Ecological restoration is being increasingly applied to reverse or mitigate biodiversity losses, re-instate ecological functions and increase the provision of ecosystem services in tropical forests. Effective assessment of the success of ecological restoration projects is critical in justifying the use of restoration, as well as improving best practice. However, there is often the assumption that once a degree of vegetation recovery occurs, diversity will increase, which equates with restoration of ecosystem functions. Since very few studies have investigated the interaction between the recovery of habitat structure, biodiversity and ecosystem functioning, this thesis aims to explore these mechanistic relationships to better understand the causal factors behind ecosystem recovery following restoration. Both mammal and dung beetle community composition was clearly progressing towards that of rainforest with increasing restoration age. Restoration increased dung beetlemediated secondary seed dispersal, leaf litter decomposition rates and decomposition multifunctionality (dung and litter decomposition). Functional trait-based metrics provided a clearer pattern of mammal recovery than traditional species-based metrics. Functional diversity metrics were also better predictors of dung beetle-mediated functionality than species diversity metrics, emphasising the need to use a variety of ecologically meaningful diversity metrics when investigating the mechanisms and patterns driving ecological recovery. In terms of vegetation structure, microhabitats were more complex and microclimatic conditions were more stable in restored sites and became more similar to rainforest with age. Faunal recovery was best explained by vegetation structure and microhabitat conditions; whereas functional recovery was explained by a combination of vegetation structure, microhabitat, soil properties and landscape context. These findings suggest that although landscape context and intrinsic site characteristics affect restoration success, they can potentially be mitigated by the establishment of a well-developed, rainforest-like habitat structure and microclimatic conditions within restoration sites. By taking a holistic approach, this thesis demonstrates that ecological restoration of tropical forests leads to the development of a structurally more complex, rainforest-like vegetation structure, a shift to more stable microclimatic conditions and increased availability of microhabitat resources. These successional changes lead to the recovery of functionally diverse, rainforest-like faunal communities and efficient ecosystem functions within a relatively short time frame (10-17 years).

Billions of people rely on wild meat (wild fish and bushmeat) for their livelihood and food security, but unsustainable harvesting is causing pan-tropic defaunation. In addition to the decline of prey populations, many harvest-dependent societies in these regions are witnessing major social and environmental changes, including; rapid urbanisation, population growth, (unequal) socio-economic development, nutritional transitions, policy alterations, habitat conversion and climatic change. Despite the potential importance of these changes for biodiversity and human well-being, our understanding of how these changes are impacting the dynamics of harvesting systems is poor. In this thesis I engage with four major knowledge gaps, in particular. The first is that we know almost nothing about the relative influence of emergent large rainforest cities on wildlife. Second, the suggested connections between defaunation and food insecurity have weak empirical foundations. Third, links between aquatic and terrestrial ecosystems are largely neglected. And fourth, despite recent recognition of their importance, social-ecological feedbacks in tropical harvesting systems remain woefully understudied. Hence, my overall aim in this thesis was to address these and related knowledge gaps by investigating the drivers and dynamics of contemporary wildlife harvesting in the Amazon, and assess the outcomes for biodiversity and rural food security. I focus on rural Amazonians, amongst who high levels of social marginalization, multi-dimensional poverty and food insecurity have been identified. These people are juxtaposed between traditional lives where seasons dictate subsistence wildlife harvesting, and a modern Amazonia in which rural livelihoods are increasingly influenced by the demands of growing urban areas, home to three out of four Amazonians. This PhD study was designed to capture the influences of the seasonal flood pulse and urban markets (metropolitan and provincial) in an extensive and largely-forested area of the Amazon floodplain. This was achieved by interviewing households in 22 communities during both the high and low water season along a 1,267 km stretch of the River Purus, the most important river for commercial fishing for Manaus, Amazonia’s largest city of over 2 million people. Food insecurity was assessed during 556 household visits, in which time nearly 600 different harvesters were interviewed about their hunting and fishing activities, including detailed catch and effort data concerning 886 fishing trips. I show that Manaus’ favourite fish species (tambaqui, Colossoma macropomum) halves in body size and catch rate (catch-per-unit-effort in biomass; CPUEb) within several hundreds of kilometres of the city, with defaunation detectable 1000 km into the rainforest wilderness (Chapter 2). Despite defaunation of the main target species, rural fishers near to Manaus managed to maintain overall fish CPUEb, and levels of food insecurity were no worse than upstream (Chapter 3). Instead, I reveal severe seasonal food insecurity among the rural population associated with falls in aggregated fishing CPUEb of 73% during high waters. I provide novel evidence that food insecurity can result from significant falls in wildlife CPUEb (Chapter 3), and that this seasonal food insecurity may drive increased bushmeat offtake (Chapter 4). I also show in Chapter 4 how seasonal water-level and market forces can dictate wildlife harvesting profiles (akin to species assemblages). I provide evidence that the mechanism driving the defaunation seen in Chapter 1 and the differing harvest profiles in Chapter 4 is the regular service of city-based boats that purchase fish and deposit ice only in communities nearer to Manaus. Chapter 5 was inspired by listening to local voices, specifically, concerns that making a living is being increasing constrained by a combination of defaunation and environmental legislation. I explored these viewpoints using a social-ecological vulnerability framework, allowing me to better understand and voice their concerns, while contributing to the poorly studied concept of social-ecological feedbacks. Through these findings, I establish and advance key links between drivers and dynamics of contemporary wildlife harvesting in the Amazon, and the outcomes for biodiversity and rural food security. As such, I emphasise the importance of taking a holistic view of the research and management of harvesting systems to help achieve sustainable food systems in the Amazon, and across the planet.

Agriculture is the greatest threat to biodiversity across the world and a major contributor to anthropogenic greenhouse gas emissions. Both pressures will increase over coming decades as populations and per capita consumption rise. How we choose to produce food will, to a large extent, determine the state of biodiversity and the wider environment in the 21st century. Balancing livestock production and environmental concerns is of particular importance: rangelands cover approximately one quarter of the world’s ice-free land and livestock consume over one third of all calories from crops. In addition, livestock, particularly ruminants are extremely inefficient and use more land, nitrogen and water than other foodstuffs, whilst producing more CO2. Finally, there is a strong relationship between wealth and meat consumption. Combined with increasing populations, this means that demand for meat is likely to continue to increase. Two alternative strategies have been proposed to minimise the environmental cost of food production: land sharing attempts to maximise biodiversity within the farmed landscape by keeping yields artificially low. Alternatively, land sparing links high yield agriculture with increased habitat protection or restoration. Previous studies have examined the relative benefits of the two strategies in different agricultural systems and regions. However, my research is the first to look at a wide range of alternative livestock production systems in a highly disturbed part of the tropics. To investigate this issue I collected data on the production and requirements of different cattle ranching systems in Yucatán, Mexico, as well as on the population densities of birds, trees and dung beetles, and carbon stocks in both natural habitats and the different ranching systems. I used novel methods to estimate the yields of my study sites and applied both previously developed density-yield functions and new scenario building methods to model how species and carbon stocks responded to increasing agricultural yields. I found that all taxa, and carbon stocks, show similar responses: rapidly declining with conversion from natural habitats to agricultural land. The populations of most species, and regional carbon stocks, were therefore maximised with a land-sparing strategy that combines high yields with forest protection or restoration. Such as strategy is broadly aligned with the goals of producers, environmental organisations and policy makers in Yucatán. However, mechanisms for active land sparing, which can link yield increases with habitat protection, will be needed to ensure that the benefits of land sparing can be realised. In addition, even with land sparing, probable 2030 production targets still resulted in forest loss, highlighting the need for demand reduction as an important part of a sustainable food strategy.

The work described in this research uses the commercial data of catch to illustrate the facilities that GIS may offer to fisheries management. Three different types of fisheries data (different in terms of original format, method of collection and geographical footprints) for Oman are presented and discussed. Data collection, manipulation, filtering and preparation for input to a relational database are discussed. The spatial integration of the data is problematic but essential. Two methods of data integration, the point and the area are used to solve this problem. Point and area data are used to display data at different levels of management and decision making. Alternative visualisations from the GIS are used including total catch distribution, catch per unit effort per unit area, fishing density, and spatio-temporal analysis. The analytical ability of the GIS is used to display the following results: preferred areas of catch, expected catch per area, areas of fish high quality habitat, unexploited areas, and areas suspected of overfishing. Hydro-biological data (marine-related data) such as sea surface temperature, dissolved oxygen, salinity, and zooplankton biomass for the year 1995 with relation to catch are presented and discussed too. The concentration of catch is found to be more within the area of the continental shelf that in the deep waters, being areas of large fish productivity as well as high zooplankton biomass. Pelagic fish occur in greater density in the northern part of the Omani waters (Gulf of Oman), and in contrast, demersal fish aggregated more at the southern part of the country (Arabian Sea). Different areas are found to have a consistently high catch throughout the years, and the catch, in general, shows no evidence of overfishing. Some areas of Omani water have not been fully utilised although this study shows them high potential catch areas.

The thesis examines community involvement in the planning of protected areas designated for biodiversity conservation. The research centres on a case study of planning at two coastal sites in Belize: Bacalar Chico and Caye Caulker. The study employs qualitative methodology to analyse forms of public participation in planning and to explore the relations and strategies of power in operation between the diverse actors in the process. The case study revealed that official public participation exercises functioned as circumscribed forms of consultation. Local stakeholders were granted some opportunity to express their opinions, but decision-making remained in the hands of the planning agencies. Certain key local actors could, however, make use of alternative channels of involvement, such as political lobbying and informal social contact. Both forms of community involvement were played out in a complex arena of power relations. The power strategies of actors drew on unevenly distributed resources such as knowledge, discourse, authority and access to state apparatus. Actors also employed a range of tactics including persuasion, compromise, manipulation, exclusion, enrolment and the formation of alliances to secure influence in the power arena. At one level of abstraction it was possible to identify a power-typology of local actors with characteristic interests, roles and relations with planners. From the two original analytical themes a third, grounded theme emerged relating to the central role played by the planning authorities. Instead of fostering meaningful participation, planners were effectively engaged in a process of containment: their actions in the power arena were geared toward avoiding or blocking disruption and maintaining control. But containment was partial, and the extent of counter-containment helped to explain differences in planning progress between the two study sites. The thesis goes on to argue that attempted containment is inherent in the planning of externally-driven, biodiversity-oriented protected areas.

Extractive operations in the Andes and Western Amazon overlap with important biodiversity sites and areas of high ecosystem services provision. On one hand, the governments of Colombia, Ecuador and Peru have been increasingly developing and exploiting oil and gas, and mining resources due to the high revenues and important contribution to their national economies. On the other, conservation priorities at regional‐ and global‐scale recognise the importance of preserving these priority sites for their biodiversity value and ecosystem services relevancy. The work contained in this thesis identifies ecosystem services that are at risk due to the extractive activities and proposes novel ways on how to evaluate these risks and prioritise conservation, ultimately contributing towards a more sustainable development of extractives in the region. As part of the research, the use of GIS techniques and modelling tools was tailored to suit the complexities of the diverse combination of variables. Datasets of mining for minerals and precious metals concessions, as well as oil and gas concessions are confronted with biodiversity and ecosystem services parameters, in order to obtain an appropriate understanding of current impacts extent and their implications. Around three quarters of the Amazon in Ecuador and Peru is covered by oil and gas concessions, whilst mining concessions are predominant in the Andes slopes. These extractive concessions overlap with protected areas and conservation priority sites. Some areas of pristine rainforest in Peru (Pacaya Samiria National Reserve) and Ecuador (Yasuni National Park) were consistently identified as high providers of carbon services, water provision and natural hazard mitigation services, as well as being home to high numbers of species of several taxonomic groups, many of them endemic to the region. This is the conflictive baseline situation of extractives and conservation priorities in the region. Modelling tools were used to establish this baseline, and from then they were applied in two ways: a) to evaluate different strategies for conservation prioritisation, and b) to create potential but realistic scenarios of future extractive development in the region. Conservation prioritisation strategies that include considerations of multiple ecosystem services, threatened biodiversity, current pressure and future threats were set to detect the topmost sites recommended for conservation. This prioritisation assessment utilised the threshold of 17% of the top areas, to resemble the Aichi target 11 for 2020. Most of the identified priority areas (77%) are already covered by current protected areas system, which helps strengthening the case to protect them, but a considerable portion (31%) is also overlapped by extractive concessions, which pose a threat to their conservation in the long term. The modelled development scenarios for extractives showed that mining operations in the Andes would cause comparatively lesser impact extent in area, but highly localised impacts that could potentially harm the means of subsistence of local populations. On the other hand, modelled oil and gas extraction in the lowlands of the Amazon is larger in extent, but may cause harm to relatively less people. Nevertheless, the pristine rainforests that would be affected hold immense value of globally‐ (e.g. carbon) and locally‐relevant (e.g. water provision) ecosystem services and constitute the habitat of unique high levels of biodiversity. Furthermore, the spatial results show how potential residuals of all modelled extractive operations could cause off‐site impacts that travel far downstream the waterways even crossing international borders. Management options for extractive development should try to find a middle ground that recognises the topmost priority areas for conservation as no‐go zones for extraction, but leaves other areas of comparatively less importance to be developed under strict environmental policy control that minimises the impacts on biodiversity and ecosystem services we all rely on.

Fish otoliths have long played an important role in sustainable fisheries management. Stock assessment models currently used rely on species specific age profiles obtained from the seasonal patterns of growth marks that otoliths exhibit. We compare methods widely used in fisheries science (elliptical Fourier) with an industry standardised encoding method (MPEG7 - Curvature-Scale-Space) and with a recent addition to shape modelling techniques (time-series shapelets) to determine which performs best. An investigation is carried out into transform methods that retain size-information, and whether the boundary encoding method is impacted be otolith age, performing tests over three 2-class otolith datasets across six discrete and concurrent age groups. Impact of segmentation methods are assessed to determine whether automated or expert segmented methods of boundary extraction are more advantageous, and whether constructed classifiers can be used at different institutions. Tests show that neither time-series shaplets nor Curvature-Scale-Space methods offer any real advantage over Fourier transform methods given mixed age datasets. However, we show that size indices are most indicative of fisheries stock in younger single-age datasets, with shape holding more discriminatory potential in older samples. Whilst commonly used Fourier transform methods generally return best results; we show that classification of otolith boundaries is impacted by the method of boundary segmentation. Hand traced boundaries produce classifiers more robust to test data segmentation methods and are more suited to distributed classifiers. Additionally we present a proof of concept study showing that high energy synchrotron scans are a new, non-invasive method of modelling internal otolith structure, allowing comparison of slices along near infinite numbers of virtual complex planes.

It is argued in this thesis that it is morally right to identify and address matters of energy justice associated with renewable energy technologies - and thus biofuels. Equity appraisals, as defined in this thesis, can help to identify social and environmental burdens caused by the implementation of these technologies and where they exist, thus helping to understand the extent to which global sustainable development ideals to reduce inequalities are being achieved. This study is the first equity appraisal of an internationally-traded (Brazil-UK) liquid biofuel (sugarcane bioethanol), across both sites of production and consumption, conducted in a manner advocated by energy justice and environmental justice theories. Furthermore, this study provides the first empirical insights in this context of the ways that principal dimensions of energy justice can interrelate and specifically how matters of procedural justice and recognition can drive distributional changes in outcomes amongst people connected and affected to a transnational liquid biofuel supply chain. Primary qualitative data collected from people living in producer and consumption localities revealed that the nature and geographical patterning of issues differed from the views of transnational governance actors and experts. Rather than the majority of burdens lying with those living in Brazil, and UK-based consumers largely indifferent and unaffected, this research found both positive and negative equity issues affecting people at both ends of the supply chain. Matters of recognition and procedural injustice were found to be affecting consumers, affecting their abilities to engage effectively with their liquid biofuels purchases that could help drive the consumption of more sustainable, just and socially acceptable biofuels. Conversely, higher levels of recognition of local communities and associated impacts in this particular Brazilian production locality were found to be improving social and environmental outcomes for residents. This research highlights the importance of situated, contextual, primary qualitative data for equity appraisals of liquid biofuels and other renewable technologies. It is argued that these types of appraisals should be conducted more systematically in the field to supplement existing forms of appraisals, support decision-making processes and improve the chances of achieving energy justice in relation to renewable energy technologies.

To mitigate potential negative impacts to marine vertebrates it is necessary to gain, and build on, knowledge and understanding of their spatial ecology. Aerial and ship based surveys, as well as satellite telemetry data, have allowed for growing insight into habitat use across a broad spectrum of migratory marine species. Furthermore, these data have often enabled characterisation of anthropogenic impacts and identified potential conservation management strategies. This thesis seeks to investigate the spatial ecology of marine vertebrates using sea turtles as a study group. Data for inter-nesting and post-nesting sea turtles are analysed, and where possible, threats investigated. The analyses presented here integrate the use of multiple spatial ecological tools, including aerial surveys, satellite tracking, remote sensing, Geographical Information Systems (GIS) and habitat modelling. Many of the analytical processes employed formulate novel methodologies, as well as build upon and develop existing techniques. For post-nesting turtles, foraging and migratory data are analysed, and observed and modelled habitat niches described. Putative threats from fisheries and climate change are investigated, and where appropriate, contextualised with data describing limits of Marine Protected Areas (MPAs). For inter-nesting turtles, at-sea distributions and coastal density patterns are explored. Vessel Monitoring System (VMS) and Automatic Identification System (AIS) data are used to elucidate shipping densities; spatial patterns of threat from fisheries, and other maritime industries are inferred. Aerial survey data are used to ascertain potential impacts to turtles on nesting beaches. Throughout this thesis spatially explicit areas are identified where concentrated conservation efforts could be applied. Furthermore, many of these analyses highlight that conservation policy must recognise the spatial extent of migratory species, and be flexible and adaptive to accommodate potential range shifts under climate change. Much of the presented analyses assimilate data from multiple sources to provide large datasets; allowing analyses to be made that would be otherwise unfeasible. Finally, this thesis demonstrates the utility of developing and applying novel analytical methodologies to these data to investigate the spatial ecology of marine vertebrates of conservation concern. As such, it is likely that many of the analytical techniques presented here could be adapted and applied to other widely dispersed marine vertebrate species to help inform global conservation planning and practice.

Small-scale fisheries (SSF) management approaches which place communities at the centre of decision making are becoming increasingly popular. Local leadership is crucial to participative methods due to increased responsibility placed on local actors. Despite its importance, an initial literature review revealed limited, focused SSFs leadership research. This thesis aimed to contribute to the emerging field of SSF leadership and increase understanding of leadership processes. My objectives were to identify key leadership research gaps, explore how leadership interacts with other important contextual conditions, decipher the influences on effective leadership, and discuss how leadership can facilitate more effective SSFs management. I used a multi-method approach to research leadership at a global scale which includes Qualitative Comparative Analysis (QCA) and interviewing. My research finds leadership to be complex, uncertain and dynamic. Leadership acts alone or in combination with other contextual conditions, to influence positive and negative, social and ecological outcomes. Numerous factors influence the propensity of an individual to engage with leadership, such as worldviews, resource constraints at the individual and community level, and interactions with other social actors. Despite the move to participatory approaches there are still numerous concerns about SSFs management, such as the sustainability of community-based organizations. A key finding is that leadership will have an increasingly important role to play in improving the longevity of community-based organizations through processes such as leaderful organizations, succession planning, and capacity building. Leadership is a new research field; therefore this work is of an explanatory nature in terms of its focus and use of novel methodologies. My research identifies important areas for further analysis, such as deciphering the influence of high level leadership on local processes, and investigating how to develop leaderful organizations. Future research should build on my findings to enhance knowledge of leadership functions and processes.