviews
88:287–309.
Marques, T. a, L. Thomas, J. Ward, N. DiMarzio, and P. L. Tyack.
2009. Estimating cetacean population density using fixed passive
acoustic sensors: an example with Blainville’s beaked whales. The
Journal of the Acoustical Society of America 125:1982–94.
McManus, M., O. Cheriton, P. Drake, D. Holliday, C. Storlazzi, P.
Donaghay, and C. Greenlaw. 2005. Effects of physical processes on
structure and transport of thin zooplankton layers in the coastal
ocean. Marine Ecology Progress Series 301:199–215.
Mead, J. G. 2009. Gastrointestinal Tract. Pages 472–477 in W. F.
Perrin, B. Würsig, and J. G. M. Thewissen, editors. Encyclopedia
of Marine Mammals. Second. Academic Press, Burlington, MA.
Mooney, T. A., W. W. L. Au, P. E. Nachtigall, and E. A. Trippel. 2007.
Acoustic and stiffness properties of gillnets as they relate to small
cetacean bycatch. ICES Journal of Marine Science 64:1324–1332.
Mooney, T. a., P. E. Nachtigall, and W. W. L. Au. 2004. Target
Strength of a Nylon Monofilament and an Acoustically Enhanced
Gillnet: Predictions of Biosonar Detection Ranges. Aquatic
Mammals 30:220–226.
Nichol, L. M., A. M. Hall, G. M. Ellis, E. Stredulinsky, M. Boogaards,
and J. K. B. Ford. 2013. Dietary overlap and niche partitioning of
sympatric harbour porpoises and Dall’s porpoises in the Salish
Sea. Progress in Oceanography 115:202–210.
Nielsen, B. K., and B. Møhl. 2006. Hull-mounted hydrophones for
passive acoustic detection and tracking of sperm whales (Physeter
macrocephalus). Applied Acoustics 67:1175–1186.
Nielsen, T., M. Wahlberg, S. Heikkilä, M. Jensen, P. Sabinsky, and T.
Dabelsteen. 2012. Swimming patterns of wild harbour porpoises
Phocoena phocoena show detection and avoidance of gillnets at
very long ranges. Marine Ecology Progress Series 453:241–248.
Northwest Straits Initiative. 2013. Species encountered by type in
4,605 derelict nets removed. Puget Sound 2002-December 31,
2013. Page 9. Bellingham, WA.
128
Nummela, S., J. G. M. Thewissen, S. Bajpai, T. Hussain, and K.
Kumar. 2007. Sound transmission in archaic and modern whales:
anatomical adaptations for underwater hearing. The Anatomical
Record 290:716–33.
Nuuttila, H. K., R. Meier, P. G. H. Evans, J. R. Turner, J. D. Bennell,
and J. G. Hiddink. 2013. Identifying Foraging Behaviour of Wild
Bottlenose Dolphins (Tursiops truncatus) and Harbour Porpoises
(Phocoena phocoena) with Static Acoustic Dataloggers. Aquatic
Mammals 39:147–161.
Ohman, M. D. 1990. The demographic benefits of diel vertical
migration by zooplankton. Ecological Monographs 60:257–281.
Owen, R. W. 1981. Fronts and Eddies in the Sea: Mechanisms,
Interactions and Biological Effects. Pages 179–233 in R. A.
Longhurst, editor. Analysis of marine Ecosystems. Academic
Press, Burlington, MA.
Palka, D. 1995. Abundance Estimate of the Gulf of Maine Harbor
Porpoise. Rep. Int.Whal. Commn.:27–50.
Palka, D. 1996. Effects of Beaufort Sea State on the Sightability of
Harbor Porpoises in the Gulf of Maine. Rep. Int. Whal. Commn
46:575–582.
Palka, D. L., and P. S. Hammond. 2001. Accounting for responsive
movement in line transect estimates of abundance. Canadian
Journal of Fisheries and Aquatic Sciences 58:777–787.
Palka, D. L., A. J. Read, A. J. Westgate, and D. W. Johnston. 1996.
Summary of current knowledge of harbour porpoises in the US
and Canadian Atlantic waters. Rep. Int. Whal. Commn. 46:559–
565.
Patterson, I. a, R. J. Reid, B. Wilson, K. Grellier, H. M. Ross, and P. M.
Thompson. 1998. Evidence for infanticide in bottlenose dolphins:
an explanation for violent interactions with harbour porpoises?
Proceedings. Biological sciences / The Royal Society 265:1167–70.
Pauly, D., A. W. Trites, E. Capuli, and V. Christensen. 1998. Diet
composition and trophic levels of marine mammals. ICES Journal
of Marine Science 55:467–481.
129
Phillips, D. L., and P. M. Eldridge. 2006. Estimating the timing of diet
shifts using stable isotopes. Oecologia 147:195–203.
Pierce, G. J., P. R. Boyle, and J. S. W. Diack. 1991. Identification of
Fish Otoliths and Bones in Faeces and Digestive Tracts of Seals.
Journal of Zoology 224:320–328.
Polacheck, T., and L. Thorpe. 1990. The swimming direction fo harbor
porpoise in relationship to a survey vessel. Rep. Int.Whal. Commn.
40:463–470.
Raum-Suryan, K. L. 1995. Distribution, abundance, habitat use, and
respriation patterns of harbor porpoise (Phocoena phocoena) of the
northern SanJuan Islands, Washington. San Jose State
University.
Raum-Suryan, K. L., and J. T. Harvey. 1998. Distribution and
Abundance of and Habitat Use by Harbor Porpoise, Phocoena
phocoena, off the Northern San Juan Islands, Washington.
Fisheries Bulletin 96:808–822.
Read, A. J. 1990. Age at Sexual Maturity an Pregnanacy Rates of
Harbour Porpoises Phocoena phocoena from the Bay of Fundy.
Canadian Journal of Fisheries and Aquatic Sciences 47:561–565.
Read, A. J. 2009. Porpoises, Overview. Pages 920–923 in W. F. Perrin,
B. Wűrsig, and J. G. M. Thewissen, editors. Encyclopedia of
Marine Mammals. Second. Academic Press, Burlington, MA.
Read, A. J., and D. E. Gaskin. 1985. Radio Tracking the Movements
and Activities of Harbor Porpoises, Phocoena phocoena (L.), in the
Bay of Fundy, Canada. Fishery Bulletin 83:543–552.
Read, A. J., and A. A. Hohn. 1995. Life in the Fast Lane: The Life
History of Harbor Porpoises from the Gulf of Maine. Marine
Mammal Science 11:423–440.
Read, A. J., K. W. Urian, B. Roberts, D. M. Waples, M. L. Burt, and C.
G. M. Paxton. 2012. Occurrence, distribution and density of
marine mammals in Camp Lejeune. Jacksonville, NC.
Reeves, R. R., and G. Notarbartolo di Sciara. 2006. The status and
distribution of cetaceans in the Black Sea and Mediterranean Sea.
Page 142. Monaco.
130
Renaud, D. L., and a N. Popper. 1975. Sound localization by the
bottlenose porpoise Tursiops truncatus. The Journal of
experimental biology 63:569–85.
Rice, D. W. 1998. Marine Mammals of the World: Systemics and
Distribution. Page 231. Marine Mammal Society.
Robertson, A. I., and R. K. Howard. 1978. Diel trophic interactions
between vertically-migrating zooplankton and their fish predators
in an ellgrass community. Marine Biology 48:207–213.
Rojas-Bracho, L., A. Jaramillo-Legoretta, G. Cardenas, E. Nieto, P. L.
de Guevara, B. Taylor, J. Barlow, T. Gerrodette, A. Henry, N.
Tregenza, R. Swift, and T. Akamatsu. 2010. Assessing Trends in
Abundance for Vaquita Using Acoustic Monitoring: Within Refuge
Plan and Outside Refuge Research Needs. Workshop Report Octover 19-23, 2009. Page 43.
Rosel, P. E., A. E. Dizon, and M. G. Haygood. 1995a. Variability of the
Mitochondrial Control Region in the populations of the harbour
porpoise, Phocoena phocoena, on ineroceanic and regional scales.
Canadian Journal of Fisheries and Aquatic Sciences 52:1210–
1219.
Rosel, P. E., M. G. Haygood, and W. F. Perrin. 1995b. Phylogenetic
Relationships among the True Porpoises (Cetacea: Phocoenidae).
Molecular Phylogenetics and Evolution 4:463–474.
Rosel, P., a Frantzis, C. Lockyer, and a Komnenou. 2003. Source of
Aegean Sea harbour porpoises. Marine Ecology Progress Series
247:257–261.
Ross, H. M., and B. Wilson. 1996. Violent Interactions between
Bottlenose Dolphins and Harbour Porpoises. Proceedings of the
Royal Society: Biological Sciences 263:283–286.
Ruilian, Y. U., Y. Xing, Z. Yuanhui, H. U. Gongren, and T. U. Xianglin.
2008. Heavy metal pollution in intertidal sediments from
Quanzhou Bay, China. Journal of Environmental Sciences 20:664–
669.
Saana, I. 2006. Coastal habitat use of harbour porpoise (Phocoena
phocoena) in Cardigan Bay Special Area of Conservation ( Wales ).
University of Jyväskylä.
131
Santos, M. B., and G. J. Pierce. 2003. The Diet of Harbour Porpoise
(Phocoena phocoena) in the Northeast Atlantic. Oceanography and
Marine Biology: An Annual Review 41:355–390.
Scheffer, V. B., and J. W. Slipp. 1948. The whales and dolphins of
Washington State with the key to the cetaceans of the west coast
of North America. The American Midland Naturalist 39:257–337.
Sekiguchi, K. 1995. Occurrence, Behavior and Feeding Habits of
Harbor Porpoises (Phocoena phocoena) at Pajaro Dunes, Monterey
Bay, California. Aquatic Mammals 21:91–103.
Simmonds, M. P., and S. J. Isaac. 2007. The impacts of climate change
on marine mammals: early signs of significant problems. Oryx
41:19.
Simon, M., H. Nuuttila, M. M. Reyes-Zamudio, and F. Ugarte. 2010.
Passive acoustic monitoring of bottlenose dolphin and harbour
porpoise, in Cardigan Bay, Wales, with implications for habitat
use and partitioning. Journal of the Marine Biological Association
of the United Kingdom 90:1539–1545.
Širović, A., J. A. Hildebrand, and S. M. Wiggins. 2007. Blue and fin
whale call source levels and propagation range in the Southern
Ocean. journal of the Acoustical Society of America 122:1208–
1215.
Smeenk, C., M. Leopold, and M. Addink. 1992. Note on the harbour
porpoise Phocoena phocoena in Mauritania, West Africa. Lutra
35:98–104.
Snohomish County PUD. 2014. Tidal | Power Supply |Snohomish
County PUD.
http://www.snopud.com/PowerSupply/tidal.ashx?p=1155.
Sousa-Lima, R. S., T. F. Norris, J. N. Oswald, and D. P. Fernandes.
2013. A Review and Inventory of Fixed Autonomous Recorders for
Passive Acoustic Monitoring of Marine Mammals. Aquatic
Mammals 39:23–53.
Stenson, G. B. 2002. Harbour porpoise (Phocoena phocoena) in the
North Atlantic: Abundance, removals, and sustainability of
removals. NAMMCO Sci. Pub. 5:271–302.
132
Sveegaard, S., and J. Teilmann. 2008. Can satellite telemetry show us
the key habitats for harbour porpoises? Pages 52–57 in P. G. H.
Evans, editor. Workshop at the 21st Annual Meeting of the ECS:
Selection Criteria for Marine Protected Areas for Cetaceans.
European Cetacean Society, San Sebastian, Spain.
Sveegaard, S., J. Teilmann, P. Berggren, K. N. Mouritsen, D. Gillespie,
and J. Tougaard. 2011a. Acoustic surveys confirm the high-density
areas of harbour porpoises found by satellite tracking. ICES
Journal of Marine Science 68:929–936.
Sveegaard, S., J. Teilmann, J. Tougaard, R. Dietz, K. N. Mouritsen, G.
Desportes, and U. Siebert. 2011b. High-density areas for harbor
porpoises (Phocoena phocoena) identified by satellite tracking.
Marine Mammal Science 27:230–246.
Thomas, J. A., and S. R. Fisher. 1986. Use of acoustic techniques in
studying whale behavior. Rep. int. Whal. Commn:121–138.
Thompson, P., S. Ingram, M. Lonergan, S. Northridge, A. Hall, and B.
Wilson. 2007. Climate change causing starvation in harbour
porpoises? Biology letters 3:533–4; discussion 535–6.
Todd, V. L. G., W. D. Pearse, N. C. Tregenza, P. A. Lepper, and I. B.
Todd. 2009. Diel echolocation activity of harbour porpoises
(Phocoena phocoena) around North Sea offshore gas installations.
ICES Journal of Marine Science 66:734–745.
Tollit, D., J. Wood, J. Broome, A. Redden, and S. Mammal. 2011.
Detection of Marine Mammals and Effects Monitoring at the NSPI
(OpenHydro) Turbine Site in the Minas Passage during 2010.
Pages 1–36. Parrsboro, Nova Scotia, Canada.
Tougaard, J., J. Carstensen, J. Teilmann, D.- Roskilde, and H. Skov.
2009. Pile driving zone of responsiveness extends beyond 20 km
for harbor porpoises (Phocoena phocoena ( L .)) ( L ). Journal of the
Acoustical Society of America 126:11–14.
Tregenza, N. C. 2013. CPOD.exe: a guide for users. Chelonia Ltd.
Tregenza, N. J. C. 1998. Site Acoustic Monitoring for Cetaceans - As
Self-Contained Sonar Click Detector. Page 5 Proceedings of the
Seismic and Marine Mammals Workshop. London, UK.
133
Trippel, E. A., M. B. Strong, J. M. Terhune, and J. D. Conway. 1999.
Mitigation of harbour porpoise (Phocoena phocoena) by-catch in
the gillnet fishery in the lower Bay of Fundy. Canadian Journal of
Fisheries and Aquatic Sciences 56:113–123.
Trites, a W., V. Christensen, and D. Pauly. 1997. Competition Between
Fisheries and Marine Mammals for Prey and Primary Production
in the Pacific Ocean. Journal of Northwest Atlantic Fishery
Science 22:173–187.
Verfusß, U. K., L. a Miller, and H.-U. Schnitzler. 2005. Spatial
Orientation in Echolocating Harbour Porpoises (Phocoena
phocoena). The Journal of Experimental Biology 208:3385–3394.
Villadsgaard, A., M. Wahlberg, and J. Tougaard. 2007. Echolocation
signals of wild harbour porpoises, </i>Phocoena phocoena</i>. The
Journal of Experimental Biology 210:56–64.
Waddell, V. G., M. C. Milinkovitch, M. Bérubé, and M. J. Stanhope.
2000. Molecular phylogenetic examination of the delphinoidea
trichotomy: congruent evidence from three nuclear loci indicates
that porpoises (Phocoenidae) share a more recent common
ancestry with white whales (Monodontidae) than they do with true
dolphins (Delp. Molecular Phylogenetics and Evolution 15:314–
318.
Walker, W. A., M. B. Hanson, R. W. Baird, and T. J. Guenther. 1998.
Food Habits of the Harbour Porpoise, Phocoena phocoena, and the
Dall’s Porpoise, Phocoenoides dalli, in the Inland Waters of British
Columbia and Washington. AFSC Processed Report 98-10. Pages
63–75.
Weijs, L., K. Das, U. Siebert, N. van Elk, T. Jauniaux, H. Neels, R.
Blust, and A. Covaci. 2009. Concentrations of chlorinated and
brominated contaminants and their metabolites in serum of
harbour seals and harbour porpoises. Environment international
35:842–50.
West, J. E. 1997. Protection and Restoration of Marine Life in the
Inland Waters of Washington State. Page 144. Olympia, WA.
Westgate, A. J., A. J. Read, P. Berggren, H. N. Koopman, and D. E.
Gaskin. 1995. Diving behaviour of harbour porpoises, Phocoena
phocoena. Canadian Journal of Fisheries and Aquatic Sciences
52:1064–1073.
134
Wiggins, S., J. Manley, E. Brager, and B. Woolhiser. 2010. Monitoring
marine mammal acoustics using Wave Glider. Pages 1,4,20–23
OCEANS 2010.
Willis, P. M., B. J. Crespi, L. M. Dill, R. W. Baird, and M. B. Hanson.
2004. Natural hybridization between Dall ’ s porpoises
(Phocoenoides dalli) and harbour porpoises (Phocoena phocoena).
Canadian Journal of Zoology 82:828–834.
Würsig, B., F. Cipriano, and M. Würsig. 1991. Dolphin Movement
Patterns: Information from Radio and Theodolite Tracking
Studies. Pages 79–111 in K. Pryor and K. S. Norris, editors.
Dolphin Societies - Discoveries and Puzzles. University of
California Press, Berkeley, CA.
Würsig, B., E. M. Dorsey, M. A. Fraker, R. S. Payne, and W. J.
Richardson. 1985. Behavior of bowhead whales, Balaena
mysticetus, summering in the Beaufort Sea: a description.
Fisheries Bulletin 83:357–377.
Zimmer, W. M. X. 2011. Passive Acoustic Monitoring of Cetaceans.
Cambridge University Press, Cambridge, UK.
135
