Animals have occupied a prominent position in geographical research for some time (Philo and Wilbert 2000, Wolch and Emel, 1998) and the importance of animals in personal narratives of identity, place and space is widely accepted (see for example Matless et al., 2005). However, such research is predominantly focussed on understandings of mammals. This thesis contests this dominance by critically investigating the significance of fish in shaping understandings of animals. The work also connects with the burgeoning geographical interest with water (see for example, Gandy, 2002; Swyngedouw, 2004; Gandy, 2004, Kaika, 2005; and Braun 2005) but through an approach which steps away from understandings of water as just a resource to recognise that it is a resource that connects different issues, scales and approaches and has a materiality that shapes, and influences understandings of people and places. It also contributes to debates surrounding nature/society as it examines the interactions between humans and non-mammalian animals to interrogate issues relating to escape, wildness, nostalgia and connects large scale ‘landscape’ approaches to close-up encounters with the more-than-human world. These connections flow from the connectivity generated through water as it connects ‘diffuse’ issues into particular organisms. This connectivity has been exploited as fish are used as indicator species for public policy. Alongside this use as indicator species, fisheries management is usually science driven. Therefore running through the thesis is a critique of the role of the natural sciences and economics in shaping understandings fish as the political ecology of which fish, in which locations, are made to count. All this was achieved through a methodology that encouraged anglers to address the complexities, inconsistencies and tensions within their angling experience.

The thesis took an interdisciplinary, problem-oriented approach to address the research problem: ’Is farmed fish capable of replacing wild-caught fish in the Malaysian diet?’ The main objectives of the project were to assess current fish consumption habits, evaluate the impacts of these on sustainability of fish stocks and determine whether aquacultured products could be a suitable substitution. A combination of data collection methods was used, namely dietary assessments, systematic literature review, market research interviews and surveys and analysis of electronic and paper-based official records. The review of the Malaysian food balance sheets showed a significant transition in diet over the last three decades, specifically with respect to an increase in animal protein by approximately 60% over this time, with fish as the major source of protein. To further explore the contribution of fish to the diet of Malaysians, and any ethnic and geographical differences in consumption, a food frequency questionnaire (FFQ) was developed to assess habitual dietary intakes within selected coastal, rural and urban populations across different cultural groups in Klang Valley. The FFQ provided a clearer picture of the quantities of fish consumed by Malaysians and enabled the characterisation of Malaysian fish consumers based on the types of fish species consumed for sustainability assessment purpose. The nutritional contribution of wild versus farmed fish was compared and contrasted. Results from this study, combined with detailed studies of perception of wild versus farmed fish by consumers, aquaculturists and wholesalers, provided a greater understanding of the factors that influence consumers’ fish buying and consumption habits. The key findings are discussed with respect to the sustainability of the current situation, potential for expanding the aquaculture sector to replace wild fish in the Malaysian diet and recommendations for future research and issues for policy makers involved in the expansion of the industry.

A problem commonly encountered in stock assessments of tropical marine resources in developing countries is data paucity, which invariably results from the lack of both human and economic capacity within the government to implement and maintain programmes for data collection and analysis. With special reference to the demersal fishery of Brunei Darussalam, this thesis examines approaches for extracting useful information from data-poor fisheries to assess the state of resources and inform fishery management actions. By using official fishery statistics, augmented by local ecological knowledge (LEK) obtained from fishers engaged in either the large-scale (LS) or small-scale (SS) fisheries in Brunei, changes in demersal fishery resources over the years were assessed. The sustainability of Brunei’s demersal capture fishery was evaluated in the face of its ongoing development and climate change. Using trophodynamic indicators such as mean trophic level (MTL), Fishing-in- Balance (FiB), trophic spectra (TS) and community structure analyses, LS fishery catches of Brunei between 2000 and 2009 revealed a deteriorating state of the coastal demersal ecosystem. Closer examination of the abundance of overall demersal finfish stocks, using the Catch-Per-Unit-Effort (CPUE) index – standardised for other factors not related to abundance – indicated a declining trend, even when total catches remained stationary, although trends in abundance of the different demersal fish families varied. This rapid significant change in recent years is further supported by fishers’ LEK on relative abundance of Brunei marine resources. The study on LEK has also revealed the ‘shifting baseline syndrome’ (SBS) among currently active fishers and their exploited populations, a phenomenon not previously reported for Brunei fisheries. Findings from the study are synthesised with other information, where a number of key issues and policy options are discussed, and recommendations for the management of the fishery are made. This thesis demonstrates that researchers in data-poor fisheries can utilise different assessment tools, given the resources at their disposal, to assist in the management of marine resources.

Young sandeels (0- and 1-group fish) form a substantial part of the total stock biomass at Shetland. As well as forming the basis for a local fishery at Shetland, they are also an extremely important prey item for many seabird species which breed around Shetland coasts. Improving our understanding of recruitment dynamics is vital for managing the Shetland fishery whilst minimizing adverse affects on breeding seabirds. Therefore, this thesis has concentrated on analyzing recruitment patterns and developing models to simulate hypothetical recruitment scenarios. Various models were developed and/or parameterised to simulate recruitment. Firstly, as there appeared to be no clearly defined stock-recruit relationship, time series models were applied to expose and improve understanding of patterns in the data. However, the relationship between stock and recruitment cannot be ignored if spawning stock biomass changes through exploitation. Therefore, Shepherd (1982) and Ricker (1954, 1975) stock-recruit models are parameterised using bootstrap analysis. For the Shepherd model, which is the more complex of the two, bootstrap analysis identified models which could be considered as "constant recruitment" models - that is no influence of spawning stock biomass on recruitment. It was found that four very low recruit years, which occurred in the late 1980s when spawning stock biomass was generally high or moderate, were strongly influencing model fit. This prompted the question: are these low recruit years typical of the range of recruitment for the given spawning stock biomasses or can they be considered outliers? In the face of uncertainty (although there is no biological reasoning) these data points were removed and the stock-recruit models re-parameterised. This enabled curves with 95% confidence intervals to be parameterised for both stock-recruit models (Shepherd & Ricker). The autoregressive model parameterised earlier was combined with the stock-recruit models so recruitment in any year was expressed as a function of SSB and of recruitment in the previous year.

This research investigates the policy response to the 2007-2010 sanitary crisis in the Chilean salmon industry, the second largest producer and exporter of salmon in the world. This industry is an emblematic case of the possible consequences of employing an intensive natural resource model for development. The research draws upon the two literatures on policy learning and policy change, and crisis and disaster management, and upon the system failure to explain the causes and consequences of the sanitary crisis in the industry. The thesis employs the qualitative method of case study and utilises primary and secondary sources of data and information. The main argument of the research is that the process of policy learning during and following catastrophic events is very different from the process of the policy learning during normal times. The main findings are, firstly, in the case of the Chilean salmon industry, the sanitary crisis disrupted the industry governance processes, including the regulatory framework of the industry, opening a window for radical institutional change. Secondly, potentially radical measures were part of a set of policies that emerged as initial responses, after which a set of more incremental policy responses were developed and applied. Moreover, the sanitary crisis was not transient nor episodic but was enduring, persistent and dynamic. Thirdly, the policy responses to the sanitary crisis destabilised the consensus in the Chilean industry causing conflict and ambiguity over policy responses. The thesis contributes to the policy learning and policy change literature in the context of catastrophic events by extending the view this literature offers on dramatic events such as crises, by considering them as dynamic and persistent situations, analysing their potential as precipitators of radical policy change, and providing a means for considering the timing and processes by which this radical policy change may occur and be directed toward better social outcomes.

Molecular ecology is one of the most important branches of evolutionary biology, and it uses the advantages of molecular techniques such as PCR-RFLP, sequencing, microsatellite analysis, and most recently the introduction of next generation sequencing, to address outstanding issues in the fields of population genetics and phylogeny. The genomic approach has been influential in providing new information relevant to traditional questions in ecology, such as genetic differentiation, speciation, species adaptation and others. The rationale of the present thesis was to incorporate the advantages of both the PCR-RFLP and sequencing techniques to gain information on the genetic variability of two commercially-important crustacean species, namely the Norway lobster Nephrops norvegicus and the giant freshwater prawn Macrobrachium rosenbergii. Nephrops norvegicus exhibits large morphological variability even between adjacent populations. The first objective of this thesis was to investigate the genetic variability of N. norvegicus from three localities, the Clyde Sea area and North Minch (North Sea) as well as Icelandic waters. The phylogeographic study found no significant differentiation between populations from the studied areas (FST: 0.01819). This finding is consistent with outcomes from previous studies that N. norvegicus populations were not geographically structured. Outcomes from the present study strongly suggest that environmental factors, rather than genetic factors, are more likely to play a more significant role in the high morphological differentiation observed in this species. A study of the most important freshwater crustacean species, Macrobrachium rosenbergii was then undertaken as a contribution to understanding the most complex biogeography in the world, the Indo Australian Archipelago (IAA). The IAA has incredible species richness and endemism and is the location of 4 out of 25 world’s biodiversity hotspots, namely the Sundaland, the Philippines, Indonesia and Wallacea. Within the IAA is the location of Wallace’s line and Huxley’s line, the most abrupt faunal transition in the world that lies between the Sunda and Sahul shelves. The studied species used in the present thesis, M. rosenbergii is an ideal model species as it has a wide geographical distribution across the IAA. The present phylogeographic study screened the COI segment using the sequencing technique to study M. rosenbergii populations collected from eight locations in Malaysia in the peninsular and east of Malaysia. These populations exhibited high genetic differentiation (FST: 0.62503) mainly due to the sample from Sabah. However, the adjacent population (Sarawak) was similar to that in Peninsular Malaysia, even though Northern Sarawak showed sub-population differentiation from the main cluster (cluster I) indicating that the genetic diversity of Northern Sarawak was more restricted. Beside, cluster II observed in the study indicated and confirmed the recent aquaculture activities of restocking the Kedah, Perak and Sarawak populations. Knowledge of the levels of genetic differentiation in N. norvegicus and M. rosenbergii could assist in the management of the species. N. norvegicus could be managed as one stock and conservation and recovery programme could be carried out based on the knowledge that all studied populations exhibited lack genetic differentiation within and between populations. In contrast, M. rosenbergii that possessed high level of genetic differentiation have to be managed separately, especially for a unique population such as the one in Sabah. The outcomes of this study could also be useful for future research in the conservation of wild population, as well as aquaculture management and product improvement purposes. The finding of Sabah as a unique population could potentially be useful for aquaculture improvement programmes. One of the most important aspects is to see whether Sabah population possessed high resistant to the disease infection. A preliminary study was conducted to investigate the susceptibility of a Malaysian wild population of M. rosenbergii to infection by the human food-poisoning bacterium V. parahaemolyticus. Nonetheless, the virulence stage of the bacterial strain, the status of the immune system of the host, the size and age of the experimental animals as well as the dose of injected bacteria might all have contributed to the inconclusiveness of the results. However, the principle of screening wild populations for disease resistance is sound, and may lead to improvements in the quality of the broodstock used in the Malaysian aquaculture industry.

The dynamics of demersal fish species assemblages and stocks on the continental shelf of Ghana, West Africa were investigated using data from trawl surveys conducted between 1963 and 1990, and in relation to marine environmental and nearshore forcing factors. Time series analysis was used to examine the dynamics of the marine environment. Classification and ordination methods were used to identify fish assemblages and to identify factors that determine assemblage formation and dynamics. The response of the identified assemblages to changes in the Ghanaian shelf ecosystem were assessed. Between 1964 and 1992, three distinct environmental time blocks (ETBs) in the continental shelf ecosystem off Ghana were recognised. In the first ETB (i. e. before 1972) sea surface and bottom temperatures were relatively high and salinity was low. The second ETB (1972 - 1982) was characterised by low temperatures, high but stable salinity, reduced rainfall and freshwater input into the sea. In the third ETB (1982 - 1992), the system bore a resemblance to the first block with rising temperatures and decreasing salinity. Six fish species assemblages were identified, namely the supra-thermocline sciaenid and lutjanid assemblages, two sparid assemblages (shallow and deep elements), the deep shelf assemblage and the upper slope assemblage. During ETB I and ETB3, the assemblages were easier to identify using ordination techniques, than during ETB2. Species diversity also decreased during ETB2 especially in waters between 10 and 50 m deep. This behaviour of the assemblagesis attributed to the increase in abundance of Balistes capriscus, a species that dominated the ecosystem for nearly 20 years from about 1971, and the observed environmental changes. The estimated density of demersal fish was higher in the upwelling season than in the thermocline season and decreased from 50 kg ha-1 in 1963-64 to 32.4 kg ha-1 in 1990. The density reached its lowest value of 22.5 kg ha-1 in the period between 1973 and 1977. The potential yield of the total demersal biomass (excluding triggerfish) is estimated to be 36,000 - 55,000 mt per annum. With landed catches exceeding 60,000 mt in some years, it is apparent that the resource is over-exploited. A number of policy options are discussed and recommendations for the management of the fishery are made.

The main aim of this thesis was to study intra- and inter-specific competition among juvenile Atlantic salmon and brown trout. The study concentrated on two critical times for survival: first, the period from shortly after emergence until later in the summer when individuals first establish territories; and second, during the winter, when many salmonid populations experience a bottleneck. In addition, the stability of social hierarchies and the influence these have on growth and survival were studied on older individuals in the summer. Experiments were carried out in a variety of conditions ranging from small arenas and artificial stream channels where behavioural interactions among fish could be observed, to the wild where it was possible to test inferences derived from laboratory observations in an ecological context. Both salmon and trout are known to undergo a behavioural shift in winter, switching from being primarily diurnal during the summer to being predominantly nocturnal in winter. Atlantic salmon and brown trout forage in slow-flowing water at night and shelter in interstitial spaces in the substrate during the day. This thesis shows that salmon and trout compete for both foraging (chapter 2) and sheltering (chapter 3) habitat in winter, illustrating that competition between the two species is not restricted to the summer months. Trout were shown to dominate salmon in competition for foraging habitat, forcing them to move into shallower water or become more diurnal (chapter 2). Intra- and inter-specific competition for shelters was equal in intensity and more dependent on arrival time (prior residency) than species identity (chapter 3). This competition could have short- and long-term survival consequences for over-wintering fish, particularly Atlantic salmon, and have implications for the carrying capacity of streams.

The potential impacts of marine aquaculture operations on the environment are reviewed. The reported effects of suspended mussel farms on the benthic environment are examined and the potential impacts discussed. A framework to assess the impacts of suspended mussel farms is presented. The use of simulation models to predict the impact of fish farm wastes on the benthic environment is discussed and the mathematical theory supporting such models is presented. The applicability of these models to mussel farming is discussed and the data required to undertake such modelling identified. The effect of increased sedimentation on the macrobenthic community, physical structure and biogeochemistry of the surficial sediment around three suspended mussel farms are examined. At one site, the benthic community was subjected to bulk sedimentation, organic enrichment and reduced macrobenthic infaunal diversity. Elevated levels of organic carbon were recorded close to the farm. At the remaining two sites, benthic impacts were less clear and not demonstrably due to the mussel farms. The settling velocity of mussel faeces and pseudofaeces was required to enable modelling of particles ejected from the farm sites. An experiment was devised to measure this parameter.- The settling velocity of mussel faeces (~0.5 cms-I ) was less than pseudofaeces (~1 cms-I ). Differences in these settling velocities were attributed to the organic content and particle size of the excreted matter. The particle tracking model DEPOMOD (Cromey et at., 2000a) was used as a platform from which to develop a simulation model predicting the benthic impact of suspended mussel farms. Parameters within the model were modified to be represent a mussel farming scenario. Data from the three sites surveyed were applied to the model. Although the model results compared favourably with the field data, the model tended to overestimate the benthic impact as measured by the Infaunal Trophic Index. The results of the model are discussed and improvements and further experiments are identified.

Perceived conflicts between piscivorous birds and commercial freshwater fisheries are common, and such a perception exists at Loch Leven, a wetland of international importance for nature conservation and a famous commercial brown trout Salmo trutta fishery, where Great Cormorants Phalacrocorax carbo have been shot in large numbers. This thesis describes the foraging behaviour of cormorants wintering on Loch Leven. It summarises changes in wintering numbers over a 32 year period, and reviews data on fish populations, fish stocking rates, angling catches and angling effort in order to seek evidence of detrimental impacts of cormorants on the fishery, and beneficial effects of large-scale cormorant shooting. Finally, this thesis considers turnover within the wintering cormorant population, and its implications for cormorant control. Cormorants spent most of the time roosting, and a single peak of feeding activity occurred early in the morning. Compared to early or late winter, cormorants showed higher foraging activity during mid-winter. Both solitary feeding and flock feeding were recorded, with flock feeding predominating. Solitary feeding was distributed over a wide area. The intensity of flock feeding was less evenly distributed, with 36% of grid squares not used. Overall, 78% of flock feeding activity took place in 23% of grid squares and 59% took place in 13% of grid squares. The best predictor of both solitary feeding and flock feeding intensity of use was mean water depth. Mean water depth was also the best predictor of winter brown trout distribution, although winter brown trout distribution was not a predictor of either solitary feeding or flock feeding intensity of use.