Hostname: page-component-7c8c6479df-27gpq Total loading time: 0 Render date: 2024-03-27T15:19:59.453Z Has data issue: false hasContentIssue false

Optimizing enforcement and compliance in offshore marine protected areas: a case study from Cocos Island, Costa Rica

Published online by Cambridge University Press:  04 August 2014

Adrian Arias*
Affiliation:
Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia.
Robert L. Pressey
Affiliation:
Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia.
Rhondda E. Jones
Affiliation:
School of Marine and Tropical Biology, James Cook University, Townsville, Australia
Jorge G. Álvarez-Romero
Affiliation:
Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia.
Joshua E. Cinner
Affiliation:
Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia.
*
(Corresponding author) E-mail adrian.arias@my.jcu.edu.au
Rights & Permissions [Opens in a new window]

Abstract

Illegal exploitation of resources is a cause of environmental degradation worldwide. The effectiveness of conservation initiatives such as marine protected areas relies on users' compliance with regulations. Although compliance can be motivated by social norms (e.g. peer pressure and legitimacy), some enforcement is commonly necessary. Enforcement is expensive, particularly in areas far from land, but costs can be reduced by optimizing enforcement. We present a case study of how enforcement could be optimized at Cocos Island National Park, Costa Rica, an offshore protected area and World Heritage Site. By analysing patrol records we determined the spatial and temporal distribution of illegal fishing and its relationship to patrol effort. Illegal fishing was concentrated on a seamount within the Park and peaked during the third year-quarter, probably as a result of oceanographic conditions. The lunar cycle in conjunction with the time of year significantly influenced the occurrence of incursions. The predictability of illegal fishing in space and time facilitates the optimization of patrol effort. Repeat offenders are common in the Park and we suggest that unenforced regulations and weak governance are partly to blame. We provide recommendations for efficient distribution of patrol effort in space and time, establishing adequate governance and policy, and designing marine protected areas to improve compliance. Our methods and recommendations are applicable to other protected areas and managed natural resources.

Type
Papers
Copyright
Copyright © Fauna & Flora International 2014 

Introduction

Offshore marine protected areas are an emerging approach to marine conservation and fisheries management (Game et al., Reference Game, Grantham, Hobday, Pressey, Lombard and Beckley2009; Graham & McClanahan, Reference Graham and McClanahan2013). Their effectiveness, as for near-shore protected areas, relies heavily on compliance by fishermen (Campbell et al., Reference Campbell, Hoey, Maynard, Kartawijaya, Cinner, Graham and Baird2012; Edgar et al., Reference Edgar, Stuart-Smith, Willis, Kininmonth, Baker and Banks2014). Even remote locations are not safeguarded from illegal fishing because fishermen will travel long distances to target commercially valuable species (Berkes et al., Reference Berkes, Hughes, Steneck, Wilson, Bellwood and Crona2006); for example, in the Chagos Archipelago, a remote area in the Indian Ocean with few inhabitants, illegal fishing has resulted in a significant decline in sharks (Graham et al., Reference Graham, Spalding and Sheppard2010).

Ensuring compliance offshore is challenging. Patrolling large and distant tracts of ocean is logistically difficult and financially expensive, and therefore detection rates for illegal activity are low. Additionally, important factors for compliance, such as social norms (e.g. peer pressure and legitimacy) and legislation, may be absent or lack support offshore. Although voluntary compliance is desirable (Ostrom, Reference Ostrom1990; Hønneland, Reference Hønneland2000; Arias & Sutton, Reference Arias and Sutton2013), not all people comply voluntarily; typically some degree of enforcement is necessary (Tyler, Reference Tyler1990; Hønneland, Reference Hønneland2000). However, enforcement is perhaps the most expensive management activity in both terrestrial and marine protected areas (Robinson et al., Reference Robinson, Kumar and Albers2010; Ban et al., Reference Ban, Adams, Pressey and Hicks2011). With limited funds for conservation, optimizing enforcement can make management more cost-effective.

There is a common misconception that enforcement involves only patrolling but it actually encompasses detection, arrest/citation, prosecution and conviction (Sutinen, Reference Sutinen1987; Akella & Cannon, Reference Akella and Cannon2004); it can be described heuristically as a four-link chain. The first link, probability of detection, is mainly technical and field-based, relying on factors such as equipment, and number and skills of wardens. The remaining links (probabilities of arrest/citation, prosecution and conviction) tend to rely progressively more on legal and political constructs. The probability of arrest depends partly on field equipment (e.g. a boat's capacity to pursue) but also on what legally constitutes non-compliance and evidence. The probability of prosecution involves the capacity of the legal and institutional system to undertake proceedings against non-compliance, underlining the importance of strong institutions and coordination between them. Conviction, and its associated penalties, rests on the judiciary and its probability depends on the quality of evidence and the overall capacity of the enforcement system (Akella & Cannon, Reference Akella and Cannon2004). Ultimately the effectiveness of all links is influenced by resources, legislation and political will.

Recognizing that all links in the chain must be strong for effective enforcement, here we focus on the probability of detection, which is a key, yet largely unstudied, aspect of natural resource management. To increase the probability of detection, patrol effort must be distributed efficiently; for this, authorities need reliable data on illegal use of resources. However, such illegal activity is typically clandestine, which poses methodological challenges for data collection (Gavin et al., Reference Gavin, Solomon and Blank2010; Arias & Sutton, Reference Arias and Sutton2013). Few methods are suitable for studying illegal use of resources offshore. Sparse populations and the presence of foreign boats reduce the practicality of social-survey techniques. Modelling, forensics and remote surveillance are expensive and require appropriate technology and/or training, inhibiting their use in developing countries. Indirect observations of illegal activity, such as derelict gear, are challenged by currents, depth and extent. Enforcement in protected areas is typically based on patrols, and patrol records are one of the few means available for assessing illegal use of resources. This approach has been used in terrestrial protected areas to provide information on factors such as patrol allocation (Leader-Williams et al., Reference Leader-Williams, Albon and Berry1990), funding of enforcement (Hilborn et al., Reference Hilborn, Arcese, Borner, Hando, Hopcraft and Loibooki2006) and levels of poaching (Knapp et al., Reference Knapp, Rentsch, Schmitt, Lewis and Polasky2010), yet patrol records for marine protected areas remain largely unanalysed (but see Davis et al., Reference Davis, Russ, Williamson and Evans2004; Mangubhai et al., Reference Mangubhai, Saleh, Suprayitno, Muljadi, Purwanto, Rhodes and Tjandra2011).

Here we present a case study of Cocos Island National Park (hereafter Cocos), Costa Rica. We focus on the probability of detection and explore options for optimizing enforcement and compliance. Cocos shares key characteristics with other offshore marine protected areas, including remoteness, presence of illegal fishing and difficulty of enforcement. We analyse illegal fishing and patrol effort by using a multi-year dataset of patrol records, expert consultation and literature. We use the concept of the enforcement chain to address two research questions: (1) How can patrol effort be optimized to match the spatial and temporal distribution of illegal fishing? (2) What are the key constraints on the subsequent links of the enforcement chain: arrest/citation, prosecution and conviction? Our methodological developments and recommendations aim to contribute to the adaptive planning and management of Cocos and other offshore marine protected areas.

Study area

Cocos is located c. 500 km south-west of Costa Rica, in the Pacific Ocean (Fig. 1). The no-take marine protected area (1,989 km2) was established in 1984 (Salas et al., Reference Salas, Ross-Salazar and Arias2012) and the only inhabitants of the Island are Park wardens and, occasionally, Coast Guard staff, researchers and volunteers. The Park is a World Heritage Site and a Ramsar site. Cocos has among the highest fish biomass in the tropics (7.8 tons ha−1), notable endemism and globally threatened species (Friedlander et al., Reference Friedlander, Zgliczynski, Ballesteros, Aburto-Oropeza, Bolaños and Sala2012) and is consequently recognized as a top international dive destination. However, it also attracts fishermen, mostly Costa Rican, who use surface longlines to target fish.

Fig. 1 Cocos Island National Park, Costa Rica, and the surrounding Seamounts Marine Management Area. The rectangle on the inset shows location of the main map in relation to mainland Costa Rica.

In 2011 a multiple-use marine protected area, Seamounts Marine Management Area (9,640 km2), was created around Cocos (Fig. 1) with the primary objective of protecting seamounts. In the Seamounts Marine Management Area bottom trawling and purse seining are prohibited and longlining is regulated. Here we focus on Cocos but also provide recommendations that relate to the Seamounts Marine Management Area.

Methods

We used a 5-year dataset (December 2005–September 2010) of patrol records from Cocos, which were compiled by Park wardens, the Coast Guard and MarViva, an NGO that assisted with patrols. The dataset contained information on 1,501 patrols and was missing data for November 2007 and July and February 2008. Data were extracted from patrol logbooks detailing hours and nautical miles patrolled and, if applicable, the confiscation of longlines. We mapped only the initial location of confiscated longlines in the Park (hereafter incursions) because final locations were not commonly given. We did not estimate catch per unit effort because soak time, which is the time a baited hook is available and an integral component of catch per unit effort (Ward et al., Reference Ward, Myers and Blanchard2004), was unknown and assumed to be highly variable.

To analyse whether and how patrols could be optimized we first examined illegal fishing in terms of catch, and the spatial and temporal distribution of incursions. We recorded (1) which fish species were commonly caught, (2) whether incursions were concentrated on specific bathymetric features and (3) whether incursions prevailed during specific months and lunar phases. Secondly, we examined the temporal distribution of patrol-days according to months and lunar phases. For these analyses we grouped all patrols for each day, giving a total of 1,078 patrol-days as input. These temporal analyses made it unnecessary to distinguish between multiple patrols on the same day. Thirdly, we tested for correlations (rs ) between (1) nautical miles and hours patrolled, (2) incursions and catch and (3) hours patrolled and incursions. For the correlations we used individual patrols, distinguishing between those occurring on the same day.

To explore bathymetry we used the GEBCO_08 Grid v. 20100927 (GEBCO, 2014) and we created a three-dimensional chart using Surfer v. 11 (Golden Software, Golden, USA). Incursions were mapped in ArcMap v. 10.1 (ESRI, Redlands, USA) and analysed for spatial patterns with Moran's I spatial autocorrelation test. We used a 5 km threshold (mean distance between incursions) within which to consider the spatial relationship between neighbouring records.

We used a logistic regression to examine the effects of time of year (i.e. year-quarters) and lunar phases (i.e. lunar-quarters) and their influence on the probability of a patrol detecting an incursion. Patrol effort was not included in the model because once an illegal incursion was detected further search effort on that day was considerably reduced, and exploratory analyses revealed that including patrol effort did not change the model´s coefficients significantly. The regressions therefore examined variations between year- and lunar-quarters in the proportion of patrol-days on which incursions were detected. Because year-quarter and lunar-quarter were categorical variables, the model yielded a perfect fit to the data and was used to provide estimates of detection probabilities for each combination of year-quarter and lunar-quarter. Data analysis was carried out using SPSS v. 20 (IBM, North Castle, USA ) and S-PLUS v. 8 (TIBCO Software, Palo Alto, USA).

To analyse data according to the lunar cycle we counted the number of days after the new moon (day 0) when an incursion was detected, and we refer to these as lunar days. We used moon phase predictions by F. Espenak (NASA, 2012). We converted lunar days to angles by multiplying each lunar day by 360 and then dividing by 29.53 (number of days in a lunar month). We refer to the phase from new moon to first quarter as first quarter, first quarter to full moon as second quarter, full moon to third quarter as third quarter, and third quarter to new moon as last quarter. Circular histograms were created using Oriana v. 4 (Kovach Computing Services, Anglesey, UK).

To analyse key constraints on arrests/citations, prosecutions and convictions related to incursions in Cocos we reviewed legislation, newspaper articles and grey literature, and interviewed five key informants, who had substantial legal and/or practical experience of enforcement in the Park. Interviews were conducted in Spanish and consisted of open-ended questions about patrols and the legal mechanisms to control illegal fishing.

Results

General characteristics of patrol effort and incursions

Three hundred incursions, with nearly 34,500 hooks, were recorded in the Park. Approximately 2,000 animals were hooked, of which 66% were tuna and 25% were sharks. The most commonly reported species were yellowfin tuna Thunnus albacares and silky shark Carcharhinus falciformis. Less common species included marlin (Istiophoridae), turtles (Chelonioidea), rays (Batoidea) and dolphins (Delphinidae). Incursions were clustered non-randomly (I = 0.301, z = 75.18, P < 0.0005) north-east of the Park, in the vicinity of a seamount (Fig. 2). The seamount has a steep wall and lies c. 15 km from the island (Fig. 2b; Lizano, Reference Lizano2012).

Fig. 2 (a) Locations of recorded incursions in Cocos Island National Park, Costa Rica. (b) Bathymetric profile of the Park. The black circle in (a) and the 3D area in (b) represent the Park. Note that north–south orientation has been inverted in both (a) and (b) to show the steep walls of the seamount in (b).

The percentage of patrol-days on which incursions were detected was 20%. The mean duration of patrols was 6.9 hours, covering 40.3 nautical miles. There was a positive correlation between total nautical miles and hours patrolled (rs (1,407) = 0.629, P < 0.0005) and between total incursions and total number of animals hooked (rs (1,474)  = 0.782, P < 0.0005). Although statistically significant, there was only a weak correlation between total hours patrolled and total incursions (rs (1,421) = 0.204, P < 0.0005).

Records revealed that illegal fishers used plastic containers and old tyres as buoys. Some boats were seen breaching the Park's boundaries repeatedly, and this persisted until more recently (SINAC-MINAET, 2012; Salas, Reference Salas2013). Using the patrol boat's radar, wardens detected multiple fishing boats frequently, some within the Park and others close to the boundary; those within the Park would flee on seeing the patrol boat. Commonly, boats failed to stop when requested, with insults and threats occasionally directed at wardens by radio. Threats have also been reported more recently (Rojas, Reference Rojas2013). Illegal fishers have reportedly been trolling (Salas, Reference Salas2013), using currents to drift their gear in and out of Cocos, fishing under cover of darkness and painting buoys in low-contrasting colours (SINAC-MINAET, 2012).

Temporal distribution of patrol effort and incursions

Time of year and lunar phase together significantly influenced the probability of encountering incursions (Table 1), which increased during the third year-quarter, with higher probabilities during the first and last lunar-quarters at this time of year (Fig. 3).

Fig. 3 Predictions of probability of encountering incursions within Cocos Island National Park (Fig. 1). Predictors are year-quarters (rows) and lunar-quarters (columns). Probabilities are given on the vertical axis.

Table 1 Analysis of deviance for the binary logistic regression. Year and lunar quarters and their interactions were tested as predictors of illegal incursions into Cocos Island National Park (Fig. 1).

Incursions peaked during the third year-quarter (Figs 3 & 4a), reaching a maximum in August (Fig. 4a). Patrol hours were more evenly distributed throughout the year than incursions (Fig. 4b). Incursions peaked during the first lunar-quarter (Fig. 4c). Patrol hours were higher during the second and third lunar-quarters (Fig. 4d).

Fig. 4 Temporal distribution of incursions and patrol effort in Cocos Island National Park. (a) Incursions by month; (b) patrol hours by month; (c) incursions by lunar cycle; (d) patrol hours by lunar cycle. Bars indicate frequencies. Labels on vertical lines represent number of recorded incursions or hours of enforcement. The radial line on each graph shows the location of the mean value. The length of the mean vector (r), a measure of variance (range 0–1), is given in the top right of each graph; larger values indicate that observations are grouped closer to the mean.

Key constraints on the probabilities of arrest/citation, prosecution and conviction

The management and conservation of marine resources in Costa Rica rely on the fisheries authority and the Environment Ministry. In 2011 the President created a commission to ‘diagnose, assess and recommend the necessary adjustments for good marine governance in Costa Rica’ (Casa Presidencial, 2011). The Commission recommended a review and restructuring of the fisheries authority (Comisión Presidencial para la Gobernanza Marina, 2012) and this was backed by environmental groups (Frente por Nuestros Mares, 2013) and the Comptroller General (CGR, 2012). There is a conflict of interests within the fisheries authority: most of the board members are representatives of the fishing industry (Quesada-Alpízar, Reference Quesada-Alpízar2006).

Probabilities of arrest or citation are low. Although only authorized vessels are allowed to enter the Park (Poder Ejecutivo, 2012), this rule is commonly flouted and is difficult to enforce. When illegal fishermen detect a patrol boat while setting or retrieving gear they cut the line and flee to avoid charges of illegal fishing (Costa Rica does not mandate gear identification). Unless wardens can intercept fleeing boats, arrests/citations are impossible. Wardens must then remove the abandoned fishing gear from the Park. If a boat is intercepted it is given a written and verbal notification stating the illegality of entering the Park. If the same boat is intercepted again within the Park it is processed for ‘disobedience to authority’, for which the captain could face a jail sentence.

Even after arrest or citation the probability of prosecution is limited. Most illegal fishing boats found in the Park lack the legal autonomy to travel further than 40 nautical miles (c. 74 km) from the coast but non-compliance with this regulation goes unsanctioned. There are reported cases of illegal fishing boats receiving fuel subsidies from the fisheries authority (Delgado, Reference Delgado2012), allowing them to reach distant waters at a lower cost. To circumvent the limitations of administrative sanctions by the fisheries authority, the Park's management goes through the judicial system. However, inefficiencies in this system can result in prosecutions being delayed by > 8 years.

At the end of the enforcement chain, convictions are rare. No longline boat or captain has been convicted with severe sanctions (e.g. jail sentences or boat confiscations) for fishing illegally or entering the Park, although one boat has been the subject of > 10 lawsuits for disobedience to authority. Of six prosecutions concluded during December 2010–March 2013 three cases were dismissed and three ended in conviction. However, in all three convictions the captains received probation.

Discussion

Increasing the probability of detection

In Cocos illegal fishing is a considerable threat to some marine species. Illegal activity takes place at targeted locations and times, and repeat offenders usually go unpunished.

Illegal fishing is focused on a seamount within the Park (Fig. 2), which is consistent with the expectation of aggregations and increased presence of predators around seamounts (Morato et al., Reference Morato, Hoyle, Allain and Nicol2010). Studies have reported derelict fishing gear and low densities of groupers at seamounts outside the Park (Starr et al., Reference Starr, Green and Sala2012) and declines in shark abundance inside the Park (Friedlander et al., Reference Friedlander, Zgliczynski, Ballesteros, Aburto-Oropeza, Bolaños and Sala2012). Fishing pressure is significant around and within Cocos and detracts from the Park's ecological integrity.

The higher incidence of illegal fishing during the third year-quarter (Figs 3 & 4a) could be attributable to oceanographic variables. For example, seasonal variations in mixed-layer depth (Fiedler & Talley, Reference Fiedler and Talley2006) can affect the distribution and catchability of yellowfin tuna (Song et al., Reference Song, Zhang, Xu, Jiang and Wang2008), potentially luring fishermen further from the coast and closer to Cocos. Fishermen commonly use remote sensing technology to find oceanographic features such as upwellings (SPC, 2011), so enforcement authorities could also exploit these tools to improve the probability of detection. However, using remotely sensed data increases the technical difficulty of analyses, possibly constraining their usage where expertise and/or funding are limited. The lunar and bathymetric data we used are freely accessible and relatively easy to explore and analyse compared to remotely sensed data on dynamic oceanographic features. Simple and effective methods for analysing patrol records can foster replication.

Fishermen have long been aware of an effect of lunar cycles on fish behaviour and catchability (Parrish, Reference Parrish1999), and the concentration of illegal fishing activity around certain lunar phases observed here is in line with the findings of other studies. Lowry et al. (Reference Lowry, Williams and Metti2007) reported peak catch of several predators, including yellowfin tuna, during the first lunar-quarter. Reduced light around the new moon could drive both prey and predators closer to the surface (Blaxter, Reference Blaxter, Evans, Bainbridge and Rackham1974), potentially increasing their catchability with surface longlines. Reduced light may also favour illegal boats by lowering the probability of detection. The observed temporal patterns of incursions are therefore not unexpected.

Our analysis of monthly trends and the lunar cycle suggests that patrols can be optimized by matching incursions more closely (Fig. 4), thereby increasing deterrence when most needed and reducing management costs by cutting unnecessary effort. However, although natural variables can explain inter- and intra-monthly fluctuations in incursions, fluctuations can also be influenced by patrol effort. If fishers became aware that detection was unlikely at a given time, their motivation for non-compliance could increase at that time and vice versa. Managers can adapt to such variations by monitoring patrol records periodically and systematically to develop approaches that maximize the probability of detection.

The role of marine protected area design in compliance

The spatial design of marine protected areas plays an important role in compliance. An example of this is the Great Barrier Reef Marine Park, Australia, where simple and easily identifiable boundaries provide clarity for users and wardens (Day et al., Reference Day, Dudley, Hockings, Holmes, Laffoley, Stolton and Wells2012). In contrast, the boundary at Cocos is defined by a radius of 12 nautical miles (Fig. 1), represented by the territorial sea, which poses problems for compliance and enforcement. The boundary is somewhat irregular because of the shape of the island and is therefore difficult to identify accurately in the field, which can cause confusion for fishermen and wardens. Modifying the Park's boundary would be difficult politically but a feasible alternative would be to zone the Seamounts Marine Management Area to create a simple polygonal buffer zone around the Park (Fig. 1), which could serve three purposes: (1) to facilitate navigation, (2) exclude longlining further from the Park, and (3) reduce the boundary effect. Fishing effort could be concentrated on the boundary areas (Kellner et al., Reference Kellner, Tetreault, Gaines and Nisbet2007) and illegal fishing would be likely to occur near boundaries because of accidental incursions or deliberate ones facilitated by easier entry and exit (Gribble & Robertson, Reference Gribble and Robertson1998). For a given shape the boundary effect is amplified in smaller marine protected areas because of larger perimeter : area ratios (Kritzer, Reference Kritzer2004), but buffers reduce this effect by reducing the perimeter per unit area. The perimeter : area ratio of the hypothetical buffer in Fig. 1 is c. 0.066 : 1, compared to 0.079 : 1 for the almost circular Park.

Constraints in subsequent links of the enforcement chain

For Cocos the probability of detection, although needing improvement, is apparently stronger than the subsequent links of the enforcement chain. The main reason for weaknesses in the three subsequent parts of the chain is an ineffective legal and governance base. Boats enter the Park continually, sometimes employing techniques to avoid detection. This suggests that some illegal fishing goes undetected and the thousands of illegal hooks that we report are probably underestimates. When illegal fishers lose their gear to wardens their economic losses are potentially offset by the high market prices for tuna and shark fins (finning has been reported in Cocos; Delgado, Reference Delgado2012). They can further minimize their losses by selecting cheap gear. Apparently in Cocos the potential gains from illegal fishing surpass the potential costs of getting caught. This also applies internationally, particularly in relation to high-valued catch (Sumaila et al., Reference Sumaila, Alder and Keith2006).

Assuming that the probabilities of detection, arrest, prosecution and conviction were each 50% the cumulative probability of being penalized would be 6.25% (Table 2, Case 1). However, at Cocos and other offshore marine protected areas these probabilities are likely to be < 50%. With 20% of patrol-days resulting in detection of incursions in Cocos it can be assumed that the overall probability of detecting incursions is lower. Some incursions are not detected because they go unseen during patrols and others occur on days with no patrols. Importantly, as the value for any one link approaches zero, overall enforcement becomes practically ineffectual (Table 2, Case 2), and weak links undermine investments in and success of stronger ones (Table 2, Cases 2 & 3). Obtaining accurate probabilities for such analysis requires systematic record keeping for each link, and/or values estimated through social surveys (Akella & Cannon, Reference Akella and Cannon2004).

Table 2 Cumulative probability of illegal fishers in Cocos Island National Park (Fig. 1) being penalized, given various hypothetical probabilities for each of the four links of the enforcement chain.

Perhaps the main gain from enforcement at Cocos has been avoiding rampant illegal fishing through the partial deterrence offered by patrols. Nevertheless, this is a limited approach that is failing to achieve strong compliance. Technology (e.g. vessel monitoring systems and radar) can increase the probability of detection and reduce management costs by directing patrols but deterrence remains low if other links are weak. Offshore marine protected areas require clear and enforceable regulations, applicable at sea and in ports (e.g. inspections, vessel blacklists). Field staff, prosecutors and judges must be trained in environmental and marine law (Akella & Cannon, Reference Akella and Cannon2004). The institutions managing compliance must collaborate and adapt to change (Hauck & Kroese, Reference Hauck and Kroese2006). Penalties must counter the illegal gains from highly valued catch and include loss of privileges (Robinson et al., Reference Robinson, Kumar and Albers2010), such as loss of access to the Seamounts Marine Management Area (Fig. 1). Graduated sanctions are a useful system in which to nest penalties (Ostrom, Reference Ostrom1990), varying sanctions according to the number and/or severity of violations (Russell, Reference Russell1990).

Although the enforcement system in place at Cocos has deficiencies it deters widespread non-compliance. However, this may not be the case in other marine protected areas. A systematic and periodic analysis of patrol records, such as that presented here, can help optimize enforcement. Our case study highlights how clear and enforceable regulations coupled with strong institutions can also help optimize enforcement. This also applies for areas beyond national jurisdiction (high seas) where, because of legal and governance weaknesses, additional national, regional and international efforts are needed to ensure adequate enforcement (Gjerde et al., Reference Gjerde, Currie, Wowk and Sack2013). Enforcement is a tool to encourage compliance, a means to an end; other tools such as social norms should also be exploited. In the process of achieving international objectives for marine conservation (e.g. effectively conserving ≥ 10% of coastal and marine areas by 2020; CBD, 2010), and with increasing pressure on global marine resources, compliance is vital. Failing to maintain compliance undermines the conservation benefits expected from effective marine protected areas and leads to metrics, such as the extent of marine protected areas, not reflecting actual outcomes for marine conservation.

Acknowledgements

We thank Área de Conservación Marina Isla del Coco, SINAC, Servicio Nacional de Guardacostas, MarViva, Comandante C. Castro, F. Quirós, S. Vega, M. Rodríguez, J. Rodríguez, Y. Peraza, J. Robinson, N. Graham, S. Walker, K. Nash, C. Magnin, FAICO, and the anonymous reviewers who helped improve this article. AA, RLP, JAR and JEC acknowledge support from the Australian Research Council. AA thanks the Australian people for support through AusAID.

Biographical sketches

Adrian Arias focuses on the planning and management of marine resources, with a particular interest in peoples’ compliance with conservation and fishery management efforts. Robert Pressey leads the Conservation Planning Programme at the Australian Research Council Centre of Excellence for Coral Reef Studies. Rhondda Jones is an ecologist with interests in tropical insects and plant–herbivore interactions. Jorge Álvarez-Romero works on integrated land–sea conservation planning. Joshua Cinner is a social scientist studying the interactions of people with the marine environment.

References

Akella, A.S. & Cannon, J.B. (2004) Strengthening the Weakest Links: Strategies for Improving the Enforcement of Environmental Laws Globally. Center for Conservation and Government at Conservation International, Washington, DC, USA.Google Scholar
Arias, A. & Sutton, S.G. (2013) Understanding recreational fishers’ compliance with no-take zones in the Great Barrier Reef Marine Park. Ecology and Society, 18, 18.CrossRefGoogle Scholar
Ban, N.C., Adams, V., Pressey, R.L. & Hicks, J. (2011) Promise and problems for estimating management costs of marine protected areas. Conservation Letters, 4, 241252.CrossRefGoogle Scholar
Berkes, F., Hughes, T.P., Steneck, R.S., Wilson, J.A., Bellwood, D.R., Crona, B. et al. (2006) Globalization, roving bandits, and marine resources. Science, 311, 15571558.CrossRefGoogle ScholarPubMed
Blaxter, J.H.S. (1974) The role of light in the vertical migration of fish, a review. In Light as an Ecological Factor II (eds Evans, G.C., Bainbridge, R. & Rackham, O.), pp. 189210. Blackwell Scientific Publications, Oxford, UK.Google Scholar
Campbell, S.J., Hoey, A.S., Maynard, J., Kartawijaya, T., Cinner, J., Graham, N.A.J. & Baird, A.H. (2012) Weak compliance undermines the success of no-take zones in a large government-controlled marine protected area. PLoS ONE, 7(11), e50074.CrossRefGoogle Scholar
Casa Presidencial (2011) Presidenta juramenta comisión para mejorar la buena gobernanza marina. San José, Costa Rica.Google Scholar
CBD (Convention on Biological Diversity) (2010) Tenth Meeting of the Conference of the Parties to the Convention on Biological Diversity. Decision X2. Strategic Plan for Biodiversity 2011–2020 and the Aichi Biodiversity Targets. UNEP, Nagoya, Japan.Google Scholar
CGR (Comptroller General) (2012) Informe sobre auditoría de carácter especial efectuada en el Instituto Costarricense de Pesca y Acuicultura (INCOPESCA) relacionado con el cumplimiento de sus funciones en materia de conservación del recurso marino. Comptroller General, San José, Costa Rica.Google Scholar
Comisión Presidencial para la Gobernanza Marina (2012) Informe. Presidencia República de Costa Rica, San José, Costa Rica.Google Scholar
Davis, K.L.F., Russ, G.R., Williamson, D.H. & Evans, R.D. (2004) Surveillance and poaching on inshore reefs of the Great Barrier Reef Marine Park. Coastal Management, 32, 373387.CrossRefGoogle Scholar
Day, J., Dudley, N., Hockings, M., Holmes, G., Laffoley, D., Stolton, S. & Wells, S. (2012) Guidelines for Applying the IUCN Protected Area Management Categories to Marine Protected Areas. IUCN, Gland, Switzerland.Google Scholar
Delgado, D. (2012) Impunidad favorece pesca en zona prohibida de la Isla del Coco. La Nación, San José, Costa Rica. Http://www.nacion.com/sucesos/Impunidad-favorece-prohibida-Isla-Coco_0_1299470117.html [accessed 18 June 2014].Google Scholar
Edgar, G.J., Stuart-Smith, R.D., Willis, T.J., Kininmonth, S., Baker, S.C., Banks, S. et al. (2014) Global conservation outcomes depend on marine protected areas with five key features. Nature, 506, 216220.CrossRefGoogle ScholarPubMed
Fiedler, P.C. & Talley, L.D. (2006) Hydrography of the eastern tropical Pacific: a review. Progress in Oceanography, 69, 143180.CrossRefGoogle Scholar
Frente por Nuestros Mares (2013) Carta a Laura Chinchilla y Gloria Abraham. San José, Costa Rica.Google Scholar
Friedlander, A.M., Zgliczynski, B.J., Ballesteros, E., Aburto-Oropeza, O., Bolaños, A. & Sala, E. (2012) The shallow-water fish assemblage of Isla del Coco National Park, Costa Rica: structure and patterns in an isolated, predator-dominated ecosystem. Revista de Biología Tropical, 60, 321338.Google Scholar
Game, E.T., Grantham, H.S., Hobday, A.J., Pressey, R.L., Lombard, A.T., Beckley, L.E. et al. (2009) Pelagic protected areas: the missing dimension in ocean conservation. Trends in Ecology & Evolution, 24, 360369.CrossRefGoogle ScholarPubMed
Gavin, M.C., Solomon, J.N. & Blank, S.G. (2010) Measuring and monitoring illegal use of natural resources. Conservation Biology, 24, 89100.CrossRefGoogle ScholarPubMed
GEBCO (General Bathymetric Chart of the Oceans) (2014) Http://www.gebco.net [accessed 18 June 2014].Google Scholar
Gjerde, K.M., Currie, D., Wowk, K. & Sack, K. (2013) Ocean in peril: reforming the management of global ocean living resources in areas beyond national jurisdiction. Marine Pollution Bulletin, 74, 540551.CrossRefGoogle ScholarPubMed
Graham, N.A.J. & McClanahan, T.R. (2013) The last call for marine wilderness? BioScience, 63, 397402.CrossRefGoogle Scholar
Graham, N.A.J., Spalding, M.D. & Sheppard, C.R.C. (2010) Reef shark declines in remote atolls highlight the need for multi-faceted conservation action. Aquatic Conservation: Marine and Freshwater Ecosystems, 20, 543548.CrossRefGoogle Scholar
Gribble, N.A. & Robertson, J.W.A. (1998) Fishing effort in the far northern section cross shelf closure area of the Great Barrier Reef Marine Park: the effectiveness of area-closures. Journal of Environmental Management, 52, 5367.CrossRefGoogle Scholar
Hauck, M. & Kroese, M. (2006) Fisheries compliance in South Africa: a decade of challenges and reform 1994–2004. Marine Policy, 30, 7483.CrossRefGoogle Scholar
Hilborn, R., Arcese, P., Borner, M., Hando, J., Hopcraft, G., Loibooki, M. et al. (2006) Effective enforcement in a conservation area. Science, 314, 1266.CrossRefGoogle Scholar
Hønneland, G. (2000) Compliance in the Barents Sea fisheries. How fishermen account for conformity with rules. Marine Policy, 24, 1119.CrossRefGoogle Scholar
Kellner, J.B., Tetreault, I., Gaines, S.D. & Nisbet, R.M. (2007) Fishing the line near marine reserves in single and multispecies fisheries. Ecological Applications, 17, 10391054.CrossRefGoogle ScholarPubMed
Knapp, E.J., Rentsch, D., Schmitt, J., Lewis, C. & Polasky, S. (2010) A tale of three villages: choosing an effective method for assessing poaching levels in western Serengeti, Tanzania. Oryx, 44, 178184.CrossRefGoogle Scholar
Kritzer, J.P. (2004) Effects of noncompliance on the success of alternative designs of marine protected-area networks for conservation and fisheries management. Conservation Biology, 18, 10211031.CrossRefGoogle Scholar
Leader-Williams, N., Albon, S.D. & Berry, P.S.M. (1990) Illegal exploitation of black rhinoceros and elephant populations: patterns of decline, law enforcement and patrol effort in Luangwa Valley, Zambia. Journal of Applied Ecology, 27, 10551087.CrossRefGoogle Scholar
Lizano, O.G. (2012) Rasgos morfológicos alrededor de la Isla del Coco y de sus montes submarinos vecinos, Pacífico de Costa Rica. Revista de Biología Tropical, 60, 4351.Google Scholar
Lowry, M., Williams, D. & Metti, Y. (2007) Lunar landings—relationship between lunar phase and catch rates for an Australian gamefish-tournament fishery. Fisheries Research, 88, 1523.CrossRefGoogle Scholar
Mangubhai, S., Saleh, M., Suprayitno, , Muljadi, A., Purwanto, , Rhodes, K.L. & Tjandra, K. (2011) Do not stop: the importance of seamless monitoring and enforcement in an Indonesian marine protected area. Journal of Marine Biology, 2011, 11.CrossRefGoogle Scholar
Morato, T., Hoyle, S.D., Allain, V. & Nicol, S.J. (2010) Seamounts are hotspots of pelagic biodiversity in the open ocean. Proceedings of the National Academy of Sciences of the United States of America, 107, 97079711.CrossRefGoogle ScholarPubMed
NASA (2012) NASA Eclipse Web Site. Http://eclipse.gsfc.nasa.gov/phase/phasecat.html [accessed 18 June 2014].Google Scholar
Ostrom, E. (1990) Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press, Cambridge, UK.CrossRefGoogle Scholar
Parrish, J.K. (1999) Using behavior and ecology to exploit schooling fishes. Environmental Biology of Fishes, 55, 157181.CrossRefGoogle Scholar
Poder Ejecutivo (2012) Decreto Ejecutivo 37023-MINAET: Reglamento de uso público del Parque Nacional Isla del Coco. MINAET, San José, Costa Rica.Google Scholar
Quesada-Alpízar, M.A. (2006) Participation and fisheries management in Costa Rica: from theory to practice. Marine Policy, 30, 641650.CrossRefGoogle Scholar
Robinson, E.J.Z., Kumar, A.M. & Albers, H.J. (2010) Protecting developing countries' forests: enforcement in theory and practice. Journal of Natural Resources Policy Research, 2, 2538.CrossRefGoogle Scholar
Rojas, P. (2013) Serias amenazas también rondan a trabajadores del Parque Nacional Isla del Coco. Crhoy, San José. Http://www.crhoy.com/serias-amenazas-tambien-rondan-a-trabajadores-del-parque-nacional-isla-del-coco/ [accessed 18 June 2014].Google Scholar
Russell, C. (1990) Game models for structuring monitoring and enforcement systems. Natural Resource Modeling, 4, 143173.CrossRefGoogle Scholar
Salas, E., Ross-Salazar, E. & Arias, A. (2012) Diagnóstico de áreas marinas protegidas y áreas marinas para la pesca responsable en el Pacífico costarricense. Fundación MarViva, San José, Costa Rica.Google Scholar
Salas, L. (2013) 20 embarcaciones pesqueras amenazan la Isla del Coco. Crhoy, San José. Http://www.crhoy.com/20-embarcaciones-pesqueras-amenazan-la-isla-del-coco/ [accessed 18 June 2014].Google Scholar
SINAC-MINAET (2012) Comunicado de prensa: Isla del Coco intensifica patrullajes en área protegida: operativos especiales dan como resultado evitar o detener actividades que atentan contra la integridad de los recursos naturales presentes en el Parque. SINAC, San José, Costa Rica.Google Scholar
Song, L.M., Zhang, Y., Xu, L.X., Jiang, W.X. & Wang, J.Q. (2008) Environmental preferences of longlining for yellowfin tuna (Thunnus albacares) in the tropical high seas of the Indian Ocean. Fisheries Oceanography, 17, 239253.CrossRefGoogle Scholar
SPC (2011) A Beginner's Guide to Using Remote Sensing for Offshore Tuna Fishing. Secretariat of the Pacific Community, Noumea, New Caledonia.Google Scholar
Starr, R.M., Green, K. & Sala, E. (2012) Deepwater fish assemblages at Isla del Coco National Park and Las Gemelas Seamount, Costa Rica. Revista de Biología Tropical, 60, 347362.Google Scholar
Sumaila, U.R., Alder, J. & Keith, H. (2006) Global scope and economics of illegal fishing. Marine Policy, 30, 696703.CrossRefGoogle Scholar
Sutinen, J.G. (1987) Enforcement of the MFCMA: an economist's perspective. Marine Fisheries Review, 49, 3643.Google Scholar
Tyler, T.R. (1990) Why People Obey the Law. Yale University Press, New Haven, USA.Google Scholar
Ward, P., Myers, R.A. & Blanchard, W. (2004) Fish lost at sea: the effect of soak time on pelagic longline catches. Fishery Bulletin, 102, 179195.Google Scholar
Figure 0

Fig. 1 Cocos Island National Park, Costa Rica, and the surrounding Seamounts Marine Management Area. The rectangle on the inset shows location of the main map in relation to mainland Costa Rica.

Figure 1

Fig. 2 (a) Locations of recorded incursions in Cocos Island National Park, Costa Rica. (b) Bathymetric profile of the Park. The black circle in (a) and the 3D area in (b) represent the Park. Note that north–south orientation has been inverted in both (a) and (b) to show the steep walls of the seamount in (b).

Figure 2

Fig. 3 Predictions of probability of encountering incursions within Cocos Island National Park (Fig. 1). Predictors are year-quarters (rows) and lunar-quarters (columns). Probabilities are given on the vertical axis.

Figure 3

Table 1 Analysis of deviance for the binary logistic regression. Year and lunar quarters and their interactions were tested as predictors of illegal incursions into Cocos Island National Park (Fig. 1).

Figure 4

Fig. 4 Temporal distribution of incursions and patrol effort in Cocos Island National Park. (a) Incursions by month; (b) patrol hours by month; (c) incursions by lunar cycle; (d) patrol hours by lunar cycle. Bars indicate frequencies. Labels on vertical lines represent number of recorded incursions or hours of enforcement. The radial line on each graph shows the location of the mean value. The length of the mean vector (r), a measure of variance (range 0–1), is given in the top right of each graph; larger values indicate that observations are grouped closer to the mean.

Figure 5

Table 2 Cumulative probability of illegal fishers in Cocos Island National Park (Fig. 1) being penalized, given various hypothetical probabilities for each of the four links of the enforcement chain.