temporal and environmental variables affecting the ... · surfacings were recorded with aid of a...

Temporal and environmental variables

affecting the foraging behaviour of minke

whales (Balaenoptera acutorostrata)

A thesis presented for the degree of Bachelor of Science Honours. School of Biological Sciences and Biotechnology, Murdoch

University, Western Australia.

Holly Greenhalgh BSc

October 2016

Declaration I declare that this thesis is a correct representation of my research. It has not been

previously submitted for a degree at any other tertiary educational institution.

……………………………….

Holly Greenhalgh

i

Abstract The patterns in the foraging behaviour of predators, both spatially and temporally, can be

influenced by environmental variables which affect their prey’s distribution and abundance.

For instance, in the marine environment varying water depth and tidal height may contribute

to where prey items are distributed, and in turn, to where predators forage. Information on

the foraging ecology of predators can be used to identify important areas and time-periods

that are critical to the fitness of these predators. This study examined the temporal, spatial

and spatiotemporal patterns in the foraging of the North Atlantic minke whale (Balaenoptera

acutorostrata) in Faxaflói Bay, Iceland, in order to understand what factors influence the

foraging probability of this population. The timing and location of individual minke whale

surfacings were recorded with aid of a theodolite from a land-based research platform and

any obvious surface feeding events that were observed were also recorded during the focal

follows.

The effects of temporal variables combined with environmental factors such as depth, tidal

height, current bearing and velocity on the foraging probability of minke whales were tested

using generalised additive models (GAMs). The full dataset and a refined dataset containing

one randomly selected position per focal follow were analysed separately, which affected the

results of the GAMs. The results indicate that the probability of minke whales foraging

decreased throughout the season, with diurnal and crepuscular peaks and increased with

depth. Spatial preferences of foraging minke whales were evident, as the core foraging area

of minke whales were mapped using the full and refined datasets. Within the core foraging

area day in the season and depth influenced their foraging probability. Therefore, this study

is indicative that temporal, spatial and spatiotemporal patterns occur in the foraging behavior

of minke whales and highlights variables that affect these patterns.

ii

Table of Contents

1 Introduction ........................................................................................................................ 1 1.1 The value of foraging in animals ............................................................................................ 1 1.2 Factors affecting foraging in animals..................................................................................... 2

1.2.1 Temporal patterns ............................................................................................................. 2 1.2.1.1 Abiotic factors ........................................................................................................... 2 1.2.1.2 Biotic factors ............................................................................................................. 2 1.2.1.3 Anthropogenic factors .............................................................................................. 3

1.2.2 Spatial patterns ................................................................................................................. 5 1.2.2.1 Abiotic factors ........................................................................................................... 5 1.2.2.2 Biotic factors ............................................................................................................. 6 1.2.2.3 Anthropogenic factors .............................................................................................. 7

1.2.3 Spatiotemporal patterns ................................................................................................... 8 1.3 The importance of baleen whales as large predators ........................................................... 9 1.4 Minke whales in Icelandic waters .......................................................................................... 9 1.5 Aims of thesis....................................................................................................................... 12

2 Methods ............................................................................................................................ 12 2.1 Field methods ...................................................................................................................... 12 2.2 Data analysis ........................................................................................................................ 14

2.2.1 Temporal patterns ........................................................................................................... 14 2.2.2 Spatial patterns ............................................................................................................... 16 2.2.3 Spatiotemporal patterns ................................................................................................. 17

3 Results ............................................................................................................................... 17 3.1 Sample size and effort ......................................................................................................... 17 3.2 Temporal patterns ............................................................................................................... 18

3.2.1 Full dataset ...................................................................................................................... 18 3.2.2 Refined dataset ............................................................................................................... 22

3.3 Spatial patterns.................................................................................................................... 25 3.4 Spatiotemporal patterns ..................................................................................................... 28

3.4.1 Full dataset ...................................................................................................................... 28 3.4.2 Refined dataset ............................................................................................................... 28

4 Discussion .......................................................................................................................... 35 4.1 Factors affecting minke whale foraging .............................................................................. 35 4.2 Spatial patterns.................................................................................................................... 40 4.3 Spatiotemporal patterns ..................................................................................................... 42 4.4 Conclusions and significance of the results ......................................................................... 43 4.5 Potential weaknesses of the study and dataset .................................................................. 44

5 References ......................................................................................................................... 46

6 Appendix A ........................................................................................................................ 53 6.1 Appendix B ........................................................................................................................... 55

iii

List of Tables

Table 1. Results of GAM for model selection in the full dataset (n=1579). Final model

selected in bold.

Table 2. Results of GAM for model selection in the refined dataset (n=240). Final

model selected in bold.

Table 3. Results of GAM for model selection using the data extracted from the 50%

core foraging area of the full dataset (n=432). The final model in bold.


core foraging area of the refined dataset (n=122). The final model in bold.

List of Figures

Fig.1. A map of Faxaflói Bay, Iceland, showing the study area. The dashed line

connects areas of equivalent geographic location.

Fig. 2. 1. Partial effect plot showing the Julian day on the foraging probability of

minke whales, based on the best fitting model (model 1 in Table 1) of the full dataset

(n=139 foraging, 1440 non-foraging).The dashed lines represent 95% confidence

The points at the top and the bottom show the distribution of foraging (1.0) and non-

foraging (0.0) whales.

Fig. 2. 2. Partial effect plot showing the effect of time of day on minke whale foraging

probability based on the best fitting model (model 1, Table 1) of the full dataset

iv

(n=139 foraging, 1440 non-foraging). The dashed lines represent 95% confidence

intervals. The points at the top and the bottom show the distribution of foraging (1.0)

and non-foraging (0.0) whales.

Fig. 2. 3. Partial effect plot showing the effect of depth (m) on the foraging probability

of minke whales, based on the best fitting model (model 1, Table 1) of the full

dataset (n=139 foraging, 1440 non-foraging). The dashed lines represent 95%

confidence intervals. The points at the top and the bottom show the distribution of

foraging (1.0) and non-foraging (0.0) whales.

Fig. 3. The effect Julian day on the foraging probability of minke whales, based on

the best fitting model (model 1 in Table 2) of the refined dataset (n=73 foraging, 167

non-foraging). The dashed lines represent 95% confidence intervals. The points at

the top and the bottom show the distribution of foraging (1.0) and non-foraging (0.0)

whales.

Fig. 4. 1. Kernel density estimates for the 50% core foraging area and 95% foraging

range of the full dataset (n=139 foraging, 1440 non-foraging), with feeding (the black

dots) and non-feeding minke whales (grey triangles).

Fig. 4. 2. Kernel density estimates for the 50% core foraging area and 95% foraging

range of the refined dataset (n=73 foraging, 167 non-foraging), with feeding (the

black dots) and non-feeding minke whales (grey triangles).

Fig. 5. The effect of Julian day on the foraging probability of minke whales, based on

the best fitting model (model 1 from Table 3) in the 50% core foraging of the full

v




Fig. 6. 1. Partial effect plot showing the effect of day Julian day on the foraging

probability of minke whales, based on the best fitting model (model 1 from Table 4)

in the 50% core foraging of the refined dataset (n=41 foraging, 81 non-foraging). The

dashed lines represent 95% confidence intervals. The points at the top and the

bottom show the distribution of foraging (1.0) and non-foraging (0.0) whales.


of minke whales, based on the best fitting model (model 1 from Table 4) in the 50%

core foraging of the refined dataset (n=41 foraging, 81 non-foraging). The dashed

lines represent 95% confidence intervals. The points at the top and the bottom show

the distribution of foraging (1.0) and non-foraging (0.0) whales.

Acknowledgements

I would like to say a huge thank you to Dr Fredrik Christiansen for providing me with

data to conduct this research, teaching about statistical modelling as well as

answering all my questions throughout the year even though he was extremely busy

with other research projects. I would also like to thank Dr Kate Sprogis for her

support and for teaching me how to use ArcMap geographical information software.

1

1 Introduction

1.1 The value of foraging in animals

Animals forage to obtain nutrients, water and energy required for survival,

reproduction and growth (Boggs 1992, Stearns 1992). In a broader sense foraging is

a crucial component of ecosystem processes (O’brien et al. 1990, Hooker & Gerber

2004). Understanding an animal’s foraging behaviour is vital in the study of

population dynamics, individual survival, nutrient cycling and energy flow through

ecosystems (O’brien et al. 1990). Knowledge and understanding of the foraging

behaviour of animals also allows conservationists to identify areas of high

conservational value (Hooker & Gerber 2004, Sutherland 1998, Hoyt 2012), explore

a species’ habitat use (Aarts et al. 2008), gain information about ecosystem

processes (Horwood 1989, Hooker & Gerber 2004, Krebs & Davies 2009), predict

consequences of environmental change (Sutherland 1998, Hooker & Gerber 2004,

Astthorsson et al. 2007) and gauge the impacts of anthropogenic activities

(Sutherland 1998, Christiansen et al. 2013a, Furness 2003). Thus, the foraging

behaviour of animals is an important component to be considered in the

conservation and management of wildlife (Cañadas et al. 2005, Sutherland 1998,

Hoyt 2012).

2

1.2 Factors affecting foraging in animals

1.2.1 Temporal patterns

1.2.1.1 Abiotic factors

Temporal patterns in the foraging behaviours of animals have been attributed to a

profusion of abiotic variables over a range of different temporal scales. Some

animals exhibit patterns in their foraging behaviour in correspondence to diel

environmental factors such as including tidal states (Irons 1998) light conditions

(Lampert 1989, Kronfeld-Schor & Dayan 2003) and lunar cycles (Horning & Trillmich

1999, Morrison 1980), whereas, over larger temporal scales, patterns in the foraging

behaviours of animals have been associated with events such as El Niño (Quiñones

et al. 2010) and the North Atlantic Oscillation (NAO) (Stenseth et al. 2004). For

instance, the Black-legged Kittiwakes’ (Rissa tridactyla) foraging behaviour

corresponds to the daily tide cycles that influence the availability of their prey

(zooplankton) (Irons 1998). This highlights that temporally dynamic climate and

environmental factors can affect the foraging behaviours of animals by influencing

the availability, distribution and abundance of their prey items.

1.2.1.2 Biotic factors

The foraging behaviours of animals are determined by natural selection and there is,

therefore, a range of contributing biotic factors which involves competition (intra- and

inter-specific), resource exploitation (Benedetti-Cecchi et al. 1998, Thomas & Fenton

1978, Ford et al. 1998) and reproduction (Merrick & Loughlin 1997), which occurs at

3

varying temporal scales. For example, animals exhibit short-term patterns in their

foraging behaviour and spatial use that correspond to daily variations in

environmental conditions and prey availabilities and over relatively greater temporal

scales there is evidence of evolution in the migration of animals to foraging habitats

(Ford et al. 1998, Alerstam et al. 2003). These sorts of behaviours are under varying

intensities of selection such as daily predation risk (Milinski & Heller 1978), annual

survival, successful mating and successful reproduction and may largely be adaptive

but not always (Kruuk 1978, Alerstam et al. 2003). For instance, in the terrestrial

environment temporal variation in the foraging behaviour of European badgers

(Meles meles L.) coincide with daily changes in local weather conditions such as

wind speed, which influences the availability of their prey species (Kruuk 1978). In

the marine environment, over a relatively large temporal scale (more than two

decades), evidence of migratory adaptation is observed in two sympatric populations

of killer whales (Orcinus orca) that exhibit temporal foraging segregation and

different prey specialisations, which has been attributed to intra-specific competition

for prey and divergent evolution (Ford et al. 1998). Thus, biotic variables influence

the foraging behaviour of animals over a range of different temporal scales whether it

be over a day or over years.

1.2.1.3 Anthropogenic factors

Anthropogenic activities and developments are known to impact animals’ natural

foraging behaviour, as animals will often exhibit abnormal temporal patterns in their

foraging behaviour concurrent to anthropogenic stress or influence (Ditchkoff et al.

2006). Anthropogenic modification of the foraging habitats of animals can influence

4

their foraging behaviours over a range of temporal scales. For instance, over a

relatively fine temporal scales factors such as traffic (Schaub et al. 2008) and the

presence of humans (Matson et al. 2005, Ditchkoff et al. 2006) can affect the

foraging behaviour of animals. Over greater temporal scales variation in the foraging

behaviour of animals can be attributed to the modification, degradation of foraging

habitats and alteration of natural resources caused by industrial, agricultural and

urban development (Ditchkoff et al. 2006, Walsh & Harris 1996). For instance, short

term disruption of the foraging behaviour of cetacean species such as the bottlenose

dolphin (Tursiops truncatus) has been attributed to boat traffic and avoidance

behaviour (Gregory & Rowden 2001, Williams et al. 2006), which illustrates that the

shifts in the foraging behaviour of animals can occur simultaneously to the presence

of humans (Matson et al. 2005). Over a greater temporal scale (i.e. over several

years), wildlife residing within habitats developed by humans may not exhibit the

same foraging behaviour as they once did prior to the developments due to their

ability to adapt to anthropogenic modification (Ditchkoff et al. 2006). For instance,

evidence of evolution in raccoons (Procyon lotor) foraging habitat selection and

foraging behaviour has occurred in suburban areas as they are less affected by

seasonal changes and have different food resources to what they once relied on

prior to the anthropogenic developments (Prange et al. 2004). Furthermore,

modification of the marine environment by humans, for instance, the establishment

of artificial reefs can influence the foraging behaviour of animals as it affects the

distribution of food resources for many animals including herbivorous fish (Einbinder

et al. 2006).

5

1.2.2 Spatial patterns

1.2.2.1 Abiotic factors

Variation in climatic and environmental conditions drives patterns in the foraging

behaviour of animals, over a range of different spatial scales, directly by limiting their

foraging ranges due to their physiological restraints (Jetz et al. 2003) or indirectly by

affecting the distribution and abundance of primary production and prey items

(Astthorsson et al. 2007). Abiotic factors such as localised weather (Johnson et al.

2001), sediment type (Nehls & Tiedemann 1993), slope (Hester et al. 1999), aspect

and topography (Moorcroft et al. 2006) vary spatially and thus the foraging behaviour

and habitat selections of animals will often correspond to these variations

(Astthorsson & Vilhjálmsson 2002, Stenseth et al. 2004). For example, over a

relatively fine spatial scale variation in snow depth, solidity and density are known to

influence the distribution and behaviour of foraging woodland caribou (Rangifer

tarandus caribou) within its foraging habitat, as it influences its food resource

availability and distribution (Johnson et al. 2001). Additionally in the marine

environment, over a relatively larger spatial scale variables such as bathymetry

(Guinet et al. 2001) and oceanic thermal fronts (Sims & Quayle 1998) are known to

influence primary production and prey availability and consequently shape spatial

patterns in marine fauna foraging behaviour. For example, the foraging habitat

selections of blue whales (Balaenoptera musculus) in California have been linked to

topographic breaks that support high primary production and high densities of

pelagic euphausiids (Croll et al. 2005). Similarly, animals such as the basking shark

(Cetorhinus maximus) actively select foraging hotspots that correspond to dynamic

6

oceanic thermal fronts and tidal currents that promote high densities of their prey

(zooplankton) (Sims & Quayle 1998).

1.2.2.2 Biotic factors

Spatial patterns in the foraging behaviour of animals occur over a range of different

spatial scales and are a product of natural selection, which has resulted in animals

foraging in areas that maximise their fitness (Werner & Hall 1974, Alerstam et al.

2003). Food availability, predation risk (Lima et al. 1985), body condition (Heithaus et

al. 2007) and competition (Ramp & Coulson 2002, Ford et al. 1998) are some of the

factors that drive spatial patterns in many animal’s foraging habitat and behaviour

(Charnov 1976, Edwards 1983, Alerstam et al. 2003). Fine scale spatial patterns in

the foraging habitat selections of green turtles (Chelonia mydas), for example, are

influenced by the risk of predation by tiger sharks (Galeocerdo cuvier) (Heithaus et

al. 2007, Dill 2007). Turtles with poor body condition will forage in highly profitable

seagrass areas where predation risk is high, whereas, turtles with good body

condition will show a preference for safer, less profitable regions (Heithaus et al.

2007). This demonstrates that the foraging habitat selection and the foraging

behaviour of animals are often determined by a trade-off between predation risk and

energy requirement that many animals face in order to maximise their fitness and

ensure their survival. A terrestrial example includes the selection of foraging sites by

the eastern grey kangaroo (Macropus giganteus) in Melbourne, Australia, which is

not only determined by food availability but also by competition for resources by

conspecifics (Ramp & Coulson 2002). On a relatively large spatial scale, animals,

such as the broadnose sevengill sharks (Notorynchus cepedianus) will often show

feeding site fidelity to predictable habitats of their prey due to seasonality and

7

subsequent prey abundance, however, they will often exhibit finer scale patterns

within this region based on prey density as well as inter- and intra-specific

competition (Barnett et al. 2011).

1.2.2.3 Anthropogenic factors

Anthropogenic modification and degradation to foraging habitats, via agriculture,

urbanisation and industrialisation are known to result in unnatural foraging behaviour

of many animals over a range of different spatial scales (from foraging habitat

degradation by urban development (Walsh & Harris 1996) and resource exploitation

(Bertrand et al. 2012) to fine scale avoidance of humans (Williams et al. 2002)).

Human developments, for instance, deforestation, urbanisation, rural developments

and conifer plantations in Britain have reduced the amount of suitable foraging

habitat for vespertilionid bats (Vespertilionidae), which in turn impacts spatial

patterns in the bats foraging behaviour (Walsh & Harris 1996, Stebbings 1988). The

bats show a preference for foraging habitats that exhibit corridors rather than

fragmented habitats, in order to reduce travel time between foraging habitats to

preserve valuable energy (Walsh & Harris 1996). Animals will often change their

foraging habitat selection and foraging behaviours due to alerted resource availability

and distribution caused by the exploitation of resources by humans (Bertrand et al.

2012, Stevens et al. 2000). For instance, in the marine environment, as fisheries

target species in specific locations, this can impact animal’s foraging behaviour

spatially, as animals that forage in habitats that overlap with fishing areas compete

with the fisheries for prey items and are often forced to seek alternative foraging

habitats (Bertrand et al. 2012). Many animals are deterred from their natural foraging

habitats by humans as they pose a threat, which demonstrates how the presence of

8

humans can influence the foraging behaviour of animals spatially (Ditchkoff et al.

2006, Schaffar et al. 2009). For instance, fine scale spatial patterns in foraging

behaviour of cetaceans in close proximity to wildlife tourism activities such whale-

watching, for example, illustrates that these anthropogenic activities can degrade the

quality of a feeding hotspots and interfere with an animal’s natural foraging

behaviours due to the animals exhibiting avoidance behaviours as they perceive

boat traffic as threatening (Williams et al. 2002, Williams et al. 2006).

1.2.3 Spatiotemporal patterns

Environmental and ecological conditions often vary spatially and temporally, giving

rise to spatiotemporal patterns in the foraging behaviour of animals (Fritz et al. 2003,

Cotté et al. 2009). These patterns in animal foraging behaviour can be adaptive and

are often influenced by factors such as food resource availability, predator avoidance

survival and the environmental features that affect these factors (Cotté et al. 2009,

Kitaysky et al. 1999, Milinski & Heller 1978, Alerstam et al. 2003). For instance, the

annual foraging behaviour of fin whales (Balaenoptera physalus) across a basin in

the Mediterranean Sea shows that their seasonal site fidelity to their feeding hotspot

corresponds to predictable habitats of their prey items (Cotté et al. 2009, Littaye et

al. 2004). Whilst at smaller mesoscales (20 km–100 km in summer) foraging patterns

of the of fin whales seem to depend on small-scale temporal variations in prey

density and availability influenced by steep thermal gradients and salinity (Cotté et

al. 2009). Therefore, dynamic environmental features drive the abundance,

distribution and availability of food resources to animals over a range of

spatiotemporal scales influencing patterns in their foraging behaviour.

9

1.3 The importance of baleen whales as large predators

Due to their large body size and high abundance baleen whales consume vast

amounts of prey and are considered to have major influences on marine ecosystems

(Bowen 1997, Hooker & Gerber 2004, Baum & Worm 2009). Changes in abundance

of baleen whales can, therefore, cause cascading effects down the food web via top-

down control of ecosystem structure, which demonstrates the importance of these

marine predators (Hooker & Gerber 2004, Bowen 1997). Their ecological

significance has been demonstrated by the consequences of industrial whaling of

baleen whale species, as killer whales (Orcinus orca) resorted to preying on

alternative species (rather than baleen whales) resulting in the collapse of seal

(Callorhinus ursinus and Phoca vitulina), sea lion (Eumetopias jubatus) and sea otter

(Enhydra lutris) populations in the North Pacific Ocean (Springer et al. 2003, Trites et

al. 2007). Because of their important role in the marine ecosystem, understanding

the factors that influence the foraging ecology of baleen whales species is of great

value for marine conservation and management (Hooker & Gerber 2004).

1.4 Minke whales in Icelandic waters

Like most baleen whales North Atlantic Minke whales (Balaenoptera acutorostrata)

are capital breeders meaning they rely on previously stored energy to finance the

cost of reproduction (Kasuya 1995). They migrate seasonally from lower latitude

(tropic) winter breeding grounds to high latitude (polar) summer feeding grounds,

where they are frequently observed foraging in the Canadian Arctic, Greenland Sea,

Gulf of St Lawrence, Norway and Iceland (Christensen et al. 1990). Within the range

10

of this region, whales will forage to gain energy required for reproduction, growth and

maintenance (Christiansen et al. 2013b, Víkingsson et al.). In Iceland, the continental

shelf waters constitute an important foraging ground for minke whales, which are

known to acquire a lot of energy when present in these waters (Christiansen et al.

2013b, Víkingsson et al.). Minke whales are most frequently encountered in Icelandic

waters during the months of summer (in the northern hemisphere) between April and

October, with a peak abundance in July (Sigurjónsson & Víkingsson 1997, Bertulli

2010).

Around Iceland the highest densities of minke whales are found in the coastal waters

of the highly productive south-west region (Faxaflói bay), where sandeels are their

main prey item, however minke whales also occur in relatively high densities in the

south-east and in the north where their diet is more diverse and includes capelin, cod

(Gadus morhua) and krill (Vikingsson & Elvarsson 2011, Víkingsson 2016, Pike et al.

2009). In 2001, a greater number of minke whales were found in the eastern waters

of Iceland in comparison to earlier surveys (Pike et al. 2009). This corresponds to the

environmental conditions on the Icelandic continental shelf which vary temporally

and spatially influencing biotic factors such as primary production and animal

species composition and abundance (Astthorsson et al. 2007, Vikingsson &

Elvarsson 2011). In Icelandic waters, the southern and western areas are constantly

bathed by the warm saline Atlantic waters, whereas, the northern and eastern waters

are more dynamic. The polar front is located in the north, which is influenced by the

influx of Atlantic and Polar waters, making the region susceptible to a considerable

amount of environmental variability, which in turn influences the dynamics of prey

species in the area (Astthorsson et al. 2007). Primary production is greater close to

shore and in regions bathed by Atlantic water and thus is more variable in the

11

northern and eastern waters of Iceland. This is reflected in the diet of minke whales

and presumably influences their foraging behaviour both spatially and temporally, as

the foraging strategy that a minke whale utilises is assumed to depend on abiotic

variables as well as dietary composition and prey abundance and distribution

(Stockin et al. 2001)

In other geographic locations, the foraging habitats and the foraging behaviours of

minke whales have been associated with oceanographic and physiographic

conditions such as tidal state (Johnson et al. 2001) sediment type, bathymetry

(Macleod et al. 2004) which promote aggregations of their prey, however these

findings need to be further researched as the majority of these studies that have

addressed these topics have elements of bias and speculation. For instance, a study

suggests that in Antarctica minke whales (Balaenoptera bonaerensis) show a

preference to forage in regions associated with sea ice edge and bathymetric slope,

however, as the sample size of individual whales of that study was restricted this

may limit the validity of these findings (Friedlaender et al. 2006). Furthermore, a

study conducted in the coastal waters of the San Juan Islands in the United States of

America illustrated that foraging strategies of minke whales vary between feeding

areas, which researchers presume to be in response to the variations in physiology

and prey abundance between local habitats, although this is solely speculation and

needs to be further investigated (Hoelzel et al. 1989).

Minke whales have an important role in shaping the marine ecosystem in Icelandic

coastal waters as they are the most abundant whale species and consume vast

amounts of prey (Sigurjónsson & Víkingsson 1997, Pike et al. 2010, Víkingsson &

Elvarsson 2010, Stefánsson et al. 1997, Borchers et al. 2009). For example, the

annual consumption of finfish by minke whales in Icelandic waters was estimated to

12

be over one million tonnes between 1973 and 1985 (Sigurjónsson & Víkingsson

1997). However, knowledge of the factors influencing foraging behaviours of minke

whales in Iceland and around the globe is very limited and it is therefore of

conservational importance to further investigate their foraging ecology.

1.5 Aims of thesis

Gaining an understanding of the factors that may cause shifts in individual

performance of minke whales and relating short-term conditions such as ocean

conditions and anthropogenic stress to their foraging ecology would allow

conservationist to gauge impacts imposed by the recent and ongoing changes in the

marine environment. As mentioned above, minke whales are ecologically important

and their abundance is declining in Icelandic waters, yet knowledge of the factors

influencing their foraging ecology is currently limited. The aim of this study is to

quantify the spatial use of a minke whale population in relation to attributes of a small

area in Faxaflói Bay, Iceland, an important feeding ground over a relatively short

period of time (two summers). Additionally, the purpose of the present study is to

investigate spatial and temporal patterns in the foraging behaviour of minke whales

and identify the temporal and environmental factors that drive these patterns.

2 Methods

2.1 Field methods

Minke whale behavioural data was collected by Dr Fredrik Christiansen and his

research team in the south-western section of Faxaflói Bay, Iceland (Fig.1), between

June and September in 2010 and 2011. A 27-meter tall lighthouse located on the

13

northern tip of Reykjanes peninsula in Garður (64°04’56’’N, 22°41’24’’W) acted as a

research platform (Fig.1). The use of this land-based field station allowed for the

behaviour of whales to be recorded without causing any disturbance to the whales.

The daily study period was between 06:00 and 18:00, given that the weather

conditions were suitable (environmental factors and Beaufort scale were visually

estimated). Data of arbitrarily selected minke whales was recorded using individual

focal follows (where multiple data points obtained from an individual, over time, were

recorded). If the focal animal was approached by another animal the follow was

ended to avoid sampling the wrong animal. Each time the focal animal surfaced, the

time and its position were recorded. A theodolite (Wild T16, Wild Heerbrugg,

Heerbrugg, Switzerland) was used to measure the position of the focal animal.

During the focal follows any observation of surface foraging (direct observation of

minke whales engulfing prey during surfacing) was also recorded and the animal at

each location (i.e. data point) was classified as either foraging or non-foraging (see

Christiansen et al. 2013 for more details).

14

Fig.1. A map of Faxaflói Bay, Iceland, showing the study area. The dashed line

connects areas of equivalent geographic locations.

2.2 Data analysis

2.2.1 Temporal patterns

A series of binomial generalised additive models (GAMs) were developed to assess

which temporal and oceanographic variables best explained the foraging behaviour

(i.e. the foraging probability) of minke whales. GAMs are an extension of generalised

linear models (GLMs), fitting a non-parametric smoothing curve throughout the data

instead of using linear terms (Christ 2009, Rigby & Stasinopoulos 2005, Crawley

2007). I used this type of model to grasp the shape of the relationship between

foraging probability and the explanatory variables without choosing a particular

15

parametric form, as I was analysing continuous explanatory variables and had no

priori reason to select a specific parametric form. Time of day and day in the season

(temporal variables), as well as depth (m) and tidal height (m) (oceanographic

variables), were included in the model selection process. Day in the season was

represented by the Julian day (a continuous count of the number of days from the

start of the year), where day 164 (13 June) and 239 (27 August) represent the first

and last day of the study season, respectively. Data on current velocity (m/s) and

bearing (°) was provided by the Icelandic Maritime Administration (www.sigling.is), at

a temporal and spatial resolution of 1 h and 0.1° latitude and 0.2° longitude,

respectively. The effects of current velocity and bearing on the foraging probability of

minke whales were investigated for the 2011 data only (data for 2010 could not be

obtained).

The analysis was performed on two datasets separately, one being the full data set

with several observations per focal follow, and the other being a restricted subset

using only one randomly selected position for each follow to account for potential

temporal or spatial autocorrelation caused by the lack of independence between

positions within a focal follow. For follows that included at least one foraging event

(data point), one foraging location was extracted as there were considerably less

foraging events within follows compared to non-foraging events.

The model selection process included a step-by-step process using Akaike’s

information criterion (AIC) to indicate the models of best fit in both datasets (the full

and the refined data set). Covariates were added to the null model (foraging

probability ~ 1) sequentially. The models were first analysed with single explanatory

variables. The covariates that best explained the foraging probability of minke

http://www.sigling.is/

16

whales (models with the lowest AIC values) were then used in a multivariate model

selection. Temporal auto-correlation (dependence between data points within

follows) was investigated by comparing a generalised additive mixed model (GAMM)

with and without a temporal auto-correlation structure within follows using AIC.

Further, the autocorrelation function (acf) and partial autocorrelation function (pacf)

were utilised to test all models for temporal autocorrelation in the full dataset,

however, there was no temporal dependence between data points and was thus left

out of the results. The models were developed using R (3.2.3) Version 0.99.892. The

chi-squared goodness of fit test was used to test the significance of the explanatory

variables in the GAMs.

Collinearity between covariates was assessed using a summary of the Pearson

correlation coefficients. In addition, as the response variable was binary (i.e. foraging

or non-foraging) over-dispersion is not possible (Crawley 2007).

2.2.2 Spatial patterns

Kernel density estimates is a common method used to quantify home ranges of

animals (i.e. nursery areas and foraging areas) (Heupel et al. 2004, Sprogis et al.

2016). To investigate spatial patterns in the foraging behaviour of minke whales, the

location data were compiled for all follows and entered into Arc.maps 10.4. The

‘kernel density’ tool was used to create kernel density estimates of foraging areas for

the population with a cell size of 250 m2, and a search radius of 1,000 m. Polygons

for the 50% and 95% kernel density estimates were created to represent the core

foraging area and the overall foraging range of the minke whales within the study

area, respectively. The total area (km2) of the 50% and 95% kernel polygons were

17

then calculated. The spatial analysis was conducted separately for each of the

datasets (the full and the refined datasets) for comparison purposes. I also

compared the core foraging areas of minke whales between the two study years

(2010 and 2011) by fitting kernels separately for each year (for the full dataset and

the refined dataset).

2.2.3 Spatiotemporal patterns

To investigate spatiotemporal patterns in the foraging behaviour of minke whales a

data subset containing the data points within the 50% core foraging area of each

datasets was created and the temporal analysis rerun as described as above

(section 2.2.1).

3 Results

3.1 Sample size and effort

Data were collected over 118 days in two seasons, between 13 June and 17 August

in 2010 and 19 June and 27 of August in 2011, with a total of 164 hours of minke

whale observations. A total of 1579 surfacings from 240 follows were recorded in the

full dataset. In the refined dataset, the number of data points was reduced to 240.

There was an uneven ratio in the activity states (foraging vs. non-foraging) of the

individual whales for both datasets particularly in the full dataset (full dataset: n=139

foraging, 1440 non-foraging, refined dataset: n=73 foraging, 167 non-foraging). The

sample size of both foraging and non-foraging minke whales in the refined dataset

was only 52.5% and 11.6% of the sample size in the full dataset, respectively. The

18

data subsets extracted from the 50% foraging core areas again reduced the sample

size (full dataset: n=66 foraging, 366 non-foraging, refined dataset: n=41 foraging, 81

non-foraging). There was still an uneven ratio in samples sizes of foraging to non-

foraging whales, subsequent to the 50% core data extraction; however, this ratio was

reduced, particularly in the refined dataset. The covariates best explaining the

foraging probability of minke whales in 2011 did not include current bearing and

velocity as variables and hence these covariates were left out the results.

3.2 Temporal patterns

3.2.1 Full dataset

Using the full dataset, the foraging probability of minke whales was best explained by

the model including time, day and depth as covariates (model 1 in Table 1). Adding a

temporal auto-correlation structure (using a GAMM) did not improve model fit

(Appendix A). Further, acf and pacf plots showed no sign of temporal auto-

correlation. For the best model, the covariates day (X21,1.001 = 10.329, p=0.00131),

time (X21.001,1.002 = 6.821, 0.00902) and depth (X2

2.634, 3.401= 9.350, 0.03729) were

significant with the model explaining 3.54% of the deviance. A summary of the

correlation coefficients illustrated that there was no correlation between the

covariates.

19

Table 1. Results of GAM for model selection in the full dataset (n=1579). Final model

selected in bold.

There was a decrease in the foraging probability of minke whales throughout the

feeding season (Fig. 2.1). The foraging probability from the start to the end of the

season ranged from approximately 0.18 to 0.01. Time also affected the foraging

probability of minke whales (Fig. 2.2), with a peak in foraging probability between

approximately 16:00 and 17:30. There were also two smaller peaks in foraging

probability, one between 07:30 and 08:00 and one around 11:00 (Fig. 2.2). The

depths where minke whales were observed foraging ranged from 12.9 to 57.8 m.

The results of the GAM showed that the foraging probability of minke whales

increases with deeper water (Fig 2.3). However, there were only two whales (both

non-foraging) recorded in regions with depths greater than 45 m, resulting in a wider

spread of the confidence intervals at greater depths. The depth range that foraging

minke whales were positioned over was between 19.06 and 43.16 m with a mean

Model number

Model D.F. (within)

AIC ΔAIC

1 FP ~ s(Time) + s(Day) + s(Depth) 6 918.9 0

2 FP ~ s(Time) 9 923.3 4.4

3 FP ~ s(Day) + s(Depth) 4 924 5.1

4 FP ~ s(Time) + s(Day) 3 928.2 9.3

5 FP ~ s(Time) + s(Depth) 4 928.5 9.6

6 FP ~ s(Day) 2 930 11.1

7 FP ~ s(Depth) 2 934.3 15.4

8 FP ~ s(Tide height) 2 942.6 23.7

FP, foraging probability

20

(±s.e.m) of 35.2±0.35 m. In comparison, the depth range was approximately 20 m

wider for non-foraging whales, with a mean of 33.7±0.15 m (range= 12.90 to 57.82

m).

Fig. 2. 1. Partial effect plot showing the Julian day on the foraging probability of

minke whales, based on the best fitting model (model 1 in Table 1) of the full

dataset (n=139 foraging, 1440 non-foraging).The dashed lines represent 95%

confidence The points at the top and the bottom show the distribution of foraging

(1.0) and non-foraging (0.0) whales.

21

Fig. 2. 2. Partial effect plot showing the effect of time of day on minke whale foraging

probability based on the best fitting model (model 1, Table 1) of the full dataset

(n=139 foraging, 1440 non-foraging). The dashed lines represent 95% confidence

intervals. The points at the top and the bottom show the distribution of foraging (1.0)

and non-foraging (0.0) whales.

22


of minke whales, based on the best fitting model (model 1, Table 1) of the full




3.2.2 Refined dataset

In comparison to the full dataset, the best fitting model for the refined dataset was a

simpler GAM including day alone as the explanatory variable (model 1 in Table 2).

23

The effect of Julian day on the foraging probability of minke whales was significant

(X21.028, 1.055= 5.346, 0.0214), and explained 1.96% of the deviance, which was 1.58

percent units less than the best fitting model in the full dataset. A summary of the

correlation coefficients illustrated that there was no correlation between the

covariates.

Table 2. Results of GAM for model selection in the refined dataset (n=240). Final

model selected in bold.

The model shows that minke whales had a higher chance of foraging at the

beginning of the season (Fig. 3), with the probability being approximately twice as

high at the beginning of the season in the refined dataset in comparison to the full

dataset. (Fig. 2.1).

Model number

Model D.F. (within)

AIC ΔAIC

1 FP ~ s(Day) 3 293.2 0

2 FP ~ s(Tide height) 2 298.4 1.3

3 FP ~ s(Time) + s(Day) + s(Depth) 5 295.2 2

4 FP ~ s(Depth) 2 297.1 2.2

5 FP ~ s(Time) 2 298.9 5.7

6 FP ~ s(Time) + s(Depth) 3 299 5.8

7 FP ~ s(Day) + s(Depth) 4 293.8 293.8

8 FP ~ s(Time) + s(Day) 4 294.9 294.9


24


the best fitting model (model 1 in Table 2) of the refined dataset (n=73 foraging, 167

non-foraging). The dashed lines represent 95% confidence intervals. The points at

the top and the bottom show the distribution of data points for foraging (1) and non-

foraging (0) whales, respectively.

25

3.3 Spatial patterns

Estimations of the size of the 50% core foraging areas and 95% foraging range

revealed that the foraging areas in the full dataset (50% = 4.68 km2 and 95% = 21.23

km2), were smaller than in the refined dataset (50% = 9.62 km2 and 95% = 24.8822

km2), particularly the 50% core area, where the total area was almost the double in

the refined dataset (Fig. 4.1, Fig. 4.2).

In the full dataset, the 50% foraging core area was condensed into two sub-areas

(Figure 4.1), whereas in the refined dataset, where the foraging locations were

sparser, the core area was split into four sub-areas (Figure 4.2). In the full dataset,

the largest sub-area of the core area was located between 22° 40’0” W and 22° 35’0”

W, which was similar to the geographical location of the largest core area in the

refined dataset.

For the 95% foraging range, the positioning was very similar between both datasets.

However, the 95% foraging range was split into eight sub-areas in the full dataset

and six sub-areas in the refined dataset. The spatial analysis was also conducted

separately for each year, however, I found that there was no obvious difference in

the location of the 50% foraging cores and 95% foraging range (Appendix B).

26

Fig. 4. 1. Kernel density estimates for the 50% core foraging area and 95% foraging range of the full

dataset (n=139 foraging, 1440 non-foraging), with feeding (the black dots) and non-feeding minke whales

(grey triangles).

27

Fig. 4. 2. Kernel density estimates for the 50% core foraging area and 95% foraging range of the refined

dataset (n=73 foraging, 167 non-foraging), with feeding (the black dots) and non-feeding minke whales

(grey triangles).

28

3.4 Spatiotemporal patterns

3.4.1 Full dataset

The best fitting GAM for the data attained from the 50% core foraging area of the full

dataset was similar to the best fitting model for the refined dataset, with Julian day

being the only explanatory variable having a significant effect on the foraging

probability of minke whales (model 1 in Table 3). Adding a temporal auto-correlation

structure (using a GAMM) did not improve model fit (Appendix A). Further acf and

pacf plots showed no sign of temporal auto-correlation. Similar to the results of the

best fitting model in the full dataset the effect of day was highly significant (X21.715,

2.12= 11.21, 0.00431), with 3.33% of the deviance being explained (which was less

than the deviance explained by the model used for the full dataset and more than the

model used in the refined dataset). The covariates best explaining the foraging

probability of minke whales did not include current bearing and velocity as

covariates. A summary of the correlation coefficients illustrated that there was no

correlation between the covariates.

29


core foraging area of the full dataset (n=432). The final model in bold.

As for the temporal analyses, the probability that minke whales will forage decreased

as the season progressed. The probability ranged from approximately 0.37 at the

start of the season to approximately 0.1 at the end of the season, which is slightly

higher at the beginning compared to the full dataset (Fig. 5. and Fig. 2.1). The rate of

the decline in foraging probability varied across the season, with the highest rate of

the decline occurring around day 180 (Fig. 5). In contrast, the rate of decrease in

foraging probability in relation to day in the season was more gradual throughout the

season in the full dataset (Fig. 2.1) and refined dataset (Fig. 3).

Model number

Model D.F. (within)

AIC ΔAIC

1 FP ~ s(Day) 3 362.5 0

2 FP ~ s(Day) + s(Depth) 5 364.1 1.6

3 FP ~ s(Time) + s(Day) 4 365.7 3.2

4 FP ~ s(Time) + s(Day) + s(Depth) 6 365.9 3.4

5 FP ~ s(Depth) 3 369 6.5

6 FP ~ s(Time) + s(Depth) 5 370.8 8.3

7 FP ~ s(Tide height) 2 373 10.5

8 FP ~ s(Time) 2 373.1 10.6


30


the best fitting model (model 1 from Table 3) in the 50% core foraging of the full




3.4.2 Refined dataset

For the data extracted from the 50% core foraging area of the refined dataset, the

best fitting model included covariates day and depth (model 1, Table 4). The effects

of day and depth on the foraging probability of minke whales were both significant

31

(day: X21, 1= 5.854, 0.0155; depth: X2

1, 1= 5.421, 0.0199), with the model explaining

9.4% of the deviance, which is at least 5.86 percent units greater than the deviance

explained by the best fitting models from the other datasets. The covariates best

explaining the foraging probability of minke whales did not include current bearing

and velocity. A summary of the correlation coefficients illustrated that there was no

collinearity between the covariates.


core foraging area of the refined dataset (n=122). The final model in bold.

Minke whales foraging probability decrease with the day in the season when using

the data extracted from the 50% core foraging area in the refined dataset (Fig. 6.1).

The foraging probability ranged from approximately 0.59 to 0.11 at the start and end

of the season, respectively (Fig. 6.1), and was very similar to the pattern observed in

Model number

Model D.F. (within)

AIC ΔAIC

1 FP ~ s(Day) + s(Depth) 3 147.12 0

2 FP ~ s(Time) + s(Day) + s(Depth) 4 149.0164 1.8964

3 FP ~ s(Day) 2 151.1954 4.0754

4 FP ~ s(Depth) 2 151.4711 4.3511

5 FP ~ s(Time) + s(Day) 3 153.0221 5.9021

6 FP ~ s(Time) + s(Depth) 3 153.1209 6.0009

7 FP ~ s(Time) 2 159.1019 11.9819

8 FP ~ s(Tide height) 2 159.7006 12.5806


32

the temporal analyses of the refined dataset (Fig. 3). The depth at the locations

where minke whales were observed ranged from 24.19 to 41.78 m (Fig. 6.2), which

is less than in the depth range in the full dataset (Fig. 2.3). Similar to the results from

the full dataset, the probability of foraging increases with depth. The foraging

probability ranged from approximately 0.05 at the shallowest point to approximately

0.6 at the deepest point (Fig. 6.2). The range of depths of the locations at which

minke whales were observed foraging within the 50% core foraging area using the

refined dataset was between 30.15 and 41.78 m, with the mean (±s.e.m) depth being

35.51±0.411 m. In comparison, the depth range was approximately 5 m wider for

non-foraging whales (ranging from 24.19 m to 40.55 m), with a mean of 33.84±

0.360 m.

33

Fig. 6. 1. Partial effect plot showing the effect of day Julian day on the foraging

probability of minke whales, based on the best fitting model (model 1 from Table 4)

in the 50% core foraging of the refined dataset (n=41 foraging, 81 non-foraging). The

dashed lines represent 95% confidence intervals. The points at the top and the

bottom show the distribution of foraging (1.0) and non-foraging (0.0) whales.

34


of minke whales, based on the best fitting model (model 1 from Table 4) in the 50%

core foraging of the refined dataset (n=41 foraging, 81 non-foraging). The dashed

lines represent 95% confidence intervals. The points at the top and the bottom show

the distribution of foraging (1.0) and non-foraging (0.0) whales.

35

4 Discussion

4.1 Factors affecting minke whale foraging

The probability of foraging decreased for minke whales throughout the feeding

season (towards the end of summer). Minke whales are opportunistic feeders and

are able to adapt their dietary preference, foraging distribution and behaviour in

accordance with prey abundance and distribution to maximise their energy gain

(Horwood 1989, Macleod et al. 2004). Baleen whales have a specialised foraging

strategy that heavily interacts with prey density and behaviour (Goldbogen et al.

2013). Thus, the effect of day in the season on the foraging probability in the present

study may suggest that the transient whales relocate to forage in alternative areas

once the concentration of their main prey item in the region, sandeels (Ammodytes

sp.), is depleted to a level where it is no longer energetically viable to capture. This

explanation is plausible as fisheries, white-beaked dolphin (Lagenorhynchus

albirostris), harbour porpoise (Phocoena phocoena) and puffins (Fratercula arctica)

compete for sandeels in Faxaflói Bay (Bertulli 2010, Sarà et al. 2009). Therefore,

further research on the Intra-seasonal movement patterns and sandeel abundance is

needed.

There have been few studies investigating intra-seasonal patterns in minke whale

foraging ecology in Icelandic coastal waters. However, Bertulli (2010) found that in

May 2009 minke whales in Faxaflói Bay were observed feeding close to the tip of the

Reykjanes Peninsula in the Garður area (the study area) whereas in June, July and

August they were observed feeding further north near Syðra-Hraun. This temporal

pattern may explain why the foraging probability of minke whales decreased

throughout the season in the study area, as they may have moved to regions near

36

Syðra-Hraun to feed towards the end of the season. Further, in 2009 and 2008

minke whale sightings per unit effort in Faxaflói Bay peaked in July according to

Bertulli (2010). This is likely to reflect the movement patterns of minke whales in

response to prey abundance and distribution in the water column (Víkingsson &

Elvarsson 2010, Bertulli 2010, Sigurjónsson & Víkingsson 1997).

Intra-seasonal patterns in minke whale foraging behaviour attributed to the

environment and prey abundance and distribution have been shown by studies in

other geographical locations. For instance, intra-seasonal variations in minke whale

foraging behaviour (i.e. habitat selection and dietary preference) were related to

abiotic variables (i.e. physiography) and prey abundance off the Isle of Mull,

Scotland (Macleod et al. 2004). Further, variations in the foraging strategies of minke

whales (represented by surfacing intervals) throughout foraging season were

observed in Mull, Coll and the Small Isles of north-west Scotland, which, most likely

represents different foraging strategies employed by minke whales to target different

prey species as the abundance of their prey species is thought to vary considerably

over the season (Stockin et al. 2001).

Alternatively, the observed effect of day in the season on the foraging probability of

minke whales corresponds to the seasonal abundance of sandeels in the water

column, which are known to burrow into the sediment in late summer (Jensen et al.

2003, Greenstreet et al. 2006). According to Hobson (1986), sandeels seek refuge

when they are unable to forage effectively (once zooplankton and phytoplankton

abundance is depleted, (Russell 1928) to avoid predation. It has also been

37

suggested that this burrowing is cued by low light intensity since sandeels rely on

sufficient amounts of light to locate their prey items (Winslade 1974). Alternatively, it

has been suggested that sandeels bury themselves in late summer for energy

preservation during periods of elevated sea surface temperatures (Greenstreet et al.

2006). Furthermore, the abundance of a copepod species (Calanus finmarchicus)

that sandeels prey on, varies intra-seasonally (Gislason et al. 2007) and this is likely

to influence sandeel abundance in the water column (Hobson 1986, Winslade 1974).

A study of other species of sandeels demonstrated that larger (older) sandeels

burrow into the sediment to commence aestivation (dormancy) earlier than smaller

sandeels (Tomiyama & Yanagibashi 2004). There have been studies that illustrate

the importance of older (larger) age-classes of sandeels (Wanless et al. 2004) in the

diet of minke whales in Faxaflói Bay, as this age class serves as a more profitable

food source than smaller sandeels (Hjörvarsdóttir 2014). Thus, it is possible that a

decrease in prey availability, particularly of the larger sandeels, towards the end of

the foraging season could be causing the observed decline in the foraging probability

of minke whales in Faxaflói Bay.

The results of the analysis of the full dataset indicated that there were diurnal

patterns the foraging probability of minke whales with a crepuscular occurring. The

fact that time was only included in the best fitting model of the full dataset may be

attributed to the fact that data from the complete focal follows were included in the

analyses, and that the measured time effect is actually an artefact of repeated

foraging events of the individual in the same location (even though there were no

signs of temporal autocorrelation in the final model). When fitting a GAMM using only

time as a covariate, including a temporal auto-correlation structure actually improved

model fit. A number of studies have shown similar results due the lack of

38

independence of data obtained in focal follow resulting in temporal auto-correlation

(Gomes et al. 2009, Visser et al. 2011, Gailey et al. 2007).

Alternatively, diurnal migrations of sandeels into the water column (Jensen et al.

2003, Winslade 1974) may be attributed to the diurnal patterns and the crepuscular

peak in the foraging probability of minke whales in the study area, as prey availability

has been linked to the diurnal pattern of minke whale foraging behaviour in other

geographical locations (Stockin et al. 2001, Johnston et al. 2005). For example, the

probability of observing a foraging minke whale in the bay of Fundy has proven to be

greatest during flood tides and lowest during the mid-ebb phases due to changes in

the concentration of prey items (Johnston et al. 2005). During summer, sandeels are

present in the water column during daylight hours when their prey items

(zooplankton and large diatoms) are present in the water column (Winslade 1974,

Christensen et al. 2008, Sarà et al. 2009). Sandeels burrow during the transition

between day and night and at this time they are considered to be most vulnerable to

(Aarts et al. 2008, Hobson 1986). Sandeels presence in the water column around the

globe has also been linked to their size (and age), light intensity, sea surface

temperature, primary productivity, physical environmental factors, hydrology and

predator avoidance (Jensen et al. 2003, Hobson 1986, Greenstreet et al. 2006,

Winslade 1974, Christensen et al. 2008). In Icelandic waters, Calanus finmarchicus

dominates the zooplankton that the sandeels prey on. C. finmarchicus are known to

make diel vertical migrations through the water column and during these times they

are considered to be most vulnerable to predation (Ohman 1988, Durbin et al.

1995b, Durbin et al. 1995a). Hence, diel migrations of C. finmarchicus and

subsequent sandeel abundance in the water column of Faxaflói Bay may be

39

contributing to the observed diurnal patterns in minke whale foraging probability,

however further research is needed to verify this.

The diurnal patterns in the foraging probability of minke whales presented in the

current study correspond to the result of previous studies of baleen whales in other

geographical locations (Lockyer 1981). According to Lockyer (1981), the majority of

baleen whales feed during the day with peaks in foraging activity generally occurring

once or twice a day. Similarly, minke whales had two to three peak foraging times

per day (Ohsumi 1979, Bushuev 1986, Armstrong & Siegfried 1991). The peak

foraging times of baleen whale are influenced by prey availability and density (John

1992, Johnston et al. 2005) as well as the energy requirements of an individual

whale (which is affected by their reproductive state, maturity class, species and body

size, (Lockyer 1981, Christiansen et al. 2013b, Watkins & Schevill 1979, Piatt et al.

1989, John 1992). In the Stellwagen Bank National Marine Sanctuary, diel patterns

in the foraging behaviour of humpback whales (Megaptera novaeangliae) have been

attributed to the distribution and abundance of their main prey item sandeels

(Friedlaender et al. 2009). Thus, sandeel abundance and density, and the energy

requirements of minke whales could be driving the observed diurnal pattern in

foraging times in this study.

In the present study, the foraging probability of minke whales increases with depth.

North Atlantic minke whales are known to forage in shallow coastal waters (Weir et

al. 2007). Correlations between the distribution of foraging minke whales and

oceanographic variables such as depth are likely to represent the influence of such

variables on prey distribution and abundances (Macleod et al. 2004, Robinson et al.

40

2009, Skern-Mauritzen et al. 2011). For instance, the presence of minke whales on a

foraging ground in northeast Scotland was correlated to a specific depth range (20-

50 m) where the environmental conditions corresponded to the typical habitat of their

prey species (Robinson et al. 2009). Bertulli (2010) indicated that in 2008 and 2009

minke whales in Faxaflói Bay were observed within a depth range that was similar to

the depth range that minke whales were observed in the present study, which may

indicate that these depths are preferable habitats of the minke whales within the

region at an inter-annual scale. Sandeels are most frequently distributed at depths

between 20 and 40 m, but they have been found as deep as 70 m (Jensen et al.

2003, Macer 1966). This corresponds to the results of the present study as the

probability of foraging is lowest in areas with depths less than 20 m. This study

showed that the positions of the foraging minke whales corresponded to the typical

depth range that sandeels inhabit, whereas, non-feeding whales were found in

depths that sandeels do not typically inhabit. Therefore, the findings of this study

indicate that physiography and prey availability influences the foraging probability of

minke whales in the study area.

The results of the analysis varied between the full and refined datasets, where

foraging probability was greater in the refined dataset. This is presumably attributed

to the variation in sample size, which influences the ratio between foraging and non-

foraging.

4.2 Spatial patterns

One of the aims of the present study was to identify spatial patterns in the foraging

behaviour of minke whales in Faxaflói Bay. The foraging core area in the full dataset

41

was smaller in comparison to the refined dataset. This size difference is presumably

attributed to the fact that individual whales within a focal follow are likely to occur

within a similar spatial proximity. Furthermore, the sample size of foraging whales

was reduced from 66 in the full dataset to 41 in the refined dataset, which would

condense the density of foraging points to a smaller core area for the full dataset.

Additionally, variation in positioning and the number of sub-areas of the kernel

density polygons (i.e. the 95% foraging area in the full dataset was divided into more

sub-areas than in refined dataset) could also be due to pseudo-replication involved

in focal follows in the full dataset.

The 50% feeding core highlights a region which seems to be a preferable foraging

habitat of minke whales within their highest density feeding ground in Icelandic

waters (Pike et al. 2009). Thus, the fact the core foraging area had such a small

spatial scale, illustrates that the region must be of high ecological significance and

perhaps high conservational value, (Pike et al. 2011, Víkingsson 2016). Similarly,

Bertulli (2010) identified a feeding hotspot close to the tip of the Reykjanes Peninsula

in the Garður area in 2009, which highlights the region’s ecological significance as it

is a popular foraging habitat for minke whales at an inter-annual scale (Bertulli 2010).

This study demonstrates that minke whales show a defined preference for specific

foraging sites within the study area. This is likely to reflect a region where a large

number of sandeels aggregate, as minke whales are known to forage in regions

where the environmental conditions promote high abundances of their prey items

(Macleod et al. 2004, Haug et al. 1995, Friedlaender et al. 2006, Hoelzel et al. 1989).

In other geographical locations, physiography such as depth and bathymetric slope

has been related to minke whale habitat preferences as it affects the distribution of

42

their prey species (Hoelzel et al. 1989, Macleod et al. 2004). This corresponds to the

results of this study as the core foraging areas are characterised by waters depths

that sandeels typically inhabit (Jensen et al. 2003).

4.3 Spatiotemporal patterns

The observed temporal patterns in foraging behaviour of minke whales (day affecting

foraging probability) were similar between analyses, irrespective if they focused on

the core foraging areas or not. However, the effect become more pronounced

(higher foraging probability) when the analyses were focused on the foraging core

area. Hence, if I did not analyse the foraging core area data subset, I would have

underestimated the foraging probability of minke whales. In addition, the reduced

sample size in the foraging core subset influenced the ratio of foraging versus non-

foraging whales which may also be causing the increase in foraging probability within

the foraging core. Many authors illustrate the importance of considering the spatial

and temporal scale when conducting ecological research (Redfern et al. 2006, Wiens

1989, Jaquet 1996). For example, the factors controlling sperm whale (Physeter

macrocephalus) distributions in several different studies contradict one another due

to poorly designed spatial and temporal scales that do not account for temporal and

spatial variations in oceanic conditions (Jaquet 1996). Thus, it comes as no surprise

that minke whales exhibit spatiotemporal patterns in their foraging behaviour within

the study area.

In contrast to the results of the best fitting GAM in the full dataset, where day in the

season, time of day and depth were the covariates that best explained the foraging

probability of minke whales, the same temporal analysis carried out on the data

43

points within the 50% core foraging area found that only day still had a significant

effect. Due to the smaller spatial scale and the relatively uniform depth (around 33

m) of the 50% foraging core area, it makes sense that depth was no longer

significant as a covariate. Similarly, Bertulli (2010) suggested that the absence of a

correlation between depth and foraging bouts of minke whales in Faxaflói Bay was

attributed to the small depth range of the sampling area she analysed. The 50%

foraging core area in the refined dataset was almost double the size of the 50%

foraging core area in the refined dataset and consequently exhibiting a greater depth

range between 30 and 40 m. This difference in depth range could explain why depth

was part of in the best fitting model in the core foraging area in the refined dataset,

but not in the foraging core area of the full dataset in the spatiotemporal analyses.

Depth was not included in the model of best fit in the refined dataset, however, when

I focused on the foraging core area depth was included in the model of best fit. This

is presumably a result of the heterogeneous environmental conditions over the larger

spatial scale of the study area and relatively uniform environmental conditions in the

small-scale foraging core area.

4.4 Conclusions and significance of the results

By analysing minke whale foraging data I was able to: (1) identify temporal and

environmental variables influencing the foraging probabilities of minke whales (2)

map their preferred foraging areas (3) and investigate spatiotemporal patterns in the

foraging behaviour of minke whales by looking at temporal patterns within the core

foraging areas. Additionally, I was able to assess the effects of pseudo-replication

within focal follows by analysing the data as described above separately for the full

dataset and a refined dataset.

44

The identification of the small-scale feeding core highlights its ecological importance,

especially because minke whale abundances have declined in Icelandic waters in

recent years. In addition, previously there has been limited knowledge about the

factors affecting spatial and temporal patterns in the foraging behaviour of minke

whales, despite their important ecological role in Iceland and globally. Thus, the

results of this study provide insights into minke whale foraging ecology that can aid

researchers to identify areas of high conservational value, and provides preliminary

findings that may help researchers to gain an understanding of the factors that affect

the foraging ecology of minke whales and predict the consequences of future climate

change and develop wildlife management frameworks.

4.5 Potential weaknesses of the study and dataset

There were several limitations to the present study due to the nature of the data and

data collection. Minke whales generally forage during daylight hours (Lockyer 1981),

however, any foraging events that may have occurred outside of the daily study

period (06:00 to 18:00) were not recorded. Although minke whales are most

frequently observed foraging in Faxaflói Bay between June and September, they

have been spotted foraging in the region as early as April and as late as November

in other years (Bertulli 2010). Thus, an extended study period may have been

beneficial to further assess seasonal and diel patterns in their foraging ecology. Sub-

setting the data into two separate years reduced the sample size of surfacing whales

(for foraging whales in 2010 in particular), therefore jeopardising the validity of

running the analysis of the data in separate years for comparison. Consequently,

inter-annual patterns were not included in the results of the present study. As the

45

dataset was collected over two seasons in 2010 and 2011 relatively long-term

patterns in the foraging behaviour of minke whales in the study area remain

undocumented. The present study could also have be improved if estimates of the

seasonal and diurnal abundances of sandeels in the water column in the study area

were compared to the foraging probability of minke whales.

The diel vertical migration of C. finmarchicus has been linked the their vulnerability to

predation, thus this may influence sandeels presence in the water column (Ohman

1988). Therefore, it would be beneficial to compare the diel migrations of

zooplankton in Faxaflói Bay to the foraging probability of minke whales to test for any

bottom-up effects on foraging behaviour. In addition, light conditions have been

linked to the abundance and distribution of zooplankton (Durbin et al. 1995a) and

sandeels (Winslade 1974) in the water column in other geographical locations.

Hence, the effects of light intensity on the behavioural ecology of minke whales in

Faxaflói Bay may be a valuable comparison to include in future studies.

Minke whales’ energy requirements and foraging are influenced by age, size, sex

and reproductive status (Christiansen et al. 2013b) thus drone footage to indicate the

size and body condition of foraging whales may be beneficial to incorporate this

information in similar succeeding studies. Christiansen’s et al. (2016) study of

humpback whales (Megaptera novaeangliae) has illustrated that photogrammetry

methods can be used to gain information on the size and body condition of baleen

whales.

Therefore, knowledge of the foraging ecology of minke whales is currently limited

and as they have an important role in the marine ecosystem, future studies on the

46

factors that influence spatial and temporal patterns in their foraging behaviour would

be beneficial for conservational and wildlife management purposes.

5 References

Aarts, G., M. Mackenzie, B. Mcconnell, M. Fedak and J. Matthiopoulos. 2008. Estimating space‐use and habitat preference from wildlife telemetry data. Ecography 31:140-160.

Alerstam, T., A. Hedenström and S. Åkesson. 2003. Long‐distance migration: evolution and determinants. Oikos 103:247-260.

Armstrong, A. and W. Siegfried. 1991. Consumption of Antarctic krill by minke whales. Antarctic Science 3:13-18.

Astthorsson, O. S., A. Gislason and S. Jonsson. 2007. Climate variability and the Icelandic marine ecosystem. Deep Sea Research Part II: Topical Studies in Oceanography 54:2456-2477.

Astthorsson, O. S. and H. Vilhjálmsson. 2002. 7 Iceland shelf LME: Decadal assessment and resource sustainability. Large Marine Ecosystems 10:219-243.

Barnett, A., K. G. Abrantes, J. D. Stevens and J. M. Semmens. 2011. Site fidelity and sex-specific migration in a mobile apex predator: implications for conservation and ecosystem dynamics. Animal behaviour 81:1039-1048.

Baum, J. K. and B. Worm. 2009. Cascading top‐down effects of changing oceanic predator abundances. Journal of Animal Ecology 78:699-714.

Benedetti-Cecchi, L., F. Bulleri and F. Cinelli. 1998. Density dependent foraging of sea urchins in shallow subtidal reefs on the west coast of Italy (western Mediterranean). Marine Ecology Progress Series 163:203-211.

Bertrand, S., R. Joo, C. Arbulu Smet, Y. Tremblay, C. Barbraud and H. Weimerskirch. 2012. Local depletion by a fishery can affect seabird foraging. Journal of Applied Ecology 49:1168-1177.

Bertulli, C. G. 2010. Minke whale (Balaenoptera acutorostrata) and white-beaked dolphin (Lagenorhynchus albirostris) feeding behaviour in Faxaflói Bay, south-west Iceland. (unpublished).

Boggs, C. 1992. Resource allocation: exploring connections between foraging and life history. Functional Ecology 6:508-518.

Borchers, D. L., D. G. Pike, T. Gunnlaugsson and G. A. Víkingsson. 2009. Minke whale abundance estimation from the NASS 1987 and 2001 aerial cue–counting surveys taking appropriate account of distance estimation errors. NAMMCO Scientific Publications 7:95-110.

Bowen, W. 1997. Role of marine mammals in aquatic ecosystems. Marine Ecology Progress Series 158:267-274.

Bushuev, S. 1986. Feeding of minke whales, Balaenoptera acutorostrata, in the Antarctic. Report of the International Whaling Commission 36:241-245.

Cañadas, A., R. Sagarminaga, R. De Stephanis, E. Urquiola and P. Hammond. 2005. Habitat preference modelling as a conservation tool: proposals for marine protected areas for cetaceans in southern Spanish waters. Aquatic Conservation: Marine and Freshwater Ecosystems 15:495-521.

Charnov, E. L. 1976. Optimal foraging, the marginal value theorem. Theoretical population biology 9:129-136.

Christ, A. 2009. Mixed Effects Models and Extensions in Ecology with R. Journal of Statistical Software 32.

Christensen, A., H. Jensen, H. Mosegaard, M. St. John and C. Schrum. 2008. Sandeel (Ammodytes marinus) larval transport patterns in the North Sea from an individual-based hydrodynamic egg and larval model. Canadian Journal of Fisheries and Aquatic Sciences 65:1498-1511.

47

Christensen, I., T. Haug and Ø. Wiig. 1990. Morphometric comparison of minke whales Balaenoptera acutorostrata from different areas of the North Atlantic. Marine Mammal Science 6:327-338.

Christiansen, F., M. Rasmussen and D. Lusseau. 2013a. Whale watching disrupts feeding activities of minke whales on a feeding ground. Marine Ecology Progress Series 478:239-251.

Christiansen, F., G. A. Víkingsson, M. H. Rasmussen and D. Lusseau. 2013b. Minke whales maximise energy storage on their feeding grounds. The Journal of experimental biology 216:427-436.

Cotté, C., C. Guinet, I. Taupier-Letage, B. Mate and E. Petiau. 2009. Scale-dependent habitat use by a large free-ranging predator, the Mediterranean fin whale. Deep Sea Research Part I: Oceanographic Research Papers 56:801-811.

Crawley, M. J. 2007. The R book. Chichester, England; Hoboken. NJ: Wiley, Imperial College London at Silwood Park, UK.

Croll, D. A., B. Marinovic, S. Benson, F. P. Chavez, N. Black, R. Ternullo and B. R. Tershy. 2005. From wind to whales: trophic links in a coastal upwelling system. Marine Ecology Progress Series 289:117-130.

Dill, L. M., And J.W. Fourqurean. 2007. State-dependent risk-taking by green sea turtles mediates top-down effects of tiger shark intimidation in a marine. Ecology 76:837-844.

Ditchkoff, S. S., S. T. Saalfeld and C. J. Gibson. 2006. Animal behavior in urban ecosystems: modifications due to human-induced stress. Urban Ecosystems 9:5-12.

Durbin, E. G., R. G. Campbell, S. L. Gilman and A. G. Durbin. 1995a. Diel feeding behavior and ingestion rate in the copepod Calanus finmarchicus in the southern Gulf of Maine during late spring. Continental Shelf Research 15:539-570.

Durbin, E. G., S. L. Gilman, R. G. Campbell and A. G. Durbin. 1995b. Abundance, biomass, vertical migration and estimated development rate of the copepod Calanus finmarchicus in the southern Gulf of Maine during late spring. Continental Shelf Research 15:571-591.

Edwards, J. 1983. Diet shifts in moose due to predator avoidance. Oecologia 60:185-189. Einbinder, S., A. Perelberg, O. Ben-Shaprut, M. H. Foucart and N. Shashar. 2006. Effects of artificial

reefs on fish grazing in their vicinity: Evidence from algae presentation experiments. Marine environmental research 61:110-119.

Ford, J. K., G. M. Ellis, L. G. Barrett-Lennard, A. B. Morton, R. S. Palm and K. C. Balcomb Iii. 1998. Dietary specialization in two sympatric populations of killer whales (Orcinus orca) in coastal British Columbia and adjacent waters. Canadian journal of zoology 76:1456-1471.

Friedlaender, A. S., P. N. Halpin, S. S. Qian, G. L. Lawson, P. H. Wiebe, D. Thiele and A. J. Read. 2006. Whale distribution in relation to prey abundance and oceanographic processes in shelf waters of the Western Antarctic Peninsula. Marine Ecology Progress Series 317:297-310.

Friedlaender, A. S., E. L. Hazen, D. P. Nowacek, P. N. Halpin, C. Ware, M. T. Weinrich, T. P. Hurst and D. N. Wiley. 2009. Diel changes in humpback whale Megaptera novaeangliae feeding behavior in response to sand lance Ammodytes spp. behavior and distribution. Marine Ecology Progress Series 395:91-100.

Fritz, H., S. Said and H. Weimerskirch. 2003. Scale–dependent hierarchical adjustments of movement patterns in a long–range foraging seabird. Proceedings of the Royal Society of London B: Biological Sciences 270:1143-1148.

Furness, R. W. 2003. Impacts of fisheries on seabird communities. Scientia Marina 67:33-45. Gailey, G., B. Würsig and T. L. Mcdonald. 2007. Abundance, behavior, and movement patterns of

western gray whales in relation to a 3-D seismic survey, Northeast Sakhalin Island, Russia. Environmental Monitoring and Assessment 134:75-91.

Gislason, A., K. Eiane and P. Reynisson. 2007. Vertical distribution and mortality of Calanus finmarchicus during overwintering in oceanic waters southwest of Iceland. Marine Biology 150:1253-1263.

Goldbogen, J. A., A. S. Friedlaender, J. Calambokidis, M. F. Mckenna, M. Simon and D. P. Nowacek. 2013. Integrative approaches to the study of baleen whale diving behavior, feeding performance, and foraging ecology. BioScience 63:90-100.

48

Gomes, C. M., R. Mundry and C. Boesch. 2009. Long-term reciprocation of grooming in wild West African chimpanzees. Proceedings of the Royal Society of London B: Biological Sciences 276:699-706.

Greenstreet, S. P., E. Armstrong, H. Mosegaard, H. Jensen, I. M. Gibb, H. M. Fraser, B. E. Scott, G. J. Holland and J. Sharples. 2006. Variation in the abundance of sandeels Ammodytes marinus off southeast Scotland: an evaluation of area-closure fisheries management and stock abundance assessment methods. ICES Journal of Marine Science: Journal du Conseil 63:1530-1550.

Gregory, P. R. and A. A. Rowden. 2001. Behaviour patterns of bottlenose dolphins (Tursiops truncatus) relative to tidal state, time-of-day, and boat traffic in Cardigan Bay, West Wales. Aquatic Mammals 27:105-113.

Guinet, C., L. Dubroca, M. A. Lea, S. Goldsworthy, Y. Cherel, G. Duhamel, F. Bonadonna and J.-P. Donnay. 2001. Spatial distribution of foraging in female Antarctic fur seals Arctocephalus gazella in relation to oceanographic variables: a scale-dependent approach using geographic information systems. Marine Ecology Progress Series 219:251-264.

Haug, T., H. Gjøs, U. Lindstrøm, K. T. Nilssen and I. Røttingen. 1995. Spatial and temporal variations in northeast Atlantic minke whale Balaenoptera acutorostrata feeding habits. Developments in marine biology 4:225-239.

Heithaus, M. R., A. Frid, A. J. Wirsing, L. M. Dill, J. W. Fourqurean, D. Burkholder, J. Thomson and L. Bejder. 2007. State‐dependent risk‐taking by green sea turtles mediates top‐down effects of tiger shark intimidation in a marine ecosystem. Journal of Animal Ecology 76:837-844.

Hester, A., I. Gordon, G. Baillie and E. Tappin. 1999. Foraging behaviour of sheep and red deer within natural heather/grass mosaics. Journal of Applied Ecology 36:133-146.

Heupel, M. R., C. A. Simpfendorfer and R. E. Hueter. 2004. Estimation of shark home ranges using passive monitoring techniques. Environmental Biology of Fishes 71:135-142.

Hjörvarsdóttir, F. Ó. 2014. The importance of sandeel (Ammodytes marinus) for minke whales (Balaenoptera acutorostrata) and white-beaked dolphins (Lagenorhynchus albirostris) in Faxaflói Bay, Iceland. Journal of Applied Ecology 33:508-518.

Hobson, E. S. 1986. Predation on the Pacific sand lance, Ammodytes hexapterus (Pisces: Ammodytidae), during the transition between day and night in southeastern Alaska. Copeia 1986:223-226.

Hoelzel, A. R., E. M. Dorsey and S. J. Stern. 1989. The foraging specializations of individual minke whales. Animal behaviour 38:786-794.

Hooker, S. K. and L. R. Gerber. 2004. Marine reserves as a tool for ecosystem-based management: the potential importance of megafauna. BioScience 54:27-39.

Horning, M. and F. Trillmich. 1999. Lunar cycles in diel prey migrations exert a stronger effect on the diving of juveniles than adult Gal pagos fur seals. Proceedings of the Royal Society of London B: Biological Sciences 266:1127-1132.

Horwood, J. W. 1989. Biology and exploitation of the minke whale. CRC press, Faculty of Aquaculture and Fish Biology, Holar University College.

Hoyt, E. 2012. Marine Protected Areas for Whales, Dolphins and Porpoises: A world handbook for cetacean habitat conservation and planning. Routledge.

Irons, D. B. 1998. Foraging area fidelity of individual seabirds in relation to tidal cycles and flock feeding. Ecology 79:647-655.

Jaquet, N. 1996. How spatial and temporal scales influence understanding of sperm whale distribution: a review. Mammal Review 26:51-65.

Jensen, H., P. J. Wright and P. Munk. 2003. Vertical distribution of pre-settled sandeel (Ammodytes marinus) in the North Sea in relation to size and environmental variables. ICES Journal of Marine Science: Journal du Conseil 60:1342-1351.

Jetz, W., J. Steffen and K. E. Linsenmair. 2003. Effects of light and prey availability on nocturnal, lunar and seasonal activity of tropical nightjars. Oikos 103:627-639.

49

John, F. P. 1992. Threshold foraging behavior of baleen whales. Marine Ecology Progress Series 84:205-210.

Johnson, C. J., K. L. Parker and D. C. Heard. 2001. Foraging across a variable landscape: behavioral decisions made by woodland caribou at multiple spatial scales. Oecologia 127:590-602.

Johnston, D., L. Thorne and A. Read. 2005. Fin whales Balaenoptera physalus and minke whales Balaenoptera acutorostrata exploit a tidally driven island wake ecosystem in the Bay of Fundy. Marine Ecology Progress Series 305:287-295.

Kasuya, T. 1995. Overview of cetacean life histories: an essay in their evolution. Developments in marine biology 4:481-497.

Kitaysky, A., J. Wingfield and J. Piatt. 1999. Dynamics of food availability, body condition and physiological stress response in breeding black‐legged kittiwakes. Functional Ecology 13:577-584.

Krebs, J. R. and N. B. Davies. 2009. Behavioural ecology: an evolutionary approach. John Wiley & Sons.

Kronfeld-Schor, N. and T. Dayan. 2003. Partitioning of time as an ecological resource. Annual Review of Ecology, Evolution, and Systematics:153-181.

Kruuk, H. 1978. Foraging and spatial organisation of the European badger, Meles meles L. Behavioral Ecology and Sociobiology 4:75-89.

Lampert, W. 1989. The adaptive significance of diel vertical migration of zooplankton. Functional Ecology 3:21-27.

Lima, S. L., T. J. Valone and T. Caraco. 1985. Foraging-efficiency-predation-risk trade-off in the grey squirrel. Animal behaviour 33:155-165.

Littaye, A., A. Gannier, S. Laran and J. P. Wilson. 2004. The relationship between summer aggregation of fin whales and satellite-derived environmental conditions in the northwestern Mediterranean Sea. Remote Sensing of Environment 90:44-52.

Lockyer, C. 1981. Growth and energy budgets of large baleen whales from the Southern Hemisphere. Food and Agriculture Organization 3:379-487.

Macer, C. 1966. Sand eels (Ammodytidae) in the south-western North Sea; their biology and fishery. Ministery of Agriculture, Fisheries and Food, Fishery Investigation, London.

Macleod, K., R. Fairbairns, A. Gill, B. Fairbairns, J. Gordon, C. Blair-Myers and E. C. Parsons. 2004. Seasonal distribution of minke whales Balaenoptera acutorostrata in relation to physiography and prey off the Isle of Mull, Scotland. Marine Ecology Progress Series 277:74.

Matson, T. K., A. W. Goldizen and D. A. Putland. 2005. Factors affecting the vigilance and flight behaviour of impalas. South African Journal of Wildlife Research 35:1-11.

Merrick, R. L. and T. R. Loughlin. 1997. Foraging behavior of adult female and young-of-the-year Steller sea lions in Alaskan waters. Canadian journal of zoology 75:776-786.

Milinski, M. and R. Heller. 1978. Influence of a predator on the optimal foraging behaviour of sticklebacks (Gasterosteus aculeatus L.). Nature 275:642-644.

Moorcroft, P. R., M. A. Lewis and R. L. Crabtree. 2006. Mechanistic home range models capture spatial patterns and dynamics of coyote territories in Yellowstone. Proceedings of the Royal Society of London B: Biological Sciences 273:1651-1659.

Morrison, D. W. 1980. Foraging and day-roosting dynamics of canopy fruit bats in Panama. Journal of Mammalogy 61:20-29.

Nehls, G. and R. Tiedemann. 1993. What determines the densities of feeding birds on tidal flats? A case study on Dunlin, Calidris alpina, in the Wadden Sea. Netherlands Journal of Sea Research 31:375-384.

O’brien, W. J., H. I. Browman and B. I. Evans. 1990. Search strategies of foraging animals. American Scientist 78:152-160.

Ohman, M. 1988. Behavioral responses of zooplankton to predation. Bulletin of Marine Science 43:530-550.

Ohsumi, S. 1979. Feeding habits of the minke whale in the Antarctic. Rep Int Whal Comm 29:473-476.

50

Piatt, J. F., D. A. Methven, A. E. Burger, R. L. Mclagan, V. Mercer and E. Creelman. 1989. Baleen whales and their prey in a coastal environment. Canadian journal of zoology 67:1523-1530.

Pike, D. G., T. Gunnlaugsson, B. Elvarsson and G. A. Víkingsson. 2011. Correcting perception bias for Icelandic aerial surveys, 2007 and 2009. North Atlantic Marine Mammal Commission (NAMMCO) SC/18/AESP/08 (unpublished).

Pike, D. G., T. Gunnlaugsson, G. A. Víkingsson and B. Mikkelsen. 2010. Estimates of the abundance of minke whales (Balaenoptera acutorostrata) from the T-NASS Icelandic and Faroese ship surveys conducted in 2007. IWC SC/62/RMP5 (unpublished).

Pike, D. G., C. G. Paxton, T. Gunnlaugsson and G. A. Víkingsson. 2009. Trends in the distribution and abundance of cetaceans from aerial surveys in Icelandic coastal waters, 1986-2001. NAMMCO Scientific Publications 7:117-142.

Prange, S., S. D. Gehrt and E. P. Wiggers. 2004. Influences of anthropogenic resources on raccoon (Procyon lotor) movements and spatial distribution. Journal of Mammalogy 85:483-490.

Quiñones, J., V. G. Carman, J. Zeballos, S. Purca and H. Mianzan. 2010. Effects of El Niño-driven environmental variability on black turtle migration to Peruvian foraging grounds. Hydrobiologia 645:69-79.

Ramp, D. and G. Coulson. 2002. Density dependence in foraging habitat preference of eastern grey kangaroos. Oikos 98:393-402.

Redfern, J., M. Ferguson, E. Becker, K. Hyrenbach, C. P. Good, J. Barlow, K. Kaschner, M. F. Baumgartner, K. Forney and L. Ballance. 2006. Techniques for cetacean–habitat modeling. Marine Ecology Progress Series 310:271-295.

Rigby, R. A. and D. M. Stasinopoulos. 2005. Generalized additive models for location, scale and shape. Journal of the Royal Statistical Society: Series C (Applied Statistics) 54:507-554.

Robinson, K. P., M. J. Tetley and E. G. Mitchelson-Jacob. 2009. The distribution and habitat preference of coastally occurring minke whales (Balaenoptera acutorostrata) in the outer southern Moray Firth, northeast Scotland. Journal of Coastal Conservation 13:39-48.

Russell, F. 1928. The Vertical Distribution of Marine Macroplankton. VII Observations on the Behaviour of Calanus finmarchicus. Journal of the Marine Biological Association of the United Kingdom (New Series) 15:429-454.

Sarà, G., M. Pirro, M. Sprovieri, P. Rumolo, H. P. Halldórsson and J. Svavarsson. 2009. Carbon and nitrogen stable isotopic inventory of the most abundant demersal fish captured by benthic gears in southwestern Iceland (North Atlantic). Helgoland Marine Research 63:309.

Schaffar, A., B. Madon, C. Garrigue and R. Constantine. 2009. Avoidance of whale watching boats by humpback whales in their main breeding ground in New Caledonia. IWC SC 61, Scientific Committee of the Interna-tional Whaling Commission, Madeira.

Schaub, A., J. Ostwald and B. M. Siemers. 2008. Foraging bats avoid noise. Journal of Experimental Biology 211:3174-3180.

Sigurjónsson, J. and G. A. Víkingsson. 1997. Seasonal abundance of and estimated food consumption by cetaceans in Icelandic and adjacent waters. Journal of Northwest Atlantic Fishery Science 22:271-287.

Sims, D. W. and V. A. Quayle. 1998. Selective foraging behaviour of basking sharks on zooplankton in a small-scale front. Nature 393:460-464.

Skern-Mauritzen, M., E. Johannesen, A. Bjørge and N. Øien. 2011. Baleen whale distributions and prey associations in the Barents Sea. Marine Ecology Progress Series 426:289-301.

Springer, A. M., J. Estes, G. Van Vliet, T. Williams, D. Doak, E. Danner, K. Forney and B. Pfister. 2003. Sequential megafaunal collapse in the North Pacific Ocean: An ongoing legacy of industrial whaling? Proceedings of the National Academy of Sciences 100:12223-12228.

Sprogis, K. R., H. C. Raudino, R. Rankin, C. D. Macleod and L. Bejder. 2016. Home range size of adult Indo‐Pacific bottlenose dolphins (Tursiops aduncus) in a coastal and estuarine system is habitat and sex‐specific. Marine Mammal Science 32:287-308.

Stearns, S. C. 1992. The evolution of life histories. Oxford University Press Oxford, London. Stebbings, R. E. 1988. The conservation of European bats. Christopher Helm London, London.

51

Stefánsson, G., J. Sigurjónsson and G. A. Víkingsson. 1997. On dynamic interactions between some fish resources and cetaceans off Iceland based on a simulation model. J. Northwest Atl. Fish. Sci. 22:357-370.

Stenseth, N. C., A. Shabbar, K.-S. Chan, S. Boutin, E. K. Rueness, D. Ehrich, J. W. Hurrell, O. C. Lingjærde and K. S. Jakobsen. 2004. Snow conditions may create an invisible barrier for lynx. Proceedings of the National Academy of Sciences of the United States of America 101:10632-10634.

Stevens, J., R. Bonfil, N. Dulvy and P. Walker. 2000. The effects of fishing on sharks, rays, and chimaeras (chondrichthyans), and the implications for marine ecosystems. ICES Journal of Marine Science: Journal du Conseil 57:476-494.

Stockin, K., R. Fairbairns, E. Parsons and D. Sims. 2001. Effects of diel and seasonal cycles on the dive duration of the minke whale (Balaenoptera acutorostrata). Journal of the Marine Biological Association of the UK 81:189-190.

Sutherland, W. J. 1998. The importance of behavioural studies in conservation biology. Animal behaviour 56:801-809.

Thomas, D. W. and M. B. Fenton. 1978. Notes on the dry season roosting and foraging behaviour of Epomophorus gambianus and Rousettus aegyptiacus (Chiroptera: Pteropodidae). Journal of Zoology 186:403-406.

Tomiyama, M. and S. Yanagibashi. 2004. Effect of temperature, age class, and growth on induction of aestivation in Japanese sandeel (Ammodytes personatus) in Ise Bay, central Japan. Fisheries Oceanography 13:81-90.

Trites, A. W., V. B. Deecke, E. J. Gregr, J. K. Ford and P. F. Olesiuk. 2007. Killer whales, whaling, and sequential megafaunal collapse in the North Pacific: a comparative analysis of the dynamics of marine mammals in Alaska and British Columbia following commercial whaling. Marine Mammal Science 23:751-765.

Vikingsson, G. and B. Elvarsson. 2011. Diet composition of minke whales (Balaenoptera acutorostrata) in Icelandic waters. ICES CM:15.

Víkingsson, G. and B. Elvarsson. 2010. Recent changes in diet composition of minke whales (Balaenoptera acutorostrata) in Icelandic waters. Document SC/17/AS/06. Agadir, Morocco: The Scientific Committee of the International Whaling Commission. 1-15 pp.

Víkingsson, G. A. 2016. Decadal changes in distribution, abundance and feeding ecology of baleen whales in Icelandic and adjacent waters. A consequence of climate change? (unpublished).

Víkingsson, G. A., G. A. Auðunsson, B. Þ. Elvarsson and Þ. Gunnlaugsson. Energy storage in common minke whales (Balaenoptera acutorostrata) in Icelandic waters 2003-2007.–Chemical composition of tissues and organs. Marine Research Institute 4:1-13.

Visser, F., K. L. Hartman, E. J. Rood, A. J. Hendriks, D. B. Zult, W. J. Wolff, J. Huisman and G. J. Pierce. 2011. Risso's dolphins alter daily resting pattern in response to whale watching at the Azores. Marine Mammal Science 27:366-381.

Walsh, A. L. and S. Harris. 1996. Foraging habitat preferences of vespertilionid bats in Britain. Journal of Applied Ecology 32:508-518.

Wanless, S., P. Wright, M. Harris and D. Elston. 2004. Evidence for decrease in size of lesser sandeels Ammodytes marinus in a North Sea aggregation over a 30-yr period. Marine Ecology Progress Series 279:237-246.

Watkins, W. A. and W. E. Schevill. 1979. Aerial observation of feeding behavior in four baleen whales: Eubalaena glacialis, Balaenoptera borealis, Megaptera novaeangliae, and Balaenoptera physalus. Journal of Mammalogy 60:155-163.

Weir, C. R., K. A. Stockin and G. J. Pierce. 2007. Spatial and temporal trends in the distribution of harbour porpoises, white-beaked dolphins and minke whales off Aberdeenshire (UK), north-western North Sea. Journal of the Marine Biological Association of the United Kingdom 87:327-338.

Werner, E. E. and D. J. Hall. 1974. Optimal foraging and the size selection of prey by the bluegill sunfish (Lepomis macrochirus). Ecology 55:1042-1052.

52

Wiens, J. A. 1989. Spatial scaling in ecology. Functional Ecology 3:385-397. Williams, R., D. Lusseau and P. S. Hammond. 2006. Estimating relative energetic costs of human

disturbance to killer whales (Orcinus orca). Biological Conservation 133:301-311. Williams, R., A. W. Trites and D. E. Bain. 2002. Behavioural responses of killer whales (Orcinus orca)

to whale-watching boats: opportunistic observations and experimental approaches. Journal of Zoology 256:255-270.

Winslade, P. 1974. Behavioural studies on the lesser sandeel Ammodytes marinus (Raitt) II. The effect of light intensity on activity. Journal of Fish Biology 6:577-586.

53

6 Appendix A

Table 1.1. Results of GAMM for model selection of the full dataset. p= Auto-regression structure (1=previous, 2=

two previous), q= Moving average (0= no moving average, 1= moving point average exists). Asterisks indicate the

best model.

Autocorrelation structure

Model None p=1,q=0 p=2,q=0 p=1,q=1 p=0,q=1

FP ~ s(time)

D.F. 3 4 5 5 4

AIC 8492.9 8488.5 8499.5 8519.9 8486.8*

ΔAIC 6.1 1.7 12.7 33.1 0

FP ~ s(Day)

D.F. 3 4 5 5 4

AIC 8560.4* 8568.5 8587.2 8615.9 8564.9

ΔAIC 0 8.0 26.8 55.4 4.5

FP ~ s(Tide height)

D.F. 3 4 5 5 4

AIC 8481.9* 8485.7 8503.5 8537.1 8482.4

ΔAIC 0 3.8 21.7 55.3 0.5

FP ~ s(Depth)

D.F. 3 4 5 5 4

AIC 8516.6 8518.1 8531.3 8537.4 8515.9*

ΔAIC 0.8 2.2 15.5 21.5 0.0

FP ~ s(Time) + s(Day) + s(Depth)

D.F. 7 8 9 9 8

AIC 8594.5* 8599.7 8616.3 8610.6 8596.6

ΔAIC 0 5.3 21.9 16.1 2.1

54

Table 2. 1. Results of GAMM for model selection using the data extracted from the 50% core foraging area of the

full dataset. p= Auto-regression structure (1= previous, 2= two previous), q= Moving average (0= no moving

average, 1= moving point average exists) Asterisks indicate the best model.

Autocorrelation structure

Model None p=1,q=0 p=2,q=0 p=1,q=1 p=0,q=1

FP ~ s(time) D.F. 3 4 5 5 4

AIC 2115.3* 2128.1 2141.2 2189.6 2125.2

ΔAIC 0 12.8 25.9 74.3 9.9

FP ~ s(Day) D.F. 3 4 5 5 4

AIC 2155.7* 2168.6 2179.2 2207.2 2166.7

ΔAIC 0 12.9 23.5 51.6 11.0

FP ~ s(Tide height) D.F. 3 4 5 5 4

AIC 2116.4* 2128.7 2141.4 2183.5 2125.9

ΔAIC 0 12.3 25.0 67.1 9.5

FP ~ s(Depth) D.F. 3 4 5 5 4

AIC 2118.1* 2136.6 2149.5 2170.8 2134.2

ΔAIC 0 18.5 31.4 52.8 16.1

FP ~ s(Time) + s(Day) + s(Depth) D.F. 7 8 9 9 8

AIC 8594.5* 8599.7 8616.3 8610.6 8596.6

ΔAIC 0 5.3 21.9 16.1 2.1

55

6.1 Appendix B

Fig. 4.1.1. Kernel density estimates for the 50% core foraging area and 95% foraging range of the full

dataset in 2010 (n= 37 foraging, 399 non-foraging), with feeding (the black dots) and non-feeding

minke whales (grey triangles).

56

Fig. 4.1.2. Kernel density estimates for the 50% core foraging area and 95% foraging range of the full

dataset in 2011 (n= 102 foraging, 1041 non-foraging), with feeding (the black dots) and non-feeding

minke whales (grey triangles).

57

Fig. 4.2.1. Kernel density estimates for the 50% core foraging area and 95% foraging range of the

refined dataset in 2010 (n=16 foraging, 38 non-foraging), with feeding (the black dots) and non-

feeding minke whales (grey triangles).

58

Fig. 4.2.2. Kernel density estimates for the 50% core foraging area and 95% foraging range of the

refined dataset in 2011 (n=57 foraging, 129 non-foraging), with feeding (the black dots) and non-

feeding minke whales (grey triangles).

temporal and environmental variables affecting the ... · surfacings were recorded with aid of a...

Documents