ArticlePDF Available

The effects of marine traffic on the behaviour of Black Sea harbour porpoises (Phocoena phocoena relicta) within the Istanbul Strait, Turkey

Authors:
  • Marine Mammals Research Association, Turkey
  • Aarhus Institute of Advanced Studies

Abstract and Figures

Marine traffic is threatening cetaceans on a local and global scale. The Istanbul Strait is one of the busiest waterways, with up to 2,500 vessels present daily. This is the first study to assess the magnitude of short- and long-term behavioural changes of the endangered Black Sea harbour porpoises (Phocoena phocoena relicta) in the presence of marine vessels within the Istanbul Strait. Markov chains were used to investigate the effect of vessel presence on the transition probability between behavioural states (diving, surface-feeding and travelling), and to quantify the effect on the behavioural budget and bout length (duration of time spent in a given state) of porpoises. Further, the changes on swimming directions of porpoises in relation to vessel speed and distance was investigated using generalized linear models. In vessel presence, porpoises were less likely to remain in a given behavioural state and instead more likely to switch to another state. Because of this, the bout length of all three behavioural states decreased significantly in the presence of vessels. The vessel effect was sufficiently large to alter the behavioural budget, with surface-feeding decreasing significantly in the presence of vessels. However, when taking into account the proportion of time that porpoises were exposed to vessels (i.e. 50%), the measured effect size was not large enough to significantly alter the animals’ cumulative (diurnal) behavioural budget. Additionally, vessel speed and distance had a significant effect on the probability of porpoises showing a response in their swimming directions. The southern and middle sections of the Istanbul Strait, which have the heaviest marine traffic pressure, had the lowest porpoise sightings throughout the year. Conversely, northern sections that were exposed to a lesser degree of marine traffic hold the highest porpoise sightings. The effect shown in this study in combination with increasing human impacts within the northern sections should be considered carefully and species-specific conservation actions, including establishment of protected areas, should be put in place to prevent the long-term consequences of marine traffic on the Black Sea harbour porpoise population
Content may be subject to copyright.
RESEARCH ARTICLE
The effects of marine traffic on the behaviour
of Black Sea harbour porpoises (Phocoena
phocoena relicta) within the Istanbul Strait,
Turkey
Aylin Akkaya Bas
1,2,3
*, Fredrik Christiansen
4
, Ayaka Amaha O
¨ztu¨rk
1,2
, Bayram O
¨ztu¨rk
1,2
,
Caley McIntosh
3
1Faculty of Fisheries, Istanbul University, Beyazit, Istanbul, Turkey, 2Turkish Marine Research Foundation,
Beykoz, Istanbul, Turkey, 3Marine Mammals Research Association, Antalya, Turkey, 4Cetacean Research
Unit, School of Veterinary and Life Sciences, Murdoch University, Murdoch, Western Australia, 6150 Australia
*akkayaaylinn@gmail.com
Abstract
Marine traffic is threatening cetaceans on a local and global scale. The Istanbul Strait is one
of the busiest waterways, with up to 2,500 vessels present daily. This is the first study to
assess the magnitude of short- and long-term behavioural changes of the endangered Black
Sea harbour porpoises (Phocoena phocoena relicta) in the presence of marine vessels within
the Istanbul Strait. Markov chains were used to investigate the effect of vessel presence on
the transition probability between behavioural states (diving, surface-feeding and travelling),
and to quantify the effect on the behavioural budget and bout length (duration of time spent in
a given state) of porpoises. Further, the changes on swimming directions of porpoises in rela-
tion to vessel speed and distance was investigated using generalized linear models. In vessel
presence, porpoises were less likely to remain in a given behavioural state and instead more
likely to switch to another state. Because of this, the bout length of all three behavioural states
decreased significantly in the presence of vessels. The vessel effect was sufficiently large to
alter the behavioural budget, with surface-feeding decreasing significantly in the presence
of vessels. However, when taking into account the proportion of time that porpoises were
exposed to vessels (i.e. 50%), the measured effect size was not large enough to significantly
alter the animals’ cumulative (diurnal) behavioural budget. Additionally, vessel speed and
distance had a significant effect on the probability of porpoises showing a response in their
swimming directions. The southern and middle sections of the Istanbul Strait, which have the
heaviest marine traffic pressure, had the lowest porpoise sightings throughout the year. Con-
versely, northern sections that were exposed to a lesser degree of marine traffic hold the
highest porpoise sightings. The effect shown in this study in combination with increasing
human impacts within the northern sections should be considered carefully and species-spe-
cific conservation actions, including establishment of protected areas, should be put in place
to prevent the long-term consequences of marine traffic on the Black Sea harbour porpoise
population.
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 1 / 20
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Akkaya Bas A, Christiansen F, Amaha
O
¨ztu¨rk A, O
¨ztu¨rk B, McIntosh C (2017) The effects
of marine traffic on the behaviour of Black Sea
harbour porpoises (Phocoena phocoena relicta)
within the Istanbul Strait, Turkey. PLoS ONE 12(3):
e0172970. doi:10.1371/journal.pone.0172970
Editor: Songhai Li, Institute of Deep-sea Science
and Engineering, Chinese Academy of Sciences,
CHINA
Received: July 19, 2016
Accepted: February 13, 2017
Published: March 15, 2017
Copyright: ©2017 Akkaya Bas et al. This is an
open access article distributed under the terms of
the Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: The original datasets
are under the supplementary files from S1 to S3
Datasets.
Funding: Only part of the named research is
funded by Istanbul University with a grant number
of 18476 which was part of a PhD thesis named
"Investigation on the interactions between
cetaceans and marine traffic in the Istanbul Strait".
However there was no additional external funding
received for this study.
Introduction
The Black Sea harbour porpoise (Phocoena phocoena relicta) is recognised as a subspecies of
the harbour porpoise (P.phocoena). The species is commonly found in shallow waters (0–200
m deep) over the continental shelf around the entire perimeter of the Black Sea, although they
may also occur further offshore within deeper waters [1]. The Black Sea harbour porpoise is
completely isolated from the nearest P.phocoena population in the North Eastern Atlantic [2],
and is endemic to the Black Sea and neighbouring waters. Their full range extends over the Black
Sea, Azov Sea, Kerch Strait, Turkish Straits System, and Northern Aegean Sea [2]. According to
the IUCN Red List of Threatened Species [1], the Black Sea subspecies is at much greater risk of
decline and thus classified as endangered under A1d and 4c,d,e categories. Although the actual
population size present within the Black Sea is unknown, the current population size is believed
to be at least several thousand animals [1,3].
Up until 1983, the main threat to the Black Sea harbour porpoises was unregulated and
uncontrolled harvesting [4]. At present, incidental mortality due to fishing nets represents the
most serious threat, followed by overfishing, habitat loss, and chemical pollution [5]. A mass
mortality event in 1982 in the Azov Sea due to gas explosion, along with two more mortality
events in 1989 and 1990, together with habitat degradation and a decline in the prey availability
(starting from the late 1980s), have also contributed to their listing as endangered [1,3]. Further-
more, vessel-cetacean collisions have been frequently reported throughout the Mediterranean
Sea [6], with a considerable skew towards mysticeti species [7]. However, small cetaceans, such
as harbour porpoises, are also at risk from vessel collisions and have been previously reported
with wounds from fatal boat strikes [8]. Even though there are no reported cases of vessel-por-
poise collisions within the Istanbul Strait, the risk of collision within this high vessel density area
should not be ignored [7].
Intense and increased use of coastal and maritime areas by humans has undoubtedly cre-
ated environmental pressure within the Turkish Straits System and Black Sea [9]. Anthropo-
genic impacts to marine life are particularly severe due to the semi-enclosed nature of the area
[10,11]. Potential effects of marine traffic on cetaceans in the Istanbul Strait and Black Sea
have been cited by a few studies [5,12,13]. These studies have stated that high marine traffic
can disrupt cetaceans within the Istanbul Strait, Black Sea and Azov Sea; however, no further
research has since been conducted to investigate the impact of marine traffic on cetaceans,
including Black Sea harbour porpoises.
Multiple studies have reported both short-term and long-term behavioural changes for sev-
eral species of cetaceans in response to increasing marine vessel pressure [1427]. Short-term
changes can manifest themselves as behavioural changes, including variations in vocalisation,
an increase in dive intervals, vertical and horizontal avoidance, and an increase in swimming
speed and a decrease in resting behaviour [25]. Long-term changes can involve population
decline and/or abandonment of an affected habitat [28,29]. Lusseau [30] noted that beha-
vioural budgets of a population can be directly related to their energy budget. Thus, changes in
an animal’s behavioural budget over extended periods of time can result in energy depletion
for that individual [30]. If a sufficiently large proportion of the population is affected, such
energetic effects can eventually lead to long-term negative effects on the population [30,31].
The Istanbul Strait (41˚13’–41˚00’ N, 29˚08’–28˚59’ E) is situated between the Black Sea
and the Mediterranean Sea. Although the Strait is an important habitat for marine life, it also
renders important economic value for commuting, shipping, fishing, and recreational activi-
ties. Commercial cargo vessels, ferries, sea buses, speed boats, and industrial and artisanal fish-
eries are common within this area, resulting in dense marine traffic. When the Montreux
Agreement was signed in 1936, the number of commercial vessels passing through the Istanbul
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 2 / 20
Competing interests: The authors have declared
that no competing interests exist.
Strait was approximately 4,500 per year, in comparison to the 46,000 vessels passing through
the Strait annually today [32]. Official statistics have reported that, on average, 130 commercial
cargo vessels and 2,500 domestic vessels pass through the Strait every day [3234].
Impact studies in other geographical locations suggest that minimising boat-cetacean inter-
actions is an important element in management of anthropogenic impacts on cetaceans. Thus,
decreasing boat pressure is vital for the protection of a species, specifically in their critical habi-
tats [35,36]. However, a sustainable management strategy requires an in-depth knowledge
and understanding of the targeted species and its vulnerability to marine traffic. In order to
establish this understanding, a behavioural impact study of marine traffic on the target species
is needed [15]. In this study, we investigated the effect of vessel traffic on the behaviour of
Black Sea harbour porpoises in the Istanbul Strait. We first compare the behavioural transition
probabilities of porpoises during impact (vessels present) and control (no vessel present) situa-
tions using Markov chain analysis, and the effect of vessel traffic on the behavioural budget
and bout duration of porpoises. Further, we tested the effect of vessel speed and distance on
the probability of changes in swimming direction of porpoises to better understand what fac-
tors might be driving their behavioural responses.
Materials and methods
Data collection
Survey platforms. Porpoise and vessel data were collected by weekly systematic land and
boat surveys between September 2011 and September 2013. Land surveys were conducted
from seven theodolite stations within four different sections of the Istanbul Strait (Fig 1). The
permission to use Ahırkapı Lighthouse has been issued by Directorate General of Coastal
Safety, while General Directorate of Cultural Heritage and Museums has issued the permis-
sions for Rumeli Castle and Hidiv Kasrı. For the rest of the observation stations, no specific
permission was required as they were accessible to the public. Each station was visited on at
least two different days each month with a daily average of 5 hours. Theodolite stations were
selected along the coastline at least 30m above the sea level. Reference points and the exact
positioning of the theodolite placement were kept constant throughout the study. The location
and behaviour of harbour porpoises and marine vessels were recorded using a theodolite
linked to the tracking software Pythagoras v. 1.2 to transform theodolite readings into geo-
graphic positions. When vessels and cetaceans were present together, coordinate points were
recorded for the vessels and the focal group alternately.
Boat-based observations covered the entire strait and were conducted on three different
days per month, independent of land survey days. A 16m gullet boat with a 185 horsepower
engine was utilised throughout the surveys. The boat was operated along pre-determined tran-
sect lines at a speed of around 4knots. Focal porpoise groups were typically followed at a dis-
tance of 50 to 400m from the side or rear. If an individual happened to approach the research
vessel closer, speed was gradually reduced until ‘idle speed’ was reached, and any sudden
movements of the vessel were avoided in order to minimise the impact of the researchers on
the animals. Any changes on the swimming direction of the focal group due to the approach
and presence of the research vessel were recorded. All sightings and effort data, as well as envi-
ronmental and survey conditions, were recorded during both land-based and boat-based
surveys.
Behavioural sampling. Group focal follow was conducted in order to determine the pre-
dominant behaviour of the harbour porpoises, i.e. the behavioural state in which >50% of the
porpoises in a group are engaged in. A group was defined as individuals engaging in similar
behaviors, with close-group cohesion (less than 50m). The behavioural state of the focal group
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 3 / 20
was sampled every 3 minutes using scan sampling methods. Behavioural states were identified
as ‘travelling’, ‘diving’, ‘surface- feeding’, ‘milling’, ‘resting’, and ‘socialising’ (Table 1) [15,25,
37,38,39]. Later, milling, resting and socializing behaviour were discarded from the analysis
due to their small sample size.
Fig 1. Sections and observation stations in the Istanbul Strait, Turkey (coverage of land-based survey is shown in light gray).
doi:10.1371/journal.pone.0172970.g001
Table 1. Definition of each behavioural state of porpoises used in this study.
Behavioural State Definition
Travelling (TR) Porpoises engage in directional movement, and make noticeable headway with
constant speed. Dive intervals are relatively short (15 sec).
Diving (DV) Coordinated, steep dives are seen in various directions. No obvious, steady
movements are recorded. Possibly linked to foraging activity.
Surface-feeding
(SU-FE)
Porpoises chase fish, majority of the behaviour takes place close to the sea
surface with rapid directional changes. Prey often observed at the sea surface,
along with ripples.
Milling (MI) Non-directional movement and frequent changes in bearing. Although the group
movement varies, group cohesion stays similar.
Resting (RE) Porpoises observed within a tight group (5m) with synchronous and steady
movements and swimming speed is low (1knot) with short dive intervals (15
sec).
Socializing (SOC) Diverse interactive events (i.e. body contacts, tail slaps, synchronise full leaps).
Aerial behavioural events are frequently observed with varied dive intervals.
doi:10.1371/journal.pone.0172970.t001
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 4 / 20
Sampling of sequential behavioural states was dependent upon the conspicuity of the
group. To illustrate, if the group was not visible in the 3 minutes after the original sampling
time, their next sighting was recorded and the sampling interval restarted with the time of the
latter sighting. If the focal group was out of sight for more than 20 minutes, the next sighting
was declared as a new group. In the case of multiple groups present at the same time, only the
first sighted group behaviour was noted and the rest of the groups were ignored.
Changes in swimming direction of porpoises in relation to the nearest vessel type were
recorded for each behavioural sampling unit and categorised as either: (a) response–when por-
poises swam away or towards a vessel, or; (b) no response–when porpoises kept a constant
direction despite vessel presence.
Marine vessel sampling. Three separate marine vessel datasets were collected during the
surveys, including (1) Marine vessel type and density: the number of marine vessels, according
to type, was counted every 10 minutes in order to estimate the marine vessel density of each
station. This type of data was collected only during land surveys, and separately from porpoise
sightings. Vessels were divided into 9 different categories; HSB (high-speed boat), FB (fishing
boat, <10m in length), FV (fishing vessel, >10m in length, usually equipped with a sonar sys-
tem), RB (research boat), FE (ferry), SB (sea bus), SCS (small commercial cargo, <200m in
length), BCS (big commercial cargo, >200m in length) and IDLE (idle speed of all the above
vessels). (2) Nearest marine vessel type to the focal group: The nearest vessel to the porpoises
was recorded for each behavioural sample in order to assess the possible impact of the nearest
vessels on swimming directional changes. The accurate distance between the nearest vessels
and the focal group was measured using Pythagoras linked to the theodolite. The nearest vessel
data from boat surveys was discarded, as the distance was estimation. Marine vessels were
placed within one of three speed categories: (a) slow vessels–idle speed up to 3knots; (b)
medium vessels– 3knots to 9knots, and; (c) fast vessels– 9knots and upwards. (3) The number
of vessels within 400m and 1,000m of the porpoises, were counted for each behavioural sam-
pling unit during land surveys.
Behavioural transitions. The number of transitions between different behavioural states
were used to create two-way contingency tables between preceding (the behavioural state
recorded at time tminutes) and succeeding (the behavioural state recorded at time t + 3 min-
utes) states during control and impact situations [15]. If no vessels were recorded for a continu-
ous period of 15 minutes between the preceding (P) and succeeding (F) behaviour, the transition
was added to the control table. If marine vessels were present within 400m of the focal group, the
transition between preceding and succeeding was added to the impact table [15]. Only focal fol-
lows containing a minimum of three transitions, during both land and boat surveys, were
included in analyses. Although the control chain represents no marine vessel presence within
the 400m zone, it was highly likely that vessels were, in fact, present beyond this distance.
Statistical analysis
Sightings. To understand the effect of seasons, sections and survey type on porpoise sight-
ings, a Poisson regression was fitted to the data. However, due to the over dispersion of the
data, negative binomial with loglink was the selected model type. While the count data of por-
poise sightings was used as the response variable, seasons (spring, summer, autumn, winter),
sections (south, middle, middle-north and north) and survey types (land and boat surveys)
were used as explanatory variables, and the survey effort in days was selected as an offset (S1
Dataset).
Markov chain and model selection on behavioural transitions. Time-discrete Markov
Chain analyses are widely applied technique to quantify the one-way dependence of an event
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 5 / 20
on the preceding event which allows the possible effect of any factor on the dependence of the
events to be assessed [15,25,30,3942]. Therefore, a contingency table (four seasons vs. four
sections vs. two marine vessel states vs. three preceding behaviours vs. three succeeding behav-
iours) was created by merging the control and impact chain for all seasons and sections.
Marine vessel (M), season (S) and section (L to avoid abbreviation confusion) effects on the
first order behavioural transitions from preceding (P) to succeeding (F to avoid abbreviation
confusion) were assessed using a log-linear analysis, as described in detail by Lusseau [15,30]
and Lusseau et al. [41]. While the model’s null hypothesis stated that succeeding behaviours
were independent of marine vessels, seasons and sections, given the preceding behaviour,
coded as PF,MSL, the fully saturated model (coded as MSLPF) stated that succeeding behav-
iours were dependent on all possible interactions of seasons, sections and vessels. Starting with
the null model, each factor was added to the initial model one by one until the saturated model
was reached. The significance of each added factor was tested by comparing the goodness-of-
fit of the initial model against its later model [15]. The best fitting model on the explanation of
behavioural transitions was selected based on their Akaike Information Criterion (AIC) [15].
Behavioural transition probabilities. Behavioural transition probability matrices were
developed by calculating transition probabilities (from preceding to succeeding behavioural
state) for both the impact and control chain [15]:
pij ¼aij
X
3
j¼1
aij
;Xpij ¼1
where pis the transition probability between preceding behavioural state iand the succeeding
behavioural state j(iand jrange from 1 to 3 due to the 3 behavioural states), and aij is the
number of transitions observed from behavioural state ito j[15]. To test the effect of vessel
interaction on the transition probability of porpoises, impact and control chains were com-
pared using a chi-square test where the observed number of transitions corresponded to the
impact contingency table and the expected number of transitions corresponded to the control
contingency table [15,42]. In addition, each control transition was compared to its corre-
sponding impact transition (33 = 9 in total), using a 2-sample test for equality of proportions
with continuity correction (S1 File) [15,42,43].
Behavioural budgets. To investigate the effect of vessel presence on the behavioural bud-
get (the proportion of time spent in different behavioural states), left eigenvectors of the domi-
nant eigenvalues of the transition matrices were calculated both for control and impact
matrices [15,30]. Due to the ergodic nature of the Markov chains, initial behavioural states
can converge toward a stationary behavioural distribution, which is proportional to left eigen-
vectors and corresponds to the behavioural budget of the population [15,30]. The differences
between the control and impact behavioural budgets were tested using a chi-square test [15,
42,43]. Each behavioural state within the control behavioural budget was compared to the cor-
responding behavioural state within the impact behavioural budget by using a 2-sample test
for equality of proportions with continuity correction. The 95% confidence intervals were cal-
culated for the estimated proportion of time spent within each behavioural state (S1 File) [15,
42].
Bout lengths. Average bout lengths (the duration of time spent in a given state) of each
behavioural states
tii were estimated for both the control and impact chain, as described by
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 6 / 20
Lusseau [15,30];
tii ¼1
1pii
with a standard error of SE ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
piið1piiÞ
ni
q
where n
i
is the number of samples with ias preceding behaviour. Later, bout lenghts were
compared between the control and impact situation using a Student’s t-test (S1 File).
Cumulative behavioural budgets. A cumulative behavioural budget is able to account for
the time porpoises spend in both the control and impact behavioural budgets. By artificially vary-
ing the proportion of time that porpoises spend with vessels per day from 0 to 100%, it is possible
to see at what level of vessel intensity the cumulative behavioural budget becomes significantly
different from the control budget, given the observed effect size (the estimated behavioural bud-
gets) and assuming that such effect size does not vary with the daytime exposure rate [30,39,42].
The effect of vessels on the daytime behavioural budget of porpoises can be investigated by com-
paring the cumulative behavioural budget with the control budget. The cumulative behavioural
budget was calculated following Lusseau [30] and Christiansen et al. [42]:
Cumulative budget ¼ ðaimpact budgetÞ þ ðbcontrol budgetÞ
where ais representative of the proportion of time porpoises spend with a marine vessel, and bis
the remaining proportion of time (1-a) spent without vessels. The difference between the cumula-
tive behavioural budget and the control budget was tested with a chi-square test and 2-sample
test for equality of proportions with continuity correction for each behavioural state (S1 File) [42,
43].
Changes in swimming direction. To investigate which vessel-related variable affects the
directional response (response vs. no response) of porpoises, a generalized linear model (GLM)
with a binomial distribution (response as a binary variable) and a logit link function were fitted
to the data collected during land surveys. The covariates investigated were distance to the near-
est vessel, the speed category of the nearest vessel (slow, medium and fast), the number of vessels
within 400m and the number of vessels within 1,000m of the porpoises. To account for tempo-
ral auto-correlation within follows, and uneven sample sizes between follows, only the first two
data point from each follow was used in the analyses. Collinearity (high correlation) between
the explanatory variables in the final model was investigated by estimating the variance inflation
factor (VIF), with an upper threshold value of three indicating collinearity. Overdispersion was
tested by dividing the residual deviance by the residual degrees of freedom, with a ratio value
(dispersion parameter, φ) above one indicating overdispersion (the mean of the variance is
larger than the mean). The best fitting model was selected using AIC (S2 File).
The level of significance for all of the above analyses was selected under 0.05 thresholds
with a 95% of confidence interval. Statistical analyses were performed using the statistical soft-
ware SPSS 20 and R 3.1.1 [44].
Results
Sightings
A total of 365 days (1928 hours) were spent searching for porpoises throughout the Istanbul
Strait. Of these, 57 days were spent at sea and 308 days on land. In total, 477 focal group follows
were undertaken over 114 days (70.6 hours), with 29 days (12 hours) being conducted from the
research vessel and 85 days (58.6 hours) from land. A group follow ranged from one sampling
unit (3 minutes) to 31 unit (93 min), with an average of 4.23 sampling units. Over the course of
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 7 / 20
the study, the behavioural data of the porpoises was recorded within 1,403 cases of scan samples,
which later corresponded to 658 behavioural transitions. Of these transitions, 364 were classi-
fied under the control chain and 294 were classified under the impact chain (S1 and S2 Tables).
Regarding the changes on the porpoise sightings based on the seasons, sections and survey
types, survey type had no significant effect (χ
2
= 0, df = 1, p = 0.99). However season (χ
2
= 22.64,
df = 3, p<0.0001) and section (χ
2
= 11.316, df = 3, p = 0.02) showed a significant effect on the
sightings. Sightings within the north section were 3.67 times higher than the south section. The
north section had the highest sightings during all seasons (winter = 3.29; spring = 3.18; and sum-
mer = 1.25 groups per survey) except autumn (0.12 groups per survey). In autumn, sightings
across all sections were below 0.5 groups per survey (Table 2,Fig 2). The south and middle sec-
tions held the lowest sightings all year round, with an average of 0.35 and 0.54 group per survey,
respectivelly.
Porpoises spent 49.6% of overall observation time (boat+land surveys) within the 400m
radius of marine vessels in the Istanbul Strait. Up to 56 vessels were recorded within 1 km of a
porpoise group, with a mean of 1.87 vessels ±0.09 SE. Regardless of porpoise presence, the 10
minute sampling interval of marine vessel data revealed an estimation of 301,247 marine ves-
sels present throughout the study period (between 2011 and 2013). The highest marine vessel
density (210,963 vessels or 70% of the total traffic) was recorded within the middle section, fol-
lowed by the south section (38,263 vessels or 13% of the total traffic), the north section (35,482
vessels or 12% of the total traffic) and the middle-north section (16,483 vessels or 5% of the
total traffic). Ferries were the most dominant vessel class in all sections, except the north sec-
tion where fishing boats were dominant vessel class. Ferries were responsible for 70% (211,444
vessels) of the total marine traffic (Fig 3). However, the majority of porpoise-vessel encounters
were recorded with cargo ships, along with the research boat (Fig 3).
Markov chain and model selection on behavioural transitions
Log-linear analysis showed that "Marine vessel (MPF, MSLP)" and "marine vessel and section
(MPF, LPF, MSLP)" models were the most supported model based on their lowest AIC values
on the variance of behavioural transitions (Fig 4). Neither the saturated model (which
Table 2. Porpoise sightings per seasons and sections within the Istanbul Strait.
Season Section Encounter in days Total group number. Survey effort in days
SPRING South 1 1 23
Middle 7 20 24
Middle-north 4 12 13
North 22 140 44
SUMMER South 7 9 22
Middle 6 6 31
Middle-north 5 16 13
North 21 65 52
AUTUMN South 2 2 30
Middle 3 6 27
Middle-north 1 3 10
North 5 5 43
WINTER South 8 26 29
Middle 8 21 23
Middle-north 4 10 11
North 20 135 41
doi:10.1371/journal.pone.0172970.t002
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 8 / 20
considers all of the interactions between vessel, season and section (MLSPF)) nor the null
model (which disregards all of the factors (PF)) provided a significant change on the beha-
vioural transitions. Starting with the null model (PF,MSLP), each factor (vessel, season and
Fig 2. Number of porpoise sightings as a function of season and section of the Istanbul Strait.
doi:10.1371/journal.pone.0172970.g002
Fig 3. Overall vessel number for each type that was present within 400m and their overall count during the study within the Istanbul Strait (The
overall count of each marine vessel type was independent of porpoise presence).
doi:10.1371/journal.pone.0172970.g003
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 9 / 20
Fig 4. Model testing for marine vessel (M) presence at 400m, season (S) and section (L) effects on behavioural transitions from preceding (P) to
succeeding (F) using log-linear analyses. Models and their respective goodness-of-fit G
2
statistics, degrees of freedom, and AIC values are shown in the
boxes. Red outlined boxes are the best fitted models. Arrows represent the flow between the models. The added factors and their significance are shown
along the arrows. Asterisks indicate an interaction term between variables (adapted from Lusseau 2003).
doi:10.1371/journal.pone.0172970.g004
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 10 / 20
section) and their interaction term was added to the following model until the saturated model
was reached (Fig 4). While vessel and section had a significant effect on the behavioural transi-
tions in each model, season factor was not significant in the explanation of behavioural transi-
tions (Fig 4).
Behavioural transition probabilities
The Markov chain analysis showed that behavioural transitions significantly changed in the
presence of marine vessels (Goodness-of-fit test, χ
2
= 158.09, df = 4, p<0.0001). Vessel pres-
ence significantly affected six of nine behavioural transitions (Fig 5). Three of the transitions,
Diving to Diving (Z-test = 9.19, p = 0.002, control = 75% 69–80 CI95%, impact 57% 51–63
CI95%,), Travelling to Travelling (Z-test = 26.62, p<0.0001, control = 65% 60–70 CI95%,
impact = 35% 30–41 CI95%) and Surface-feeding to Surface-feeding (Z-test = 4.7, p = 0.03,
control = 48% 42–53 CI95%, impact = 19% 14–24 CI95%), significantly decreased in the pres-
ence of vessels. On the other hand, the probability of changing from Diving to Travelling (Z-
test = 12.76, p<0.0001, control = 21% 17–26 CI95%, impact = 42% 36–48 CI95%), Surface-
feeding to Diving (Z-test = 6.04, p = 0.014, control = 20% 16–25 CI95%, impact = 52% 50–58
CI95%) and Travelling to Diving (Z-test = 33.51, p<0.0001, control = 27%, 23–32 CI95%,
impact = 61% 55–66 CI95%) significantly increased (Fig 6).
Behavioural budgets
In the absence of vessels, porpoises spent most of their time diving, followed by travelling and
surface-feeding (Fig 7). The behavioural budget was significantly affected by the presence of ves-
sels (Goodness of fit test, χ
2
= 14.59, df = 2, p<0.0001). The proportion of surface-feeding was sig-
nificantly lower in the impact budget (Z-test = 10.53, p = 0.001, control = 9%, impact = 2%).
Nonetheless, the proportion of time spent diving (Z-test = 3.13, p = 0.07) and travelling (Z-test =
0.01, p = 0.9) did not differ between control and impact situations.
Bout lengths
The average bout lengths (min.) of all three behavioural states showed a significant decline in
the presence of vessel traffic (Fig 8). The diving bout length was reduced from 12.14±0.1 SE
during control situations to 7.02±0.14 SE during impact situations (Student t-test = 30.512,
df = 278, p<0.0001), while surface-feeding was also reduced from 5.17±0.24 SE to 3.68±0.22
(Student t-test = 5.94, df = 65, p<0.0001). Travelling bout length also significantly decreased
Fig 5. Transition matrices for control chain (C) and impact chain (I). Behavioural states were diving (DV), travelling (TR), and surface-feeding (SU-FE).
The numbers represents probabilities. While the green text shows significant decline in the presence of vessels, the red text shows a significant increase.
doi:10.1371/journal.pone.0172970.g005
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 11 / 20
from 8.55±0.12 during control situations to 4.62±0.11 during impact situations (Student t-
test = 24.257, df = 309, p<0.0001). The diving and travelling bout durations were reduced by
over 40%, along with a 36% decline in surface-feeding bout duration, during vessel presence.
Cumulative behavioural budgets
At the current vessel exposure level (49.6%), the cumulative behavioural budget was not signifi-
cantly different from the control behavioural budget of porpoises (χ
2
= 2.928, df = 2, p = 0.23).
When effects were built linearly, only surface-feeding behaviour demonstrated significant differ-
ences when the vessel exposure reached up to 62% of daytime hours (Fig 9). The diving and travel-
ling states did not show any significant difference between the cumulative and control budget, even
if the porpoises were to spend all their daylight hours in the presence of marine vessels (Fig 9).
Changes in swimming direction
The best fitting GLM showed a significant effect of vessel distance (P<0.001, n = 305) and ves-
sel speed (P<0.001, n = 305) on the response (directional changes) probability of porpoises.
The number of vessels did not affect the response of porpoises towards vessels. The model
Fig 6. Differences in behavioural transitions between the control and impact chain (p
ij(impact)-
p
ij(control)
). The vertical line separates each
preceding behavioural state, while the succeeding behavioural state is represented by bars. Asterisks indicate significant behavioural transitions
(p<0.05).
doi:10.1371/journal.pone.0172970.g006
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 12 / 20
Fig 7. Behavioural budget of control and impact chain. Error bars represent 95% confidence intervals. An asterisk
indicates significant differences between behavioural transitions (p<0.05).
doi:10.1371/journal.pone.0172970.g007
Fig 8. Bout lengths of each behavioural state during the control (white) and impact (gray) situations. Error bars
represent 95% confidence intervals. Asterisks indicate significant behavioural transitions (p<0.05).
doi:10.1371/journal.pone.0172970.g008
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 13 / 20
explained 15.8% of the deviance (pseudo-R2) in the data. There was no collinearity between
the explanatory variables in the best fitting model and no sign of overdispersion (φ= 1.06).
The probability of porpoises showing directional responses to vessels decreased with the dis-
tance to the nearest vessel (logit scale: probability = -0.008, SE = 0.001) (Fig 10). The response
was strongest for fast moving vessels (logit scale: probability = 1.253, SE = 0.404), compared to
medium (logit scale: probability = 0.658, SE = 0.396) and slow moving vessels (logit scale:
probability = -0.381, SE = 0.424) (Fig 10). The effect of distance did not differ between vessels
from different speed categories, thus there was no interaction term in the model. At close dis-
tances (<50m), the response probability was around 40, 65 and 80% for slow, medium and fast
moving vessels, respectively. As the distance to the nearest vessel increased, the probability of
porpoises showing response decreased rapidly, to around 20, 40 and 60% at 100m and around
10, 30 and 40% at 200m, respectively (Fig 10). Beyond 400m, the response probability of por-
poises was less than 10%, irrespective of the speed of the vessel (Fig 10) (S2 Dataset).
Discussion
Assessing the effects of marine vessels on cetaceans have been the focus of many studies over
the past two decades, in response to the global increase of marine traffic [5,1316,21,2628,
30,39,45,46]. Although various studies have focused on the effect of whale and dolphin
watching tourism on cetaceans [31,37,40,42,4749], fewer studies have discussed local
marine traffic and international maritime impact on harbour porpoises [5053]. The current
study revealed that Black Sea harbour porpoises spend half of their daylight time within the
Fig 9. Effect of marine vessels on the cumulative behavioural budget of harbour porpoises during differentlevels
of exposure. The y-axis represents the p-value of the difference between the cumulative behavioural budget and the control
behavioural budget for the three behavioural states (see legend) at different vessel exposure levels. The dashed red line
represent the statistical level of significance (p <0.05). The solid red line indicates the current exposure level of porpoises to
marine vessels in Istanbul Strait.
doi:10.1371/journal.pone.0172970.g009
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 14 / 20
400 m range of marine vessels throughout the Istanbul Strait, and that marine traffic induces
significant changes not only on swimming direction but also on behavioural transitions. In the
close proximity of high speed vessels (<50m), porpoises changed their swimming direction up
to 80% of the observations, yet this percentage dropped to 10% when vessel distance was over
400m. Our results on distance-response relationship are in line with previous studies [13,54
58]. Further, vessel speed might lead to injuries, which is clear that the severity of injuries
caused by an impact is likely to increase with vessel speed [59].
The average time porpoises spent in a behavioural state dropped for all the behaviours in
the presence of vessels. Porpoises also had a reduced probability of remaining in the same
behavioural state. They were more likely to shift their behaviour to diving in the vessel pres-
ence. The behavioural transitions were large enough to affect their behavioural budget, with
surface-feeding showing a noticeable drop in the presence of vessels. However the relative
time that they spent in each state overall did not change enough to alter the activity budget for
diving and travelling.
It is well established that a decrease in surface-feeding behaviour can reduce energy intake
and ultimately cause a long-term behavioural consequences as in reduce an animal’s health,
survival and reproductive success [15,37,49,54]. Even though a significant decrease of surface
feeding in the budget raises concerns, the current vessel exposure was not sufficiently large to
alter the porpoises’ cumulative behavioural budget. Concerning the reliability of our results, all
behavioural transitions occurred at least five times, with the exception of Diving to Surface-
feeding, which only occurred once during impact situation. However because the transition
Fig 10. Probability of porpoises showing a response on their swimming direction towards vessels as a function
of the distance to the nearest vessel for slow (solid line), medium (dashed line) and fast (dotted line) moving
vessels. The lines represent the fitted values of the best fitting generalized linear model. The distribution of distance
values for porpoises showing a response and no response are shown by the top and bottom rug plots, respectively.
n = 305.
doi:10.1371/journal.pone.0172970.g010
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 15 / 20
probability between Diving and Surface-feeding was low both during control (0.04) and
impact situations (0.01), it is unlikely that the low sample size during impact situations would
have significantly influenced our results.
Despite the significant behavioural changes under vessel presence, the cumulative beha-
vioural budget of porpoises wasn’t significantly changed in the current exposure level (50%).
The unchanged cumulative budgets might be linked to the area preference and/or behavioural
adjustments of the animals. Porpoises might be compensating for reduced feeding opportuni-
ties during daytime by feeding more at night, when vessel activity is lower. A passive acoustic
monitoring study in the middle-north section of the Strait detected the most click trains of del-
phinids and porpoises, indeed, at night [60], likely associated with foraging behaviour. How-
ever, further research into the nighttime behaviour throughout the Strait is needed to clarify
the possible diurnal behavioural changes of porpoises.
Regarding the area avoidance behaviour, porpoises in the Istanbul Strait might be able to
reduce their overall exposure to vessels, by spending more time in areas with lower and slower
vessel traffic, represented by the northern sections in the strait. Our study provides evidence
that porpoise sightings were indeed concentrated within the northern sections. The southern
and middle sections had the lowest sightings throughout the year and have the heaviest marine
traffic pressure, characterised by a disproportionally high number of high speed vessels. How-
ever, imperfect visual detectability of porpoises must be taken into account on the accuracy of
area preferences. Seasonal area avoidance behaviour was also recorded, with a sharp decline in
autumn sightings in the north and middle-north section.
Temporal area avoidance of dolphins during the high vessel activities was also documented
in Australia [61]. Autumn in the Istanbul Strait is characterised by the pelagic fish migration
and the start of the industrial fishing season. During this time, the north and middle-north sec-
tions was exposed to heavy fishing vessel pressure, with over 50 fishing vessels (purse seines)
recorded simultaneously in 1km
2
. The south and middle sections are closed to fishing due to
the risk of collision between fishing vessels and daily marine traffic. Although fishing vessel
pressure was absent in the south and middle section, the lack of corresponding increase in
autumn sightings rate indicates a probable lack of movement to these areas. It is possible that
high fishing vessel density elicits a seasonal avoidance response from the entire Istanbul Strait,
even at the expense of foraging during high prey density. Increased and consisting behavioural
compensation on their area replacement and/or seasonal area avoidance, may lead to long-
term energy depletion for affected individuals, thus potentially destabilising the entire popula-
tion. Istanbul Strait serves as the only migration corridor for cetaceans between the Aegean
Sea and the Black Sea [62]. Thus, increasing marine traffic might eventually act as a barrier
between the Black Sea and the Aegean Sea.
Current study provided the first in-depth investigation of the vessel-porpoise interactions
within the Istanbul Strait in order to implement effective and viable conservation actions for
the Black Sea harbour porpoises. Despite it’s one of the busiest waterway of the world, the
Istanbul Strait lacks any kind of conservation and management measures for the porpoises
that are listed as at risk. The proven behavioural transitions and avoidance responses of por-
poises in response to the marine traffic, along with increasing human impacts on the north
and middle-north sections, highlight the need for immediate conservation actions to mitigate
the negative vessel impacts on the porpoise population. Lastly, regular surveys of the local pop-
ulation should be conducted to monitor the behavioural and biological changes under yearly
varying marine traffic in the strait.
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 16 / 20
Conclusion
Behavioural changes demonstrated by Black Sea harbour porpoises were related to marine ves-
sel presence within the Istanbul Strait, and the effect on behavioural budgets is already signifi-
cant. Surface-feeding was the only behaviour significantly affected by vessel presence within
the budget. While slow speed vessels do not evoke a significant change on swimming direc-
tions, high speed vessels not only elicit a strong response, but could also lead to active area
avoidance on a larger spatial scale. There is currently high marine traffic throughout the Istan-
bul Strait, with the same area pinpointed as one of the busiest international waterways, spe-
cies-specific conservation measures and management strategies ought to be put in place
immediately to avoid the long-term biological consequences. Such controls should consider
vessel-free regions for the core zones of harbour porpoise habitats, enforced speed limits,
marine vessel density limitations, and special channels specific for ferries within the Istanbul
Strait.
Supporting information
S1 File. R codes for Markov Chain analysis.
(R)
S2 File. R codes for directional changes on porpoise swimming under the vessel speed, dis-
tance and density.
(R)
S1 Table. Control contingency table.
(XLS)
S2 Table. Impact contingency table.
(XLS)
S1 Dataset. Data on the porpoise sightings.
(XLS)
S2 Dataset. Original data on the swimming directional changes of porpoises.
(XLS)
S3 Dataset. Original data used during Markov Chain analysis and model selections.
(XLS)
Acknowledgments
We would like to thank all the research assistants, fishermen and villagers who helped us
throughout this project with the collection of the data. We also thank Directorate General of
Coastal Safety for permission to use the Ahırkapı Lighthouse. Moreover, we would like to give
a special thank you to Anna M. Meissner for all of her valuable advice, and Ophelie Humphrey
for her time to proof read our manuscript.
Author Contributions
Conceptualization: AAB AAO
¨BO
¨.
Data curation: AAB.
Formal analysis: FC.
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 17 / 20
Funding acquisition: AAB AAO
¨BO
¨.
Investigation: AAB.
Methodology: AAB.
Project administration: AAB AAO
¨BO
¨.
Resources: AAB AAO
¨BO
¨.
Supervision: AAB AAO
¨BO
¨.
Validation: AAB FC AAO
¨BO
¨.
Visualization: AAB FC.
Writing – original draft: AAB FC CM.
Writing – review & editing: AAB FC AAO
¨BO
¨CM.
References
1. Birkun Jr AA, Frantzis A. Phocoena phocoena ssp. relicta. The IUCN Red List of Threatened Species
[Internet]. 2008 [Downloaded on 27 January 2016]. Available from: e.T17030A6737111.
2. Frantzis A, Gordon J, Hassidis G, Komnenou A. The enigma of harbour porpoise presence in the Medi-
terranean Sea. Marine Mammal Science. 2002; 17(4):937–943.
3. Reeves RR, Notarbartolo Di Sciara G. The status and distribution of cetaceans in the Black Sea and
Mediterranean Sea. IUCN Centre for Mediterranean Cooperation, Malaga, Spain. 2006.
4. IWC. Annex L. Report of the Sub-committee on Small Cetaceans. Journal of Cetacean Research and
Management. 2004; 6(Suppl.):315–334.
5. Birkun A Jr. Interaction between cetaceans and fisheries: Black Sea. In: di Sciara G. Notarbartolo (ed.),
Cetaceans of the Mediterranean and Black Seas: State of Knowledge and Conservation Strategies.
ACCOBAMS Secretariat, Monaco. 2002. pp. 98–107.
6. Panigada S, Pesante G, Zanardelli M, Capoulade F, Gannier A, Weinrich MT. Mediterranean fin whales
at risk from fatal ship strikes. Marine Pollution Bulletin. 2006; 52:1287–1298. doi: 10.1016/j.marpolbul.
2006.03.014 PMID: 16712877
7. Laist DW, Knowlton AR, Mead JG, Collet AS, Podesta M. Collisions between ships and whales. Marine
Mammal Science. 2001; 17(1):35–75
8. Van Waerebeek K, Baker AN, Felix F, Iniguez M, Sanino GP, Secchi E, et al. Vessel collisions with
small cetaceans worldwide and with large whales in the Southern Hemisphere: an initial assessment.
Latin American Journal of Aquatic Mammals. 2007; 6(1):43–69
9. European Commission. Commission staff working paper: Executive summary of the impact assess-
ment. European Commission, Brussels, Belgium [Internet]. 2013. Available at: http://ec.europa.eu/
maritimeaffairs/policy/maritime_spatial_planning/documents/swd_2013_64_en.pdf.
10. Hoyt E. Sustainable ecotourism on Atlantic islands, with special reference to whale watching, Marine
Protected Areas and sanctuaries, for cetaceans. Biology & Environment: Proceedings of the Royal Irish
Academy. 2005; 105(3):141–154.
11. Notarbartolo di Sciara G, Birkun Jr A. Conservation needs and strategies. In G. Notarbartolo di Sciara,
ed. Cetaceans of the Mediterranean and Black Seas: State of knowledge and conservation strategies.
A report to the ACCOBAMS Secretariat, Monaco, 2002 Feb; 21 pp.
12. Dede A, Amaha O
¨ztu¨rk A, Tonay MA. Cetacean Surveys in the Istanbul (Bosphorus) Strait In 2006,
22nd Annual Conference of European Cetacean Society, Egmond aan Zee, Hollanda, 10–12 Mart
2008, 22:22.
13. BaşAA, O
¨ztu¨rk AA, O
¨ztu¨rk O
¨ztu¨rk B. Selection of critical habitats for bottlenose dolphins (Tursiops trun-
catus) based on behavioural data, in relation to marine traffic in the Istanbul Strait, Turkey. Marine Mam-
mal Science. 2015; 31:979–997.
14. Nowacek SM, Wells RS, Solow AR. Short-term effects of boat traffic on bottlenose dolphins, Tursiops
truncatus, in Sarasota bay, Florida. Marine Mammal Science. 2001; 17(4):673–688.
15. Lusseau D. Effects of tour boats on the behaviour of bottlenose dolphins: using Markov chains to model
anthropogenic impacts. Conservation Biology. 2003; 17(6):1785–1793.
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 18 / 20
16. Lusseau D. Residency pattern of bottlenose dolphins Tursiops spp. in Milford Sound, New Zealand, is
related to boat traffic. Marine Ecology Progress Series. 2005; 295:265–272.
17. Lusseau D. The short-term behavioural reactions of bottlenose dolphins to interactions with boats in
doubtful sound, New Zealand. Marine Mammal Science. 2006; 22(4):802–818.
18. Williams RM, Lusseau D, Hammond PS. Estimating relative energetic costs of human disturbance to
killer whales (Orcinus orca). Biological Conservation. 2006; 133(3):301–311.
19. Gordon J, Leaper R., Hartley FG, Chappell O. Effects of whale-watching vessels on the surface and
underwater acoustic behaviour of sperm whales off Kaikoura, New Zealand. Science and Research
Series 52. 1992.
20. Barr K, Slooten E. Effects of tourism on dusky dolphins at Kaikoura. International Whaling Commission
Scientific Committee. 1998. SC/50/WW10. 30 pp.
21. Bejder L, Dawson SM, Harraway JA. Responses by Hector’s dolphins to boats and swimmers in Por-
poise Bay, New Zealand.” Marine Mammal Science. 1999; 15(3):738–750.
22. Van Parijs SM, Corkeron PJ. Boat traffic affects the acoustic behaviour of Pacific humpback dolphins,
Sousa chinensis. Journal of the Marine Biological Association of the UK. 2001; 81(03):533–538.
23. Hastie GD, Wilson B, Tufft LH, Thompson PM. Bottlenose dolphins increase breathing synchrony in
response to boat traffic. Marine Mammal Science. 2003; 19(1):74–84
24. Lemon M., Lynch TP, Cato DH, Harcourt RG. Response of travelling bottlenose dolphins (Tursiops
aduncus) to experimental approaches by a powerboat in Jervis Bay, New South Wales, Australia. Bio-
logical Conservation. 2006; 127(4):363–372.
25. Stockin KA, Lusseau D, Binedell V, Wiseman N, Orams MB. Tourismaffects the behavioural budget of
the common dolphin Delphinus sp. in the Hauraki Gulf, New Zealand. Marine Ecology Progress Series.
2008; 355:287–295.
26. Christiansen F, Rasmussen MH, Lusseau D. Inferring activity budgets in wild animals to estimate the
consequences of disturbances. Behavioural Ecology. 2013; 24(6):1415–1425
27. Christiansen F, Lusseau D. Linking behaviour to vital rates to measure the effects of non-lethal distur-
bance on wildlife. Conservation Letters. 2015; 8(6):424–431.
28. Bejder L, Samuels A, Whitehead H, Gales N. Interpreting short-term behavioural responses to distur-
bance within a longitudinal perspective. Animal Behaviour. 2006; 72(5):1149–1158.
29. Currey RJC, Dawson SM, Slooten E. An approach for regional threat assessment under IUCN red list
criteria that is robust to uncertainty: the Fiordland bottlenose dolphins are critically endangered. Biologi-
cal Conservation. 2009; 142(8):1570–1579.
30. Lusseau D. The hidden cost of tourism: detecting long-term effects of tourism using behavioural infor-
mation. Ecology and Society. 2004; 9(1):2.
31. Christiansen F, Lusseau D. Understanding the ecological effects of whale watching on cetaceans.
Pages 177–192 in Higham J, Bejder L, Williams R, editors. Whale-watching, sustainable tourism and
ecological management. Cambridge University Press, Cambridge, UK. 2014.
32. Anonymous. Annual report (in Turkish). T. C. Ulastirma, Denizcilik ve Haberlestirme Bakanlığı [Inter-
net]. 2013. Available at: http://www.kiyiemniyeti.gov.tr.
33. Istanbul Port Authority. Domestic marine traffic guide [Internet]. 2011. Available at: http://www.
istanbullimangovtr/tr/indexphp#.
34. Denizcilik Mu¨steşarlığı TC. Tu¨rk Boğazları Gemi Trafik Yo
¨netim ve Bilgi Sistemi (GTYBS)—Gemi Trafik
Hizmetleri (GTH) Projesi [Internet]. 2005. Available at: htpp://www.denizcilik.gov.tr Turkish.
35. Simmonds M, Dolman S, Weilgart L. Oceans of noise: A WDCS science report. Whale and Dolphin
Conservation Society, Plymouth, MA. 2004.
36. Anonymous. Identification and protection of special areas and particularly sensitive sea areas. IMO
Assembly, 61th Session. MEPC. 2010. 61/9. p 14.
37. Constantine R., Brunton DH, Dennis T. Dolphin-watching tour boats change bottlenose dolphin (Tur-
siops truncatus) behaviour. Biological Conservation. 2004; 117(3):299–307.
38. Neumann DR, Orams MB. Impacts of ecotourism on short-beaked common dolphins (Delphinus del-
phis) in Mercury Bay, New Zealand. Aquatic Mammals. 2006; 32(1):1–9
39. Meissner AM, Christiansen F, Martinez EM, Pawley MDM, Orams MB, Stockin KA. Behavioural effects
of tourism on oceanic common dolphins, Delphinus sp., in New Zealand: the effects of Markov analysis
variations and current tour operator compliance with regulations. PLos ONE 10. 2015.
40. Bejder L, Samuels A. Evaluating impacts of nature-based tourism on cetaceans. Pages 229–256 in
Gales N, Hindell M, Kirkwood R, editors. Marine Mammals: Fisheries, Tourism and Management
Issues. Collingwood: CSIRO. 2003.
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 19 / 20
41. Lusseau D, Bain DE, Williams R, Smith JC. Vessel traffic disrupts the foraging behaviour of southern
resident killer whales Orcinus orca. Endangered Species Research. 2009; 6(3):211–221.
42. Christiansen F, Lusseau D, Stensland E, Berrgren P. Effects of tourist boats on the behaviour of Indo-
Pacific bottlenose dolphins off the south coast of Zanzibar. Endang Species Res. 2010; 11:91–99.
43. Fleiss JL. The measurement of interrater agreement. Statistical methods for rates and proportions.
1981; 2:212–236.
44. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna, Austria. 2014. Available from http://www.r-project.org/
45. Hildebrand J. Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecological Pro-
gess Series. 2009; 395:5–20
46. McDonald MA, Hildebrand JA, Wiggins SM. Increases in deep ocean ambient noise in the Northeast
Pacific west of San Nicolas Island, California. The Journal of the Acoustical Society of America. 2006;
120:711–718. PMID: 16938959
47. Bain DE, Trites AW, Williams R. A model linking energetic effects of whale watching to killer whale (Orci-
nus orca) population dynamics. Friday Harbour Laboratories, University of Washington, Friday Har-
bour, Washington. 2002.
48. Bain DE, Smith JC, Williams R, Lusseau D. Effects of vessels on behaviour of Southern Resident killer
whales (Orcinus spp.) 2003–2005. NMFS Contract report No. AB133F05SE3965. National Marine Fish-
eries Service, National Marine Mammal Laboratory, Seattle, Washington. 2006.
49. Lusseau D, Slooten L, Currey RJ. Unsustainable dolphin-watching tourism in Fiordland, New Zealand”.
Tourism in Marine Environments. 2006; 3(2):173–178.
50. Polacheck T, Thorpe L. The swimming direction of harbour porpoise in relationship to a survey vessel.
Report of the International Whaling Commission. 1990; 40:463–470.
51. Dyndo M, Wisniewska DM, Rojano-Donate L, Madsen PT. Harbour porpoises react to low levels of high
frequency vessel noise. Scientific reports. 2015; 5.
52. Kastelein RA, Steen N, De Jong C, Wensveen PJ, Verboom WC. Effect of broadband-noise masking
on the behavioural response of a harbour porpoise (Phocoena phocoena) to 1-s duration 6–7 kHz sonar
up-sweeps. The Journal of the Acoustical Society of America. 2011; 129(4):2307–2315 doi: 10.1121/1.
3559679 PMID: 21476686
53. Hermannsen L, Beedholm K, Tourgaard J, Madsen PT. High frequency components of ship noise in
shallow water with a discussion of implications for harbour porpoises (Phocoena phocoena). The Jour-
nal of the Acoustical Society of America. 2014; 136(4):1640–1653. doi: 10.1121/1.4893908 PMID:
25324068
54. Dans SL, Crespo EA, Pedraza SN, Degrati M, Garaffo GV. Dusky dolphin and tourist interaction: effect
on diurnal feeding behaviour. Marine Ecology Progress Series. 2008; 369:287–296.
55. Bryant PJ, Lafferty CM, Lafferty SK. Reoccupation of Laguna Guerrero Negro, Baja California, Mexico,
by gray whales. Pages 375–387 in Jones ML, Swartz SL, Leatherwood S, editors. The Gray whale
(Eschrichtius robustus). Academic Press, Orlando, FL, USA. 1984.
56. Richardson WJ, Finley KJ, Miller GW, Davis RA, Koski WR. Feeding, social and migration behaviour of
bowhead whales, Balaena mysticetus, in Baffin Bay vs. the Beaufort Sea—regions with different
amounts of human activity. Marine Mammal Science. 1995; 11(1):1–45.
57. Notarbartolo di Sciara G. The Ligurian Sea international sanctuary for Mediterranean cetaceans: ratio-
nale, history and current status. Proceedings of the workshop Protected Areas for Cetaceans. Euro-
pean Cetaceans Newsletter. 2001; 38:28–30.
58. Aguilar de Soto N, Johnson M, Madsen PT, Tyack PL, Bocconcelli A, Borsani, JF. Does intense ship
noise disrupt foraging in deep-diving Cuvier’s beaked whales (Ziphius cavirostris). Marine Mammal Sci-
ence. 2006; 22(3):690–699.
59. Martin J, Sabatier Q, Gowan TA, Giraud C, Gurarie E, Calleson CS, Ortega-Ortiz JG, Deutsch CJ,
Rycyk A, Koslovsky SM. A Quantitative Framework for Investigating Risk of Deadly Collisions between
Marine Wildlife and Boats. Methods in Ecology and Evolution.2015.
60. Dede A, O
¨ztu¨rk AA, Akamatsu T, Tonay AM, O
¨ztu¨rk B. Long-term passive acoustic monitoring revealed
seasonal and diel patterns of cetacean presence in the Istanbul Strait, Journal of the Marine Biological
Association of the United Kingdom. 2013. 1–8.
61. Bejder L, Samuels A, Whitehead H, Gales N, Mann J, Connor R, Heithaus M, Watson-Capps J, Flaherty
C, Kruetzen M. Decline in relative abundance of bottlenose dolphins exposed to long-term disturbance.
Conservation Biology. 2006: 20(6), 1791–1798. doi: 10.1111/j.1523-1739.2006.00540.x PMID:
17181814
62. O
¨ztu¨rk B, O
¨ztu¨rk AA. On the biology of the Turkish Straits System. Bulletin de Institute Oceanographi-
que, Monaco. CIESM Science Series No 2. 1996. pp 205–221.
Marine traffic effect on the Black Sea harbour porpoises within the Istanbul Strait
PLOS ONE | DOI:10.1371/journal.pone.0172970 March 15, 2017 20 / 20
... To quantify the effect of DWB and PB interactions on the behaviour of Istrian bottlenose dolphins, time-discrete Markov Chain analyses were applied (see Akkaya Bas et al., 2017 for details). Markov Chains have been widely used to quantify the effect of marine vessels on cetacean behaviour in the past (Lusseau 2003;Christiansen et al., 2010;Meissner et al., 2015;Akkaya Bas et al., 2017) and were found to be best suited for this study. ...
... To quantify the effect of DWB and PB interactions on the behaviour of Istrian bottlenose dolphins, time-discrete Markov Chain analyses were applied (see Akkaya Bas et al., 2017 for details). Markov Chains have been widely used to quantify the effect of marine vessels on cetacean behaviour in the past (Lusseau 2003;Christiansen et al., 2010;Meissner et al., 2015;Akkaya Bas et al., 2017) and were found to be best suited for this study. Firstly, using two-way contingency tables, a first-order Markov Chain was used to determine the probability of transitioning from the preceding to the succeeding behavi oural state, in both the control and impact chain as per Lusseau (2003) and Akkaya Bas et al. (2017): ...
... Markov Chains have been widely used to quantify the effect of marine vessels on cetacean behaviour in the past (Lusseau 2003;Christiansen et al., 2010;Meissner et al., 2015;Akkaya Bas et al., 2017) and were found to be best suited for this study. Firstly, using two-way contingency tables, a first-order Markov Chain was used to determine the probability of transitioning from the preceding to the succeeding behavi oural state, in both the control and impact chain as per Lusseau (2003) and Akkaya Bas et al. (2017): ...
Article
Full-text available
Several studies indicate that unregulated nautical tourism can have negative implications on cetacean behaviour. In recent years, dolphin watching activities (DWA) have increased off the West coast of Istria, Croatia, a region in which the NATURA 2000 site: ‘Akvatorij zapadne Istre’ has been proposed to be designated for bottlenose dolphins (Tursiops truncatus M.). For data collected between 2016 and 2019, we compared dolphin group behaviours from this region during impact (presence of nautical tourism boats (NTBs)) and control (absence of NTBs) scenarios, as well as providing descriptive analysis on the displacement of individuals in the presence of NTBs. Throughout the study years, 48.5 % of NTBs were observed within 15m of the dolphin focal groups and 97 % were observed within 50 m distance. The greatest rates of displacement in dolphin focal groups occurred when NTB numbers were greatest per individual dolphin. Markov chain analysis were used to quantify the short-term effects of NTB presence on dolphin behaviour. In the presence of NTBs, dolphins were more likely to spend time milling and less time foraging. Cumulative behavioural budgets, derived by accounting for the time bottlenose dolphins spent in the presence or absence of NTBs, indicated that vessel exposure levels of 14 % and 25 % were enough to statistically affect milling and foraging behaviours respectively. To lessen the lack of sustainable DWA, the implementation of relevant guidelines, e.g. Global Best Practice Guidance for Responsible Whale and Dolphin Watching (50 m no approach and 300 m caution zone) is therefore crucial to mitigate any long-term consequences the actions of NTBs may have on this key species. To date, 162 bottlenose dolphins have been photo-identified off West coast of Istria and cumulative interference to this population could affect direct ecosystem functioning.
... Ships and fisheries threaten cetaceans in these areas either directly through causing stress, by-catch, entanglements, and collisions, or indirectly by affecting their prey and depleting their food stocks. Noise pollution from marine traffic and other anthropogenic sources may alter cetaceans' diving and movement patterns, affect their communication, and deter them from critical habitats [5,6,8,10,14]. ...
... The Istanbul Strait represents one such area where dolphins live in close contact with heavy marine traffic and coastal development [12,14,15]. As a site of local maritime activity, major fisheries operations, and international shipping, it is one of the most heavily trafficked straits in the world [13][14][15][16]. ...
... The Istanbul Strait represents one such area where dolphins live in close contact with heavy marine traffic and coastal development [12,14,15]. As a site of local maritime activity, major fisheries operations, and international shipping, it is one of the most heavily trafficked straits in the world [13][14][15][16]. Thousands of local vessels, including ferries, fishing boats, and private craft, crowd the strait daily [13] as well as larger maritime vessels shipping between the Mediterranean and the Black Sea states. ...
Conference Paper
Common dolphins in the Mediterranean have seen declines in recent decades, with increased fragmentation and patchiness, generally attributed to a range of anthropogenic threats. Despite this, few attempts have been made to identify critical areas of dolphins. One such area with heavy overlap of human threats and cetaceans is the Istanbul Strait. The current study presents the findings of a long-term, continuous survey effort that provides the first fine-scale, seasonal distribution and encounter rates of common dolphins for the Istanbul Strait. Land-and Boat-based surveys were conducted systematically between 2011 and 2013 with opportunistic boats surveys completed in 2020. A custom Python script divided the Strait into 500m 2 grid cells, before compiling the total hours of survey effort in each cell to enable calculation of seasonal encounter rate. Dolphins were distributed throughout the Strait with distribution varying seasonally. The majority of the strait was used by dolphins in the summer, with only the north used in winter. Encounter rate also varied by season, with the highest mean encounter rate across grid cells of 1.9 individuals/hour in the summer and just 0.07 individuals/hour in winter. The results provide baseline seasonal data for the area and identify the northern area of the Strait as important across all seasons. As previous studies on cetaceans have identified vessel type, vessel speed, intensity of marine traffic, and distance from the dolphin group having an effect on cetaceans, protection is recommended in the north with seasonal management suggested in the central region in spring and summer.
... Panel (b) shows the smallest proportion of each network that would have to be affected by human disturbance to increase the average epidemic size when affected individuals have their behavioral or immunological competence increased by 1. 6. Panel (c) shows the smallest number of affected individuals that must be initially exposed and infected with a pathogen to increase the probability of an epidemic to 50%, when competence is increased by 1.6 and 15% of the network is affected by human disturbance. [52,66,67,101,133,145,147,150,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,65,170,171,172,173,174,175,23,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217] Decreased time spent in habitat Higher edge vulnerability (Increased Susceptibility) ...
... Tursiops truncatus, Tursiops aduncus, Phocoena phocoena, Balaenoptera acutorostrata, Arctocephalus pusillus doriferus, Delphinus delphis, Sousa chinensis, Orcinus orca, Platanista gangetica gangetica, Stenella longirostris Belize, Australia, Turkey, Wales, Ireland, South Africa, Hong Kong, New Zealand, India, Hawaii (USA)[69,153,158,160,165,173,176,191,194,218,219] Reorientation or changes to swimming patternsTursiops truncatus, Tursiops aduncus, Ornicus orca, Sousa chinensis, Arctocephalus pusillus, Megaptera novaeangliae, Eubalaena australis, Stenella longirostris, Arctocephalus pusillus doriferus Florida (USA), Canada, Hong Kong, Australia, New Caledonia, Argentina, Ecuador, Hawaii (USA), Argentina, Texas (USA), Tanzania [64, 188, 189, 198, 205, 214, 215, 220, 221, 222, 223] Tursiops anduncus, Eubalaena glacialis, Balaenoptera physalus, Megaptera novaeangliae, (Balaenoptera musculus, Sousa chinensis, Orcaella brevirostris Australia, Canada, Northern Pacific Ocean, Northern Atlantic Ocean, Malaysia, Increased contact or larger group sizes associated with animals engaged in human associated foraging Higher degree (Increased Individual Exposure) Increased behavioral competence Tursiops truncatus, Sousa chinensis, Physeter catodon Georgia (USA), Hong Kong, Florida (USA), Canada Tursiops truncatus, Tursiops aduncus Australia, Georgia (USA), North Carolina (USA), Florida (USA), Brazil, Italy Tursiops truncatus, Zalophus californianus, Ursus maritimus, Phoca vitulina, Phocoena phocoena, Trichechus manatus manatus, Physeter catodon, Globicephala macrorhynchus, Megaptera novaeangliae, Balaenoptera physalus, Balaenoptera musculus, Balaenoptera brydei, Balaenoptera acutorostrata, Odobenus rosmarus, Orcinus orca, Stenella coeruleoalba California (USA), United Kingdom, British Isles, Australia, Gulf of Mexico, Sea of Cortez, Mexico, Spain, Greenland, Portugal ...
Chapter
Full-text available
Humans have been altering wildlife habitatsHabitat and wildlife behavior worldwide at an accelerated pace in recent decades. While it is well-understood how human-induced behavioral changes affect infectious disease risk in terrestrial wildlife, less is known in marine life. Here we examine this link in marine mammalMarine mammals populations by (1) conducting a systematic literature review to determine how human disturbancesDisturbance change marine mammal behavior in ways that can impact disease spread, and (2) using a mathematical modeling framework to examine how these behavioral changes might influence potential epidemics. Human disturbances can influence marine mammal behavior in ways that increase their exposure and susceptibility to pathogens, as well as their infectivity, or ability to effectively shed pathogens and infect conspecifics. When these changes to exposure, susceptibility, and infectivity are applied in four different marine mammal case studies (California sea lionsSea lion, Zalophuscalifornianus; Australian humpback dolphinsHumpback dolphin, Australian, Sousa sahulensisSousa sahulensis; killer whalesKiller whale, Orcinus orcaOrcinus orca; Indo-Pacific bottlenose dolphinsBottlenose dolphin, Indo-Pacific, Tursiops aduncusTursiops aduncus), epidemics are predicted to be larger and more likely to occur. Considering the rate at which human disturbanceDisturbance is increasing in the marine environmentEnvironment and the large number of marine mammal species and populations that are endangeredEndangered or on the verge of extinctionExtinction, we advocate for the careful consideration of the direct and indirect impact of human disturbance on marine mammalMarine mammalshealthHealth.
... By rerouting or reducing vessel speed within these areas, the collision risk and noise pollution for sperm whales and beaked whales could be considerably reduced with minimal inconvenience for the shipping industry (Vanderlaan and Taggart, 2009;Frantzis et al., 2019). While ship strikes may be less of a direct threat to the bottlenose dolphins' population, increased noise and pollution levels may come with direct and indirect consequences to the dolphin populations such as habitat shifts and behavioral alterations (Papale et al., 2012;Akkaya Bas et al., 2017). ...
Article
Full-text available
Marine traffic has been identified as a serious threat to Mediterranean cetaceans with few mitigation strategies in place. With only limited research effort within the Eastern Basin, neither baseline species knowledge nor the magnitude of threats have been comprehensively assessed. Delineating the extent of overlap between marine traffic and cetaceans provides decision makers with important information to facilitate management. The current study employed the first seasonal boat surveys within the Eastern Mediterranean Sea of Turkey, incorporating visual and acoustic survey techniques between 2018 and 2020 to understand the spatial distribution of cetacean species. Additionally, marine traffic density data were retrieved to assess the overlap with marine traffic. Encounter rates of cetaceans and marine traffic density were recorded for each 100 km 2 cell within a grid. Subsequently, encounter and marine traffic density data were used to create a potential risk index to establish where the potential for marine traffic and cetacean overlap was high. Overall, eight surveys were undertaken with a survey coverage of 21,899 km 2 between the Rhodes and Antalya Basins. Deep diving cetaceans (sperm and beaked whales) were detected on 28 occasions, with 166 encounters of delphinids of which bottlenose, striped and common dolphins were visually confirmed. Spatially, delphinids were distributed throughout the survey area but encounter rates for both deep diving cetaceans and delphinids were highest between the Rhodes and Finike Basins. While sperm whales were generally detected around the 1000m contour, delphinids were encountered at varying depths. Overall, two years of monthly marine traffic density were retrieved with an average density of 0.37 hours of monthly vessel activity per square kilometer during the study period. The mean density of vessels was
... The Istanbul Strait is the sea passage located in the Turkish Straits Region and connects the Black Sea to the Maramara. The Strait is the busiest waterway in the world after the Strait of Malacca in terms of the number of ships crossing, and it is the only waterway that stands out with the danger of maritime accidents among its peers on the world maritime trade network (Köse, Başar, Demirci, Güneroǧlu, and Erkebayet, 2003;Rodrigue, 2004;Görçün and Burak, 2015;Akkaya Bas, Christiansen, Amaha Öztürk, Öztürk, and McIntosh, 2017;Altan and Otay, 2017;Korçak and Balas, 2020). Maritime accidents increase in parallel with the increasing ship traffic (Weng et al., 2020;Görçün and Burak, 2015;Ulusçu et al., 2009). ...
Article
Full-text available
The current study predicts the future course of ships passing through the İstanbul Strait. In this direction, considering that the world economy is the biggest factor in the demand for maritime transport, the relationship between the GDPs and trade volumes of the Black Sea states, and ship traffic is analyzed using regression estimation in two separate models. The included countries are Bulgaria, Georgia, Romania, Russia, Turkey and Ukraine. Then, considering the growth forecasts published by the IMF for the relevant states, it was estimated how much an increase in traffic would be in 2026 compared to 2020. Considering the coefficients obtained from the two models, in 2026, the GDP model proposes a 20.02% increase, and the trade volume model proposes a 28.8% increase in ship tonnage passing the strait. These results reveal the importance and necessity of strategies and projects developed to regulate the rise in strait traffic.
... As odontocetes, harbor porpoises rely on sound for orientation, predation and intraspecific communication (Clausen et al., 2010;Wisniewska et al., 2016;Sørensen et al., 2018), which is why they are particularly vulnerable to underwater noise and any hearing impairment should be prevented. Beside impulsive noise events, the global underwater soundscape is largely dominated by shipping noise proven to have negative effects on harbor porpoise behavior (Hermannsen et al., 2014;Dyndo et al., 2015;Akkaya Bas et al., 2017;Wisniewska et al., 2018b). ...
Article
Full-text available
The North Sea is one of the most heavily used shelf regions worldwide with a diversity of human impacts, including shipping, pollution, fisheries, and offshore constructions. These stressors on the environment can have consequences for marine organisms, such as our study species, the harbor porpoise ( Phocoena phocoena ), which is regarded as a sentinel species and hence has a high conservation priority in the European Union (EU). As EU member states are obliged to monitor the population status, the present study aims to estimate trends in absolute harbor porpoise abundance in the German North Sea based on almost two decades of aerial surveys (2002–2019) using line-transect methodology. Furthermore, we were interested in trends in three Natura2000 Special Areas of Conservation (SACs), which include the harbor porpoise as designated feature. Trends were estimated for each SAC and two seasons (spring and summer) as well as the complete area of the German North Sea. For the trend analysis we applied a Bayesian framework to a series of replicated visual surveys, allowing to propagate the error structure of the original abundance estimates to the final trend estimate and designed to deal with spatio-temporal heterogeneity and other sources of uncertainty. In general, harbor porpoise abundance decreased in northern areas and increased in the south, such as in the SAC Borkum Reef Ground. A particularly strong decline with a high probability (94.9%) was detected in the core area and main reproduction site in summer, the SAC Sylt Outer Reef (−3.79% per year). The overall trend for the German North Sea revealed a decrease in harbor porpoise abundance over the whole study period (−1.79% per year) with high probability (95.1%). The assessment of these trends in abundance based on systematic monitoring should now form the basis for adaptive management, especially in the SAC Sylt Outer Reef, where the underlying causes and drivers for the large decline remain unknown and deserve further investigation, also in a regional North Sea wide context.
Article
Full-text available
Cetacean behavior has long attracted scientific attention as humans endeavor to discover what makes these mammals so emotive and engaging. To date, much of this research has focussed on abundant and widely distributed cetacean species such as bottlenose dolphins (Tursiops truncatus) and humpback whales (Megaptera novaeangliae). As an endangered and often evasive species, research regarding Irrawaddy dolphin (Orcaella brevirostris) behavior is limited. This study uses data collected by The Cambodian Marine Mammal Conservation Project, to investigate the behavioral responses of Irrawaddy dolphins towards a dead conspecific. During a routine boat survey of Cambodia's Kep Archipelago, the carcass of an adult female Irrawaddy dolphin was recovered and attached to the stern of the research vessel and promptly towed to the research island for further examination. During this survey, there was a four-fold increase in the number of Irrawaddy dolphin groups observed compared to the seasonal average (post-monsoon), in addition to an atypically positive response towards the research vessel and an atypical increase in the number of behavioral events observed. These behavioral variations were believed to be in response to the towed dead conspecific. The authors propose future dedicated research to assess the complexity of wild Irrawaddy dolphin behavior, cognition, and awareness to robustly exemplify the species' apparent sentience and intelligence.
Article
Full-text available
Speed regulations of watercraft in protected areas are designed to reduce lethal collisions with wildlife but can have economic consequences. We present a quantitative framework for investigating the risk of deadly collisions between boats and wildlife.We apply encounter rate theory to demonstrate how marine mammal-boat encounter rate can be used to predict the expected number of deaths associated with management scenarios. We illustrate our approach with management scenarios for two endangered species: the Florida manatee Trichechus manatus latirostris and the North Atlantic right whale Eubalaena glacialis. We used a Monte Carlo simulation approach to demonstrate the uncertainty that is associated with our estimate of relative mortality.We show that encounter rate increased with vessel speed but that the expected number of encounters varies depending on the boating activities considered. For instance, in a scenario involving manatees and boating activities such as water skiing, the expected number of encounters in a given area (in a fixed time interval) increased with vessel speed. In another scenario in which a vessel made a transit of fixed length the expected number of encounters decreases slightly with boat speed. In both cases the expected number of encounters increased with distanced travelled by the boat. For whales, we found a slight reduction (~0.1%) in the number of encounters under a scenario where speed is unregulated; this reduction, however, is negligible, and overall expected relative mortality was ~30% lower under the scenario with speed regulation. The probability of avoidance by the animal or vessel was set to 0 because of lack of data, but we explored the importance of this parameter on the model predictions. In fact, expected relative mortality under speed regulations decreases even further when the probability of avoidance is a decreasing function of vessel speed.By applying encounter rate theory to the case of boat collisions with marine mammals, we gained new insights about encounter processes between wildlife and watercraft. Our work emphasizes the importance of considering uncertainty when estimating wildlife mortality. Finally, our findings are relevant to other systems and ecological processes involving the encounter between moving agents.This article is protected by copyright. All rights reserved.
Article
Full-text available
Cetaceans rely critically on sound for navigation, foraging and communication and are therefore potentially affected by increasing noise levels from human activities at sea. Shipping is the main contributor of anthropogenic noise underwater, but studies of shipping noise effects have primarily considered baleen whales due to their good hearing at low frequencies, where ships produce most noise power. Conversely, the possible effects of vessel noise on small toothed whales have been largely ignored due to their poor low-frequency hearing. Prompted by recent findings of energy at medium-to high-frequencies in vessel noise, we conducted an exposure study where the behaviour of four porpoises (Phocoena phocoena) in a net-pen was logged while they were exposed to 133 vessel passages. Using a multivariate generalised linear mixed-effects model, we show that low levels of high frequency components in vessel noise elicit strong, stereotyped behavioural responses in porpoises. Such low levels will routinely be experienced by porpoises in the wild at ranges of more than 1000 meters from vessels, suggesting that vessel noise is a, so far, largely overlooked, but substantial source of disturbance in shallow water areas with high densities of both porpoises and vessels.
Article
A ship-based transect survey was conducted for Phocoena phocoena in the W Bay of Fundy during summer 1988. Results suggest that harbor porpoises exhibit avoidance behaviour. -from Authors