This article has been accepted for publication and undergone full peer review but has not been through the
copyediting, typesetting, pagination and proofreading process, which may lead to differences between this
version and the Version of Record. Please cite this article as doi: 10.1111/1365-2664.13525
This article is protected by copyright. All rights reserved
MR JAMES JEFFREY WAGGITT (Orcid ID : 0000-0003-3172-7096)
PROFESSOR JAN GEERT G HIDDINK (Orcid ID : 0000-0001-7114-830X)
DR CAMILO SAAVEDRA (Orcid ID : 0000-0003-1032-647X)
MR HENRIK SKOV (Orcid ID : 0000-0001-8835-3166)
MR RUBEN CHRISTIAAN FIJN (Orcid ID : 0000-0002-6233-2954)
Article type : Research Article
Handling Editor: Andre Punt
Distribution maps of cetacean and seabird populations in
the North-East Atlantic
Waggitt, J.J.
1
; Evans P.G.H.
1, 2
; Andrade, J.
3
; Banks, A.N
4
. ; Boisseau, O.
5
; Bolton, M.
6
; Bradbury, G.
7
;
Brereton, T.
8
; Camphuysen, C.J.
9
; Durinck, J.
10
; Felce, T.
11
; Fijn, R.C.
12
, Garcia-Baron, I.
13
; Garthe, S.
14
;
Geelhoed, S.C.V
15
; Gilles, A.
16
; Goodall, M.
17
; Haelters, J.
18
; Hamilton, S.
19
; Hartny-Mills, L.
20
; Hodgins, N.
21
; James, K.
2
; Jessopp, M.
22,23
; Kavanagh, A.S.
22
; Leopold, M.
15
; Lohrengel, K.
2
; Louzao, M.
13
; Markones,
N.
14
Martinez-Cediera, J.
24
; O’Cadhla, O.
25
; Perry, S.L.
26
; Pierce, G.J.
27
; Ridoux, V.
28
; Robinson, K.P.
29
;
Santos, M.B.
30
; Saavedra, C.
30
; Skov, H
31
; Stienen E.W.M.
32
; Sveegaard, S.
33
; Thompson, P.
34
; Vanermen,
N.
32
; Wall, D.
35
; Webb, A.
36
; Wilson, J.
37
; Wanless, S.
38
; Hiddink J.G.
1
1
School of Ocean Sciences, Bangor University, Menai Bridge, UK
2
Sea Watch Foundation, Amlwch, Isle of Anglesey, UK
Accepted Article
Journal of Applied Ecology
This article is protected by copyright. All rights reserved
3
Sociedade Portuguesa para o Estudo das Aves, Lisboa, Portugal
4
Natural England, Exeter, UK.
5
Marine Conservation Research, Kelvedon, UK
6
Royal Society for the Protection of Birds, Sandy, UK
7
Wildfowl and Wetlands Trust Consulting, Slimbridge, UK
8
Marinelife, Bridport, UK
9
Royal Netherlands Institute for Sea Research, Texel & Utrecht University, Texel, Netherlands
10
Marine Observers, Snedsted, Denmark
11
Manx Whale and Dolphin Trust, Peel, Isle of Man
12
Bureau Waardenburg, Culemborg, Netherlands
13
AZTI Fundazioa, Pasaia, Spain
14
Research and Technology Centre (FTZ), University of Kiel, Büsum, Germany
15
Wageningen Marine Research, Den Helder, The Netherlands
16
Institute for Terrestrial and Aquatic Wildlife Research, University of Veterinary Medicine Hannover Foundation, Büsum, Germany
17
Cornwall Wildlife Trust, Truro, UK
18
Royal Belgian Institute of Natural Sciences, Ostend, Belgium
19
ORCA, Portsmouth, UK
20
Hebridean Whale and Dolphin Trust, Isle of Mull, UK
21
Whale and Dolphin Conservation, Chippenham, UK
22
MaREI Centre, Environmental Research Institute, University College Cork, Cork, Ireland
23
School of Biological, Earth & Environmental Science, University College Cork
24
Coordinadora para o Estudo dos Mamíferos Mariños, Pontevedra, Spain
25
Science and Biodiversity Section, National Parks & Wildlife Service, Galway, Ireland
26
Cardigan Bay Marine Wildlife Centre, The Wildlife Trust of South and West Wales, New Quay, UK
27
Instituto de Investigacións Mariñas, Vigo, Spain
28
Observatoire PELAGIS, Université de La Rochelle, La Rochelle, France
29
Cetacean Research and Rescue Unit, Banff, UK
30
Instituto Español de Oceanografía, Centro Oceanográfico de Vigo, Vigo, Spain
31
DHI, Hørsholm, Denmark
32
Research Institute for Nature and Forest, Brussels, Belgium
33
Department of Bioscience, Aarhus University, Roskilde, Denmark
34
Lighthouse Field Station, University of Aberdeen, Cromarty, UK
35
Irish Whale and Dolphin Group, Kilrush, Ireland
36
HiDef Aerial Surveying Ltd., Cumbria, UK
37
Marine Scotland Science, Aberdeen, UK
38
Centre for Ecology & Hydrology, Penicuik, Midlothian, UK
Accepted Article
This article is protected by copyright. All rights reserved
ABSTRACT
1. Distribution maps of cetaceans and seabirds at basin and monthly scales are needed for
conservation and marine management. These are usually created from standardised and
systematic aerial and vessel surveys, with recorded animal densities interpolated across study
areas. However, distribution maps at basin and monthly scales have previously not been
possible because individual surveys have restricted spatial and temporal coverage.
2. This study develops an alternative approach consisting of: (1) collating diverse survey data to
maximise spatial and temporal coverage, (2) using detection functions to estimate variation in
the surface area covered (km
2
) among these surveys, standardising measurements of effort
and animal densities, and (3) developing species distribution models (SDM) that overcome
issues with heterogeneous and uneven coverage.
3. 2.68 million km of survey data in the North-East Atlantic between 1980 and 2018 were
collated and standardised. SDM using Generalized Linear Models and General Estimating
Equations in a hurdle approach were developed. Distribution maps were then created for 12
cetacean and 12 seabird species at 10 km and monthly resolution. Qualitative and
quantitative assessment indicated good model performance.
4. Synthesis and applications. This study provides the largest ever collation and standardisation
of diverse survey data for cetaceans and seabirds, and the most comprehensive distribution
maps of these taxa in the North-East Atlantic. These distribution maps have numerous
applications including the identification of important areas needing protection, and the
quantification of overlap between vulnerable species and anthropogenic activities. This study
demonstrates how the analysis of existing and diverse survey data can meet conservation and
marine management needs.
Keywords: Species distribution models, detection function models, North Sea, Celtic Sea, Bay of
Biscay, English Channel, Irish Sea, Hebrides
Introduction
Accepted Article
This article is protected by copyright. All rights reserved
Environmental change (Hoegh-Guldberg & Bruno, 2010) and anthropogenic activities
(Halpern et al., 2015, 2008) can have profound impacts on marine ecosystems. In many cases,
assessing these impacts requires an understanding of species distributions. For instance, knowing
species distributions helps identify the proportion of populations interacting with anthropogenic
activities, information that can explain declines (Boivin et al., 2016) and/or be used to develop
appropriate mitigation and management solutions (Wood, 2003). Information on species
distributions at monthly and basin scales is needed in marine ecosystems, where large numbers
of species routinely move hundreds or thousands of kilometres in migratory or dispersive
movements (Hays & Scott, 2013).
As apex-predators, cetaceans and seabirds have important ecological roles including the
top-down regulation of lower trophic levels (Hunt & McKinnell, 2006) and the transport of
nutrients (Doughty et al., 2016). They are also charismatic species of socio-economic importance,
due to their cultural appeal and focus for eco-tourism (Higham & Lück, 2007). However, these
taxa face numerous anthropogenic threats including bycatch, habitat-loss, energy extraction,
noise disturbance, prey reductions, pollution and vessel traffic (Avila, Kaschner, & Dormann,
2018; Croxall et al., 2012). Since their conservation is of importance for regulatory bodies, the
need for distribution maps at monthly and basin scales has been recognised by the European
Union (Habitats Directive: 92/43/EEC, Birds Directive: 2009/147/EC, Marine Strategy Framework
Directive: 2008/56/EC).
Distribution maps of cetaceans and seabirds are usually produced from transects using
humans/cameras on moving platforms to record animals (Buckland et al., 2012; Camphuysen,
Fox, Leopold, & Petersen, 2004; Evans & Hammond, 2004). Animal densities (individuals per km
2
)
are then estimated along transects (Buckland et al., 2001; Thomas et al., 2010), before being
interpolated across study areas (Hammond et al., 2013). In most cases, transects are performed
using similar platforms and observation methods, providing comparable measurements of
surface area covered and animal densities. Systematic transect-designs are also used, providing
homogeneous and even survey effort. However, due to financial and logistical constraints,
surveys covering whole basins occur at decadal intervals (Hammond et al., 2002, 2013) whilst
Accepted Article
This article is protected by copyright. All rights reserved
those covering seasonal cycles focus on relatively small areas (Gilles et al., 2016). Therefore,
distribution maps at monthly and basin scales are lacking, and their provision demands an
alternative approach.
This study develops an alternative approach to provide distribution maps for 12 cetacean
and 12 seabird species (Table 1) at 10 km and monthly resolution in the North-East Atlantic. This
approach consists of three stages. First, effort in time and space is maximised by collating survey
data from as many different sources and suppliers as possible (Mannocci et al., 2018; Paxton,
Scott-Hayward, Mackenzie, Rexstad, & Thomas, 2016; Roberts et al., 2016). Second, differences
among surveys linked with platform-type (aircraft versus vessel, low versus high), transect-design
(line-transect versus strip-transect), observation method (human versus camera) and weather
(sea state) are accounted for by calculating variations in the surface area effectively covered
(Buckland et al., 2001). Finally, species distribution models (SDM) (Elith & Leathwick, 2009) are
used to overcome problems with the heterogeneous and uneven effort in collations of survey
data (Paxton et al., 2016).
Materials and Methods
2.1 COLLATION
Aerial and vessel survey data were collated from the North-East Atlantic between 1980
and 2018. The North-East Atlantic was considered here to represent areas spanning between
Norway and Iberia on a north-south axis, and Rockall to the Skagerrak on an east-west axis. Only
survey data collected using dedicated human observers (i.e. not performing other duties) or
cameras to record animals were used. Survey data also needed to include information for the
calculation of variations in the surface area covered among surveys; namely platform-type,
platform-height, transect-design and recording method. Survey data were screened for
typographical and positional errors. Platforms and sightings recorded as being on land (i.e.
incorrect coordinates) were removed. Platforms recorded as travelling at unrealistic speeds were
also removed. To do so, mean (µ) speeds were calculated for each platform. For each vessel,
Accepted Article
