A population estimate of Heaviside’s dolphins in the southern end of their range

Get Complete Project Material File(s) Now! »

Chapter 2 Near-shore distribution of Heaviside’s (Cephalorhynchus heavisidii) and dusky dolphins (Lagenorhynchus obscurus) at the southern limit of their range in South Africa, interspecific interactions and potential conflicts with fisheries.

Abstract

The range of Heaviside’s dolphins off South Africa overlaps with several fisheries and the dusky dolphin. Using data collected during small boat photo-ID surveys of Heaviside’s dolphins from 1999-2001 along 390km of coastline north of Cape Town (340 S), we investigate the distribution of Heaviside’s and dusky dolphins with respect to environmental conditions and inshore fishing effort. Although not feeding diurnally near-shore where data were collected, longshore concentrations of Heaviside’s dolphins were consistent between years and tended to be higher adjacent to areas which over the long term had higher availability of small hake Merluccius capensis (their principal prey). Preference was also shown for areas of higher swell and wave activity and to some extent areas with sandy rather than rocky shores. Heaviside’s dolphins were found in significantly shallower water than dusky dolphins and both species were sighted in significantly cooler waters than in the environment generally. Dusky dolphin sighting rates varied considerably between years, but were generally higher in areas with sandy shores (mostly straighter coastline). No evidence of a nursery area was found as both mixed groups and exclusively mother-calf groups were seen throughout the study area. Very large groups of 50-500 dusky dolphins were only seen in St Helena Bay, which is the site of a wind-driven upwelling zone. Very large groups were possibly aggregations of several smaller feeding groups, suggesting variability in feeding strategy throughout their range. Near-shore fishing activity (line fishing, crayfishing and setnetting) was higher in the northern half of the study area and clustered around harbours. Set netting, the most threatening fishery type for dolphins, occurred in two main areas only (Yzerfontein and St Helena Bay) but due to an industry collapse is currently thought to be a low threat to the population. The opportunity for interactions between Heaviside’s and dusky dolphins was high but those observed were usually neutral and sympatry appears to be mediated by differences in overall range and the type and size of prey species taken by the two species.

Introduction

Within the range of most species there is considerable variation in the number and density of animals at a variety of scales (Begon et al. 2005). Marine mammals are no exception and variation in their abundance has been associated with both environmental and biological influences including depth, sea floor gradient, oceanographic features, predator avoidance and competitive exclusion (Goodall et al. 1995; Tynan 1997; Davis et al. 1998; Elwen & Best 2003; Hastie et al. 2005; Heithaus & Dill 2006). The relationships between abundance and covariates may change with the scale of the study, and it is these changes which can illustrate the ecological relationships involved (Benoit-Bird & Au 2003; Johnston et al. 2005). Along the west coast of southern Africa, Heaviside’s and dusky dolphins are known to be sympatric within the coastal environment but knowledge of their distribution is currently limited to broad scale descriptors. Findlay et al. (1992) describe the general range of Heaviside’s dolphins as “west of Cape Point (18.50E)… possibly into southern Angola…all sightings in waters shallower than 200m, the highest densities being inshore of the 100m isobath”, and dusky dolphins as entirely sympatric but with wider, although not entirely known limits extending to 190E (east of Cape Point into False Bay), northwards into Angola to at least 120S and offshore to at least 500m depth and possibly as much as 2000m of water. Such a broad scale overlap of ranges between potentially competing predators may be more clearly differentiated by habitat selection at finer spatial (Parra 2006; Heinrich 2006; Goodall et al.1995) or temporal (Thompson et al. 2004) scales. To date, no studies have investigated either the environmental factors influencing niche or habitat selection or the sympatry of Heaviside’s and dusky dolphins at finer spatial scales.Habitat or niche selection by animals is assumed to reflect the optimal choice (or
compromise) of a suite of conditions, resources and influences both biotic (e.g. the distribution of predators, prey & competitors) and abiotic (e.g. depth, temperature, oceanographic features) within their potential range given the constraints of time, space and physiology. The ecological link between physical or oceanographic features and cetacean distribution patterns is frequently their association with prey which have either themselves been physically aggregated by oceanographic features or subsequently attracted to aggregations of their own smaller prey (Wolanski & Hamner 1988; Johnston et al. 2005; Tynan 1997). Heaviside’s dolphins are known to move offshore at night where they are thought to feed on vertically migrating prey and to spend daylight hours (when the data used in this study were collected) closer to shore resting and not feeding (Elwen et al. 2006; Chapter 1). In animals that are not feeding, influences on distribution can be far less clear,but most evidence suggests that when resting or otherwise unengaged, cetaceans choose environments that are safe, from both predators and harsh environmental conditions(Whitehead & Moore 1982; Smultea 1994; Elwen & Best 2003; Lammers 2004, Heithaus &Dill 2006). Thus, even though Heaviside’s dolphins appear to not be feeding while near shore they may still exhibit a preference for certain habitat types where predation risk is lower(inshore generally) or conditions are more conducive to resting and socialising.
A further consideration in the ecology of Heaviside’s dolphins is potential competition with the slightly larger, sympatric dusky dolphin. The observable consequences of interspecific competition between sympatric predators vary from occasional harassment to potentially devastating local extinction level influences on the ‘weaker’ species (Linnel & Strand 2000; Creel & Creel 1996). Interactions between competing predators might not be obvious and may only be seen as avoidance by the weaker (almost always the physically smaller species,in this case the Heaviside’s dolphin) of the stronger species. This may result in habitat differentiation and occasionally the counter-intuitive result of the weaker species having a higher survival rate in areas of lower prey abundance, if these areas act as refuges from competition (Linnel & Strand 2000; Durant 1998). Spatial competitive exclusion of Burmeister’s porpoises has been suggested in Golfo San José, Argentina with porpoises occupying intermediate depths between the preferred ranges of dusky and bottlenose(Tursiops truncatus) dolphins (Goodall et al. 1995). Temporal (and possibly spatial) segregation of harbour porpoises (Phocoena phocoena) and bottlenose dolphins has been observed in areas of the Moray Firth, Scotland (Thompson et al. 2004). Competition between sympatric cetaceans may be mediated more subtly by resource partitioning including differences in fine scale habitat selection and prey species taken (Bearzi 2005; Heinrich 2006;Parra et al. 2005). There is some evidence to suggest that the two species in the current study occupy different niches as they respond to different environmental cues (Chapter 1) and eat different types and sizes of prey (Sekiguchi 1994) although there is a significant overlap of prey species taken.In this paper we use the sightings data gathered during inshore photo-ID surveys to investigate how environmental variables affect the distribution of Heaviside’s and dusky dolphins in the near-shore environment. We also investigate if their overlapping distributions (Findlay et al. 1992) are differentiated at finer scales. We further compare the observed patterns of distribution with that of near shore fishery activity observed during the same period.

READ  Géologie du socle cristallin de Madagascar par le PGRM 2012

Methods

Field data

The spatial data used in this study were not collected specifically to answer questions of distribution and habitat modelling but rather as a by-product of a 3 yr photo-ID survey focussed on questions of individual movements and range along the southern west coast of South Africa (Fig. 2.1). In the first year of the study 1999, effort was restricted to a 20km long stretch of coast around Britannia Bay aimed at photographically identifying all animals in the area, in the 2nd and 3rd years of the study (2000 and 2001) we searched the full ~390km study area in an effort to recapture those animals identified in the first year. Although we attempted to search as much of the coastline within the overall study area as possible, effort was uneven along the coast (Fig. 2.2) and generally higher closer to the harbours from which the research boat was launched. Also, collecting photo-ID data entails closing with a group of animals and spending considerable time with them, increasing the likelihood of seeing other animals nearby or having them attracted to the boat. Thus a varied approach was taken to analysis, with a model being generated for all sightings with all available data and a broader scale approach looking at the relationships between dolphin density and environmental factors between binned 10km sections of the coastline.All data were collected from a 6m RIB fitted with twin 40hp outboard motors and an elevated observation platform (putting eyes at approx 3m ASL). The boat was launched daily, weather permitting and used to run coastwise searches (dependent on previously searched areas and prevailing winds), parallel to the breaker line where densities of Heaviside’s dolphins are known to be highest during the day, at a search speed of 6-8kn. Upon encounter, dolphins were followed until photography was regarded as complete or until the dolphins were lost. At each sighting sea surface temperature (SST), depth (from on board fish finder) and GPS position were noted, also maximum, minimum and best group size estimates and group composition (adult or calf ) as well as noting unusual behaviours (e.g. evasiveness, boat attraction, mating, inter-specific interactions). Some individuals were strongly attracted to the boat for bow-riding and when necessary, a sustained burst of speed of up to 15kn was used when moving away from a completed group to prevent them following the boat and being
counted repeatedly. Such data were collected for all cetaceans encountered, but for those not part of the photo-ID study (including dusky dolphins) encounters were considerably shorter and accuracy of counts could thus be slightly affected by this. Furthermore, as group sizes were generally larger in dusky than Heaviside’s dolphins, counts of very large groups were inevitably less precise than of small groups, with numbers becoming obviously rounded off above about 30 animals.All search effort in this study occurred within a relatively narrow band close to the shore and was well defined along-shore by the collection of GPS waypoints for the start and finish of each search leg. The boat track was essentially parallel to the coast and within 1 km of it. To define the offshore limit to the study area (or “strip width”) we used the furthest distance offshore of any dolphin sighting made under normal search conditions and regarded that as the furthest reasonable distance offshore that a dolphin sighting could be made, thereby defining a 2km wide strip of ocean adjacent to the shore. This 2km wide strip of coast was then split into 36 bins (or blocks) roughly 10km long each (a scale which created enough variation for differences to be seen but was not so small as to have very large variations in search effort), within which the encounter rates per kilometre searched (for groups, dolphins and calves) were calculated. A fairly fine scale digital coastline (1:150 000) was used and due to the convolutions of the coast, measuring blocks precisely 10km long was not feasible and in some instances borders were pushed slightly to align with headlands if appropriate.

Environmental Data

Both depth and SST were collected (from on board fish-finder) at the site of each dolphin encounter as well as at the start and stop points of each search leg. The measured depth values were combined with data from the digitised 1:150 000 South African Naval Hydrographers navigation charts for the study region (SAN 117-119) and were used to build a TIN – a 3D sea floor model, in Arcview GIS 3.3. From this model we estimated the slope of the sea floor at each dolphin sighting and the slope and depth for large area averages. Exposure to swell was calculated using a 12 yr average (1990-2001) of offshore data (17-190E by 33-350S) collected from voluntary observer ships and kept in the CSIR EMATEK Marclim database(following Elwen & Best 2003). This average effectively represents the offshore swell climate, in the absence of any land to break it up. The majority of swell off southern Africa originates in the southwest quarter resulting in the corners of most bays on the south and west coasts being fairly protected while straight sections of coastline were mostly exposed to open ocean swell. Areas that were protected from more than 30% of this swell (east of the SSWNNE line) were regarded as partly-protected, while areas protected from more than 60% of swell (south and east of the WSW-ENE line) were regarded as “protected”. The shore type of the coast (which can be reasonably assumed to represent the near-shore substrate, Elwen & Best 2003) was defined following the Coastal Sensitivity Atlas of Southern Africa (Jackson &Lipschitz 1984); the coastline within the study area consisted of three shore types, namely“fine sandy beaches”, “wave cut rocky platforms” and “exposed rocky headlands”, with a few small sections defined as “Estuarine” (usually muddy).

Chapter 1 Near-shore diurnal movements and behaviour of Heaviside’s dolphins(Cephalorhynchus heavisidii), with some comparative data for dusky dolphins (Lagenorhynchus obscurus).
Abstract 
Introduction 
Methods 
Results 
Discussion 
Literature Cited 
Figures 
Chapter 2 Near-shore Distribution of Heaviside’s (Cephalorhynchus heavisidii) and dusky dolphins (Lagenorhynchus obscurus) at the southern limit of their range in South Africa, interspecific interactions and potential conflicts with fisheries.
Abstract 
Introduction 
Methods 
Results 
Discussion 
Literature Cited 
Figures 
Chapter 3 Site fidelity in a dynamic environment: Range, dispersal and social structure of Heaviside’s dolphin Cephalorhynchus heavisidii.
Abstract 
Introduction 
Methods 
Results 
Discussion 
Literature Cited 
Figures 
Chapter 4 A population estimate of Heaviside’s dolphins in the southern end of their range.
Abstract 
Introduction 
Methods 
Results 
Discussion 
Literature Cited 
Figures 
Appendix 1 – Example images of marks used to identify Heaviside’s dolphins 
Chapter 5 Range and movements of female Heaviside’s dolphins Cephalorhynchus heavisidii, as determined by satellite-linked telemetry.
Abstract
Introduction 
Methods 
Results 
Discussion 
Literature Cited 
Figures
Chapter 6 Synthesis
Literature Cited 
Appendix: Reprint of: Range and movements of female Heaviside’s dolphins Cephalorhynchus heavisidii as determined by satellite-linked telemetry.Journal of Mammalogy 87(5):866–877.

GET THE COMPLETE PROJECT

Related Posts