The efficacy of a Marine Protected Area (MPA) is contingent on it having a design appropriate for the species it is intended to protect. Hector�s dolphin (Cephalorhynchus hectori), a coastal delphinid endemic to New Zealand, is endangered due to bycatch in gillnets. Analyses of survival rate and population viability suggest that the Banks Peninsula population is most likely still declining despite the presence of the Banks Peninsula Marine Mammal Sanctuary (BPMMS), where gillnetting is regulated. More data on distribution and movements of dolphins are therefore required to improve the design of the BPMMS. On aerial surveys of Hector�s dolphin distribution at Banks Peninsula over three years, sightings were made up to 19 n.mi. offshore. On average, 19% of dolphins were sighted outside the BPMMS�s 4 n.mi. offshore boundary in summer, compared to 56% in winter. On similar surveys of the South Island�s west coast, all dolphins were sighted within 6 n.mi. of the coast and there was no seasonal change in distribution. At each location, Mantel tests indicated that distance offshore had the strongest and most consistent effect on distribution. However, a logistic regression model using the combined datasets suggested that distribution was most strongly defined by water depth, with all sightings made inside the 90 m isobath. Boat surveys were carried out at Banks Peninsula (2002 to 2006) to continue the long-term photo-ID project. Using the 22 year dataset, alongshore home-range of the 20 most frequently sighted dolphins was estimated by univariate kernel methods. Mean alongshore range was 49.69 km (SE = 5.29), 60% larger than the previous estimate. Fifteen percent of these individuals had ranges extending beyond the northern boundary of the BPMMS. An acoustic data logger, the T-POD, was trialled for passive acoustic monitoring of Hector�s dolphins. Simultaneous T-POD/theodolite surveys revealed that T-PODs reliably detected dolphins within 200m. No detections were made beyond 500m. To monitor inshore habitat use, T-PODs were deployed in three locations at Banks Peninsula (n = 431 days). A GLM analysis of Detection Positive Minutes (DPM) per day indicated that season had the largest effect on detection rate, with over twice as many DPMs per day in summer (x̄ = 99.8) as winter (x̄ = 47.6). The new findings on Hector�s dolphin distribution and ranging can be used to improve the design of the BPMMS. It is recommended that the offshore boundary of the BPMMS is extended to 20 n.mi. (37 km), the northern boundary is moved 12 km north and recreational gillnetting is prohibited year round. In areas where distribution of Hector�s dolphin has not been studied, the offshore boundary of MPAs should enclose the 100 m isobath.

Species Distribution Modelling (SDM) plays a key role in a number of biological applications: assessment of temporal trends in distribution, environmental impact assessment and spatial conservation planning. From a statistical perspective, this thesis develops two methods for increasing the accuracy and reliability of maps of density surfaces and provides a solution to the problem of how to collate multiple density maps of the same region, obtained from differing sources. From a biological perspective, these statistical methods are used to analyse two marine mammal datasets to produce accurate maps for use in spatial conservation planning and temporal trend assessment. The first new method, Complex Region Spatial Smoother [CReSS; Scott-Hayward et al., 2013], improves smoothing in areas where the real distance an animal must travel (`as the animal swims') between two points may be greater than the straight line distance between them, a problem that occurs in complex domains with coastline or islands. CReSS uses estimates of the geodesic distance between points, model averaging and local radial smoothing. Simulation is used to compare its performance with other traditional and recently-developed smoothing techniques: Thin Plate Splines (TPS, Harder and Desmarais [1972]), Geodesic Low rank TPS (GLTPS; Wang and Ranalli [2007]) and the Soap lm smoother (SOAP; Wood et al. [2008]). GLTPS cannot be used in areas with islands and SOAP can be very hard to parametrise. CReSS outperforms all of the other methods on a range of simulations, based on their fit to the underlying function as measured by mean squared error, particularly for sparse data sets. Smoothing functions need to be flexible when they are used to model density surfaces that are highly heterogeneous, in order to avoid biases due to under- or over-fitting. This issue was addressed using an adaptation of a Spatially Adaptive Local Smoothing Algorithm (SALSA, Walker et al. [2010]) in combination with the CReSS method (CReSS-SALSA2D). Unlike traditional methods, such as Generalised Additive Modelling, the adaptive knot selection approach used in SALSA2D naturally accommodates local changes in the smoothness of the density surface that is being modelled. At the time of writing, there are no other methods available to deal with this issue in topographically complex regions. Simulation results show that CReSS-SALSA2D performs better than CReSS (based on MSE scores), except at very high noise levels where there is an issue with over-fitting. There is an increasing need for a facility to combine multiple density surface maps of individual species in order to make best use of meta-databases, to maintain existing maps, and to extend their geographical coverage. This thesis develops a framework and methods for combining species distribution maps as new information becomes available. The methods use Bayes Theorem to combine density surfaces, taking account of the levels of precision associated with the different sets of estimates, and kernel smoothing to alleviate artefacts that may be created where pairs of surfaces join. The methods were used as part of an algorithm (the Dynamic Cetacean Abundance Predictor) designed for BAE Systems to aid in risk mitigation for naval exercises. Two case studies show the capabilities of CReSS and CReSS-SALSA2D when applied to real ecological data. In the first case study, CReSS was used in a Generalised Estimating Equation framework to identify a candidate Marine Protected Area for the Southern Resident Killer Whale population to the south of San Juan Island, off the Pacific coast of the United States. In the second case study, changes in the spatial and temporal distribution of harbour porpoise and minke whale in north-western European waters over a period of 17 years (1994-2010) were modelled. CReSS and CReSS-SALSA2D performed well in a large, topographically complex study area. Based on simulation results, maps produced using these methods are more accurate than if a traditional GAM-based method is used. The resulting maps identified particularly high densities of both harbour porpoise and minke whale in an area off the west coast of Scotland in 2010, that might be a candidate for inclusion into the Scottish network of Nature Conservation Marine Protected Areas.

The Eastern Scottish population of bottlenose dolphins (Tursiops truncatus) is the northernmost population of this species. The resident core of this population consists of 120 to 150 different individuals. This small size and its geographical isolation from other populations raises questions about its viability and whether the population has behavioural patterns that differ from those common to other populations of the same species. Microsatellite genetic diversity was low and mitochondrial DNA genetic diversity values were lowest in East Scotland compared to other populations worldwide and to neighbouring populations around UK waters. It has been well documented, from four different field sites worldwide, that male bottlenose dolphins form alliances with preferred male associates. These alliances can last for several years and the males involved show association coefficients similar to those of mothers and calves (0.8-1.0). These alliances appear to be of great importance in obtaining matings for the males. In the Eastern Scottish population males do not form alliances. No evidence of strong associations between individuals of either sex was found and there was no correlation between association and relatedness patterns. I suggest that the isolation and small size of the population together with reduced genetic diversity affects the pressure of kin selection for altruistic behaviours. There is no gain in competing or associating with close relatives for access to mates and it might be more important to avoid inbreeding by dispersing. Although evidence of gene flow between East Scotland and its neighbouring populations was not confirmed with Bayesian clustering analysis, a small set of individuals from Wales were found to be closely related to individuals from the East Coast of Scotland. In general the dynamics found in UK water populations resemble those of the Western North Atlantic with sympatric populations of coastal as well as pelagic individuals.

South Pacific humpback whales were devastated by commercial whaling in their Antarctic feeding areas during the 20th century. Understanding migratory connections and current abundance of these isolated breeding stocks is crucial for the allocation of historical Antarctic catches in population dynamic models used to assess current recovery. However, only a small number of migratory connections have been documented between Oceania breeding stocks within the South Pacific and feeding areas in the Antarctic. In addition, little is known about abundance of these stocks which encompass a vast oceanic region. For this thesis I first used mixed-stock analysis (MSA) to allocate migratory connections from four Antarctic feeding areas (n=142) to seven South Pacific breeding stocks (n=1,373), including four in Oceania, based on genetic marker frequencies. The use of this method was justified by the breeding stocks showing genetic differentiation at the haplotype level with an F[subscript ST] value of 0.027 (p-value <0.001). The results showed a relatively strong connection of Western Australia to Antarctic Area IV, Tonga to the border of Antarctic Area VI/I, Colombia to the Antarctic Peninsula, and a split allocation of Eastern Australia and New Caledonia to Antarctic Area V. This study provides the first population-level information supporting previous individual-based studies that humpback whale migration may not necessarily be direct north south. Next, utilizing capture-recapture methodology of unique humpback whale fluke photographs, I estimated abundance of one of the least studied Oceania breeding stocks, French Polynesia, a stock which also showed no significant migratory allocation using MSA. Taking into consideration the possible advantages of using Quality Control (QC) photographs to minimize bias in matching, estimates were generated using the complete photo catalogue and also using only photographs adhering to QC criteria. I found that the choice of using QC has an effect on the abundance generated and discuss the implications of this finding. Despite the photo catalogue used, the French Polynesia stock is estimated to number less than 1,900 individuals. Lastly, to provide additional information on the French Polynesia stock I used photo-identification to compare French Polynesia whales to whales in the Antarctic Peninsula and Strait of Magellan (Antarctic Area I), a possible migratory connection suggested by previous microsatellite genotyping. No conclusive matches were found. Although this does not discount the possibility of a few migrants traveling between these regions it does indicate the Antarctic Peninsula and the Strait of Magellan are not primary feeding areas of French Polynesia. This new information regarding abundance and migration of French Polynesia whales is important for the Comprehensive Assessment of Southern Hemisphere humpback whales. This document is currently being completed as the International Whaling Commission considers the next critical steps in recovery for Oceania humpback whales stocks. / Graduation date: 2010

In conservation ecology, abundance estimates are an important factor from which management decisions are based. Methods to estimate abundance of cetaceans from visual detections are largely developed, whereas parallel methods based on passive acoustic detections are still in their infancy. To estimate the abundance of cetacean species using acoustic detection data, it is first necessary to correctly identify the species that are detected. The current automatic PAMGUARD Whistle Classifier used to automatically identify whistle detection of cetacean species is modified with the objective to facilitate the use of these detections to estimate cetacean abundance. Given the variability of cetacean sounds within and between species, developing an automated species classifier with a 100% correct classification probability for any species is unfeasible. However, through the examples of two case studies it is shown that large and high quality datasets with which to develop these automatic classifiers increase the probability of creating reliable classifiers with low and precise misclassification probability. Given that misclassification is unavoidable, it is necessary to consider the effect of misclassified detections on the number of observed acoustic calls detected and thus on abundance estimates, and to develop robust methods to cope with these misclassifications. Through both heuristic and Bayesian approaches it is demonstrated that if misclassification probabilities are known or estimated precisely, it is possible to estimate the true number of detected calls accurately and precisely. However, misclassification and uncertainty increase the variance of the estimates. If the true numbers of detections from different species are similar, then a small amount of misclassification between species and a small amount of uncertainty in the probabilities of misclassification does not have a detrimental effect on the overall variance and bias of the estimate. However, if there is a difference in the encounter rate between species calls associated with a large amount of uncertainty in the probabilities of misclassification, then the variance of the estimates becomes larger and the bias increases; this in return increases the variance and the bias of the final abundance estimate. This study despite not bringing perfect results highlights for the first time the importance of dealing with the problem of species misclassification for cetacean if acoustic detections are to be used to estimate abundance of cetaceans.

The ranging behaviour, habitat preferences, genetic structure, and demographic parameters of bottlenose dolphins living in the Azores were studied using data collected from 1999 to 2004. Only 44 dolphins out of 966 identified were frequently sighted within and between years and showed strong site fidelity. The remaining individuals were either temporary migrants from within or outside the archipelago, or transients. Estimates of home range size were three times larger than previously reported for this species, possibly as a result of the lower availability of food resources. Mitochondrial DNA sequences showed very high gene and nucleotide diversity. There was no evidence of population structuring within the Azores. The Azorean population was not differentiated from the pelagic population of the Northwest Atlantic, suggesting the "unproductive" waters of the Atlantic do not constitute a barrier to dispersal. Population size, survival and temporary emigration rates were estimated using open-population models and Pollock's robust design. A few hundreds of dolphins occur in the area on a given year, though the majority should use it temporarily, as suggested by the high emigration rates. Bottlenose dolphins preferentially used shallow areas with high bottom relief. Temporal and spatial persistence of dolphin-habitat associations documented in this study further supports the idea of a close relationship between certain bathymetric features and important hydrographic processes and suggests the occurrence of prey aggregations over these areas may be, to some extent, predictable. Several results of this study suggest there are no reasons for concern about the status of this population. Yet, the resident group may be negatively affected by increasing pressure from the whale watching activity. Although the proposed Marine Park constitutes important habitat for resident dolphins, at present, the area is clearly insufficient to satisfy their spatial requirements and its conservation value may be limited.

Sotalia guianensis is listed as “Data Deficient” by the IUCN and as “Vulnerable” in Colombia. This study aimed to advance understanding of the ecology of this species and its habitats, and to provide information to conservation management in the southern Gulf of Morrosquillo, Colombia. Systematic boat-based surveys were conducted during 395 days in 2002-2006 and 2009-2010, following established routes. Total survey effort was 15,199 km in an area covering ~ 310km². Fine scale habitat use and behavioural modelling, photo-identification and mark-recapture techniques were used to analyze the ecological patterns for this species. The most recent abundance estimate of dolphins using the study area during dry and rainy seasons, varied from 225 (CV = 0.34; 95% CI: 118-426) to 232 (CV = 0.32; 95% CI: 127-246). Annual survival rate is estimated at 0.948 (95% CI = 0.876-0.980). Overall density was 0.74/km². Dolphins were present year-round in the whole study area. Results indicate that they do not use the study area uniformly and that the use of particular zones is related to eco-geographic variables. Dolphins showed a preference for waters greater than 3m in depth with a slightly increased preference for waters about 5m and 15-25m deep. The average group size was nine individuals. Some individuals show long-term high site fidelity to some zones within the study site boundaries. Even though the site fidelity to feeding areas varied individually, all the individuals focused primarily on one specific area. Foraging was among one of the most predominant behaviours observed. The individual movements show that some dolphins use both bay and gulf waters. Dolphins show a range of surface cooperative foraging and feeding strategies. These cooperative behaviours were influenced by zone, group size and prey type. Based on these results an area of special management for the species will be created in Colombia.

Thesis (Ph.D.)--University of Aberdeen, 2005. / Includes bibliographical references. Available in PDF format via the World Wide Web.