Poaching in marine protected areas: drivers of and responses to illegal fishing

Conservation is primarily reliant on people's compliance with rules and regulations. Yet, non-compliance is widespread throughout the world, and regularly negates expected conservation outcomes. The effects of non-compliance are especially apparent in protected areas, where even short bursts of poaching (defined in this thesis as fishing in no-take zones, or no-fishing marine reserves) can rapidly negate the effects of protection that often take decades to produce (e.g. Russ & Alcala 2010; Wittemyer et al. 2014). In addition, efforts to reduce poaching often fail due to a lack of capacity, resources, and/or general understanding of the drivers or influences on the behaviour (e.g. Gill et al. 2017). Thus, curtailing poaching depends on gathering detailed information to guide specifically targeted behavioural and management interventions. However, gathering information on poaching is inherently difficult because it is an illegal, clandestine, and often socially unacceptable activity. This thesis therefore addresses this considerable knowledge gap by applying specialized techniques to measure, assess, and understand poaching. Many methods and approaches can be used to assess and measure poaching, but no single method is a panacea, because each is limited in the information it can provide and subject to different types of bias (Gavin et al. 2010; Arias 2015; Bergseth et al. 2015). Reliably assessing and measuring poaching therefore necessitates a triangulation approach that utilizes complimentary methods to provide a holistic picture of poaching. Chapter 1 therefore provides a general introduction about the methods used to estimate compliance levels, as well as the different theoretical disciplines and approaches that can be used to understand poaching. The focus of Chapter 2 is to answer the first research question of my thesis: "How can poaching be measured given its cryptic nature?" In this chapter, I combine social surveys with a complementary field-based method to 'ground-truth' and assess the prevalence poaching by recreational fishers in the Great Barrier Reef Marine Park (GBRMP). I use three specialized questioning techniques [Self-administered questioning (SAQ); the Randomised Response Technique (RRT; Warner 1965); the Unmatched Count Technique (UCT; Droitcour et al. 1991)] to estimate poaching rates. Concurrently, I develop and examine the potential of two theoretically grounded, proxy indicators of non-compliance (perceived prevalence of poaching and personally knowing a poacher). Both of these indicators yield higher estimates than those provided by specialised techniques, which suggests that specialised questioning techniques are still subject to underreporting. Furthermore, my findings indicate that the false consensus effect and social learning are causing poachers and their associates to overestimate the prevalence of poaching. I also quantify the accumulation rate of derelict fishing gears on fringing coral reefs inside and outside no-fishing reserves of two nearby inshore island groups. Surprisingly, I find no difference in the accumulation rate of derelict gears between areas open and closed to fishing. Overall, I demonstrate that a substantial amount of poaching is occurring in inshore island groups once thought to be among the best-enforced areas in the GBRMP, which has important management implications. Understanding the drivers or influences on recreational fisher's poaching decisions is a critical component of designing management and behavioural interventions to curtail poaching. While numerous disciplines explore the complexities of human decision-making and resultant behaviours, most investigations of poaching suffer from 'disciplinary silo-thinking' which inhibits a comprehensive understanding of poaching behaviours (Von Essen et al. 2014). Thus, the focus of Chapter 3 is to develop a multi-disciplinary approach to comprehensively examine the social dimensions of poaching. In this chapter, I integrate pertinent theories from a range of disciplines including sociology, criminology, wildlife biology, and psychology to develop 29 potential drivers of compliance behaviour. I then explore the prevalence and distribution of these potential drivers. I find that most fishers perceive high levels of legitimacy for management agencies and see poaching as personally and socially unacceptable. However, my research findings also suggest that two additional (mis)perceptions (pluralistic ignorance and a perceived lack of deterrence) are likely operative and at least partially responsible for the continuation of poaching in the GBRMP. Lastly, I reveal that fishers perceived two primary motivations to poach: better fishing in no-fishing reserves, and a low probability or risk of detection. These results suggest that extolling certain benefits of no-fishing reserves (i.e. they hold bigger and more fish) where enforcement capacity is low could actually lead to the perverse outcome of encouraging noncompliance. Based on these results, I highlight tools such as social norms approaches, strengthened coercive deterrence, and fear arousing communication that can be used to address these misperceptions and increase compliance. In Chapter 4, I ask: "how are different poaching measure or proxies related to the potential drivers of compliance behaviour?" Specifically, I collated and condensed 29 potential influences and drivers of poaching that were identified in Chapter 3, and empirically examined their influence on three indicators or proxies of poaching: the RRT, perceived levels of poaching, and personally knowing a poacher. The RRT model performed particularly poorly at identifying behavioural drivers of poaching, likely because the admitted level of poaching was lower than the intentionally introduced statistical noise (to obscure respondent's answers and ensure confidentiality). However, 66% of the drivers that were significant for one proxy indicator of poaching were also significant for the second proxy. When considered in light of the results from Chapter 2, this suggests that these proxies could be further integrated to estimate and understand poaching, especially in contexts where poaching is socially unacceptable and fairly rare. Yet, many of the worlds marine protected areas suffer from a critical lack of capacity (Gill et al. 2017) and subsequent low levels of compliance, which negate expected outcomes. Accordingly, natural resource management agencies are increasingly attempting to bolster compliance by engaging the latent surveillance and enforcement potential of resource users (GBRMPA 2016; Green 2016; Kohn 2016). However, little is known about the conditions or institutions that encourage this behaviour. In Chapter 5 I fill this knowledge gap by collating and analysing more than 2000 fisher surveys from 55 MPAs in 7 countries to fill this knowledge gap by answering four research questions: 1) How many fishers have previously observed poaching?: 2) How do they respond to observed non-compliance?; 3) Why do fishers remain inactive?; and 4) What are the effects of institutions and conditions on fisher's responses to poaching? I found that nearly half of these fishers had previously observed poaching, but the most common response was inaction, typically due to conflict avoidance, a sense that it was not their responsibility or concern, and the perception that poaching was a survival strategy. Furthermore, I quantified how institutional design elements relate to fisher's responses to poaching, and highlight avenues to responsibly engage fishers while mitigating risk. This body of work advances the current state of knowledge in compliance management, particularly in regard to the approaches necessary for fully measuring and understanding the prevalence and drivers of poaching. The findings provided by my multidisciplinary approach advance our understanding and management of human behaviour in four concepts that are critical for effective conservation: 1) further consideration of how people process information; 2) re-conceptualizing how people behave; 3) developing communication strategies to bolster compliance; and 4) designing rules and interactions to shape behaviour. In addition, this thesis demonstrates the necessity of triangulating sources of compliance information, and delivers findings that have been directly adapted for communication and outreach strategies employed by one of the world's most iconic marine parks