ThesisPDF Available

Cumulative Exposure of Sperm Whales to Whale Watching Boats using Spatially-Explicit Capture- Recapture Models

Authors:

Abstract and Figures

There has been a globally growing interest in cetacean-based tourism resulting in increasing cumulative impacts on many wildlife species. In the Azores archipelago, the local whale watching industry has drastically evolved over the last two decades, especially targeting a population of sperm whales that use the habitat as important feeding ground and can be found year-round. Some individuals return annually to the area and stay over several weeks to breed, which makes them especially vulnerable to human-induced disturbance. Hence, this study aimed to contribute to the establishment of a conservation framework to ensure the future sustainability of this industry by including the comprehensive evaluation of cumulative exposure of sperm whales to whale watching boats in the area into the management of human activities. For the first time spatially-explicit capture-recapture (SECR) models have been developed to quantify the cumulative interaction time between photo-identified individual sperm whales and whale watching boats in the area. The study provided baseline estimates of sperm whale encounter probabilities which were integrated together with the whale watching intensity in the area to estimate spatial variations of exposure on individual-level. Model estimates revealed that whale watching activities were mainly concentrated in two distinct areas north of Faial and south of Pico island, consequently exposure levels were found to be significantly higher in respective ‘core’ areas of disturbance. Furthermore, results indicated seasonal variations in daily individual exposure levels with a peak in June, where the most repeated interactions between the same individual and whale watching boats took place. The present findings stress the importance of taking the individual exposure into consideration when it comes to the management of potentially harmful human activities and support a precautionary approach.
Content may be subject to copyright.
University of Hamburg
Faculty of Mathematics, Informatics and Natural Sciences
Department of Biology
Institute for Marine Ecosystem and Fishery Sciences
Cumulative Exposure of Sperm Whales to Whale
Watching Boats using Spatially-Explicit Capture-
Recapture Models
Chiara Mandl
Master thesis submitted to the University of Hamburg to obtain the master’s degree in
Marine Ecosystem and Fisheries Sciences (MARSYS)
!
Horta, March 2020
!
"!
!
#!
Cumulative Exposure of Sperm Whales to Whale
Watching Boats using Spatially-Explicit Capture-
Recapture Models
Chiara Mandl
Under the supervision of:
!
PhD Sergi Pérez-Jorge
Centro do IMAR
Departamento de Oceanografia e Pescas
OKEANOS
Universidade dos Açores
Prof. Dr. Christian Möllmann
Institute for Marine Ecosystem and Fishery Sciences
Department of Biology
University of Hamburg
!
$!
This MSc project was supported by:
!
Research project ACORES-01-0145-FEDER-00057
In collaboration with:
!
%!
Table of contents
Abstract!.............................................................................................................................!7!
Deutsche Zusammenfassung!.............................................................................................!8!
1. Introduction!..................................................................................................................!9!
1.1 The human-wildlife conflict: Main impacts of non-lethal human disturbance on cetacean
species!........................................................................................................................................!9!
1.1.1 Sources of disturbance in marine environments!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!'!
1.1.2 Influence of tourism on wildlife species!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!'!
1.1.3 Short and long-term effects of whale watching tourism on cetaceans!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!()!
1.2 Spatial and temporal variation and how to cope with it: A SECR approach to estimate
cetacean densities!.....................................................................................................................!11!
1.2.1 Heterogeneity over space and time!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!((!
1.2.2 Spatially-explicit mark-recapture models!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!("!
1.3 Shifted perspectives of a long-known industry: A sperm whale case study!.......................!13!
1.3.1 The development of whale watching on the Azores!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!(#!
1.3.2 Research motivation of the present study!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!($!
1.3.3 Main objectives and key methods!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!(%!
2. Materials and Methods!................................................................................................!16!
2.1 Study area!..........................................................................................................................!16!
2.2 Data collection!....................................................................................................................!17!
2.3 Whale watching intensity!...................................................................................................!19!
2.4 Individual encounter probabilities!.....................................................................................!21!
2.4.1 Data structure!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!"(!
2.4.2 Model development!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!""!
2.4.3 The state model!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!"#!
2.4.4 The observation model!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!"$!
2.4.5 Estimating spatial encounter probabilities!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!"%!
2.5 Individual exposure to whale watching boats!.....................................................................!25!
3. Results!.........................................................................................................................!26!
3.1 Sampling effort!...................................................................................................................!26!
3.2 Individual encounter probabilities!.....................................................................................!28!
3.3 Whale watching intensity!...................................................................................................!30!
3.4 Individual exposure to whale watching boats!.....................................................................!33!
4. Discussion!....................................................................................................................!37!
4.1 Main findings!.....................................................................................................................!37!
4.1.1 Individual exposure to whale watching boats!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#*!
4.1.2 Individual encounter probabilities!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#'!
4.1.3 Whale watching intensity!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!$(!
4.2 Future research!..................................................................................................................!42!
4.3 Conservation implications!..................................................................................................!43!
5. Conclusion!...................................................................................................................!44!
Acknowledgements!..........................................................................................................!46!
References!........................................................................!+,-.,/0!1,2345/6,!789-3!:,;878,/3&!
Appendix!.........................................................................................................................!58!
!
!
<!
List of figures
Figure 1. Map of the study area.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!(<!
Figure 2. Shape extraction of sperm whale fluke in Phlex.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!(=!
Figure 3. Shape match of sperm whale fluke in Match.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!(=!
Figure 4. Sampling effort and sperm whale encounters.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!"=!
Figure 5. Spatial mean encounter probability of encountered sperm whales.!&&&&&&&&&&&&&&&&&&&&&&&&!"'!
Figure 6. Seasonal variation in daily number of whale watching trips.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#(!
Figure 7. Spatial variation in whale watching intensity. .!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#"!
Figure 8. Spatial variation in individual exposure to whale watching boats.!&&&&&&&&&&&&&&&&&&&&&&&&&!##!
Figure 9. Seasonal variation in individual exposure to whale watching activities.!&&&&&&&&&&&&&&&&!#$!
Figure 10. Density histograms of the cumulative number of encounters.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#<!
Appendix
Figure A1. Boat activity in the sampling area between May and October based on AIS B
data.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<%!
List of tables
Table 1.!List of boats with and without AIS B and corresponding boat effort.!&&&&&&&&&&&&&&&&&&&&&&!")!
Table 2. Summary of sampling effort, sperm whale encounters, and re-sightings.!&&&&&&&&&&&&&&&&!"*!
Table 3. Estimated model parameters estimated based on spatial capture-recapture data.!&&&&&!"'!
Table 4. Summary of whale watching intensity, estimated encounters, and individual
cumulative interaction time.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!#%!
Appendix
Table A1. Summary of monthly whale watching effort by boat for April.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!%=!
Table A2. Summary of monthly whale watching effort by boat for May.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!%=!
Table A3. Summary of monthly whale watching effort by boat for June.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!%'!
Table A4. Summary of monthly whale watching effort by boat for July.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!%'!
Table A5. Summary of monthly whale watching effort by boat for August.!&&&&&&&&&&&&&&&&&&&&&&&&&&!<)!
Table A6. Summary of monthly whale watching effort by boat for September.!&&&&&&&&&&&&&&&&&&&&!<)!
Table A7. Summary of monthly whale watching effort by boat for October.!&&&&&&&&&&&&&&&&&&&&&&&!<(!
Table A8. Summary of all SECR models, detection function, parameters.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<(!
Table A9. Fitted SECR models for May.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<"!
Table A10. Fitted SECR models for June.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<"!
Table A11. Fitted SECR models for July.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<#!
Table A12. Fitted SECR models for August.!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<#!
Table A13. Fitted SECR models for September!&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!<$!
!
!
!
*!
Abstract
There has been a globally growing interest in cetacean-based tourism resulting in increasing
cumulative impacts on many wildlife species. In the Azores archipelago, the local whale
watching industry has drastically evolved over the last two decades, especially targeting a
population of sperm whales that use the habitat as important feeding ground and can be found
year-round. Some individuals return annually to the area and stay over several weeks to breed,
which makes them especially vulnerable to human-induced disturbance. Hence, this study
aimed to contribute to the establishment of a conservation framework to ensure the future
sustainability of this industry by including the comprehensive evaluation of cumulative
exposure of sperm whales to whale watching boats in the area into the management of human
activities. For the first time spatially-explicit capture-recapture (SECR) models have been
developed to quantify the cumulative interaction time between photo-identified individual
sperm whales and whale watching boats in the area. The study provided baseline estimates of
sperm whale encounter probabilities which were integrated together with the whale watching
intensity in the area to estimate spatial variations of exposure on individual-level. Model
estimates revealed that whale watching activities were mainly concentrated in two distinct areas
north of Faial and south of Pico island, consequently exposure levels were found to be
significantly higher in respective ‘core’ areas of disturbance. Furthermore, results indicated
seasonal variations in daily individual exposure levels with a peak in June, where the most
repeated interactions between the same individual and whale watching boats took place. The
present findings stress the importance of taking the individual exposure into consideration when
it comes to the management of potentially harmful human activities and support a precautionary
approach.
Keywords: sperm whales, cumulative exposure, spatially-explicit capture-recapture (SECR)
models, photo-identification, non-lethal disturbance, cetacean conservation, Azores
!
=!
Deutsche Zusammenfassung
Das weltweit steigende Interesse und somit auch der stetig wachsende Tourismus, basierend
auf der Beobachtung von marinen Säugetieren, resultiert in einer deutlichen Zunahme
kumulativer Einflüsse auf viele Wildtier-Arten. In dem Archipel der Azoren befindet sich eine
“Whale Watching” Industrie, die sich über die letzten zwei Jahrzehnte stark entwickelt hat und
insbesondere auf eine Pottwal Population richtet, für die das Habitat ein bedeutsamer
Nahrungsgrund darstellt, wo sie das ganze Jahr über vorzufinden sind. Einige Individuen
kehren jährlich in das Gebiet zurück und halten sich dort über mehrere Wochen auf, um ihre
Nachkommen zu gebären, wodurch sie besonders angreifbar für Belastungen durch
menschlichen Einflüsse werden. Das Ziel dieser Studie war es folglich, zur Entwicklung eines
Rahmenkonzeptes des Artenerhalts mittels eines integrierten Ansatzes beizutragen, der eine
umfassende Gesamtbewertung der kumulativen Einflüsse der “Whale Watching” Aktivitäten
auf Pottwale beinhaltet, um die zukünftige Nachhaltigkeit dieser Industrie zu sichern und
menschliche Einflüsse zu regulieren. Erstmals wurden sogenannte georeferenzierte SECR
Modelle entwickelt, um die kumulative Interaktionszeit zwischen einzelnen foto-identifizierten
Pottwalen und kommerziellen “Whale Watching” Booten in dem Gebiet zu quantifizieren. Die
Studie lieferte grundlegende Schätzungen der Wahrscheinlichkeiten, etwaige Individuen
vorzufinden, die im Folgenden zusammen mit der Intensität der “Whale Watching” Aktivitäten
integriert wurden, um räumliche Veränderungen der Einwirkung durch menschliche Einflüsse
darzustellen, basierend auf den dokumentierten Individuen-Sichtungen. Modelschätzungen
enthüllten, dass sich die “Whale Watching” Aktivitäten hauptsächlich auf zwei Gebiete
konzentrierten, nördlich von Faial und südlich von der Insel Pico. Folglich war das Niveau der
Belastung der Boote auf einzelne Pottwale in diesen Gebieten deutlich erhöht. Des Weiteren
wiesen die Ergebnisse darauf hin, dass die täglichen Belastungen der einzelnen Pottwale
saisonalen Schwankungen, mit einem deutlichen Hoch im Juni unterliegen, in dem die meisten
wiederholten Interaktionen zwischen demselben Individuum und kommerziellen “Whale
Watching” Booten stattfanden. Die vorliegenden Ergebnisse heben die Bedeutung der
Betrachtung der Gesamtbelastung durch menschliche Einflüsse auf Individuen hervor, die unter
Berücksichtigung eines konservativen Vorgehens in einem integrierten Ansatz ins Management
miteinbezogen werden sollte.
!
'!
1. Introduction
1.1 The human-wildlife conflict: Main impacts of non-lethal human disturbance on
cetacean species
1.1.1 Sources of disturbance in marine environments
Marine megafauna represents a major role in many marine ecosystems due to nutrient and
biomass transfer and large-scale distribution of sediments from feeding or feces, spreading them
across the world’s oceans (Doughty et al., 2016; Roman et al., 2014). In addition, there are
evidences that recovering whale populations influence the structure of pelagic ecosystems due
to either direct trophic interactions via top down control on prey stocks or indirectly by
enhancing local primary production (Surma & Pitcher, 2015; Witteveen et al., 2015). However,
in the recent years, these marine giants have been increasingly exposed to a variety of purposive
and non-targeted human activities, making them especially vulnerable when occurring
repeatedly. Therefore, a comprehensive understanding of these impacts is essential for the
conservation and management of marine megafauna.
Over the last decades, several stressors associated with human activities raised attention, such
as shipping (Gomez et al., 2017; Marley et al., 2017), military sonar exercises (DeRuiter et al.,
2013; Harris et al., 2018), commercial oil and gas processing (Goddard et al., 2013), and
installations of off-shore wind farms producing renewable energy i.e. pile-driving (Dähne et
al., 2013; David, 2006). Although they are not purposively targeting cetacean populations, they
have shown negative impacts on several species. In contrast to many of these non-targeted
human activities, ecotourism represents a directional source of disturbance bearing potential
negative effects for targeted cetacean species (Christiansen & Lusseau, 2014; García-Cegarra
& Pacheco, 2017; Parsons, 2012; Sprogis et al., 2017; Trave et al., 2017; Visser et al., 2011).
1.1.2 Influence of tourism on wildlife species
Wildlife-based tourism is evolving globally, associated with an increasing pressure of
anthropogenic influences on wildlife populations (Bejder et al., 2009; Messmer, 2000).
Assessing the population-level consequences of disturbance is a recent objective of many
ecologists developing an ecosystem-based framework of managing human activities to
accomplish a sustainable conservation of wildlife species (Bejder et al., 2009; Pirotta et al.,
2018). However, animal responses to human stimuli can be diverse and it remains a major
!
()!
challenge for modern conservation and management to quantitatively link non-lethal
disturbance to population dynamics (Gill et al., 2001). A comprehensive understanding of
human impacts affecting wildlife populations is required, especially when it comes to the
management of long-living, wide-ranging species, for which empirical data is often limited to
relatively small spatial and temporal scales (National Research Council, 2005).
Animal behavioral responses to anthropogenic stressors are similar to their responses to
predation risk (Beale & Monaghan, 2004; Frid & Dill, 2002). Hence, the presence of non-lethal
disturbance can lead to severe alterations in animal physiology and a range of ecological and
biological features, such as home range size, habitat use, foraging behavior, and reproductive
success among others (Bejder et al., 2009; Pirotta et al., 2018). Behavioral or physiological
changes can directly affect survival probabilities, for example, increased risk of predation or
consequences of injuries decrease an individuals’ chances of survival (Hooker et al., 2012).
However, such changes can also affect individual vital rates indirectly by impairing life
functions (Pirotta et al., 2018).
Assessing population-level effects, requires a long-term monitoring of target species associated
with further information on varying exposure between sites, populations, and individuals
(Lusseau & Bejder, 2007). Significant influences on vital rates, due to added energetic
constraints of the individuals’ response, are likely to affect viability of the offspring population.
In order to cope with disturbance, animals evaluate energetic trade-offs in relation to costs
(Lusseau & Bejder, 2007). When costs of tolerance exceeding benefits of remaining in their
once suitable habitat, short-term evasive strategies lead to long-term habitat shifts to less-
disturbed areas (Bejder et al., 2006). Therefore, it has been the objective of many recent studies
to assess the cumulative effects of multiple stressors based on their temporal and spatial
variations (Christiansen et al., 2015; Halpern et al., 2015; Pérez-Jorge et al., 2017; Pirotta et al.,
2018; Pirotta et al., 2019). The spatially-explicit assessments of cumulative impacts are an
important step towards understanding complex interactions inherited by human-induced
disturbance and further effects on economic and ecological key components, such as cetacean
species (Maxwell et al., 2013).
1.1.3 Short and long-term effects of whale watching tourism on cetaceans
Cetaceans have always been a desirable target to exploit, for both ecotourism and hunting, due
to their economic value. In the light of history, the commercial whale hunting once led to local
extinction and collapses of populations. Since whaling was banned globally by the International
Whaling Commision (IWC) in the 1970s, there has been a shift from their extractive to non-
!
((!
extractive use driven by the growth of marine tourism focusing on cetacean observation
(Mazzoldi et al., 2019). Rapid development has often outpaced management, resulting in
concerns over the long-term sustainability of this industry (New et al., 2015). Previous research
effort aiming to assess the impact of whale watching on cetaceans found evidences of short-
term behavioral responses, such as variations in respiration patterns or alteration in path
directedness, resulting from horizontal and vertical avoidance tactics (Frid & Dill, 2002). These
short-term responses can lead to biologically relevant effects with long-term consequences for
the life history of cetaceans, such as survival, reproduction or population size (Bejder et al.,
2006; Lusseau & Bejder, 2007).
Sperm whale populations off New Zealand showed a reduction in surface periods, number of
blows, intervals between blows, and the frequency of foraging dives in the presence of whale
watching boats (Gordon et al., 1992). Such variations in their behavioral patterns are often
triggered by a stress response in the animals and can lead to a significant increase of energy
expenditure (Christiansen et al., 2014). Potential long-term effects on body conditions have
been associated with decreased reproductiveness and survival of minke whales when such
human-induced disruptions by whale watching boats occurred frequently (Christiansen et al.,
2014; Christiansen et al., 2013).
1.2 Spatial and temporal variation and how to cope with it: A SECR approach to
estimate cetacean densities
1.2.1 Heterogeneity over space and time
Non-lethal anthropogenic disturbances are assumed to vary over space and time (Christiansen
et al., 2015). Just as human activities take place on distinct temporal scales, diurnal or seasonal
patterns are also common for cetacean species, further determining their distribution (Garcia-
Cegarra et al., 2019; Pirotta et al., 2019). Likewise, spatial variation can be found in both, the
heterogeneous distribution of the targeted species in space and the inconsistent location of
human activities. For instance, global shipping routes illustrating analogues to terrestrial roads,
with concentrated vessel movement resulting in direct and indirect impacts on the adjacent
environment due to pollution, collision, and ship noise, with higher intensity closer to the
‘marine roads’ (Pirotta et al., 2019).
The temporal and spatial overlap between the stressors and the cetacean population determine
an individuals’ level of exposure (i.e. duration and intensity during a given period of time) in
relation to the human-induced disturbance (Hazen et al., 2017). Additionally, it is important to
!
("!
consider the time period individuals are staying in affected areas further depending on the size
of individual home ranges, migration patterns, and habitat use (Pirotta et al., 2019), which could
be determined by an individual’s age, sex, or social status (Vandeperre et al., 2016). Therefore,
the incorporation of such spatial variability of disturbance inherit to both human activities and
affected animals, is an important next step to quantify the cumulative effects of non-lethal
disturbance on wildlife populations.
1.2.2 Spatially-explicit mark-recapture models
Earlier approaches of capture-recapture methods (CRM) estimating population size were based
on capture-frequency data obtained from photo identified individuals (Karanth & Nichols,
1998; Otis et al., 1978). Subsequently, the effective trapping area is derived from the distance
between the locations of recaptures to estimate animal densities (Athreya et al., 2013). Ideally,
these densities estimates are based on data obtained over a large spatial and temporal range, but
added cost and demanding logistics often limit CRM studies (Boys et al., 2019). However,
density estimates are sensitive to the pre-defined effective sampling area and can further be
biased by individual heterogeneity in detection probability due to variations in distance between
capture devices and animal home range (Gimenez et al., 2019). The analysis of standard CRM
is non-spatial, individual capture probabilities can be modeled time-specific based on capture
experience but there is no association to location, movement or the detector type (Otis et al.,
1978).
Spatially explicit capture-recapture (SECR) complemented conventional CRM by
incorporating geo-referenced information on capture locations directly into the modeling and
the estimation of capture probabilities (Gopalaswamy et al., 2012; Royle et al., 2009a). In
comparison, it is a more fine-scaled and realistic approach of the capture process as it avoids
negative effects of unmodeled spatial heterogeneity in non-spatial models by including both
individual and trap location into the model (Borchers & Efford, 2008; Efford, 2004; Efford et
al., 2009a; Efford et al., 2005; Royle & Young, 2008). In many datasets, unbalanced data arises
from spatial heterogeneity in sampling effort due to weather conditions or logistical constraints.
SECR models account for such spatial variation in sampling effort and additionally allowing
for temporal variation of effort in capture probability by ad hoc integration (Efford et al., 2013).
Unbiased estimates of population density are obtained from a variety of different detectors, that
are distinguished according to their ability of retaining an animal in traps, or not (proximity
detectors) (Efford et al., 2019). They can further be classified whether they are able to detect
!
(#!
the same individual more than once within a sampling occasion (‘count’ proximity detector) or
not (binary detector) (Efford et al., 2009b).
As SECR models yield robust density estimates and maximize precision (Efford et al., 2005),
and they have lately been applied post hoc to a variety of capture-recapture datasets (Efford,
2011; Efford & Fewster, 2013; Gopalaswamy et al., 2012; Thompson et al., 2012). SECR
models can be a useful tool to assess individual movement and heterogeneity arising from
variations in capture probability, especially for highly mobile and wide-ranging species, like
cetaceans. However, a lot of research effort has been put into estimating densities of larger
carnivores (see Athreya et al., 2013; Gimenez et al., 2019; Royle et al., 2009b; Russell et al.,
2012), but only few studies have been applying SECR models on cetacean species (Cheoo,
2019; Marques et al., 2012).
1.3 Shifted perspectives of a long-known industry: A sperm whale case study
1.3.1 The development of whale watching in the Azores
The Azores archipelago is characterized by a wide range of different habitat types such as
shallow seamounts, submarine canyons, steep island slopes, narrow island shelves and vast
areas of abyssal plain (Morton et al., 1998) . In this high diversity of habitats, a total of 28
cetacean species have been documented, from several species that can be observed year-round,
such as the common bottlenose dolphin Tursiops truncatus (Montagu, 1821), and the sperm
whale Physeter macrocephalus Linnaeus, 1758, to other migrant species, such as the blue whale
Balaenoptera musculus (Linnaeus, 1758), and the Atlantic spotted dolphin Stenella frontalis
(Cuvier, 1829) (Gomes Pereira, 2008; Silva et al., 2014; Silva, 2007; Steiner 2017). During
several decades the resident sperm whale population of the Azores was target of intense
commercial whaling, becoming an important economic sector, until it was finally banned in
1984 (Gomes Pereira, 2008; Mazzoldi et al., 2019). Local enterprises established whale
watching (ww) as sustainable alternative, starting their business in 1993 with only two boats,
rapidly evolving to a fleet including 44 boats operated by 23 licensed ww companies in 2008
(Sequeira et al., 2009). With around 45,000 annual clients this industry yielding an approximate
income of 1.4 million Euros every year, considering exclusively the sale of tickets (Sequeira et
al., 2009). Although, current legislation restricting the number of boats, their velocity and
approaching distances, there are no effective regulations or implemented management
measures yet aiming to limit the growth (in number of tourists) of this economic sector (Oliveira
et al., 2007a). Considering the fact, that the number of licensed boats has already reached its
!
($!
mandatory threshold, it can be assumed that Azores have grown close to their carrying capacity
in terms of operating ww vessels (Oliveira et al., 2007a; Sequeira et al., 2009). The rapid growth
of this industry and increasing tourist demand arise the urgent need of improving current
management regulations to ensure future sustainability along with the economic development
of the region. Since the early 2000s, the companies located on São Miguel island began to
operate year-round in response to the steady increasing tourism. In comparison, the ww
activities in the Azores central group are highly seasonal and mainly conducted between
April/May and October (Oliveira et al., 2007b; Sequeira et al., 2009). They are mainly
concentrated on the area around Faial and Pico island and the channel between Pico and São
Jorge island. Recently, ww business is also being established on Terceira island (Governo
Regional dos Açores, 2020). From lookout points on land (Vigias), experienced observers use
binoculars to search for cetacean activity in the ocean before and during trips. Sighting positions
are passed on to the ww boats, directing them towards the animals, in order to minimize fuel
expenditure and maximize the tourist satisfaction.
1.3.2 Research motivation of the present study
For many sperm whales groups, the Azores provide an important feeding and mating ground.
Especially abundant witihin the central group, they can be observed year-round in this area,
which makes them especially vulnerable to human activities (Oliveira, 2014; Silva et al., 2014).
Mark-recapture data from individually photo-identified sperm whales in the study area have
been used in the past to estimate abundance and population size (Matthews et al., 2001) and
identify migration patterns (Steiner et al., 2012) but so far, no effort was made to estimate their
encounter probabilities using spatially-explicit models.
In Azores central group, the main human activities within the marine environment are
concentrated on commercial fishery, transportation, and ecotourism, such as cetacean
observation, as well as recreational diving and fishing. Considering the fact that nowadays,
ecotourism occupies an increasingly important socio-economic role in this region (Neves-
Graça, 2004; Silva, 2015), there is a need to assess the potential negative effects of ww activity
on cetacean species (Magalhães et al., 2002; Visser et al., 2011). Previous studies on sperm
whales in the area identified short-term effects of ww activities, such as alterations in swimming
speed and aerial displays during interactions with ww boats. Although these studies only
indicated minimal effects (Brasseur et al., 2010; Magalhães et al., 2002), others found evidence
of potentially harmful effects of ww on target cetacean species, such as evasive strategies and
alteration in daily resting patterns leading to increased energy expenditure (Neves-Graça, 2004;
!
(%!
Visser et al., 2011). If previously reported indications of short-term avoidance tactics can
potentially lead to long-term impacts on the population dynamics is yet unclear for sperm
whales.
However, little has been done to investigate cumulative exposure of individuals to non-lethal
human disturbance arising from the repeated interactions between sperm whales and ww boats
during the ww season. SECR models can be used to estimate such cumulative exposure on
sperm whales, quantifying the interaction time between whales and ww vessels, as shown by
Christiansen et al. (2015). He developed a SECR model to assess spatial variations in exposure
on individual-level and to evaluate potentially negative effects of whale watching activities in
Faxaflói Bay, Iceland. In the study, the cumulative exposure of minke whales and variations in
individual exposure to the source of disturbance were estimated based on individual encounter
probabilities in the area integrated with the intensity of whale watching activity. Finally,
Christiansen et al. (2015) concluded that added energy constraints due to repeated disturbance
during the feeding season could influence individual vital rates of minke whales. If such
cumulative effects caused by ww activity could ultimately be linked to reduced fitness, habitat
displacement or via physiological constraints on individual-level is still in urgent need of
investigation for sperm whales.
1.3.3 Main objectives and key methods
This study aims to provide estimates of the cumulative exposure of sperm whales (Physeter
macrocephalus) to ww activities in the Azores central group using a spatially explicit capture-
recapture modelling approach. The final estimates provide the basis for evaluating potential
disturbances caused by the interaction with whale watching boats to further contribute to the
development of a conservation framework.
The spatial encounter probability of sperm whale individuals identified in the study area is
estimated using spatial boat survey data. Followed by a quantification of the ww intensity in
the same area over the time period of the study. The individual exposure of the sperm whales
to ww boats is estimated over space and time based on the probability of detecting individuals
and the total ww intensity. Eventually, the cumulative interaction time that sperm whales spent
with the ww boats is calculated using the obtained exposure estimates. The final estimates aim
to shed light on the question whether ww activities yield potentially harmful effects for the
targeted sperm whale population.
!
(<!
2. Materials and Methods
2.1 Study area
!
The Azores archipelago is located between 36°55’-39°43’N and 24°46’-31°16’W in the North
Atlantic, extending along the northwest to southeast with more than 600km crossing the Mid-
Atlantic Ridge (Morton et al., 1998). Its nine volcanic islands are divided into three main
groups: the eastern group (composed of the islands of São Miguel and Santa Maria), the central
group (comprising the islands of Faial, Pico, Terceira, São Jorge and Graciosa) and the western
group (islands Flores and Corvo). This study focuses on the central group, and is restricted to
an area of 12064 km2 comprising the islands of Faial, Pico, São Jorge and the channel between
Faial and Pico (Figure 1).
Figure 1. Map of the study area located within Azores central group in the North Atlantic.
The blue dots represent the locations of recorded sperm whale encounters between March
and October 2018. The light grey dots indicate the possible locations of activity centers
(habitat mask). The black dots represent the main ports in the study area.
.
!
(*!
2.2 Data collection
Photo-identification of the fluke of sperm whales is considered to be a valuable tool in capture-
mark-recapture (CMR) studies allowing naturally marked individuals to be recognized over the
years (Childerhouse et al., 1996; Dufault & Whitehead, 1998; 1995). The objective of this
technique is to estimate demographic parameters based on individual encounter histories that
are generated over time from photo-identification data. In this study, photos of sperm whales
were obtained from commercial ww boats operating in the main study area between March and
October 2018. Individual sperm whales were identified based on the number and location of
nicks and scallops along the trailing edge of their fluke (Steiner et al., 2017). These natural
marks ensure reliable re-identification due to their stability and longevity over time. The photos
were analyzed by extracting the shape of each sperm whale fluke using the software Phlex
version 1.3 (Beekmans et al., 2005).The shape of each lobe of the fluke was analyzed separately,
and extracted manually or automatically within the selected extraction area (Figure 2).
Subsequently, the software Match 1.2 (Beekmans et al., 2005) was used to compare each photo
and its associated shape to all other photos of the database and their respective shapes. The
program allows for visual comparison of photographs and a detailed verification of the
potentially matching shapes (Figure 3).
!
(=!
!
Figure 2. Shape extraction of the left lobe of a sperm whale fluke using the software Phlex
version 1.3. Each lobe is analyzed separately with manual or automatic extraction of its
shape.
Figure 3. Example of a match between two photos of the same individual sperm whale
using the software Match 1.2. Detailed comparison of fluke shapes of the two images are
shown in the bottom left panel.
>?8@5!A3,87,/!
>?8@5!A3,87,/!
!
('!
Photos were classified according to Arnbom (1987) to assess the quality of images (Ranges:
Q1 = bad quality to Q5 = excellent quality) irrespective of the degree of distinctiveness of the
fluke markings. Images of low quality (Q<3) where the individual could not be identified (<8%)
were discarded. Based on Dufault & Whitehead (1995) and Childerhouse et al. (1996), each
individual was assigned a distinctiveness value based on certainty of future re-identification by
fluke markings and shape (D0 = no markings to D5 = missing portion of fluke). Only
individuals with sufficiently distinct marks to ensure further recognition (D 3) were included
in the database. Selecting reliable photos based on the criteria listed above aims to minimize
heterogeneity in captures that can be caused by misidentification of non-distinctive flukes
(Boys et al., 2019). In total, 1406 photos from four different companies were used for photo-
identifying individual sperm whales, 108 images were discarded. The photo-identification data
provides the base to create an encounter history of the individually identifiable whales. The
position of each encounter was recorded using Global Positioning System (GPS).
2.3 Whale watching intensity
The Automatic Identification System (AIS) Class B, designated for small vessels, was used to
record the survey track of the ww boats in the study area during each trip and to account for
spatial heterogeneity in boat effort. As three of the eleven whale watching boats operating in
the area (~25% of the fleet) did not have an onboard AIS B system, extrapolation of their survey
tracks was required for the whole study period to estimate the total whale watching effort in the
study area. The new survey tracks were created based on the number of daily trips, including
the time of the day, provided by the ww companies and the existing tracks recorded from the
boats with AIS system used as a reference. As some tour operators used up to four boats to
carry out their trips, AIS information from the respective company was preferably chosen to
estimate the new survey tracks. In case AIS data from the same company was not available for
the date or time of the day, AIS information from boats with similar characteristics was used
(Table 1). Experienced observers operated from distinct land-based lookout points to spot the
animals with binoculars (Steiner 20x80 mm) before and during trips leading the ww boats on
the shortest way to the sighting. Therefore, estimated tracks are representative for the track the
unmonitored boats could take during the specific sampling occasion. Whale watching activities
are mainly limited to a designated area within 12 nautical miles around the islands as most
companies are not licensed to go further or simply avoid unnecessary fuel expedition (Sequeira
et al., 2009).
!
")!
!
Table 1.!Summary table of whale watching boats operating in the area during the season 2018
that either a.) provide AIS B information or b.) whose boat effort was estimated by taking the
available AIS information recorded by the ww boats of the respective company as reference.
The total number of whale watching trips that were conducted between May and October 2018
is included below.
a.) Whale watching boats with AIS B
!
Boat
Company
Location
Total number
of trips
Physeter
Noberto Diver
Faial
80
Faidoca
Noberto Diver
Faial
149
Risso
Azores Experience
Faial
208
Kometa
Cetacean Watching
Pico
177
Bicho Do Mar
Naturalist
Faial
204
Reefcat
Pico Sport
Pico
124
Maisha
Pico Sport
Pico
98
José Azevedo
Peterzee
Faial
98
!
!
b.) Whale watching boats without AIS B
!
Boat
Company
Primary
reference boat
Secondary
reference boat
Total number of
extrapolated trips
BWA
Cetacean
Watching
Kometa
Risso
64
Chá-Preto
Noberto Diver
Faidoca
Physeter
130
Gonçalo
Noberto Diver
Faidoca
Physeter
90
A basic concern when working with AIS data are the occasional time gaps ranging from several
minutes up to various hours, especially when it comes to month with dense AIS information.
In order to increase spatial accuracy of the ww intensity and improve the resolution within the
pre-defined 1x1km2 track grid cells, the AIS B boat data including the extrapolated trips was
used as a base to interpolate track positions in between time intervals. Representing the amount
of time expressed in standard units of effort in hours, the ww intensity was estimated based on
the spatial information from the boat data including all ww boats and further interpolated to
continuous positions of a known time interval. The R software version 3.5.3 (R Development
!
"(!
Core Team, 2019) and the R package ‘trip’ version 1.6.0. (Sumner et al., 2019) was used to
estimate positions in a time interval of every minute and smooth out the boat tracks creating a
stable vector over time that was easily summarized by grid cell. The total effort was calculated
for each grid cell k of the track grid in boat hours T0, this approach allowed to infer the total
boat effort using the interpolated spatial data. The generated matrix containing the spatial ww
intensity Iskw equals boat effort T0 and is defined by the time the boat spent in each grid cell k
during trip w in month s.
2.4 Individual encounter probabilities
2.4.1 Data structure
Spatially explicit capture-recapture (SECR) models are most commonly applied to capture-
recapture data obtained from an array of ‘detectors’ such as live-capture traps, mist nets, camera
traps or ‘hair snairs’ (Royle & Young, 2008). Recent studies applied this modeling approach to
search-encounter data (Athreya et al., 2013; Royle et al., 2009a; Russell et al., 2012; Thompson
et al., 2012). A key element of the sampling design is in any case its distinct spatial structure
defined by fixed detectors that can be set up randomly or systematically to draw georeferenced
individual encounter histories linked to detector locations. Such spatial encounter histories
provide valuable information about the position in space and time where an individually
identifiable sperm whale was photo-identified. The main deficiency of closed population
models is that, space and movement do not have an explicit manifestation, which means that
the estimated population size N is not linked to any spatial context (Royle et al., 2009a). In
addition to standard CMR methods using individual encounter histories, SECR models include
the auxiliary spatial information from encounter locations directly in the density estimation
process (Athreya et al., 2013).
A key element of the data structure in the present study, was that individual encounter data was
not created from discrete trap locations, such as ‘hair snares’, camera traps, or mist nets that are
often used in standard CMR studies (Russell et al., 2012). Instead, individual encounters arise
from boat survey data that was continuously distributed over space with a non-discrete
structure, which did not provide a precise characterization of space equal to a fixed trap array.
To resolve this, a grid-like array of detectors, was set up over the study area as distinct sampling
units. The idea is to use grid cells as conceptional traps to create encounter histories based on
the presence or absence of an individual in each grid cell during a sampling occasion (Russell
et al., 2012). Regarding the spatial resolution, Thompson et al. (2012) proposed grid cells to be
!
""!
(i) small enough to display capture heterogeneity in presence-absence over space although (ii)
large enough in context to their biological meaning. Christiansen et al. (2015) used a grid cell
size of 1km2 to capture the relevant movement of the whales yet providing meaningful estimates
of the animals’ activity range to estimate the exposure of individual minke whales to ww
vessels. Accordingly, a habitat mask of 1km2-grid cells was set up in the study area including
a 12-km buffer zone around it (Figure 1), which represent the potentially suitable habitat for
sperm whales (Efford, 2004). For the SECR models, reference coordinates were obtained from
the center point of each grid cell. The state of each grid cell was considered to be ‘active’ when
vessels enter the cell during a certain sampling occasion or ‘inactive’ whenever they were not
visited, respectively. Similarly, the presence-absence of the individuals was linked to the
coordinates of the grid cells’ center point rather than its actual sample position (Russell et al.,
2012; Thompson et al., 2012). Resulting in a georeferenced encounter history, assigning each
capture to the centroid of a specific grid cell.
2.4.2 Model development
An important key assumption of SECR models is that the probability of encountering an animal
is higher closer to the core area that the animal most frequently occupies (i.e., activity center)
(Russell et al., 2012). The aim is to estimate the animals’ activity centers based on individual
encounter histories and additionally estimating their activity range given by the probability of
its spatial distribution around the center of activity. There are two types of SECR models that
were recently developed, from which Bayesian models are preferably used over likelihood
based models as they do not rest on asymptotic assumption (Athreya et al., 2013).
In this study, a hierarchical SECR model was developed composed of an observation model
(the trap locations) that is conditional on the state model (set of latent random variables). The
models were fitted based on the encounter histories and the spatial sampling effort applying a
maximum likelihood approach using the R package ‘secr’ version 4.1.0. (Efford, 2019). The
latent (i.e. unobserved) variables representing the hypothetical activity centers of each
individual of the population. The locations of individuals during the sampling occasions are
part of a partially observed random variable (Royle & Young, 2008). Removing them from the
conditional likelihood by integration yields inference and estimates of the absolute density of
activity centers within the trapping array (Royle et al., 2009a). Heterogeneity models using
likelihood methods normally accomplish that by direct likelihood integration. In this study, a
Bayesian analysis was applied to the hierarchical model using straightforward methods of
!
"#!
MCMC (Markov chain Monte Carlo) integration to remove unobserved locations from the
conditional likelihood. The MCMC approach aims to avoid the necessity of explicit integration
(Royle & Young, 2008). Consequently, the best fitting model for each month was selected
based on Akaike’s Information Criterion (AIC) (Akaike, 1974). SECR models accounting for
individual heterogeneity in the activity range width parameter (σ) yielded the best overall fit,
which supports the assumption that there is a strong variation in detection probability among
individuals within the present dataset. Therefore, a finite mixture model was chosen for the
data, assuming two latent classes differing in detection parameters with estimated mixing
proportions (weighted p-mix) in each class (Borchers & Efford, 2008; Pledger, 2000). The
likelihood uses a weighted sum over the two classes to estimate p-mix. Although unmodeled
heterogeneity is maintained in σ, the model yields identical encounter intensities and densities
for each class. The actual abundance can be obtained by multiplying the weighted p-mix with
the estimated density (Efford, 2015).
2.4.3 The state model
The main objective of the state model is to estimate an unobservable (latent) process based on
observed data, which can be obtained from the georeferenced animal encounter histories (Sibert
et al., 2003). Consequently, the observation parameters based on the encounter histories are
estimated conditionally on the animals’ activity center Xi. It is assumed to be fixed during the
sampling occasion representing the center point of the animals’ movement in space. Rather than
being associated to its biological meaning, Xi describes the probability of detecting an
individual i modeled in terms of distance from the traps to its unobserved location. Assuming
a uniform distribution over space, the sum of estimated detection probabilities over a specified
region S equal to the effective sampling area within an array of given traps (Marques et al.,
2013).
Xi ~ Uniform(S)
Setting up a framework for spatial models of density are based on point process procedures.
Hence, the animals’ activity centers Xi is described by a two-dimensional coordinate (x1i, x2i).
Assuming that the positions of Xi is determined by a 2-dimensional distribution, defined by a
homogeneous Poisson point process with intensity D, the population density. As some animals
have their Xi closer to the traps they are more likely to be captured than others, hence SECR
accounts for heterogeneity in detection probabilities by including the trap location into the
!
"$!
estimating procedure. The animals’ activity center Xi is specific to the study area and the time
period covered by the survey aiming to ensure a sufficient coverage to estimate individual
exposure to ww activity. The study was carried out over six months, with each month being
modeled as a single sampling occasion with respective exposure levels for every month.
2.4.4 The observation model
The observation model was developed conditionally on Xi and yielded an estimation of the
expected number of encounters given as function of distance (dk) from the animals’ activity
center Xi (Efford et al., 2013; Efford et al., 2009). The observed encounter frequency nisk for
each individual i during sampling occasion s at grid cell k can be described by a Poisson model:
nisk ~ Poisson(!isk)
where !isk is the expected number of encounters. As mentioned above, each month was modeled
as single sampling occasion using a count proximity detector which permits multiple encounters
in one or more grid cells during one sampling occasion (Efford et al., 2009; Royle et al., 2009).
A key assumption of the SECR model is the independence of encounters (Efford et al., 2009),
therefore only the first encounter of a day contributed to the animals’ encounter history. The
probability of encountering an individual i declines with increasing distance from its activity
center (Efford, 2004). The negative exponential encounter intensity function is denoted as:
!"#$ % & '#$
'(
&)&!*+), -.$/0"1
2&
The baseline encounter frequency is !0 for all animals in the study area, the decline rate in
encounter intensity in relation to the animals’ activity center Xi is σ, the standard unit of effort
(in boat hrs) is given by T0 and the sampling effort Tsk on sampling occasion s (i.e., month) in
grid cell k. Following Efford et al. (2013), heterogeneity in the sampling effort can be accounted
for by including it as an offset variable in the detection function. Each month is treated as a
single sampling occasion, therefore the distinct model is fitted for the respective month. Hence,
there is no time-specificity within the monthly sampling data.
!
"%!
2.4.5 Estimating spatial encounter probabilities
The probability mass function Zism(Xi) is derived from the activity center Xi of each individual
within a certain sampling occasion (Efford et al., 2013). It describes the probability that the
activity center Xi of the individual is located in a specific grid cell k during sampling occasion
s. The spatial encounter probability Pr(3isk = 1) of detecting individual i at least once in grid cell
k per unit effort T0 (in boat hours) during sampling occasion s is given by:
45/3"#$%&61&%& 7"#8&)& 6&9&+), -!(:#&+), -.$;
2#
&
<
8=>
&
where m indicates the modeled activity center on the habitat mask with total K number of grid
cells (K = 12,064). The parameters !0 and σ are estimates of the encounter intensity function,
the sum of the encounter probabilities for all potential activity centers contributes to the spatial
encounter probability.
2.5 Individual exposure to whale watching boats
The exposure Eisw of an encountered individual i to ww boats on each day w within month s is
estimated by summarizing individual encounter probabilities Pr(3isk = 1) for each grid cell k in
month s per unit effort T0. In other words, the probability of encountering individual i on day w
during month s is:
?"#@ % & 45 3"#$ % &6 &)&A#$@
<
$=>
The encounter data is analyzed by sighting with a restriction of one possible encounter for each
whale per day. Hence, Eisw is limited to values between 0 and 1, replacing all values > 1 by 1.
In this study, sightings were distinguished based on (i) time interval and (ii) distance between
encounters. Previous research has shown that sperm whales are able to communicate in a
distance of 10 to 16 kilometers underwater (Madsen et al., 2002; Watkins, 1980). Together with
the assumption that the mean encounter duration is 1 hour each encounter was assigned to a
distinct sighting aiming to avoid several encounters of the same individual within one trip.
!
"<!
The monthly interaction frequency fis for each encounter i with ww boats during the month is
determined by the summarizing exposure probabilities of encountering an individual i on day
w during month s which can be seen as a Bernoulli process with binary outcomes from all days
W in the specific month:
B
"# % & C+5DEFGGH/?"#@1
I
@=>
It was assumed that sperm whale abundance in the study area remains relatively constant during
the time period of the study (April-October) comprising present individuals and temporary
emigrants (Boys et al., 2019). In order to obtain a mean individual monthly interaction
frequency, a total of 1000 fis values for all individuals i during each month s were estimated
using a Monte Carlo approach. Based on estimated fis values, the cumulative interaction time
with ww boats (encounter duration time in hours) during the month was determined by
randomly allocating an interaction time to each encounter, based on the distribution of
interaction durations obtained from observer boats.
3. Results
3.1 Sampling effort
Photo identification data was taken from observer boats, Risso and Physeter, between March
and November 2018, covering a total time period of nine months. As respective AIS data from
these boats was only fully available for May to October 2018, photos taken outside this study
period were excluded from the analysis. Within the study period, a total of 288 trips comprising
a sampling effort of 1024 hours sampling effort over 134 days were conducted (Table 2).
Sampling effort stayed relatively constant during May and August (189-222 boat hours),
slightly decreasing to the end of the season with 141 hours in September, and a remarkably low
effort for October (43 hours). Although the spatial distribution of the sampling effort was
relatively similar over the study period, differences were identified between months (Figure 4).
The highest boat effort was detected in the north and south of Faial island and southwest of
Pico island. The sampling area covered by the observer boats comprised a total of 933 km2 in
May, reaching its maximum in August with 1328 km2 and decreasing rapidly to 381 km2 in
October.
!
"*!
The number of sperm whale encounters (detections) with individually identifiable animals
reported by the observer boats, ranged between 16 encounters in October up to 181 animals in
August (Table 2). From a minimum 49 individuals within 75 encounters in May, the number of
individuals increasing to a maximum of 132 individuals within 181 encounters in August. In
October, all individuals identified (n = 16) were from different encounters. The number of
unique individuals that could be photo identified over the whole season 2018 was 401. A total
of 357 individuals (89%) were sighted only once during the study period, 54 individuals were
sighted twice (<14%) and only 29 individuals were re-sighted more than twice (<8%). Within
each month, the sightings per unit effort (SPUE) varied between 0.36 and 0.84 encounters/boat
hour over the study period. It is an indicator of the efficiency with which the study area was
sampled (Range 0-1), as increased boat effort can yield low SPUE values, when the number of
encounters is low corresponding to the sampling effort (May and July).
!
Table 2. Summary table including sampling effort, total number of sperm whale encounters,
number of individuals encountered, and number of individuals in each category of re-sightings
(recaptures) between May and October 2018 in the study area, Azores central group. The values
represent sampling data from the observer boats (Risso and Physeter). The SPUE indicates the
sightings per unit effort [encounters/boat hours]. The number of unique individuals identified
over the whole season 2018 was 401.
!
Number of re-sightings
Month
Sampling
effort [h]
Grid
cells
surveyed
Grid cells
with
encounters
Number of
encounters
SPUE
Number of
individuals
encountered
0
1
2
3
4
5
May
211
933
49
75
0.36
49
39
5
2
1
0
2
June
189
969
71
142
0.75
78
46
17
7
3
3
0
July
222
1157
68
128
0.58
94
87
11
1
0
0
0
August
216
1328
111
181
0.84
132
113
15
2
3
0
0
September
141
997
52
95
0.67
67
56
6
4
1
0
0
October
43
381
15
16
0.37
16
16
0
0
0
0
0
Total
1,022
5,765
366
637
0.62
436
357
54
16
8
3
2
!
"=!
3.2 Individual encounter probabilities
The best fitting model for each month was based on a half-normal detection function further
allowing for individual heterogeneity in the activity range parameter 2 (for detailed description
see: Appendix Table A8-A13). The estimated baseline encounter probability !0 varied between
months from 0.0204 and 0.0670 encounters/boat hour (Table 3). The density estimates should
be treated with caution as the population was not closed during each month and the study design
was not set up randomly. The estimated activity centers of the animals are located mainly north
and south of Faial and Pico island over a concentrated ‘core’ area corresponding to the areas of
higher sampling effort (Figure 5). Hence, the estimated individual encounter probability is
higher around these hotspot areas reaching up to 0.0136 per hour (Table 3). Regarding the mean
estimated encounter probability, variations are given between months.
!
"'!
Table 3. Summary of the model parameters estimated based on the sperm whale spatial capture-
recapture data from May to September 2018 in the study area, Azores central group. The density
D of the individuals is given in number of whales/km2, !0 represents the baseline encounter
probability per boat hours, and the movement parameter 2, providing a measure of activity
range width in km.
!
Month
D
SE
J0
SE
K
SE
May
0.0694
0.0298
0.0204
0.0058
8.3988
1.0851
June
0.0385
0.0089
0.0589
0.0108
7.3802
1.1939
July
0.1867
0.1314
0.0276
0.7958
3.5307
1.0830
August
0.2033
0.1005
0.0338
0.0089
3.4456
1.0034
September
0.1156
0.0601
0.0670
0.0203
3.0840
0.8099
Figure 5. Raster map showing the spatial mean encounter probability (see color scale) of
detected sperm whales between May and September 2018 in the study area, Azores central
group. Activity center of the individually identifiable sperm whales are indicated with white
filled circles. The map is based on model estimates.
!
#)!
3.3 Whale watching intensity
The number of trips offered per day varied within the year with a maximum number of 20 trips
per day, reported for August (Figure 6). During May and June the ww intensity ranges widely
between 5 and 15 trips per day, as it is expected for activities that largely depend on weather
conditions, tourist demand and the operation conditions of the boats. The remarkable seasonal
peak of ww intensity is reached in the summer months, July and August, with 15-20 daily trips,
decreasing in September to a mean number of 10 trips/day.
The spatial intensity of whale watching activities around Faial and Pico island varied
significantly within each month but also between months (Figure 7). A total of 1404 ww trips
were estimated throughout the study period (May-September), using the track data recorded
from the AIS B positioning system, and the supplementary information provided by the local
ww companies. The approximate duration of each ww trip followed a normal distribution with
a mean time of 3.025 hours (SD = 0.776, min = 1.004, max = 8.069). The average time for ww
activities reported by the companies is about 2-3.5 hours, corresponding to the estimated values.
The estimated monthly ww intensity between May-September (in boat hours) indicate inter-
mensal variations, as well as heterogeneity over space (Figure 7). The highest estimated
intensities were concentrated to an area north of Pico island and northeast of Faial island,
respectively. Elevated intensities were also identified around the main harbors of Horta (Faial)
and Madalena (Pico) extending further towards the south and southeast.
!
#(!
Figure 6. Seasonal variation in the number of ww trips per day between April and October
2018 in the study area, Azores central group. The dashed vertical lines correspond to the first
and last day of the study period (May-September). Sperm whales can be observed in the area
during this time. The figure is based on AIS B positioning data and auxiliary information on
the daily number of trips, provided by local ww companies. A smoothing spline (red solid
line) was fitted to the season with degrees of freedom set to 20 including the standard error
(grey). The days on the x-axis are given as ordinal dates.
!
#"!
Figure 7. Raster map showing the spatial variation in ww intensity (see color scale), the amount
of time spent by ww boats in each grid cell within the study area, between May and September
2018 in the study area, comprising the islands of Faial, Pico, and S. Jorge (black). Sperm whales
can be found in the study area during the entire season (Silva et al., 2014). The map is based on
model estimates.
!
##!
3.4 Individual exposure to whale watching boats
The estimated spatial individual exposure to ww boats relatively consistent between months,
although the overall exposure differed, with June having the highest level of exposure based on
the ww boat encounter probability (Figure 8). In addition, the estimated daily exposure varied
throughout the study period (May-September), when sperm whales were assumed to stay in the
study area (Figure 9), corresponding to monthly variations in ww intensity (Figure 7). The
distribution of the probability estimates changed based on variations in spatial ww intensity,
with increasing intensity leading to higher probability estimates and increased individual
heterogeneity.
Figure 8. Raster map showing the spatial variation in individual exposure of sperm whales to
ww boats between May and September 2018 in the study area, Azores central group. The values
(see color scale) indicate the probability of a detected sperm whale to encounter a ww boat per
boat hour during the season 2018 (May-September), given the ww intensity for respective month.
The figure is based on model estimates.
!
#$!
The estimated cumulative number of encounters and encounters per individual varied between
months (Figure 10). Based on individual means, estimated cumulative interaction time per
month varied between the months (Table 4). The lowest estimated mean number of encounters
was only 61 in July, with most individuals being encountered only once (Figure 10). The highest
estimated mean number of encounters with individual identifiable sperm whales was in June
with about 213 encounters and more than 3 encounters per individuals (Table 4). In this month,
the maximum cumulative interaction time was nearly 9 hours, which doubles the individual
exposure in May and August. In July and September individuals were estimated to interact only
once or twice with ww boats corresponding even to a 4-fold increase of individual exposure
compared to June (Figure 10).
!
#%!
Corresponding to the estimated number of encounters per individual, the cumulative interaction
time with the ww boats was highest in May and June with 2.8-3.6 hours per month. Sperm
whales had a comparatively lower interaction time with 0.7-1.4 hours during the month in July,
August, and September. Variations between encountered individuals in their cumulative
exposure to ww boats were given in all months. However, the individual with the highest
exposure (8.76 hours), only spent 1.2 % of its time with ww boats on the study area (Table
4).
Table 4. Summary of ww intensity, total number of encounters, encounters per individual,
individual cumulative interaction time in hours during each month, and the percentage of time
spent with ww boats. All values are based on model estimates, assuming that sperm whales are
found in the study area during each month of the season (Silva et al., 2014).
!
Whale
watching
intensity
___________
Total number of
encounters
______________
Encounters
per
individual
____________
Individual cumulative
interaction time [h] per month
__________________________
Percentage [%] of
time spent with whale
watching boats
___________________
Month
Trips
hours
Mean
SD
Mean
SD
Mean
SD
Min.
Max.
Mean
Min.
Max.
May
220
688
86.053
9.090
2.531
1.175
2.837
1.321
0.454
5.030
0.381
0.061
0.676
June
222
638
213.546
12.423
3.187
1.871
3.578
2.099
0.352
8.763
0.497
0.049
1.217
July
338
933
61.429
7.443
0.731
0.425
0.821
0.474
0.075
2.071
0.114
0.010
0.288
August
399
1163
141.89
11.266
1.419
1.089
1.597
1.228
0.066
3.966
0.214
0.009
0.533
September
225
643
68.242
7.396
1.219
0.585
1.371
0.661
0.240
2.702
0.190
0.033
0.375
!
#<!
Figure 10. Density histograms of the cumulative number of encounters between detected
individual sperm whales and ww boats during the season 2018 (May-September) in the study
area, Azores central group. The distributions are based on the estimated mean encounter
frequencies of individual sperm whales from 1,000 bootstrap simulations.
!
#*!
4. Discussion
4.1 Main findings
4.1.1 Individual exposure to whale watching boats
The study was aiming to develop a spatial model of cumulative exposure rates of sperm whales
to ww boats around the islands of Faial, Pico, and São Jorge, to characterize how their
individual exposure varies over space and time. The estimated exposure levels indicated
seasonal variations, with the most exposed individuals in June spending up to 9 hours during
the month with ww boats, which equals 1.2 % of their time within this month (Table 4). In
comparison, recent findings in minke whales estimated a maximum individual exposure of 2.24
hours (0.05 %) over the entire feeding season (April-October) (Christiansen et al., 2015).
Hence, the estimates of the present study suggested significantly more disruptions during the
season caused by ww activities. Such non-lethal human activities disrupting the animals during
feeding or resting can cause stress responses when occurring repeatedly, which could strongly
be linked to increased energy expenditure in several cetacean species (Christiansen et al., 2014;
Lesage et al., 2017; Scheidat et al., 2004; Srinivasan et al., 2018; Williams et al., 2006).
Although the results indicated relatively low exposure rates, there is so far nothing known about
the population-level effects of ww activities on sperm whales in the study area. Therefore, more
research is urgently needed, focusing on potentially negative long-term effects of ww boats on
the local sperm whale population.
The exposure rates were estimated based on individual encounter histories which revealed that
most of the individuals were only seen once during the study period (Table 2). The estimated
encounter probabilities partially depend on the number of re-sightings within one sampling
occasion (i.e. month). Hence, the month with the highest percentage of re-sighted individuals
(June, 38%; Table 2), yielded significantly higher encounter probabilities (Figure 5) and
consequently exposure levels (Figure 8). As sperm whale abundance is assumed to be relatively
lower during June (Silva et al., 2014), the probability of encountering the same individual more
than once is higher than in the other months, when more animals can be found in the study area.
The overall low rate of re-sighted individuals and consequently exposure levels might depend
on the fact that trips were designed to primarily satisfy the clients. Therefore, tour operators
prefer focusing on the animals encountered directly at the location of the sighting, rather than
actively searching the area for more individuals. Moreover, including additional information
!
#=!
from ww boats operating in the area that currently do not provide AIS B data could positively
influence the exposure levels of some individuals. Similarly, boat survey data from ww
companies based in Lajes (Pico island) was not available during the time period of the study,
leading to an underestimation of the ww intensity in the area. The total ww intensity and
consequently the level of exposure of some individuals that were targeted by the boats without
AIS B data are likely to be higher than present estimates assumed.
Corresponding to the estimated spatial encounter probabilities, the estimated exposure is higher
for those individuals that have their activity center closer to the areas where the major part of
the ww activity is concentrated, or their activity range is smaller (Figure 8). As mentioned
above, individual heterogeneity in encounter probabilities or activity range (σ) is likely to be
present in the current dataset and should therefore be considered in the modelling process. For
individuals with a smaller activity range and/or an activity center closer to the area of increased
disturbance, exposure estimates are likely to be negatively biased. The estimated activity range
varied between 3.1 km in September to 8.4 km in May (Table 3) suggesting a higher movement
range for sperm whales than it is known for other cetacean species, like minke whales, which
have an activity range between 1.9-4.6 km (Christiansen et al., 2015). Although mink whales’
activity range was estimated to be smaller, their movement pattern also varied substantially
between the years. Similarly, bottlenose dolphins showed large inter-annual fluctuations in their
movement pattern (2.2-8.2 km) within an area of increased disturbance due to boat traffic
(Pirotta et al., 2015). Large variations in the animals’ activity range further indicate differences
in their movement patterns between the month which should be analyzed in detail using open
model scenarios that account for such temporal and spatial variations.
A minority of photos (< 8%) were discarded because the individuals could not clearly be
identified due to poor weather conditions, low image quality, a non-perpendicular angle
between observer and the sperm whale fluke, or a large distance between the observer boat and
the whale (Stevick et al., 2001). The unidentified individuals were not included in the present
SECR models which could have negatively biased encounter probability estimates and
individual exposure rates, respectively. Additionally, there is a possibility of an on-board
observer not detecting an individual, although it was encountered and therefore exposed to ww
boats. This is due to their social behavior, groups of 5-8 individuals spread within a distance of
1-2 km (Antunes et al., 2009) and might therefore be exposed without being photographed. A
failure of detecting an exposed individual would also lead to an underestimation of monthly
!
#'!
exposure rates. There are methods of incorporating such unidentifiable individuals and non-
detection encounters into the modelling process (Sollmann et al., 2013). Unfortunately, this is
quite a challenge, considering the fact that sperm whales were identified based on distinct fluke
marks distinguishing the individuals, which are difficult to be recognized in low quality
photographs.
4.1.2 Individual encounter probabilities
The estimated encounter probabilities of sperm whales in the study area were overall relatively
low (Figure 5), which might also be influenced by migration on and off the area during the time
period of the study. The activity range parameter was estimated to be increased for sperm
whales (Table 3) as it is expected for highly mobile species, such as cetaceans (Thompson et
al., 2012). Together with the low number of re-sightings (Table 2), estimates supporting the
assumption that individual movement ranged outside the study area. The closure of the
population constitutes a key assumption of standard mark-recapture analysis and spatially-
explicit modelling. As present findings strongly refer to a non-closure of the sperm whale
population in the study area, the estimated densities should be taken with caution (Table 3).
Past mark-recapture studies in the area confirmed that the sperm whale population is subject to
temporary emigration, most individuals are assumed to stay for a few days, and only a minority
are seen over a period of several months (Matthews et al., 2001). Silva et al. (2014) found that
large female groups spend an average time of two weeks in the area. A recent study using non-
spatial multi-state open robust design (MSORD) models and accounting for heterogeneity, due
to temporary emigration, also assumed a short residence time for this population on the study
area (Boys et al., 2019). The estimated population size ranging from 275 to 367 individuals
during the summer month. In comparison, lagged identification models estimated the annual
number of sperm whales visiting the Azores central group to approximately 700 individuals
(Silva et al., 2006) or even a few thousand animals within the entire archipelago (Silva et al.,
2014). Latter estimates were obtained from land-based observation and boat surveys and
correspond to the estimated densities yielded by the SECR model in this study. Nevertheless,
these densities should be taken with caution, as SECR models were designed based on the
assumption that there is no movement on and off the study area within each sampling occasion
(i.e. month). As a result, density estimates could also be potentially biased by individual
heterogeneity, although the encounter probability estimates should be unaffected, as they were
obtained from the observed encounters over the same temporal unit (i.e. month). Assuming that
some individuals are more likely to return to the sampling site than others, individual encounter
!
$)!
rate might be biased. For individuals spending more time outside the study area, the probability
of encounter would be lower, whereas it would be higher for individuals that stay relatively
longer in the study area. Individual heterogeneity in site fidelity is difficult to quantify and
needs a more profound knowledge about the seasonal movement patterns of sperm whales. Still,
it needs to be considered, as it could potentially bias the exposure estimates in both directions.
An approach that overcomes the limitation of demographically closed population modelling
should be considered. As an extension of conventional spatial models to open model scenarios,
it accounts for such spatially-induced individual heterogeneity in capture probabilities (Efford
& Schofield, 2019; Pledger et al., 2010).
The variability in exposure rate was mainly attributed to variations in the individual encounter
probabilities between the months. Large fluctuations of the estimated baseline encounter
frequencies (!0) might be influenced by changes in the way the animals use their habitat or
environmental factors that further determine their distribution (Pirotta et al., 2015). For
instance, their abundance also depends on the distribution, availability and abundance of their
prey, especially when they have specialized diets (Silber et al., 2017). This would imply that
animals could expand their movement range (σ) when resources are limited. In order to cope
with these small-scale variations, the animals would spend less time inside the study area and
their probability of being (re-)encountered (!0) decreases. Their environmental nice partitioning
in the region is not yet fully understood, but it is assumed that sperm whale abundance depends
on a combination of several factors related to different life history requirements (Tobeña et al.,
2016). Herein, primary production has shown to mainly contribute to variations in their
distribution pattern. The model approach used, accounted for individual heterogeneity in σ, but
was constant across individuals for !0. Allowing for heterogeneity, either between social groups
or individuals, would improve its biological realism and precision of the encounter probability
estimates (Pirotta et al., 2015).
The estimated activity center indicated different usage in space between the month (Figure 5),
in terms of its relative location, but also the animals’ movement range (σ) around it (Table 3).
Although the model approach assumed homogenous encounter probabilities across individuals,
spatial heterogeneity might be attributed to the estimation procedure. For instance, with every
iteration in the modelling process, an uncertainty of the relative position of the animals’ activity
center is incorporated in the parameter estimation. Further resulting in a more precisely
definition of the activity center of an individual, whose activity center was always estimated to
!
$(!
be at the same location and vice versa. Similarly, a higher number of re-sightings contribute to
an increased precision when it comes to the definition of the animals’ activity center (Pirotta et
al., 2015). Consequently, the estimated exposure was partially driven by variations in the spatial
usage of individuals.
The estimated individual encounter probabilities were highest in distinct ‘core’ areas north and
south of Faial and Pico island, where most of the activity centers were concentrated (Figure 5).
It is likely that individuals aggregate in these hotspot areas to feed, as it is known that larger
groups of up to 50 sperm whales can be found in concentrated small areas on foraging grounds
(Silva et al., 2014). The Azores further provide an important calving and mating ground for
sperm whales (Clarke, 1956), in the central group the peak of the breeding season takes place
between April-June and the highest number of newborn calves was recorded in August (Silva
et al., 2014). Highest estimated exposure to ww boats were recorded during the peak of the
breeding season (Table 3). In these month the estimated number of ww trips increased up to
10-15 daily trips. Additionally, a second peak of 20 daily trips is given in August, indicating a
seasonal variation of ww activity with a maximum during the summer months (Figure 7). The
increased level of disturbance could potentially affect the young calves and disrupt resting or
feeding activity during the time period when they are especially vulnerable to human activity
(Schaffar et al., 2013). Previous studies found strong evidence of short-term changes in the
behavior of mother and calf and reduced resting time due to interference with ww boats that
might negatively impact cetaceans when disruption occur repeatedly (Morete et al., 2007;
Richter et al., 2006). However, a long-term study off New England did not indicate negative
effects on calf survival (Weinrich & Corbelli, 2009). Although studies are contradictory, there
is so far nothing known about the influences of ww on calving rates and calf survival rates
considering the local sperm whale population. The temporal overlap between the highest
breeding activity and an increased number of daily trips (Figure 6) along with the spatial overlap
between potential feeding hotspots and the areas where most ww activity was concentrated
(Figure 7), should therefore be taken into consideration when it comes to conservation and
management of this species.
4.1.3 Whale watching intensity
Another important key element of SECR models is the placement of the traps, which are
assumed to be located randomly in relation to the animals’ activity centers (Efford et al., 2009c).
In the present study, the survey design was not random. Hence, heterogeneity in sampling effort
!
$"!
would positively bias the estimated densities without influencing encounter probabilities
(Thompson et al., 2012). Further, the best fitting model for each month assumed a linear
relationship between the sampling effort and the encounter rates, which is appropriate for boat
survey data with spatial variations in sampling effort but might be inappropriate for scenarios
with higher ww intensity (Christiansen et al., 2015). Additionally, the number of possible
encounters was a priori determined, allowing each individual to be seen only one time per day.
Although the majority of whales were only seen once, there were few cases where observer
boats encountered the same individual twice during a second trip on the same day. As some
individuals have their center of activity closer to the spots of increased ww intensity, they have
a higher probability to be encountered than others. A model allowing for more than one
encounter per day would probably yield slightly higher encounter probabilities for those
individuals.
4.2 Future research
As previously discussed, an important key assumption of the SECR model was violated because
the population was not closed during the time period of the study. Most recently, Efford &
Schofield (2019) described a spatial open population capture-recapture model, distinguishing
between mortality and emigration, allowing for temporal heterogeneity in the spatial extent of
sampling, and accounting for individual heterogeneity in encounter probabilities. The
movement between sampling times is included into the modelling process, which could reduce
the bias in encounter probabilities, and consequently the cumulative exposure levels, due to
individual movement on and off the study area. Moreover, the seasonal variability in ww
intensity is accounted for as the open SECR model includes temporal heterogeneity in sampling
effort. An implementation of this model would further avoid running a separate analysis for
each month but would also require a larger number of recaptured animals (Pirotta et al., 2015).
Overall, it would be a worthwhile effort to extend the present SECR model to an open
population scenario in order to increase the precision of the parameter estimates and reduce
potential bias.
For several cetacean species, additional energetic costs due to non-lethal human disturbance
were quantitatively determined (Bain, 2002; Farmer et al., 2018; Isojunno & Miller, 2015;
Rechsteiner et al., 2013; Srinivasan et al., 2018). Such bioenergetic models estimate
consumption rates based on the energetic requirements for reproduction, growth, metabolism,
!
$#!
and waste (Brodie et al., 2016). These models are a useful tool when it comes to species
conservation, as they contribute to an evaluation of the biological significance of human
disturbance on a population-level. Appropriate longitudinal data is required together with the
corresponding ww intensity and the cumulative exposure rates to assess variations in the
distribution and reproduction of sperm whales in the area (Richter et al., 2006). The additional
information about bioenergetic costs of disturbance during interactions with ww boats could be
combined with the present model to estimate exposure rates and establish a link to the
alterations of body conditions and ultimately survival (Athreya et al., 2013).
The overall low estimated encounter probabilities, and correspondingly exposure, seemed to be
further influenced by the low number of re-sighted animals within a sampling occasion (i.e.
month). In order to increase the number of re-sightings and the precision of the model estimates,
future studies could adapt the present SECR model by treating each year as a single sampling
occasion (Christiansen et al., 2015). Extended comparison of spatial variations in exposure
between the years could give additional information on the cumulative impact of ww activities
on sperm whales illustrating changes over a larger temporal scale. As the exposure was only
estimated for one year in this study, although the source of disturbance has been evolving over
several decades, it is likely that the observed spatial distribution could have already been a
response to the ww activity in the region. Therefore, future studies should investigate, if the
low rate of re-sightings in the area already indicates avoidance strategies due to increased
energetic costs in the presence of ww boats and if this could cause long-term habitat shifts.
4.3 Conservation implications
Sperm whales are the basis for a commercially important ww industry in the Azores central
group. A comprehensive understanding of the linkage between short-term behavioral responses
and long-term effects on individual vital rates are needed to address the management of non-
lethal human disturbance and the conservation of the sperm whale population in the area.
Therefore, this study aimed to quantitatively estimate the cumulative impact of ww activities
on individual-level, emphasizing the importance of assessing individual exposure in order to
contribute to management implications. So far, past research only focused on changes in inter-
breathing intervals to characterize short-term effects of ww activities in sperm whales in the
area (Magalhães et al., 2002). Meanwhile, more recent studies off Norway, found that sperm
whales significantly increase their surface time (75%), as they were almost seven times more
!
$$!
likely to perform near-surface-events (NSEs) in the presence of ww boats (Cosentino, 2016).
This time will be no longer available for foraging or resting activities. Therefore, it is important
to account for NSEs and respiration dynamics in such impact studies, as they are suitable
indicators of human disturbance. Aiming to reduce exposure and to avoid negative
consequences of cumulative impacts, ww boats should avoid targeting the same individual and
animals indicating signs of disturbance, such as NSEs (Cosentino, 2016).
The present study estimated a mean cumulative interaction time with ww boats of less than 3.6
hours per month for the most exposed individual. This ‘small’ effect sizes may not have a
biological significant impact on the population. However, the estimates were based on data
from a small-temporal scale and direct measurements of seasonal energetic costs were not
available. In addition, individual exposure rates are likely to be underestimated, as AIS B
information was not available for all ww boats that operated in the area during the time period
of the study. Hence, a precautionary management approach is recommended, including a long-
term monitoring of the population, aiming to establish a link between individual survival rate
and potential negative effects of ww activities in the future (Booth et al., 2020). Moreover, the
sources of anthropogenic disturbances in the area are diverse. Therefore, it is recommended to
evaluate the impact of ww activities in relation to other human activities that might affect sperm
whales (e.g. commercial fisheries, marine traffic) to enhance management decisions.
The relative locations of the animals’ activity center, as well as their movement range around
it, were subject to inter-mensal variations, implying that some individuals that have their
activity center closer to the areas of increased ww intensity and/or a more spatially concentrated
movement around it, are more exposed than others. Although the European Union implemented
the assessment of cumulative effects on individual-level to human disturbance into legislation
(Cooper & Sheate, 2002), such spatial differences require the establishment of a more flexible
management scheme. Overall it can be concluded, that results emphasize the importance of
incorporating analytical approaches accounting for individual heterogeneity in behavior and
habitat use into exposure assessment (Pirotta et al., 2015). For instance, individuals that are
more vulnerable to effects of disturbance (mother, calves) or animals whose spatial distribution
coincide with areas of increased human disturbance fulfill the need of special conservation.
!
!
$%!
!
5. Conclusion
In conclusion, present results indicated that both, the cumulative exposure of sperm whales to
ww activities (Figure 8) and the ww intensity (Figure 7) varied spatially between months. In
May and June, where most individuals were re-sighted at least twice (Table 2), the cumulative
interaction time was significantly higher (Table 4) and led to a positive skewness in the
estimated frequency distributions for the encountered individuals (Figure 10). This suggests
that the detection of significant mean exposure levels to ww activity implied severe effects on
at least some individuals. Results emphasize the importance of a population-level management
under the consideration of the complete cumulative exposure including impacts on individuals
rather than the mean. Although the overall cumulative effects seemed to be relatively low in
the current state of exposure, an efficient long-time monitoring is needed together with seasonal
data of ww intensity to ensure an appropriate conservation of the local sperm whale population.
Concerning the management of ww activities, a precautionary approach is suggested, as the
ecotourism sector is continuously growing while long-term effects on sperm whales in the area
are yet poorly understood. Despite facing some limitation, the present SECR model provided
the first robust baseline estimates for cumulative exposure of sperm whales to ww activities in
the area. Furthermore, the model allows for extension with auxiliary data on bioenergetic costs,
as well as open model scenarios accounting for individual heterogeneity in capture
probabilities. Encouraging future research to further improve the precision of exposure
estimates and comprehensively evaluate potential long-term effects to guarantee the
sustainability of this industry.
!
$<!
Acknowledgements
!
Above all I would like to express my greatest gratitude to PhD Sergi Pérez-Jorge for his constant
support during the entire project. He gave me the unique opportunity to get an insight into
several interesting research fields and invited me to participate in the group meetings, always
helping me to advance my skills and improve my knowledge.
Moreover, I would like to thank Prof. Dr. Christian Möllmann for giving me the opportunity to
accomplish my master’s thesis under his supervision.
A special thanks goes to PhD Rui Prieto, who further gave me the opportunity to participate in
the “Watch it” project, as well as PhD Mónica A. Silva and all other scientists of the working
group who supported my research with inspiring ideas and critical questions, constantly giving
me helpful advices on the analysis of my data.
I am highly appreciating the outstanding collaboration of the many tour operators which
provided their data. The following contributed to the establishment of a comprehensive photo-
identification data base: Norberto Diver (Lisa Steiner, Alba Iglesia), Horta Cetáceos (Pedro F.
Cedeu, Anna Sánchez Mora, Sanne), Peterzee (Miguel Gonçalves), Naturalist (José N. Pereira,
Gisela Dionísio), Cetacean Watching (Enrico Villa), and Dive Azores (Tiago Castro, Joana
Vaz-Pereira). The following shared GPS and sighting information: Cetacean Watching (Enrico
Villa), Aqua Azores (Paula Dutra), Horta Cetáceos (Pedro F. Cedeu, Anna Sánchez Mora,
Sanne), Naturalist (José N. Pereira, Gisela Dionísio), and Peterzee (Rita). The following shared
daily trip information: Norberto Diver (Sophia Fontes).
I am especially thankful to Frederick Christiansen who contributed to this work by sharing his
data and giving me important hints on the statistical analysis. I would also like to thank Murray
G. Efford and Arjun M. Gopalaswamy for their time and helpful feedback on the SECR models.
Without my family who always supported and believed in me and my dearest life companion,
Miguel Silva who has never left my side despite endless ups and downs, I would have never
been able to accomplish this work, I am overly grateful for having them in my life.
!
$*!
References
Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on
Automatic Control, 19(6), 716–723.
Alves, F., Alessandrini, Anita, Servidio, A., Mendonça, A. S., Hartman, K. L., Prieto, R., …
Aguilar de Soto, N. (2019). Complex biogeographical patterns support an ecological
connectivity network of a large marine predator in the north-east Atlantic. Diversity and
Distributions, 25(2), 269–284. https://doi.org/10.1111/ddi.12848
Antunes, R., Schulz, T., Gero, S., Whitehead, H., Gordon, J., & Rendell, L. (2009).
Individually distinctive acoustic features in sperm whale codas. Animal Behaviour,
81(4), 723-730.
Arnbom, T. (1987). Individual identification of sperm whales. Report of the International
Whaling Commission, 37(20), 1–204.
Athreya, V., Odden, M., Linnell, J. D. C., Krishnaswamy, J., & Karanth, U. (2013). Big Cats
in Our Backyards: Persistence of Large Carnivores in a Human Dominated Landscape in
India. PLoS ONE, 8(3), 2–9. https://doi.org/10.1371/journal.pone.0057872
Bain, D. E. (2002). A model linking energetic effects of whale watching to killer whale
(Orcinus orca) population dynamics. Friday Harbor Laboratories University of
Washington Friday Harbor Washington, 23.
Beale, C. M., & Monaghan, P. (2004). Human disturbance: People as predation-free
predators? Journal of Applied Ecology, 41(2), 335–343. https://doi.org/10.1111/j.0021-
8901.2004.00900.x
Beekmans, B. W. P. M., Whitehead, H., Huele, R., Steiner, L., & Steenbeek, A. G. (2005).
Comparison of Two Computer-Assisted Photo-Identification Methods Applied to Sperm
Whales (Physeter macrocephalus). Aquatic Mammals, 31(2), 243–247.
https://doi.org/10.1578/am.31.2.2005.243
Bejder, L., Samuels, A., Whitehead, H., Finn, H., & Allen, S. (2009). Impact assessment
research: Use and misuse of habituation, sensitisation and tolerance in describing
wildlife responses to anthropogenic stimuli. Marine Ecology Progress Series, 395, 177–
185. https://doi.org/10.3354/meps07979
Bejder, Lars, Samuels, A., Whitehead, H., & Gales, N. (2006). Interpreting short-term
behavioural responses to disturbance within a longitudinal perspective. Animal
Behaviour, 72(5), 1149–1158. https://doi.org/10.1016/j.anbehav.2006.04.003
Bejder, Lars, Samuels, A., Whitehead, H., Gales, N., Mann, J., Connor, R., … Krützen, M.
!
$=!
(2006). Decline in relative abundance of bottlenose dolphins exposed to long-term
disturbance. Conservation Biology, 20(6), 1791–1798. https://doi.org/10.1111/j.1523-
1739.2006.00540.x
Booth, C. G., Sinclair, R. R., & Harwood, J. (2020). Methods for Monitoring for the
Population Consequences of Disturbance in Marine Mammals: A Review. Frontiers in
Marine Science, 7(2), 1–18. https://doi.org/10.3389/fmars.2020.00115
Borchers, D. L., & Efford, M. G. (2008). Spatially explicit maximum likelihood methods for
capture-recapture studies. Biometrics, 64(2), 377–385. https://doi.org/10.1111/j.1541-
0420.2007.00927.x
Boys, R. M., Oliveira, C., PérezBJorge, S., Prieto, R., Steiner, L., & Silva, M. A. (2019).
MultiBstate open robust design applied to opportunistic data reveals dynamics of wideB
ranging taxa: the sperm whale case. Ecosphere, 10(3), e02610.
https://doi.org/10.1002/ecs2.2610
Brasseur, S., Brensing, K., Clarke, M., Duck, C., S., G., J., H., & Jepson, P. (2010). Report of
the Working Group on Marine Mammal Ecology (WGMME). Horta, Azores.
Cheoo, G. V. (2019). Estimation of bowhead whale (Balaena mysticetus) population density
using spatially explicit capture-recapture (SECR) methods.
Childerhouse, S. J., Dawson, S. M., & Slooten, E. (1996). Stability of fluke marks used in
individual photoidentification of male sperm whales at Kaikoura, New Zealand. Marine
Mammal Science, 12(3), 447–451. https://doi.org/10.1111/j.1748-7692.1996.tb00597.x
Christiansen, F., Bertulli, C. G., Rasmussen, M. H., & Lusseau, D. (2015). Estimating
cumulative exposure of wildlife to non-lethal disturbance using spatially explicit capture-
recapture models. Journal of Wildlife Management, 79(2), 311–324.
https://doi.org/10.1002/jwmg.836
Christiansen, F., & Lusseau, D. (2014a). Understanding the ecological effects of whale-
watching on cetaceans. Whale-watching: Sustainable Tourism and Ecological
Management, 177–192.
Christiansen, F., Rasmussen, M. H., & Lusseau, D. (2014b). Inferring energy expenditure
from respiration rates in minke whales to measure the effects of whale watching boat
interactions. Journal of Experimental Marine Biology and Ecology, 459, 96–104.
https://doi.org/10.1016/j.jembe.2014.05.014
Christiansen, F., Rasmussen, M., & Lusseau, D. (2013). Whale watching disrupts feeding
activities of minke whales on a feeding ground. Marine Ecology Progress Series, 478,
239–251. https://doi.org/10.3354/meps10163
!
$'!
Clarke, R. (1956). Sperm Whales of the Azores (28:237-98.). Discovery Reports.
Cooper, L. M., & Sheate, W. R. (2002). Cumulative effects assessment: A review of UK
environmental impact statements. Environmental Impact Assessment Review, 22(4),
415–439. https://doi.org/10.1016/S0195-9255(02)00010-0
Cosentino, A. M. (2016). Effects of whale-watching vessels on adult male sperm whales off
Andenes, Norway. Tourism in Marine Environments, 11(4), 215–227.
https://doi.org/10.3727/154427316X14580612748560
Dähne, M., Gilles, A., Lucke, K., Peschko, V., Adler, S., Krügel, K., … Siebert, U. (2013).
Effects of pile-driving on harbour porpoises (Phocoena phocoena) at the first offshore
wind farm in Germany. Environmental Research Letters, 8(2).
https://doi.org/10.1088/1748-9326/8/2/025002
David, J. A. (2006). Likely sensitivity of bottlenose dolphins to pile-driving noise. Water and
Environment Journal, 20(1), 48–54. https://doi.org/10.1111/j.1747-6593.2005.00023.x
Deruiter, S. L., Southall, B. L., Calambokidis, J., Zimmer, W. M. X., Sadykova, D., Falcone,
E. A., … Tyack, P. L. (2013). First direct measurements of behavioural responses by
Cuvier’s beaked whales to mid-frequency active sonar. Biology Letters, 9(4), 0–4.
https://doi.org/10.1098/rsbl.2013.0223
Doughty, C. E., Roman, J., Faurby, S., Wolf, A., Haque, A., Bakker, E. S., … Svenning, J. C.
(2016). Global nutrient transport in a world of giants. Proceedings of the National
Academy of Sciences of the United States of America, 113(4), 868–873.
https://doi.org/10.1073/pnas.1502549112
Dufault, S., & Whitehead, H. (1998). Regional and Group-Level Differences in Fluke
Markings and Notches of Sperm Whales. Journal of Mammalogy, 79(2), 514–520.
https://doi.org/10.2307/1382982
Dufault, Susan, & Whitehead, H. (1995). An assessment of changes with time in the marking
patterns used for of individual. Marine Mammal Science, 11(3), 335–343.
Efford, M. (2004). Density estimation in live-trapping studies. Oikos, 106(3), 598–610.
https://doi.org/10.1111/j.0030-1299.2004.13043.x
Efford, M. (2015). Finite mixture models in secr 2.10. 7. Retrieved from
http://www.otago.ac.nz/density/pdfs/secr-finitemixtures.pdf
Efford, M. G. (2011). Estimation of population density by spatially explicit capture-recapture
analysis of data from area searches. Ecology, 92(12), 2202–2207.
https://doi.org/10.1890/11-0332.1
Efford, M. G. (2019). secr: Spatially explicit capture-recapture models. Retrieved from
!
%)!
https://cran.r-project.org/package=secr
Efford, M. G., Borchers, D. L., & Byrom, A. E. (2009a). Density Estimation by Spatially
Explicit Capture-Recapture: Likelihood-Based Methods. Environmental and Ecological
Statistics 3.
Efford, M. G., Borchers, D. L., & Mowat, G. (2013). Varying effort in capture-recapture
studies. Methods in Ecology and Evolution, 4, 629–636.
Efford, M. G., & Boulanger, J. (2019). Fast evaluation of study designs for spatially explicit
capture–recapture. Methods in Ecology and Evolution, 10(9), 1529–1535.
https://doi.org/10.1111/2041-210X.13239
Efford, M. G., Dawson, D. K., & Borchers, D. L. (2009b). Population density estimated from
locations of individuals on a passive detector array. Ecology, 90(10), 2676–2682.
Retrieved from http://www.esajournals.org/doi/abs/10.1890/08-1735.1
Efford, M. G., & Fewster, R. M. (2013). Estimating population size by spatially explicit
capture-recapture. Oikos, 122(6), 918–928. https://doi.org/10.1111/j.1600-
0706.2012.20440.x
Efford, M. G., & Schofield, M. R. (2019). A spatial open-population capture--recapture
model. Biometrics, 0–1. https://doi.org/10.1111/biom.13150
Efford, M. G., Warburton, B., Coleman, M. C., & Barker, R. J. (2005). A field test of two
methods for density estimation. Wildlife Society Bulletin, 33(2), 731–738.
Farmer, N. A., Noren, D. P., Fougères, E. M., Machernis, A., & Baker, K. (2018). Resilience
of the endangered sperm whale Physeter macrocephalus to foraging disturbance in the
Gulf of Mexico, USA: A bioenergetic approach. Marine Ecology Progress Series,
589(Hoyt 2012), 241–261. https://doi.org/10.3354/meps12457
Frid, A., & Dill, L. (2002). Human-caused disturbance stimuli as a form of predation risk.
Ecology and Society, 6(1). https://doi.org/10.5751/es-00404-060111
García-Cegarra, A. M., & Pacheco, A. S. (2017). Whale-watching trips in Peru lead to
increases in tourist knowledge, pro-conservation intentions and tourist concern for the
impacts of whale-watching on humpback whales. Aquatic Conservation: Marine and
Freshwater Ecosystems, 27(5), 1011–1020. https://doi.org/10.1002/aqc.2754
Garcia-Cegarra, A. M., Villagra, D., Gallardo, D. I., & Pacheco, A. S. (2019). Statistical
dependence for detecting whale-watching effects on humpback whales. Journal of
Wildlife Management, 83(2), 467–477. https://doi.org/10.1002/jwmg.21602
Gill, J. A., Norris, K., & Sutherland, W. J. (2001). Why behavioural responses may not reflect
the population consequences of human disturbance. Biological Conservation, 97(2),
!
%(!
265–268. https://doi.org/10.1016/S0006-3207(00)00002-1
Gimenez, O., Gatti, S., Duchamp, C., Germain, E., Laurent, A., Zimmermann, F., &
Marboutin, E. (2019). Spatial density estimates of Eurasian lynx (Lynx lynx) in the
French Jura and Vosges Mountains. Ecology and Evolution, 9(20), 11707–11715.
https://doi.org/10.1002/ece3.5668
Goddard, M., Buckton, C., Leahy, L., & Lindsay, M. (2013). A community survey of
humpback whales, Megaptera novaeangliae, near the site of the proposed James Price
Point Browse Liquefied Natural Gas precinct. (March).
Gomes Pereira, J. N. D. S. (2008). Daily species checklist from whale-watching - Studying
the research potential with an Azorean case study. Journal of the Marine Biological
Association of the United Kingdom, 88(6), 1283–1288.
https://doi.org/10.1017/S0025315408001070
Gomez, C., Lawson, J., Kouwenberg, A. L., Moors-Murphy, H., Buren, A., Fuentes-Yaco, C.,
… Wimmer, T. (2017). Predicted distribution of whales at risk: Identifying priority areas
to enhance cetacean monitoring in the Northwest Atlantic Ocean. Endangered Species
Research, 32(1), 437–458. https://doi.org/10.3354/esr00823
Gopalaswamy, A. M., Royle, J. A., Hines, J. E., Singh, P., Jathanna, D., Kumar, N. S., &
Karanth, K. U. (2012). Program SPACECAP: Software for estimating animal density
using spatially explicit capture-recapture models. Methods in Ecology and Evolution,
3(6), 1067–1072. https://doi.org/10.1111/j.2041-210X.2012.00241.x
Gordon, J., Leaper, R., Hartley, F. G., & Chappell, O. (1992). Effects of whale watching
vessels on the surface and underwater acoustic behaviour of sperm whales off Kaikoura,
New Zealand. In Science and research series.
Governo Regional dos Açores (2020). Atividade Marítimo-Turística. (March, 3). Retrieved
from http://www.azores.gov.pt/Portal/pt/entidades/srtop-
drt/textoImagem/Actividade_Mar%C3%ADtimo-Tur%C3%ADstica.htm
Halpern, B. S., Frazier, M., Potapenko, J., Casey, K. S., Koenig, K., Longo, C., … Walbridge,
S. (2015). Spatial and temporal changes in cumulative human impacts on the world’s
ocean. Nature Communications, 6(May), 1–7. https://doi.org/10.1038/ncomms8615
Harris, C. M., Thomas, L., Falcone, E. A., Hildebrand, J., Houser, D., Kvadsheim, P. H., …
Janik, V. M. (2018). Marine mammals and sonar: Dose-response studies, the risk-
disturbance hypothesis and the role of exposure context. Journal of Applied Ecology,
55(1), 396–404. https://doi.org/10.1111/1365-2664.12955
Hazen, E. L., Palacios, D. M., Forney, K. A., Howell, E. A., Becker, E., Hoover, A. L., …
!
%"!
Bailey, H. (2017). WhaleWatch: a dynamic management tool for predicting blue whale
density in the California Current. Journal of Applied Ecology, 54(5), 1415–1428.
https://doi.org/10.1111/1365-2664.12820
Hooker, S. K., Fahlman, A., Moore, M. J., Aguilar de Soto, N., Bernaldo de Quirós, Y.,
Brubakk, A. O., … Tyack, P. L. (2012). Deadly diving? physiological and behavioural
management of decompression stress in diving mammals. Proceedings of the Royal
Society B: Biological Sciences, 279(1731), 1041–1050.
https://doi.org/10.1098/rspb.2011.2088
Isojunno, S., & Miller, P. J. O. (2015). Sperm whale response to tag boat presence:
biologically informed hidden state models quantify lost feeding opportunities.
Ecosphere, 6(1). https://doi.org/10.1890/ES14-00130.1
Karanth, K. U., & Nichols, J. D. (1998). Estimation of tiger densities in India using
photographic captures and recaptures. Ecology, 79(8), 2852–2862.
https://doi.org/10.1890/0012-9658(1998)079[2852:EOTDII]2.0.CO;2
Lesage, V., Omrane, A., Doniol-Valcroze, T., & Mosnier, A. (2017). Increased proximity of
vessels reduces feeding opportunities of blue whales in the St. Lawrence Estuary,
Canada. Endangered Species Research, 32(1), 351–361.
https://doi.org/10.3354/esr00825
Lusseau, D., & Bejder, L. (2007). The long-term consequences of short-term responses to
disturbance experiences from whalewatching impact assessment. International Journal
of Comparative Psychology, 20, 228–236.
Magalhães, S., Prieto, R., Silva, M. A., Gonçalves, J., Afonso-Dias, M., & Santos, S. (2002).
Short-term reactions of sperm whales (Physeter macrocephalus) to whale-watching
vessels in the Azores. Aquatic Mammals, 28(May 2014), 267–274.
Marley, S. A., Salgado Kent, C. P., Erbe, C., & Parnum, I. M. (2017). Effects of vessel traffic
and underwater noise on the movement, behaviour and vocalisations of bottlenose
dolphins in an urbanised estuary. Scientific Reports, 7(1), 1–14.
https://doi.org/10.1038/s41598-017-13252-z
Marques, T. A., Thomas, L., Martin, S. W., Mellinger, D. K., Jarvis, S., Morrissey, R. P., …
DiMarzio, N. (2012). Spatially explicit capture-recapture methods to estimate minke
whale density from data collected at bottom-mounted hydrophones. In Journal of
Ornithology (Vol. 152). https://doi.org/10.1007/s10336-010-0535-7
Marques, T. A., Thomas, L., Martin, S. W., Mellinger, D. K., Ward, J. A., Moretti, D. J., …
Tyack, P. L. (2013). Estimating animal population density using passive acoustics.
!
%#!
Biological Reviews, 88(2), 287–309. https://doi.org/10.1111/brv.12001
Matthews, J. N., Steiner, L., & Gordon, J. (2001). Mark-recapture analysis of sperm whale
(Physeter macrocephalus) photo-id data from the Azores ( 1987-1995 ). Journal of
Cetacean Research and Management, 3(3), 219–226.
Maxwell, S. M., Hazen, E. L., Bograd, S. J., Halpern, B. S., Breed, G. A., Nickel, B., …
Costa, D. P. (2013). Cumulative human impacts on marine predators. Nature
Communications, 4(October). https://doi.org/10.1038/ncomms3688
Mazzoldi, C., Bearzi, G., Brito, C., Carvalho, I., Desiderà, E., Endrizzi, L., … MacDiarmid,
A. (2019). From sea monsters to charismatic megafauna: change in perception and use of
large marine animals. In PLoS One.
Messmer, T. A. (2000). The emergence of human-wildlife conflict management: Turning
challenges into opportunities. International Biodeterioration and Biodegradation, 45(3–
4), 97–102. https://doi.org/10.1016/S0964-8305(00)00045-7
Morete, M. E., Bisi, T. L., & Rosso, S. (2007). Mother and calf humpback whale responses to
vessels around the Abrolhos Archipelago, Bahia, Brazil. Journal of Cetacean Research
and Management, 9(3), 241–248.
Morton, B., Britton, J. C., & Martins, A. D. F. (1998). Ecologia costeira dos Açores. Ponta
Delgada: Sociedade Afonso Chaves.
National Research Council. (2005). Marine mammal populations and ocean noise:
Determining when noise causes biologically significant effects. Washington D.C.: The
National Academies Press.
Neves-Graça, K. (2004). Revisiting the tragedy of the commons: Ecological dilemmas of
whale watching in the Azores. Human Organization, 63(3), 289–300.
https://doi.org/10.17730/humo.63.3.tae1vdl7573hrbt7
New, L. F., Hall, A. J., Harcourt, R., Kaufman, G., Parsons, E. C. M., Pearson, H. C., …
Schick, R. S. (2015). The modelling and assessment of whale-watching impacts. Ocean
and Coastal Management, 115, 10–16. https://doi.org/10.1016/j.ocecoaman.2015.04.006
Oliveira, C., Filla, G., Gonçlaves, J., Silva, M., Prieto, R., Magalhães, A., & Santos, R.
(2007a). A social-economic perspective of the whale watching activity in the Azores.
International Whaling Commission Conference, Anchorage, Alaska, (June 2016), 1–8.
https://doi.org/SC/59/WW8
Oliveira, C., Gonçalves, J., Magalhães, S., Prieto, R., Silva, M., & Santos, R. (2007b). Whale
watching management in the Azores: An updated review of the regulations.
International Whaling Commission Conference, Anchorage, Alaska, (June 2016), 1–4.
!
%$!
https://doi.org/SC/59/ww7
Oliveira, C. I. B. (2014). Behavioural ecology of the sperm whale (Physeter macrocephalus)
in the North Atlantic Ocean. 119. https://doi.org/10.13140/RG.2.1.2461.6486
Otis, D. L., Green, R. F., Burnham, K. P., White, G. C., & Anderson, D. R. (1978). Statistical
Inference From Capture Data on Closed Animal Populations. Journal of the American
Statistical Association, 76(374), 498. https://doi.org/10.2307/2287873
Parsons, E. C. M. (2012). The Negative Impacts of Whale-Watching. Journal of Marine
Biology, 2012, 1–9. https://doi.org/10.1155/2012/807294
Pérez-Jorge, S., Louzao, M., Oro, D., Pereira, T., Corne, C., Wijtten, Z., … Christiansen, F.
(2017). Estimating the cumulative effects of the nature-based tourism in a coastal
dolphin population from southern Kenya. Deep-Sea Research Part II: Topical Studies in
Oceanography, 140, 278–289. https://doi.org/10.1016/j.dsr2.2016.08.011
Pirotta, E., Thompson, P. M., Cheney, B., Donovan, C. R., & Lusseau, D. (2015). Estimating
spatial, temporal and individual variability in dolphin cumulative exposure to boat traffic
using spatially explicit capture-recapture methods. Animal Conservation, 18(1), 20–31.
https://doi.org/10.1111/acv.12132
Pirotta, Enrico, Booth, C. G., Costa, D. P., Fleishman, E., Kraus, S. D., Lusseau, D., …
Harwood, J. (2018). Understanding the population consequences of disturbance. Ecology
and Evolution, 8(19), 9934–9946. https://doi.org/10.1002/ece3.4458
Pirotta, V., Grech, A., Jonsen, I. D., Laurance, W. F., & Harcourt, R. G. (2019).
Consequences of global shipping traffic for marine giants. Frontiers in Ecology and the
Environment, 17(1), 39–47. https://doi.org/10.1002/fee.1987
Pledger, S. (2000). Unified maximum likelihood estimates for closed capture-recapture
models using mixtures. Biometrics, 56(2), 434–442. https://doi.org/10.1111/j.0006-
341X.2000.00434.x
Pledger, S., Pollock, K. H., & Norris, J. L. (2010). Open capture-recapture models with
heterogeneity: II. Jolly-Seber model. Biometrics, 66(3), 883–890.
https://doi.org/10.1111/j.1541-0420.2009.01361.x
Prieto, R., Silva, M. A., Magalhães, S. M., Cabecinhas, R., & Santos, R. S. (2000).
Desenvolvimento do ‘Whale Watching’ nos Açores.
R Core Development Team. (2019). R: A Language and Environment for Statistical
Computing. Retrieved from https://www.r-project.org/
Rechsteiner, E. U., Rosen, D. A. S., & Trites, A. W. (2013). Energy requirements of Pacific
white-sided dolphins (Lagenorhynchus obliquidens) as predicted by a bioenergetic
!
%%!
model. Journal of Mammalogy, 94(4), 820–832. https://doi.org/10.1644/12-mamm-a-
206.1
Richter, C., Dawson, S., & Slooten, E. (2006). Impacts of commercial whale watching on
male sperm whales at Kaikoura, New Zealand. Marine Mammal Science, 22(1), 46–63.
https://doi.org/10.1111/j.1748-7692.2006.00005.x
Roman, J., Estes, J. A., Morissette, L., Smith, C., Costa, D., McCarthy, J., … Smetacek, V.
(2014). Whales as marine ecosystem engineers. Frontiers in Ecology and the
Environment, 12(7), 377–385. https://doi.org/10.1890/130220
Royle, J. A., Karanth, K. U., Gopalaswamy, A. M., & Kumar, N. S. (2009a). Bayesian
inference in camera trapping studies for a class of spatial capture-recapture models.
Ecology, 90(11), 3233–3244. https://doi.org/10.1890/08-1481.1
Royle, J. A., Nichols, J. D., Karanth, K. U., & Gopalaswamy, A. M. (2009b). A hierarchical
model for estimating density in camera-trap studies. Journal of Applied Ecology, 46,
118–127.
Royle, J. A., & Young, K. V. (2008). A hierarchical model for spatial capture-recapture data.
Ecology, 89(8), 2281–2289. https://doi.org/10.1890/10-1440.1
Russell, R. E., Royle, J. A., Desimone, R., Schwartz, M. K., Edwards, V. L., Pilgrim, K. P., &
McKelvey, K. S. (2012). Estimating abundance of mountain lions from unstructured
spatial sampling. Journal of Wildlife Management, 76(8), 1551–1561.
https://doi.org/10.1002/jwmg.412
Schaffar, A., Madon, B., Garrigue, C., & Constantine, R. (2013). Behavioural effects of
whale-watching activities on an Endangered population of humpback whales wintering
in New Caledonia. Endangered Species Research, 19(3), 245–254.
https://doi.org/10.3354/esr00466
Scheidat, M., Castro, C., Gonzalez, J., & Williams, R. (2004). Behavioural responses of
humpback whales (Megaptera novaeangliae) to whalewatching boats near Isla de la
Plata, Machalilla National Park, Ecuador. Journal of Cetacean Research and
Management, 6(1), 63–68.
Sequeira, M., Elejabeitia, C., Silva, M. A., Dinis, A., de Stephanis, R., Urquiola, E., …
Freitas, L. (2009). Review of whalewatching activities in mainland Portugal, the Azores,
Madeira and Canary archipelagos and the Strait of Gibraltar. Journal of Cetacean
Research Management, SC61/WW11(January), 1–40.
Sibert, J. R., Musyl, M. K., & Brill, R. W. (2003). Horizontal movements of bigeye tuna
(Thunnus obesus) near Hawaii determined by Kalman filter analysis of archival tagging
!
%<!
data. Fisheries Oceanography, 12(3), 141–151. https://doi.org/10.1046/j.1365-
2419.2003.00228.x
Silber, G. K., Lettrich, M. D., Thomas, P. O., Baker, J. D., Baumgartner, M., Becker, E. A.,
… Waples, R. S. (2017). Projecting marine mammal distribution in a changing climate.
Frontiers in Marine Science, 4(DEC). https://doi.org/10.3389/fmars.2017.00413
Silva, L. (2015). How ecotourism works at the community-level: the case of whale-watching
in the Azores. Current Issues in Tourism, 18(3), 196–211.
https://doi.org/10.1080/13683500.2013.786027
Silva, M. A., Magalhães, S. M., Prieto, R., Quérouil, S., Pinela, A., Seabra, M. I., & Santos,
R. S. (2006). Ecologia e estrutura populacional dos roazes e cachalotes nos Açores:
relação com as características do habitat. Arquivos DOP Á Série Estudos, 4.
Silva, M. A., Prieto, R., Cascão, I., Seabra, M. I., Machete, M., Baumgartner, M. F., &
Santos, R. S. (2014). Spatial and temporal distribution of cetaceans in the mid-Atlantic
waters around the Azores. Marine Biology Research, 10(2), 123–137.
https://doi.org/10.1080/17451000.2013.793814
Silva, M. S. A. (2007). Population Biology of Bottlenose Dolphins in the Azores Archipelago.
School of Biology, University of St. Andrews, 254.
Sollmann, R., Gardner, B., Parsons, A. W., Stocking, J. J., McClintock, B. T., Simons, T. R.,
… O’Connell, A. F. (2013). A spatial mark–resight model augmented with telemetry
data. Ecology, 94(3), 553–559.
Sprogis, K. R., Raudino, H. C., Hocking, D., & Bejder, L. (2017). Complex prey handling of
octopus by bottlenose dolphins (Tursiops aduncus). Marine Mammal Science, 33(3),
934–945. https://doi.org/10.1111/mms.12405
Srinivasan, M., Swannack, T. M., Grant, W. E., Rajan, J., & Würsig, B. (2018). To feed or not
to feed? Bioenergetic impacts of fear-driven behaviors in lactating dolphins. Ecology and
Evolution, 8(2), 1384–1398. https://doi.org/10.1002/ece3.3732
Steiner, L., Lamoni, L., Plata, M. A., Jensen, S. K., Lettevall, E., & Gordon, J. (2012). A link
between male sperm whales, Physeter macrocephalus, of the Azores and Norway.
Journal of the Marine Biological Association of the United Kingdom, 92(8), 1751–1756.
https://doi.org/10.1017/S0025315412000793
Steiner, L., Machete, M., & Hammerschlag, M. (2017). Photo-identification and surveys of
cetaceans in the central group of the Azores islands. Biosphere Expeditions, 1–100.
Stevick, P. T., Palsbøll, P. J., Smith, T. D., Bravington, M. V., & Hammond, P. S. (2001).
Errors in identification using natural markings: Rates, sources, and effects on capture-
!
%*!
recapture estimates of abundance. Canadian Journal of Fisheries and Aquatic Sciences,
58(9), 1861–1870. https://doi.org/10.1139/cjfas-58-9-1861
Sumner, M. D., Luque, S., Fischbach, A., & Hengl, T. (2019). trip: Tools for the Analysis of
Animal Track Data.
Surma, S., & Pitcher, T. J. (2015). Predicting the effects of whale population recovery on
Northeast Pacific food webs and fisheries: An ecosystem modelling approach. Fisheries
Oceanography, 24(3), 291–305. https://doi.org/10.1111/fog.12109
Thompson, C. M., Royle, J. A., & Garner, J. D. (2012). A framework for inference about
carnivore density from unstructured spatial sampling of scat using detector dogs. Journal
of Wildlife Management, 76(4), 863–871. https://doi.org/10.1002/jwmg.317
Tobeña, M., Prieto, R., Machete, M., & Silva, M. A. (2016). Modeling the potential
distribution and richness of cetaceans in the Azores from fisheries observer program
data. Frontiers in Marine Science, 3(OCT). https://doi.org/10.3389/fmars.2016.00202
Trave, C., Brunnschweiler, J., Sheaves, M., Diedrich, A., & Barnett, A. (2017). Are we killing
them with kindness? Evaluation of sustainable marine wildlife tourism. Biological
Conservation, 209, 211–222. https://doi.org/10.1016/j.biocon.2017.02.020
Vandeperre, F., Aires-da-Silva, A., Lennert-Cody, C., Serrão Santos, R., & Afonso, P. (2016).
Essential pelagic habitat of juvenile blue shark (Prionace glauca) inferred from
telemetry data. Limnology and Oceanography, 61(5), 1605–1625.
https://doi.org/10.1002/lno.10321
Visser, F., Hartman, K. L., Rood, E. J. J., Hendriks, A. J. E., Zult, D. B., Wolff, W. J., …
Pierce, G. J. (2011). Risso’s dolphins alter daily resting pattern in response to whale
watching at the Azores. Marine Mammal Science, 27(2), 366–381.
https://doi.org/10.1111/j.1748-7692.2010.00398.x
Weinrich, M. T., & Corbelli, C. (2009). Whale watching in Southern New England does not
negatively impact humpback whale (Megaptera novaeangliae) calving rates or calf
survival. Biological Conservation, 142(12), 2931–2940.
Williams, R., Lusseau, D., & Hammond, P. S. (2006). Estimating relative energetic costs of
human disturbance to killer whales (Orcinus orca). Biological Conservation, 133(3),
301–311. https://doi.org/10.1016/j.biocon.2006.06.010
Witteveen, B. H., De Robertis, A., Guo, L., & Wynne, K. M. (2015). Using dive behavior and
active acoustics to assess prey use and partitioning by fin and humpback whales near
Kodiak Island, Alaska. Marine Mammal Science, 31(1), 255–278.
https://doi.org/10.1111/mms.12158
!
%=!
Appendix
!
Table A1. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during April 2018.
Trip numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
3
18
354
221
473
127
1
0
Faidoca
3
10
196
185
207
11
1
0
Risso
4
11
158
110
207
42
1
0
Kometa
7
20
172
112
218
40
1
0.55
Bicho
Do Mar
5
14
169
139
197
28
1
0.5
Reefcat
0
0
0
0
0
0
0
0
Maisha
0
0
0
0
0
0
0
0
José
Azevedo
0
0
0
0
0
0
0
0
Table A2. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during May 2018.
Trip numbers and boat effort were estimated based on AIS B data.
!
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
17
69
244
116
484
119
1
0.43
Faidoca
20
62
186
68
253
48
1
0.58
Risso
27
86
192
61
233
37
2
0.51
Kometa
20
57
172
82
234
43
1
0.32
Bicho
Do Mar
27
89
198
126
251
30
1
0.51
Reefcat
22
76
208
169
256
23
1
0.47
Maisha
0
0
0
0
0
0
0
0
José
Azevedo
18
58
194
142
226
25
1
0.41
!
%'!
Table A3. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during June 2018.
Trip numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
23
82
213
135
413
70
1
0.46
Faidoca
30
160
319
71
1437
405
2
0.61
Risso
34
107
189
142
242
21
1
0.5
Kometa
24
71
178
62
605
99
1
0.69
Bicho
Do Mar
32
105
198
94
445
55
1
0.48
Reefcat
19
63
199
128
265
34
1
0.24
Maisha
4
10
152
68
224
65
1
0
José
Azevedo
12
35
176
108
246
44
2
0.53
Table A4. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during July 2018. Trip
numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip
duration [min]
Daily number of
trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
17
70
246
144
422
98
1
0.43
Faidoca
42
126
180
86
414
49
2
0.58
Risso
51
152
179
116
256
30
2
0.58
Kometa
47
132
168
98
273
36
2
0.68
Bicho
Do Mar
48
144
180
60
234
34
2
0.55
Reefcat
34
104
183
70
351
55
2
0.67
Maisha
23
67
175
81
221
31
1
0.46
José
Azevedo
19
53
168
60
236
49
1
0.4
!
<)!
Table A5. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during August 2018.
Trip numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
12
48
238
128
440
116
1
0.3
Faidoca
39
117
181
119
350
41
2
0.58
Risso
58
169
174
95
248
31
2
0.6
Kometa
54
183
204
63
554
132
2
0.81
Bicho
Do Mar
57
177
186
71
247
34
2
0.57
Reefcat
31
90
173
88
253
45
1
0.5
Maisha
47
133
170
65
263
34
2
0.67
José
Azevedo
27
88
196
139
248
26
1
0.39
Table A6. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during September
2018. Trip numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
10
42
252
142
423
95
1
0.46
Faidoca
16
46
173
70
490
96
1
0.44
Risso
33
99
180
92
224
32
1
0.51
Kometa
30
128
256
76
558
167
1
0.44
Bicho
Do Mar
34
108
190
92
260
34
2
0.6
Reefcat
18
53
178
68
296
60
1
0.47
Maisha
23
68
178
65
276
46
1
0.42
José
Azevedo
21
63
179
125
231
28
1
0.51
!
<(!
Table A7. Whale watching boat effort [hours], number of monthly and daily trips and estimated
mean duration of trips from boats with AIS B operating in the study area during October 2018.
Trip numbers and boat effort were estimated based on AIS B data.
Boat
Number
of whale
watching
trips
Whale
watching
boat effort
[h]
Whale watching trip duration
[min]
Daily number
of trips
Mean
Min.
Max.
SD
Mean
SD
Physeter
1
3
196
196
196
0
1
0
Faidoca
2
5
152
147
158
7
1
0
Risso
5
16
198
162
237
31
1
0.5
Kometa
2
5
136
122
150
20
1
0
Bicho
Do Mar
6
20
202
139
307
62
1
0
Reefcat
0
0
0
0
0
0
0
0
Maisha
1
3
166
166
166
0
1
0
José
Azevedo
1
2
92
92
92
0
1
0
!
Table A8. Summary of SECR models, detection function, model parameters, and brief model
description
Model
Detection
function
SECR function
Model description
M0
halfnormal
D~1 g0~1 2~1
Null model, g0 const. across individuals,
sampling occasions and detectors
M1
hazardrate
D~1 g0~1 2~1 z~1
Null model, g0 const. across individuals,
sampling occasions and detectors
M2
exponential
D~1 g0~1 2~1
Null model, g0 const. across individuals,
sampling occasions and detectors
M3
halfnormal
D~1 g0~1 2~1
Model accounts for effort,
Function: g(d)=g0 x exp[-d2/(22 2)]
M4
hazardrate
D~1 g0~1 2~1 z~1
Model accounts for effort,
Function: g(d)=g0 x [1-exp[-(d/&2)-z]]
M5
exponential
D~1 g0~1 2~1
Model accounts for effort,
Function: g(d)=g0 x exp(-d/2)
M6
halfnormal
D~1 g0~h2 2~1
pmix~h2
Individual heterogeneity in g0,
Model accounts for effort
M7
hazardrate
D~1 g0~h2 2~1 z~1
pmix~h2
Individual heterogeneity in g0,
Model accounts for effort
M8
exponential
D~1 g0~h2 2~1
pmix~h2
Individual heterogeneity in g0,
Model accounts for effort
M9
halfnormal
D~1 g0~1 2~h2
pmix~h2
Individual heterogeneity in 2,
Model accounts for effort
M10
hazardrate
D~1 g0~1 2~h2 z~1
pmix~h2
Individual heterogeneity in 2,
Model accounts for effort
!
<"!
M11
exponential
D~1 g0~1 2~h2
pmix~h2
Individual heterogeneity in 2,
Model accounts for effort
M12
halfnormal
D~1 g0~h2 2~h2
pmix~h2
Individual heterogeneity in g0 and 2,
Model accounts for effort
M13
hazardrate
D~1 g0~h2 2~h2 z~1
pmix~h2
Individual heterogeneity in g0 and 2,
Model accounts for effort
M14
exponential
D~1 g0~h2 2~h2
pmix~h2
Individual heterogeneity in g0 and 2,
Model accounts for effort
!
Table A9. Fitted SECR models for May 2018 based on corresponding sampling effort data
from observer boats and photo-identified sperm whale individuals during the month.
Model
Detection
function
npar
logLik
AIC
AICc
dAICc
AICcwt
M9
halfnormal
5
-253.3561
516.712
518.855
0.000
0.4166
M11
exponential
5
-253.9056
517.811
519.954
1.099
0.2405
M10
hazardrate
6
-253.1632
518.326
521.437
2.582
0.1146
M12
halfnormal
6
-253.3134
518.627
521.738
2.883
0.0986
M14
exponential
6
-253.7571
519.514
522.625
3.773
0.0633
M13
hazardrate
7
-252.4984
518.997
523.304
4.449
0.0453
M7
hazardrate
6
-255.6028
523.206
526.317
7.462
0.0114
M6
halfnormal
5
-257.6262
525.252
527.395
8.542
0.0058
M5
exponential
3
-260.3243
526.649
527.449
8.594
0.0057
M8
exponential
5
-258.4243
526.849
528.991
10.136
0
M4
hazardrate
4
-260.2215
528.443
529.822
10.967
0
M3
halfnormal
2
-265.7007
535.401
535.789
16.934
0
M1
hazardrate
4
-290.2174
588.435
589.814
70.959
0
M2
exponential
3
-292.9470
591.894
592.694
73.839
0
M0
halfnormal
3
-293.8325
593.665
594.465
75.61
0
Table A10. Fitted SECR models for June 2018 based on corresponding sampling effort data
from observer boats and photo-identified sperm whale individuals during the month.
Model
Detection
function
npar
logLik
AIC
AICc
dAICc
AICcwt
M3
halfnormal
3
-519.9562
1045.912
1046.293
0.000
0.2481
M9
halfnormal
5
-517.8296
1045.659
1046.643
0.350
0.2083
M10
hazardrate
6
-516.8758
1045.752
1047.152
0.859
0.1615
M12
halfnormal
6
-517.1730
1046.346
1047.746
1.453
0.1200
M13
hazardrate
7
-516.3398
1046.680
1048.578
2.285
0.0792
M5
exponential
3
-521.3158
1048.632
1049.013
2.720
0.0637
M4
hazardrate
4
-520.5903
1049.181
1049.826
3.533
0.0424
M6
halfnormal
5
-519.7568
1049.514
1050.497
4.204
0.0303
M11
exponential
5
-520.0138
1050.028
1051.011
4.718
0.0235
M14
exponential
6
-519.6203
1051.241
1052.641
6.348
0.0104
M8
exponential
5
-521.1352
1052.270
1053.254
6.961
0.0076
M7
hazardrate
6
-520.3534
1052.707
1054.107
7.814
0.0050
M2
exponential
3
-605.5663
1217.133
1217.514
171.221
0
M0
halfnormal
3
-605.6553
1217.311
1217.692
171.399
0
M1
hazardrate
4
-605.9889
1219.978
1220.623
174.330
0
!
<#!
Table A11. Fitted SECR models for July 2018 based on corresponding sampling effort data
from observer boats and photo-identified sperm whale individuals during the month.
Model
Detection
function
npar
logLik
AIC
AICc
dAICc
AICcwt
M3
halfnormal
3
-461.8182
929.636
929.936
0.000
0.2973
M5
exponential
3
-462.1492
930.298
930.598
0.662
0.2135
M9
halfnormal
5
-460.6154
931.231
932.000
2.064
0.1059
M10
hazardrate
6
-459.5789
931.158
932.249
2.313
0.0935
M13
hazardrate
7
-458.5681
931.136
932.610
2.674
0.0781
M4
hazardrate
4
-462.2706
932.541
933.047
3.111
0.0628
M6
halfnormal
5
-461.6184
933.237
934.006
4.070
0.0389
M11
exponential
5
-461.6917
933.383
934.153
4.217
0.0361
M12
halfnormal
6
-460.6131
933.226
934.317
4.381
0.0333
M8
exponential
5
-462.1492
934.298
935.068
5.132
0.0228
M14
exponential
6
-461.6775
935.355
936.446
6.510
0.0115
M7
hazardrate
6
-462.2706
936.541
937.632
7.696
0.0063
M1
hazardrate
4
-533.5958
1075.192
1075.698
145.762
0
M2
exponential
3
-535.5643
1077.129
1077.429
147.493
0
M0
halfnormal
3
-535.8125
1077.625
1077.925
147.989
0
!
Table A12. Fitted SECR models for August 2018 based on corresponding sampling effort
data from observer boats and photo-identified sperm whale individuals during the month.
Model
Detection
function
npar
logLik
AIC
AICc
dAICc
AICcwt
M9
halfnormal
5
-611.1907
1232.381
1233.020
0.000
0.2542
M12
halfnormal
6
-610.4271
1232.854
1233.757
0.737
0.1759
M11
exponential
5
-611.6896
1233.379
1234.018
0.998
0.1544
M14
exponential
6
-610.6735
1233.347
1234.250
1.230
0.1374
M8
exponential
5
-612.3089
1234.618
1235.256
2.236
0.0831
M10
hazardrate
6
-611.7201
1235.440
1236.343
3.323
0.0483
M6
halfnormal
5
-613.1763
1236.353
1236.991
3.971
0.0349
M13
hazardrate
7
-610.9586
1235.917
1237.135
4.115
0.0325
M5
exponential
3
-615.4999
1237.000
1237.250
4.230
0.0307
M3
halfnormal
3
-615.8189
1237.638
1237.888
4.868
0.0223
M7
hazardrate
6
-612.7987
1237.597
1238.501
5.481
0.0164
M4
hazardrate
4
-615.5402
1239.080
1239.501
6.481
0.0100
M2
exponential
3
-681.4651
1368.930
1369.180
136.160
0
M1
hazardrate
4
-681.3863
1370.773
1371.194
138.174
0
M0
halfnormal
3
-682.5381
1371.076
1371.326
138.306
0
.
!
<$!
Table A13. Fitted SECR models for September 2018 based on corresponding sampling effort
data from observer boats and photo-identified sperm whale individuals during the month
Model
Detection
function
npar
logLik
AIC
AICc
dAICc
AICcwt
M9
halfnormal
5
-374.6309
759.262
760.462
0.000
0.2334
M12
halfnormal
6
-373.9062
759.812
761.527
1.065
0.1370
M10
hazardrate
6
-374.2387
760.477
762.192
1.730
0.0983
M11
exponential
5
-375.5412
761.082
762.282
1.820
0.0939
M3
halfnormal
3
-378.0426
762.085
762.547
2.085
0.0823
M5
exponential
3
-378.1264
762.253
762.714
2.252
0.0757
M6
halfnormal
5
-375.8231
761.646
762.846
2.384
0.0709
M13
hazardrate
7
-373.4839
760.968
763.301
2.839
0.0564
M7
hazardrate
6
-374.8698
761.740
763.454
2.992
0.0523
M14
exponential
6
-375.0210
762.042
763.756
3.294
0.0450
M8
exponential
5
-376.4767
762.953
764.153
3.691
0.0369
M4
hazardrate
4
-378.4031
764.806
765.591
5.129
0.0180
M2
exponential
3
-406.5182
819.036
819.498
59.036
0
M0
halfnormal
3
-406.7849
819.570
820.031
59.569
0
M1
hazardrate
4
-406.7541
821.508
822.293
61.831
0
!
!
!
<%!
Figure A1. Boat activity in the sampling area between May and October 2018 based on AIS
B data. Boat effort is indicated in hours (see color scale).
!
<<!
Hiermit versichere ich an Eides statt, dass ich die Arbeit eigenständig verfasst habe. Ich habe
keine anderen als die angegebenen Quellen und Hilfsmittel benutzt sowie wörtliche und
sinngemäße Zitate kenntlich gemacht. Die Arbeit hat in gleicher oder ähnlicher Form noch
keiner Prüfungsbehörde vorgelegen und die eingereichte schriftliche Fassung entspricht der auf
dem elektronischen Speichermedium. Ich bin mit einer Veröffentlichung meiner Arbeit
einverstanden.

Horta, 01.03.2020
Chiara Mandl
!
!
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Assessing the non-lethal effects of disturbance and their population-level consequences is a significant ecological and conservation challenge, because it requires extensive baseline knowledge of behavioral patterns, life-history and demography. However, for many marine mammal populations, this knowledge is currently lacking and it may take decades to fill the gaps. During this time, undetected population declines may occur. In this study we identify methods that can be used to monitor populations subject to disturbance and provide insights into the processes through which disturbance may affect them. To identify and address the knowledge gaps highlighted above, we reviewed the literature to identify suitable response variables and methods for monitoring these variables. We also used existing models of the population consequences of disturbance (PCoD) to identify demographic characteristics (e.g., the proportion of immature animals in the population, or the ratio of calves/pups to mature females) that may be strongly correlated with population status and therefore provide early warnings of future changes in abundance. These demographic characteristics can be monitored using established methods such as visual surveys combined with photogrammetry, and capture-recapture analysis. Individual health and physiological variables can also inform PCoD assessment and can be monitored using photogrammetry, remote tissue sampling, hands-on assessment and individual tracking. We then conducted a workshop to establish the relative utility and feasibility of all these approaches for different groups of marine mammal species. We describe how future marine mammal monitoring programs can be designed to inform population-level analysis.
Article
Full-text available
Marine megafauna has always elicited contrasting feelings. In the past, large marine animals were often depicted as fantastic mythological creatures and dangerous monsters, while also arousing human curiosity. Marine megafauna has been a valuable resource to exploit, leading to the collapse of populations and local extinctions. In addition, some species have been perceived as competitors of fishers for marine resources and were often actively culled. Since the 1970s, there has been a change in the perception and use of megafauna. The growth of marine tourism, increasingly oriented towards the observation of wildlife, has driven a shift from extractive to non-extractive use, supporting the conservation of at least some species of marine megafauna. In this paper, we review and compare the changes in the perception and use of three megafaunal groups, cetaceans, elasmobranchs and groupers, with a special focus on European cultures. We highlight the main drivers and the timing of these changes, compare different taxonomic groups and species, and highlight the implications for management and conservation. One of the main drivers of the shift in perception, shared by all the three groups of megafauna, has been a general increase in curiosity towards wildlife, stimulated inter alia by documentaries (from the early 1970s onwards), and also promoted by easy access to scuba diving. At the same time, environmental campaigns have been developed to raise public awareness regarding marine wildlife, especially cetaceans, a process greatly facilitated by the rise of Internet and the World Wide Web. Currently, all the three groups (cetaceans, elasmobranchs and groupers) may represent valuable resources for ecotourism. Strikingly, the economic value of live specimens may exceed their value for human consumption. A further change in perception involving all the three groups is related to a growing understanding and appreciation of their key ecological role. The shift from extractive to non-extractive use has the potential for promoting species conservation and local economic growth. However, the change in use may not benefit the original stakeholders (e.g. fishers or whalers) and there may therefore be a case for providing compensation for disadvantaged stakeholders. Moreover, it is increasingly clear that even non-extractive use may have a negative impact on marine megafauna, therefore regulations are needed.
Article
Full-text available
Abstract Obtaining estimates of animal population density is a key step in providing sound conservation and management strategies for wildlife. For many large carnivores however, estimating density is difficult because these species are elusive and wide‐ranging. Here, we focus on providing the first density estimates of the Eurasian lynx (Lynx lynx) in the French Jura and Vosges mountains. We sampled a total of 413 camera trapping sites (with two cameras per site) between January 2011 and April 2016 in seven study areas across seven counties of the French Jura and Vosges mountains. We obtained 592 lynx detections over 19,035 trap days in the Jura mountains and 0 detection over 6,804 trap days in the Vosges mountains. Based on coat patterns, we identified a total number of 92 unique individuals from photographs, including 16 females, 13 males, and 63 individuals of unknown sex. Using spatial capture–recapture (SCR) models, we estimated abundance in the study areas between 5 (SE = 0.1) and 29 (0.2) lynx and density between 0.24 (SE = 0.02) and 0.91 (SE = 0.03) lynx per 100 km2. We also provide a comparison with nonspatial density estimates and discuss the observed discrepancies. Our study is yet another example of the advantage of combining SCR methods and noninvasive sampling techniques to estimate density for elusive and wide‐ranging species, like large carnivores. While the estimated densities in the French Jura mountains are comparable to other lynx populations in Europe, the fact that we detected no lynx in the Vosges mountains is alarming. Connectivity should be encouraged between the French Jura mountains, the Vosges mountains, and the Palatinate Forest in Germany where a reintroduction program is currently ongoing. Our density estimates will help in setting a baseline conservation status for the lynx population in France.
Article
Full-text available
1. Spatially explicit capture–recapture methods use data from the detection of marked animals at known points in space to estimate animal population density without bias from edge effects. Detection is by means of stationary devices such as traps, automatic cameras or DNA hair snags. Data collection is often expensive, and it is not obvious how to optimize the frequency of sampling and the spatial layout of detectors. Results from a pilot study may be extrapolated by simulation to predict the effectiveness of different configurations of multiple detectors, but simulation is slow and requires technical expertise. 2. Another approach for evaluating novel designs is to compute intermediate variables such as the expected number of detected individuals E(n) and expected number of recapture events E(r), and to seek relationships between these variables and quantities of interest such as precision and power. 3. We present formulae for the expected counts and power. For many scenarios therelative standard error (RSE) of estimated density is close to 1∕√min{E(n),E(r)}, and for maximum precision E(n) ≈ E(r). We compare the approximation for RSE( ̂ D) with more rigorous results from simulation. 4. Computation of E(n) and E(r) is deterministic and much faster than simulation, so it is readily included in interactive software for designing studies with enough power to answer ecological questions. The related approximation for RSE( ̂ D) is adequate for many purposes
Article
Full-text available
Capture-mark-recapture methods have been extensively used to estimate abundance, demography, and life history parameters of populations of several taxa. However, the high mobility of many species means that dedicated surveys are logistically complicated and expensive. Use of opportunis-tic data may be an alternative, if modeling takes into account the inevitable heterogeneity in capture probability from imperfect detection and incomplete sampling, which can produce significant bias in parameter estimates. Here, we compare covariate-based open Jolly-Seber models (POPAN) and multi-state open robust design (MSORD) models to estimate demographic parameters of the sperm whale population sum-mering in the Azores, from photo-identification data collected opportunistically by whale-watching operators and researchers. The structure of the MSORD also allows for extra information to be obtained, estimating temporary emigration and improving precision of estimated parameters. Estimates of survival from both POPAN and MSORD were high, constant, and very similar. The POPAN model, which partially accounted for heterogeneity in capture probabilities, estimated an unbiased super-population of ~1470 whales, with annual abundance showing a positive trend from 351 individuals (95% CI: 234-526) in 2010 to 718 (95% CI: 477-1082) in 2015. In contrast, estimates of abundance from MSORD models that explicitly incorporated imperfect detection due to temporary emigration were less biased, more precise, and showed no trend over years, from 275 individuals (95% CI: 188-404) in 2014 to 367 (95% CI: 248-542) in 2012. The MSORD estimated short residence time and an even-flow temporary emigration, meaning that the probability of whales emigrating from and immigrating to the area was equal. Our results illustrate how failure to account for transience and temporary emigration can lead to biased estimates and trends in abundance, compromising our ability to detect true population changes. MSORD models should improve inferences of population dynamics, especially when capture probability is low and highly variable, due to wide-ranging behavior of individuals or to non-standardized sampling. Therefore, these models should provide less biased estimates and more accurate assessments of uncertainty that can inform management and conservation measures.
Article
Full-text available
Aim The knowledge of a species biogeographical patterns greatly enhances our understanding of geographical ecology, which can improve identifying key conservation needs. Yet, this knowledge is still scarce for many marine top predators. Here, we aim to analyse movement patterns and spatial structuring of a large predator, the short‐finned pilot whale Globicephala macrorhynchus, over a wide geographical area. Location North‐east Atlantic, in Macaronesian archipelagos (Azores, Madeira and Canaries) and Iberian Peninsula (Sagres). Methods We used likelihood techniques to estimate residency times and transition probabilities and carried out social analysis from individual photographic identification data, and analysed year‐round distribution from effort‐related sightings, collated between 1999 and 2015. Results The best‐fitting models included emigration and reimmigration and showed different residency times within each archipelago. A total of 26 individual movements from 21 individuals (from a sample of >2,300 individuals) were recorded between Madeira and the neighbouring archipelagos, and heterogeneous transition probabilities were estimated within and between areas. A social network diagram showed associations from animals with distinct residency patterns. Higher significant sighting rates were recorded during autumn in the Azores and Madeira. Main conclusions The variation in site fidelity and year‐round occupancy among areas of the Macaronesia is consistent with some degree of population structuring, which combined with a connectivity network and seasonal inflows from animals inhabiting offshore waters, support the development of a complex social and geographical ecology in short‐finned pilot whales. The combination of techniques applied in this study was an effective way to estimate parameters of movement, which could be a good practice to be used for other scenarios and species.
Article
Full-text available
Shipping routes in the ocean are analogous to terrestrial roads, in that they are regularly used thoroughfares that concentrate the movement of vessels between multiple locations. We applied a terrestrial road ecology framework to examine the ecological impacts of increased global shipping on “marine giants” (ie great whales, basking sharks [Cetorhinus maximus], and whale sharks [Rhincodon typus]). This framework aided in identifying where such “marine roads” and marine giants are likely to interact and the consequences of those interactions. We also reviewed known impacts of shipping routes on these species, and then applied the road ecology framework to detect unknown and potentially threatening processes. In the marine environment, such a framework can be used to incorporate knowledge of existing shipping impacts into management practices, thereby reducing the detrimental effects of future expansion of shipping routes on marine giants.