Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: The Galapagos Marine Reserve

Mauricio Castrejón; Anthony Charles

doi:10.1371/journal.pone.0228094

. 2020 Jan 23;15(1):e0228094. doi: 10.1371/journal.pone.0228094

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: The Galapagos Marine Reserve

Mauricio Castrejón ^1,^¤,^*, Anthony Charles ²

Editor: Heather M Patterson³

PMCID: PMC6977758 PMID: 31971982

Abstract

Assessments of the effectiveness of marine protected areas (MPAs) usually assume that fishing patterns change exclusively due to the implementation of an MPA. This assumption increases the risk of erroneous conclusions in assessing marine zoning, and consequently counter-productive management actions. Accordingly, it is important to understand how fishers respond to a combination of the implementation of no-take zones, and various climatic and human drivers of change. Those adaptive responses could influence the interpretation of assessment of no-take zone effectiveness, yet few studies have examined these aspects. Indeed, such analysis is often unfeasible in developing countries, due to the dominance of data-poor fisheries, which precludes full examination of the social-ecological outcomes of MPAs. In the Galapagos Marine Reserve (Ecuador), however, the availability of long-term spatially explicit fishery monitoring data (1997–2011) for the spiny lobster fishery allows such an analysis. Accordingly, we evaluated how the spatiotemporal allocation of fishing effort in this multiple-use MPA was affected by the interaction of diverse climatic and human drivers, before and after implementation of no-take zones. Geographic information system modelling techniques were used in combination with boosted regression models to identify how these drivers influenced fishers’ behavior. Our results show that the boom-and-bust exploitation of the sea cucumber fishery and the global financial crisis 2007–09, rather than no-take zone implementation, were the most important drivers affecting the distribution of fishing effort across the archipelago. Both drivers triggered substantial macro-scale changes in fishing effort dynamics, which in turn altered the micro-scale dynamics of fishing patterns. Fishers’ adaptive responses were identified, and their management implications analyzed. This leads to recommendations for more effective marine and fishery management in the Galapagos, based on improved assessment of the effectiveness of no-take zones.

Introduction

There is a growing recognition worldwide that marine protected areas (MPAs), in combination with co-management regimes and the allocation of spatially-exclusive fishing rights to local fishing communities, can be an effective solution for rebuilding depleted marine populations and conserving key biodiversity areas [1–4]. This trend has encouraged an increasing number of governments to adopt this spatially-explicit management tool to promote the recovery of small-scale fisheries and conserve marine biodiversity [5,6]. Unfortunately, typically very limited or no human and economic resources are allocated to monitor the performance of MPAs, notably in developing countries [7]. Consequently, in some regions, such as Latin America and the Caribbean, there are few empirical examples that demonstrate the long-term socio-ecological outcomes generated by the adoption of MPAs [7,8], particularly those designed for multiple use (i.e., those MPAs that allow extractive use in a regulated way, generally under marine zoning schemes, which may include no-take zones). On the other hand, even in those cases in which an assessment of the performance of MPAs can be carried out, research is usually biased to the bio-ecological aspects of the coastal social-ecological systems, leading to poor understanding of human (social, economic, cultural and institutional) dimensions that influence the effectiveness of MPAs to accomplish conservation and fishery management objectives [9–11].

Major areas of analysis about human dimensions of MPAs include how fishers deal with their displacement from traditional fishing grounds as a result of MPA implementation, and the management implications associated with adaptation of their fishing patterns (i.e., variations in selection of fishing grounds, fishing methods, target species, organizational and marketing processes) [12–14]. There are also analyses of how, in many cases, fishing effort tends to aggregate around MPA boundaries, an effect known as “fishing the line”, indicating that MPA location is of interest to fishers either because they have traditionally fished around those areas or because a spillover of adults, or larval export, to nearby fished areas has occurred [15,16].

While the need to understand the implications of MPAs for fisheries is clear [17,18], it is important to recognize that other factors affect fishing patterns in addition to MPAs. Recent studies suggest that spatiotemporal allocation of fishing effort is not only influenced by the location of MPAs, but also by factors such as distance of fishing grounds to the nearest port, weather and oceanographic conditions, habitat features, fishing method employed, travel costs, product price and expected revenues [14,19,20]. Nevertheless, to our knowledge, no study has yet examined, in a quantitative way, how fishers respond to those situations in which they have to cope simultaneously with implementation of a multiple-use MPA and with diverse climatic and human drivers of change–usually ignored or neglected in MPA management effectiveness assessments–such as extreme climatic events (e.g., El Niño), the globalization of markets, and the boom-and-bust exploitation of alternative fisheries.

Each driver of change can produce “cascade effects” on the socioeconomic dynamics of fishing communities, whether through changes in the availability and accessibility of target species or variations in environmental and market conditions. This leads fishers to adapt their fishing patterns to prevent or mitigate the damage to their livelihoods [12]. If the main reasons behind these adaptations are not well understood, a bias in the interpretation of the observed patterns could be produced, leading to errors in planning, implementing and assessing MPAs. This is relevant for fisheries management, particularly in those cases in which the main assumption in assessing the effectiveness of an MPA is that any adaptation in fishing patterns was caused exclusively by the MPA, rather than by the combined impact of different human and climatic drivers of change.

The Galapagos Marine Reserve (GMR) represents a unique biodiversity and climate change hotspot in Latin America and the Caribbean, which provides an excellent case study to illustrate how the interaction of various large-scale human and climatic drivers around a network of no-take zones influences fishing patterns. In this multiple-use MPA, marine zoning was implemented between 2000 and 2006, in combination with a co-management regime and the allocation of exclusive fishing rights to local small-scale fishers, to mitigate the impacts of human activities on sensitive ecological areas and to ensure the sustainability of Galapagos small-scale fisheries [21,22]. However, decisions to locate no-take zones in areas of low abundance of the most lucrative fishery resources, in combination with a lack of effective enforcement and a high rate of non-compliance [23], severely limited the effectiveness of Galapagos marine zoning for shellfish fisheries management purposes [2]. Despite these shortcomings, spiny lobster (Panulirus penicillatus and P. gracilis) stocks showed an unexpected and remarkable recovery after a period of overexploitation [24]. Previous studies suggest that the Galapagos spiny lobster fishery recovery was caused by the combined effect of market forces and favorable environmental conditions, rather than no-take zone implementation [2,25]. However, this hypothesis has not been tested yet by a long-term impact assessment of the effectiveness of no-take zones. This type of assessment could be influenced by fishers’ adaptive responses, so a proper assessment can only be made given a better understanding of how local fishing communities coped with the interactions of human and climatic drivers, before and after marine zoning implementation.

Using geographic information system (GIS) modelling techniques, in combination with boosted regression models, this paper evaluates how the spatiotemporal allocation of fishing effort in the Galapagos spiny lobster fishery was affected by the interactions of human and climatic drivers over a 15-year period (1997–2011). Based on the analysis of changes in fishing patterns, we build an understanding of the main drivers and factors influencing fishers’ adaptive responses to drivers of change, before and after implementation of marine zoning, including their links to the geographic and socioeconomic features of fishing communities, and their implications for fisheries management. We integrated this knowledge to provide a series of recommendations to improve the design and effectiveness of Galapagos marine zoning to reconcile conservation and fishery management objectives.

Materials and methods

Study area

The Galapagos Islands is comprised of approximately 234 islands, islets and rocks with a total land area and coastline of ca. 7 985 km² and 1667 km, respectively [26]. According to Edgar et al. [27], this volcanic archipelago is divided into five marine biogeographical regions, named as far-Northern, Northern, South-Eastern, Western and Elizabeth (Fig 1). Each one shows particular assemblages of fish and macro-invertebrate species, whose abundance and distribution are strongly affected by the El Niño Southern Oscillation [25,28].

Fig 1 — Red circles indicate the location of the three main fishing ports: Puerto Villamil (PV), Puerto Ayora (PA) and Baquerizo Moreno (BM). Black areas indicate the location of no-take zones.

Only 4% of the total land area is inhabited by ca. 25,144 residents (Table 1) distributed on five islands (Santa Cruz, Baltra, San Cristobal, Isabela, and Floreana). The remaining land area is protected as a national park. There are three main fishing ports (Baquerizo Moreno, Puerto Ayora and Villamil; Fig 1) that display specific geographic and socioeconomic features, particularly in terms of population density, number of fishers, composition of the fishing fleet, and available land-based tourism infrastructure (Table 1). There are 1084 license holders and 416 vessels registered in Galapagos, although only 37% of them remain active in the spiny lobster fishery (Table 1). Each fishing license provides its owner the right to fish any type of shellfish and finfish species commercially permitted. Approximately 97% of active vessels are smaller than 9.6 m long (fiber glass or wooden made) and equipped with outboard engines (15–200 HP). Only 13% consist of large wooden boats (8 to 18 m long) equipped with inboards engines (30–210 HP). These “mother boats” are used as storage, resting and towing units for up to four small vessels [29]. Most harvesting activities usually last one or two days, although mother boats are able to operate for a maximum of 12 days.

Table 1. General features of the three main fishing ports of the Galapagos Islands, including a summary of the fishery information analyzed in this study, for each sampling method used.

	San Cristobal	Santa Cruz	Isabela	Total
Fishing port	Baquerizo Moreno	Ayora	Villamil	3
Main landing sites	1	2	1	4
Population¹	7495	15393	2256	25144
Coastline (km)²	~ 156	~ 170	~ 617	~ 944
Hotel capacity (beds)³	449	990	193	1632
Restaurants and bars³	35	61	18	114
License holders (active/registered)⁴	174/552	136/293	100/239	410/1084
Small vessels (active/registered)⁴	59/163	44/87	48/107	151/357
Mother boats (active/registered)⁴	2/32	1/19	1/8	4/59
Cooperatives	2	1	1	4
Interview based data⁵ (1997–2011)	4387	4246	6727	15360
Fishery observer based data⁵ (2000–2006)	1058	719	586	2363

Open in a new tab

¹ Galapagos census 2010 conducted by INEC.

² PNG [30].

³ Epler [31].

⁴ Reyes and Ramírez [32].

⁵ Participatory Programme of Fisheries Monitoring and Research; no data were collected in 2007.

The most valuable shellfish species in Galapagos are the red and green spiny lobsters (P. penicillatus and P. gracilis), and the sea cucumber Isostichopus fuscus, harvested exclusively by artisanal hookah and skin divers mostly in sub-tidal rocky habitats. The fishing season since 1999 for sea cucumbers usually lasts from June to August (two months) and for spiny lobsters from September to December (four months), although slight variations have occurred through the years. The number of landing sites along the coast is quite limited (Table 1), facilitating the systematic and reliable collection of fishery-related data at each port since 1997.

In March 1998, the Galapagos archipelago and its surrounding open ocean were enclosed in a multiple-use MPA of nearly 138,000 km², the GMR, through the enacting of the Galapagos Special Law [22,33]. This law decreed an institutional shift from a hierarchical (top-down) to a co-governance mode, and from an open access to a common property regime. Since then, large-scale fishing was prohibited inside the reserve and fishing licenses and permits were exclusively allocated to local small-scale fishers.

The Galapagos’ marine zoning was created and implemented between 2000 and 2006 [22]. It comprised 76 no-take zones distributed across the archipelago, covering 17% of the coastline (Fig 1). The dimensions of the zones range from small offshore islets to 22.8 km of coastline [34]. The total area per management zone is unknown, as the offshore boundaries were not legally established [22]. Fishing and tourist activities, such as snorkelling and scuba diving, are prohibited inside 14 no-take zones, known as “conservation zones”. In the remaining 62 no-take zones, known as “tourism zones”, only tourist activities are permitted. No buffer zones were established. Therefore, in some regions, conservation and tourism zones are contiguous, constituting “no-take networks” (i.e., interconnected groups of individual no-take zones). The largest ones are distributed in Fernandina, Santiago, Santa Cruz and Floreana islands (Fig 1).

Before and after creation of the marine zoning, the spiny lobster fishery was impacted by diverse climatic and anthropogenic drivers of change, most of which acted simultaneously on various spatio-temporal scales. The most relevant are indicated in Table 2, which shows the category of the driver, according to the classification defined by Hall [35], the specific form of the change, and the corresponding time scale.

Table 2. Main climatic and human drivers that potentially affected the spiny lobster fishery from the Galapagos Islands between 1997 and 2011.

Category	Drivers of change	Temporal scale
Climate and environment	El Niño 1997/1998	April 1997-June 1998 (~14 months)
Governance	Co-governance and common property period	March 1998-onwards
International trade and globalization of markets	Boom and bust exploitation of the sea cucumber fishery by roving bandits	April 1999- (decades)
Governance	Marine zoning	April 2000-onwards (decades)
International trade and globalization of markets	Global financial crisis	December 2007- June 2009 (~18 months)

Open in a new tab

Fishing effort data

Fishing effort data for the period 1997–2011 were gathered from the Participatory Programme of Fisheries Monitoring and Research (PIMPP, in Spanish). Before this period, there are no spatially-explicit fishing effort data available for analysis. Only annual aggregated fishing effort data (in fishing days) per island for the period 1974–1979, published by Reck [36], remain for comparative purposes.

Fishery-related data were collected from 17,764 fishing trips, equivalent to 20,203 fishing effort records per fishing ground, either by interviewers (1997–2011) or observers onboard (2000–2006), at the three main ports of Galapagos (Puerto Ayora, Baquerizo Moreno, and Villamil) on a daily basis over each fishing season (Table 1 and S1 Table). Geographical positioning systems were usually used by observers to collect spatially-explicit data onboard fishing vessels, including position of fishing grounds, fishing method, effective fishing hours, number of divers, vessel type and name, departure and landing port, departure and arrival date, and catch per spiny lobster species.

The same types of data were collected by interviewers using semi-structured questionnaires. However, in this case, fishing grounds’ names visited per fishing trip were obtained instead of the exact geographical location where fishing activity took place. To make this subset of data spatially explicit, we added the geographic coordinates published by Chasiluisa and Banks [37], who defined reference positions (latitude and longitude) for 320 fishing grounds identified and distributed across the archipelago.

The PIMPP dataset was extensively reviewed, standardized and cleaned before being filtered. Data collected by interviewers and observers onboard were included in this study. However, analyses were restricted to those spiny lobster fishing trips conducted in small vessels (fiber glass and wooden made) by one or two hooka divers, where the number of effective fishing days, at a single fishing ground, ranged from one to seven. The final dataset accounts for 17,723 fishing effort data units (Table 1), representing 88% of the original dataset. Approximately 78% of these data are georeferenced, i.e., they include the exact, or reference, position of each fishing ground visited per fishing trip. The remaining 22% of the data simply specify the islands in which fishing activity took place.

To determine the representativeness of the data selected, for each sampling method used, we estimated the sampling effort of small fishing vessels and fishing effort, measured in diver-hours, registered by the PIMPP between 1997 and 2011 (S1 Table). According to our results, sampling effort by interviews was on average 67% and 37% for small fishing vessels and fishing effort, respectively. In contrast, sampling effort by observers on board was 19% and 5% for small fishing vessels and fishing effort, respectively (S1 Table). These results suggest that interview-based data are more representative of the spatiotemporal dynamics of the fishing fleet than data collected by observers onboard. However, they could also be less reliable, if fishers provided inaccurate information about the locations of their fishing grounds. To account for this type of uncertainty, both data sources were analyzed in most cases separately to compare the fishing patterns identified.

Data analysis

Fishing effort data were grouped into a suitable number of time periods to evaluate how the spatio-temporal dynamics of fishing patterns in the spiny lobster fishery were affected by the potential drivers of change described in Table 2. However, as the temporal scale of each driver is different, and most of them occurred simultaneously, it was not feasible to divide the data available evenly between periods. Accordingly, we defined six time periods (Table 3), based on the following logic:

Table 3. Periods defined to evaluate the spatio-temporal dynamic of fishing patterns in the spiny lobster fishery, based on the most relevant climatic and human drivers occurring between 1997 and 2011.

Period	Acronym	Temporal scale
1. Co-governance and El Niño	CoM-EN	June 1997-December 1998
2. Sea cucumber re-opening phase	RovBan1	September 1999- December 2000
3. Sea cucumber expansion phase	RovBan2	September 2001- December-2002
4. Sea cucumber overexploitation phase and marine zoning	MarZon	September 2003-December 2005
5. Sea cucumber collapse phase and global financial crisis	Crisis	September-December 2006 and September-December 2008
6. Spiny lobster recovery	Recovery	September-December 2011

Open in a new tab

We subdivided the boom-and-bust exploitation period of the sea cucumber fishery into four phases (re-opening, expansion, overexploitation and collapse) to evaluate their specific impact on the spatial allocation of fishing effort in the spiny lobster fishery. The sea cucumber overexploitation phase and the marine zoning implementation were grouped together because both occurred simultaneously. In this case, we assumed that marine zoning was implemented after a transition period of three years (2000–2003) once the moratorium on the entry of new fishers was put in place, fishing regulations were decreed, enforcement capacity increased, and fishers were aware of zoning boundaries and legal framework.
We grouped the sea cucumber collapse phase and the global financial crisis 2007–09 together because both drivers affected the profitability of fishing activity [2,38], allowing us to evaluate their combined impact on fisher’s behaviour. Data from 2009 and 2010 were excluded from this period due to a lack of georeferenced data.
The unexpected and remarkable recovery of the spiny lobster fishery was considered an additional period. Such an event does not represent a driver itself, at least not in the short term, but a social-ecological impact potentially caused by climatic and human drivers that occurred in earlier periods.

Interaction between sea cucumber and spiny lobster fisheries

We performed a Pearson’s correlation analysis to evaluate how the fishing effort capacity in the spiny lobster fishery was affected by the different phases (re-opening, expansion, overexploitation and collapse) of the boom-and-bust exploitation of the sea cucumber fishery, using as variables the number of active fishers, small-vessels and mother boats in both fisheries between 1997 and 2011. The information was obtained from PIMPP, Galapagos National Park Service fishing registry, Moreno et al. [39] and Reyes and Ramírez [32]. The correlation analysis was conducted in the R statistical programming language, version 3.1.2 (R Development Core Team 2014).

Spatiotemporal analysis of fishing patterns

The spatiotemporal allocation of fishing effort across the archipelago was evaluated using GIS modelling techniques with ArcGIS 10.2.2 (ESRI) software. We calculated standard deviation ellipses (SDE) polygons by point pattern statistics [40] to determine the core areas and distribution ranges of the fishing fleets based in the three ports (Baquerizo Moreno, Puerto Ayora and Villamil) during the six periods defined.

In this study, SDE represent graphical summaries of the central tendency, dispersion and directional trends of fishing fleets. Core areas and distribution ranges refer to those areas covering 68% (1 SDE) and 95% (2 SDE) of the full spatial extent of fishing fleet distribution, respectively. Furthermore, to determine if the same areas have been reused by fishers from different ports at different periods, we estimate an index of reuse (IOR), following the procedure described by Morrisey and Gruber [41] and Horta e Costa et al. [42]. Small vessels core areas and distribution ranges were used to estimate IOR, by the following formula [41]:

I O R = \frac{O V (A_{1} + A_{2})}{A_{1} + A_{2}}

where [OV (A₁+A₂)] refers to the overlapping area between two core areas (or distribution ranges), and (A₁+A₂) to the total area of both core areas (or distribution ranges). IOR values range from 0 (both areas do not overlap) to 1 (both areas overlap completely). One and two-way ANOVAs were employed to test the null hypothesis of absence of differences in core areas, distribution range and IOR between different periods, and between ports and sampling methods (interviews vs fishery observers). A Bartlett’s test was performed prior to all analyses to test the assumption of homogeneity of variances among treatments. When data were heteroscedastic, or did not fulfill the normality assumption, transformations were conducted.

We also performed a hotspot analysis using area pattern statistics [40] to evaluate if the areas where most fishing effort is concentrated (i.e., hotspots) have varied across each period and to determine if the fishing patterns identified vary according to the sampling method employed. Based on this analysis, we determined the spatial distribution of hotspots before and after marine zoning implementation, allowing us to evaluate if fishers were displaced from their traditional fishing grounds and if fishing effort concentrates around no-take zones, producing a “fishing the line” effect.

We aggregated fishing effort data per period and sampling method and performed a single hotspot analysis for each possible combination (nine in total). The following procedure was applied to each combination:

A grid with a 2.25 km² cell size was superimposed over the entire archipelago. Such resolution was selected considering the size of the study area, as well as the precision and resolution required to evaluate the fine-scale distribution of fishing fleets;
a buffer of 2.5 km was delimited around the coastline of each island, islet and rock, defined based on the dispersion of data and a maximum bathymetry of 40 m, so as to contain the area where the spiny lobster fishery takes place. Grid cells located outside this buffer, including the land area, were removed; then, we proceeded to eliminate those resulting grid cells smaller than 3% of the original grid cell size;
total fishing effort (diver-hours) per grid cell was summarized and a measure of effort density (diver-hours km^-2) was calculated by dividing the total sum of fishing effort per cell by the original grid cell size (2.25 km²);
a spatial weights matrix was generated using the k-nearest neighbors (k = 8) as the conceptualization of the spatial relationship among data (i.e., small-vessels). The latter method was selected considering the extensive and uneven spatial distribution of our data across the study area and the skewed distribution of fishing effort values;
finally, a hotspot analysis was performed, using effort density as the input field. Such analysis identified statistically significant spatial clusters of high effort density values (hot spots) and low effort density values (cold spots) across the archipelago, based on the Getis/Ord Gi* statistic [43], producing a Z-score and p-values as measures of statistical significance.

The null hypothesis in this case is that the spatial allocation of fishing effort is the result of random spatial processes, which is rejected if the Z-score ≥ 1.96 and p < 0.05 with a 95% confidence level. A high Z-score and small p-value indicates a hotspot, while a low negative z-score and small p-value indicates a cold spot. The higher (or lower) the Z-score, the more intense the clustering, while a Z-score near zero indicates no apparent spatial clustering [44].

Climatic and human drivers affecting spatial fishing effort allocation

We defined, for the set of fishing georeferenced data, a diverse suite of explanatory variables potentially having an influence on the spatial allocation of fishing effort, measured in diver-hours. Geographic, oceanographic and socioeconomic variables were selected based on the human and climatic drivers identified as relevant for this study. Each was categorized as either temporally static or temporally dynamic, based on whether it changes over time [19].

The first category includes latitude, longitude, bioregion, homeport, distance to home port (DistHP), and distance to the nearest no-take zone (NearNTZ). Here, we defined homeport as the port from which a vessel primarily operates, regardless of its registry. To calculate the shortest effective distance between each fishing record and the corresponding vessel’s home port, we conducted a cost-distance analysis using the spatial analyst extension in ArcGIS 10.2.2. The same analysis was used to calculate the shortest effective distance between each fishing record and the nearest no-take zone.

The second category includes historic period (Period), month, average ex-vessel price per year (ExVesPrice), lobster catch obtained in previous fishing trips (PrevCatch), average sea cucumber revenues obtained the fishing season before the beginning of lobster season (SeaCucRev), vessel type, and the Oceanic Niño Index (ONI). The ONI represents the month moving average of ERSST.v3b SST anomalies in the Niño 3.4 region (i.e., west of the GMR 5°N– 5°S, 120°-170°W), based on centered 30-year base periods updated every 5 years. ONI is the main indicator used by NOAA for monitoring El Niño and La Niña, which are opposite phases of the climate pattern called the El Niño-Southern Oscillation. Data were obtained from the NOAA Climate Prediction Centre at www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt

Analysis of fishing effort hotspots and fishing fleet distribution ranges and core areas were used to evaluate how spatiotemporal distribution changes in relation to the external drivers analyzed in this study. Then, boosted regression trees (BRTs) were used to identify the factors that explain such spatiotemporal patterns. The goal was to predict fishing effort, measured in diver-hours per port and period, as a function of geographic, oceanographic and socioeconomic variables described above.

BRT models can be defined as flexible additive regression models in which individual terms are simple trees, created by recursive binary splits constructed from predictor variables and combined to optimize predictive performance, which are fitted in a forward, stagewise fashion [45,46]. Unlike general linear models and general additive models, BRT models accommodate missing values in continuous or categorical predictors, are able to handle outliers, collinear variables, interactions between variables, and nonlinear relationships between predictor and response variables, showing additionally similar, or even stronger, predictive performance [19,46,47].

BRT model fitting requires the definition of three parameters: (1) learning rate (lr), also known as the shrinking parameter, which determines the contribution of each tree to the growing model (i.e., controlling the rate at which the model converges on a solution); (2) tree complexity (tc), which refers to the number of nodes (or splits) in a tree (i.e., the ability of model interactions); and (3) the two previous parameters are used to estimate the optimal number of trees (nt) required to increase performance prediction. In addition, to improve accuracy and reduce overfitting, we introduced stochasticity to the BRT model through a “bag fraction”, which specifies the proportion of data to be selected at each step [46]. The BRT model was fit to allow interactions using a tree complexity of 2 and a learning rate of 0.005 and a bag fraction of 0.6. Ten-fold, cross-validation of training data was used to determine the optimal number of trees necessary to minimize deviance and maximize predictive performance to independent test data. Model performance was assessed based on predictions made using the independent testing set that was withheld during cross-validation.

Deviance explained and Pearson’s correlation coefficient (r) were used to assess the predictive performance of BRT models. Furthermore, variable importance (VI) was estimated by averaging the number of times a variable is selected for splitting and the squared improvement resulting from these splits [48,49]. VI scores provide a measure of the relative influence of predictor variables used to build the model [19]. Values are scaled so that the sum adds to 100, with higher numbers indicating a stronger influence on the response variable. Following Soykan et al. [19], a random number (RN) between 1 and 100 was added to identify useful variables for modeling a response. Useful variables in predicting fishing effort were those that had higher VI scores than RN. Finally, for interpreting BRT models results, we generated a partial dependence plot for each predictor variable. Such graphs show the effect of a variable on the response after accounting for the average effects of all other variables in the model, including the RN [19]. BRT model fitting was conducted in the R statistical programming language, version 3.1.2 (R Development Core Team 2014) using the “gbm” and “dismo” libraries complemented with the brt.functions code developed by Elith et al. [46].

Results

Interaction between sea cucumber and spiny lobster fisheries

The analysis of active fishing capacity from 1999 to 2011 showed that large-scale changes in fishing effort dynamics for the spiny lobster fishery occurred during the boom-and-bust exploitation of the sea cucumber fishery and the global financial crisis 2007–09. The active number of fishers, small-vessels and mother boats increased in the spiny lobster fishery, reaching a maximum value during the re-opening and expansion phase of the sea cucumber fishery, corresponding to the RovBan1 and RovBan2 periods (Fig 2A–2C). A similar pattern was observed in the sea cucumber fishery, although in this case the number of fishers and small-vessels reached a maximum during the overexploitation phase of the sea cucumber fishery and marine zoning implementation, corresponding to the MarZon period (Fig 2A–2C). During this period the active fishing capacity in both fisheries showed a strong positive linear trend in Puerto Ayora, Baquerizo Moreno and Puerto Villamil (Fig 2D–2F).

Fig 2 — a) active fishers per year; b) active small-scale vessels per year; c) active mother boats per year; d) relationship between active lobster and sea cucumber fishers per port; e) relationship between active lobster and sea cucumber small-vessels per port; and f) relationship between active lobster and sea cucumber mother boats per port. BM: Baquerizo Moreno; PA: Puerto Ayora; PV: Puerto Villamil; CoM-EN: Co-governance and El Niño; RovBan1: Sea cucumber re-opening phase; RovBan2: Sea cucumber expansion phase; MarZon: Sea cucumber overexploitation phase and marine zoning; Crisis: Sea cucumber collapse phase and global financial crisis; Recovery: Spiny lobster recovery.The sea cucumber fishery was closed five years. In 2001, there was an unsuccessful attempt to implement an individual quota system in the sea cucumber fishery, which led to a temporal reduction in fishing effort [50,51].

The active fishing capacity in the spiny lobster fishery decreased gradually since 2000 (RovBan1). This trend intensified after the total closure of the sea cucumber fishery occurred in 2006 and during the global financial crisis 2007–09 (Fig 2A–2C). Between 2000 and 2010, the number of fishers decreased 80.3%, from 1183 to 233, while the number of small vessels and mother boats decreased 55.3%, from 286 to 128 (Fig 2A and 2B). However, the most remarkable decrease was observed in the number of mother boats, which decreased 88.0%, from 42 to 5 during the same period (Fig 2C). A similar decreasing trend in active fishing capacity was observed in the sea cucumber fishery, which was reopened in three occasions after 2006 (Fig 2A–2C). The total number of fishers and small vessels in both fisheries increased slightly during the recovery period of the spiny lobster fishery (Fig 2A and 2B), while the number of mother boats remained at very low numbers (Fig 2C). These results suggest that a significant number of fishers from Puerto Ayora, Baquerizo Moreno and Puerto Villamil responded to the economic perturbations caused by the collapse of the sea cucumber fishery and the global financial crisis 2007–09, by abandoning the spiny lobster and the sea cucumber fisheries. This macro-scale change in fishing capacity influenced the micro-scale spatiotemporal dynamic of fishing patterns in the spiny lobster fishery, as described in the next section.

Spatiotemporal analysis of fishing patterns

The spatiotemporal allocation of fishing effort showed different patterns between ports and periods. These did not vary, in most analyses, according to the sampling method used (Figs 3 and 4). According to port-based interviews, Puerto Ayora and Baquerizo Moreno’s fishing fleets showed larger core areas and distribution ranges than Puerto Villamil (S2 Table; Fig 3). However, such differences were not significant between ports (core areas: H = 2.667; d.f. = 2; p = 0.264; distribution ranges: H = 1.906; d.f. = 2; p = 0.385). Similar results were shown by observer onboard data (core areas: H = 5.955; d.f. = 2; p = 0.051; distribution ranges: H = 5.067; d.f. = 2; p = 0.079, although in this case Baquerizo Moreno’s fishing fleet showed larger core areas and distribution ranges than Puerto Ayora (S2 Table; Fig 4).

Fig 3 — Co-governance and El Niño (CoM-EN); Sea cucumber re-opening phase (RovBan1); Sea cucumber expansion phase (RovBan2); Sea cucumber overexploitation phase and marine zoning (MarZon); Sea cucumber collapse phase and global financial crisis (Crisis); and Spiny lobster recovery (Recovery).

Fig 4 — Results are shown for three time periods: Sea cucumber expansion phase (RovBan2); Sea cucumber overexploitation phase and marine zoning (MarZon); Sea cucumber collapse phase and global financial crisis (Crisis).

Both sampling methods showed that Galapagos fishing fleets had, on average, low degrees of overlap in their fishing activity spaces, particularly in relation to their core areas, as it was denoted by IOR values close to zero, meaning that fishing spaces between fishing fleets do not overlap (S3 Table). Puerto Ayora and Baquerizo Moreno’s fishing fleets had the highest similarity of core areas and distribution ranges in comparison with any other combination of ports (S3 Table, Figs 3 and 4). In contrast, Baquerizo Moreno and Puerto Villamil’s fishing fleets showed the lowest degree of overlapping of their core areas and distribution ranges (S3 Table, Figs 3 and 4). Nevertheless, both sampling methods showed that the central tendency, dispersion and directional trends of the Baquerizo Moreno, Puerto Ayora and Puerto Villamil fishing fleets’ core areas and distribution ranges showed large variations among periods, changing from no overlap to large overlap (Figs 3 and 4).

According to port-based interviews, Baquerizo Moreno fishing fleet’s core area and distribution range were located exclusively around San Cristobal Island from 1997 to 2005, corresponding to CoM-EN, RovBan1, RovBan2 and MarZon periods (Fig 3). However, the core area expanded towards Santa Fe, western and southern parts of Santa Cruz and Santiago Islands during the Crisis and Recovery periods, overlapping with Puerto Ayora’s fishing fleet core area and distribution range (Fig 3). Baquerizo Moreno’s fishing fleet distribution range showed a similar but larger expansion pattern toward Española and the eastern part of Isabela Island, reaching the western part of Floreana during the recovery of the spiny lobster fishery (Fig 3). Observer onboard data showed a similar pattern, although in this case Baquerizo Moreno fishing fleet’s core area and distribution range showed a larger expansion during the RovBan2 period (Fig 4). However, both sampling methods showed that spiny lobster fishing grounds located along San Cristobal Island are used exclusively by fishers from Baquerizo Moreno, although Puerto Ayora’s fishing fleet did expand temporally part of its distribution range toward San Cristobal during RovBan1 and RovBan2 periods (Fig 3).

Puerto Ayora fishing fleet registered also a large variation in its spatiotemporal distribution between 1997 and 2011 (Figs 3 and 4). According to port-based interviews, the core area and distribution range of this fishing fleet covered exclusively Santa Cruz and Santa Fe Islands during the CoM-EN period, showing no overlapping positions with Puerto Villamil and Baquerizo Moreno (Fig 3). However, Puerto Ayora’s core area expanded to Santiago and the west and eastern parts of Isabela Island, while the distribution range extended practically to the entire archipelago during the RovBan1 and RovBan2 periods (Fig 3). A similar pattern was shown by observer onboard data, although in this case the core area extended until the far-northern islands of Darwin and Wolf during the RovBan2 period (Fig 4). In contrast, the core area and distribution range contracted remarkably during MarZon and Crisis periods, until reaching similar dimensions to those observed during the CoM-EN period (Figs 3 and 4). However, core area and distribution range re-expanded again during the Recovery period, until reaching dimensions similar to those observed during the MarZon period (Fig 3).

Unlike Puerto Ayora and Baquerizo Moreno, Puerto Villamil´s fishing fleet showed minimum variations of its core area and distribution range through the years, denoting a remarkable fidelity of Puerto Villamil’s fishers to their traditional fishing grounds (Figs 3 and 4). Our results showed that spiny lobster fishing grounds located in the southern and western part of Isabela Island were used exclusively by fishers from Puerto Villamil. The only exception occurred during the reopening phase of the sea cucumber fishery (RovBan1), when Puerto Villamil and Puerto Ayora fishing fleets’ core areas slightly overlapped (Figs 3 and 4). Puerto Villamil fishing fleet’s distribution range showed a consecutive expansion and contraction pattern, similar to that described for Puerto Ayora during the same periods. However, unlike Puerto Ayora’s fishers, who expanded their distribution range beyond Santa Cruz Island, Puerto Villamil’s fishers remained fishing exclusively along the coastline of their home island, Isabela (Figs 3 and 4).

Hotspot analysis revealed the existence of significant fishing clusters across the archipelago and throughout the six periods analyzed (Figs 5 and 6). One of the the most relevant patterns identified by this analysis was that aggregation of fishing effort was not detected around the boundaries of no-take zones (Figs 5 and 6). Hotspots of fishing effort did not show large variations before and after the implementation of no-take zones (MarZon), suggesting that fishers were not displaced from their traditional fishing grounds nor attracted to no-take zone boundaries by a “fishing the line” effect.

Fig 5 — Six-time periods are shown: Co-governance and El Niño (CoM-EN); Sea cucumber re-opening phase (RovBan1); Sea cucumber expansion phase (RovBan2); Sea cucumber overexploitation phase and marine zoning (MarZon); Sea cucumber collapse phase and global financial crisis (Crisis); and Spiny lobster recovery (Recovery).

According to port-based interviews, fishing effort showed high densities exclusively in the southern part of Isabela and San Cristobal Islands, near Baquerizo Moreno and Puerto Ayora, during CoM-En period (Fig 5). However, new hotspots appeared along the western and eastern parts of Santa Cruz Island, the western and southwestern parts of Isabela Island, the eastern part of San Cristobal Island and the southeastern part of Genovesa Island during RvBan1 and RVBan2 periods (Fig 5). Since then hotspot location patterns have shown few variations. Some hotspots disappeared sporadically during and after the sea cucumber overexploitation and marine zoning implementation phase, particularly those located in the western and southwestern part of Isabela Island (Fig 5). However, most hotspots have remained in the same locations through the years, particularly those located near fishing ports. Only during the spiny lobster recovery period, a single hotspot appeared in the southwestern part of Isabela Island, which had not been registered in previous periods, suggesting that fishing effort aggregated in new fishing grounds in 2011 (Fig 5).

Similar fishing patterns were identified by analysis of onboard observer data (Fig 6), although some slight variations were detected. According to this source of data, during the expansion phase of the sea cucumber fishery (RvBn2), there was a group of hotspots in the northwestern part of Marchena and Pinta Islands, as well as in the western and eastern part of Floreana Island, which were not detected by port-based interview data analysis. Despite these minor differences, the general fishing patterns identified by both sampling methods were quite similar (Figs 5 and 6).

Climatic and human drivers affecting spatial fishing effort allocation

The BRT analysis was carried out with a focus on evaluating the relevance of no-take zones as a fishing effort predictor. However, NearNTZ was a relevant explanatory variable only for Puerto Villamil, and even in this case, the ranking of NearNTZ was very low (Table 4). This result suggests little if any effect of no-take areas on fishing patterns, at least for two of the three communities analyzed (Puerto Ayora and Baquerizo Moreno).

Table 4. For each predictor variable, the variable importance score (summing to 100) and the ranking is shown, for regional results and for each port.

Variable	Regional	Ranking	Puerto Villamil	Ranking	Puerto Ayora	Ranking	Baquerizo Moreno	Ranking
DistHP	22.4	1	22.2	1	17.5	1	7.4	7
Latitude	9.0	4	10.5	3	15.7	3	8.4	6
Longitude	7.2	5	8.2	5	16.4	2	17.1	1
Bioregion	2.6	12	2.4	10	5.5	9	0.1	13
NearNTZ	4.7	9	6.5	7	4.7	10	4.0	10
Period	11.1	3	7.5	6	7.3	5	9.8	5
Month	1.9	13	1.3	12	2.1	11	5.7	9
ExVesPrice	4.1	10	2.0	11	6.2	7	2.3	11
Vessel	1.0	14	0.8	13	1.7	12	0.6	12
ONI	5.9	7	6.0	8	6.3	6	13.6	2
SeaCucRev	6.1	6	10.3	4	1.3	13	12.3	3
PrevCatch	15.2	2	18.3	2	9.6	4	11.7	4
HomePort	4.1	11	NA	NA	NA	NA	NA	NA
RN	4.8	8	4.0	9	5.8	8	7.0	8
Sum of static variables importance	50.0		49.8		59.8		37.0
Sum of dynamic variables importance	45.3		46.2		34.5		56.0
Deviance explained (%)	29.47		35.73		32.66		15.74
Pearson’s correlation coefficient (r)	0.55		0.59		0.54		0.44

Open in a new tab

Bold numbers: predictor performed better than random numbers (RN). Shown at the bottom of the table are summary values for regional and homeport BRT models, i.e. the sums of static and dynamic VI scores, the deviance explained, and the Pearson’s correlation coefficient. DistHP: distance to homeport; NearNTZ: distance to the nearest no-take zone; ExVesPrice: average ex-vessel price per year; PrevCatch: lobster catch obtained in previous fishing trip; SeaCucRev: average sea cucumber revenues obtained before the beginning of lobster fishing season season; ONI: Oceanic Niño Index; NA: No Applicable.

According to the regional BRT model, the most important fishing effort predictors were (Table 4): (1) DistHP, (2) PrevCatch, and (3) Period, followed by (4) latitude, (5) longitude, (6) SeaCucRev and (7) ONI. The remaining predictors were not useful in predicting fishing effort, as they performed worse than RN. Homeport BRT models showed, in most cases, similar patterns to the regional BRT model (Table 4). However, the contribution of fishing effort predictors varied among homeports, particularly in Baquerizo Moreno (Table 4). Based on the sum of the VI scores, static variables contributed very slightly more to the regional BRT model than dynamic predictor variables (Table 4, static: 50.0% dynamic: 45.3%). Likewise, the importance of static variables was higher than dynamic variables both for Puerto Villamil and Puerto Ayora BRT models (Table 4, PV: static (49.8%), dynamic (46.2%); PA: static (59.8%), dynamic (34.5%)). In contrast, dynamic predictor variables contributed more than static predictor variables in the BRT model for Baquerizo Moreno (Table 4, static: 37.0%; dynamic: 56.0%). According to the BRT model for Puerto Villamil, fishing effort was mostly influenced by eight variables, with DistHP, PrevCatch, latitude, and SeaCucRev, being the four most important predictors (Table 4).

The most important predictors contributing to the BRT model for Puerto Ayora were similar overall to Puerto Villamil, and DistHP was again the most important predictor. However, there were differences in other rankings (Table 4). Unlike Puerto Villamil, spiny lobster ex-vessel price performed better than RN in the BRT model for Puerto Ayora, while NearNTZ and SeaCucRev performed worse. The three most important predictors contributing to the BRT model for Baquerizo Moreno were, in contrast to Puerto Villamil and Puerto Ayora, longitude, ONI, and SeaCucRev (Table 4).

Performance statistics for the BRT models suggest that all models showed a good predictive performance to independent test data (Table 4). The regional BRT model explained 29.47% of the deviance in the data, while the Pearson’s correlation coefficient was 0.55 (Table 4). Homeport BRT models showed, in most cases, better predictive performance than the regional BRT model (Table 4). Specifically, the BRT models for Puerto Villamil, Puerto Ayora and Baquerizo Moreno explained 35.73%, 32.66% and 15.74% of deviance in the data, respectively, while their Pearson’s correlation coefficients were 0.59, 0.54 and 0.44, respectively.

Fishing effort was affected in different ways by the most influential variables identified by BRT models (Figs 7–10). Even though similar results were obtained by the regional and homeport BRT models, different patterns were observed among ports. All BRT models showed that fishing effort, measured as diver-hours, increased with distance to homeport, albeit with some variation (Figs 8–10). In Puerto Villamil, fishing effort increased gradually between 30 and 70 km from the homeport, leveling off subsequently (Fig 8). In contrast, for Puerto Ayora, fishing effort increased gradually between 20 and 150km, then levelled off until 320 km, reaching maximum values after this distance (Fig 9). In other words, fishers from Puerto Ayora tend to fish farther away from their homeport in comparison with their peers from Puerto Villamil. In Baquerizo Moreno, while distance to homeport performed better than RN, its importance as a fishing effort predictor was much lower in comparison with Puerto Ayora and Puerto Villamil (Table 4; Fig 10).

Fig 7 — The response variable (diver-hours) has been centered by subtracting its mean. Variable importance scores are shown in parentheses. Rug plots indicate the distribution of observations in relation to the predictor variable. CoM-EN: Co-governance and El Niño; RovBan1: Sea cucumber re-opening phase; RovBan2: Sea cucumber expansion phase; MarZon: Sea cucumber overexploitation phase and marine zoning; Crisis: Sea cucumber collapse phase and global financial crisis; Recovery: Spiny lobster recovery.

Fig 10 — The response variable (diver-hours) has been centered by subtracting its mean. Variable importance scores are shown in parentheses. Rug plots indicate the distribution of observations in relation to the predictor variable. CoM-EN: Co-governance and El Niño; RovBan1: Sea cucumber re-opening phase; RovBan2: Sea cucumber expansion phase; MarZon: Sea cucumber overexploitation phase and marine zoning; Crisis: Sea cucumber collapse phase and global financial crisis; Recovery: Spiny lobster recovery.

Fig 8 — The response variable (diver-hours) has been centered by subtracting its mean. Variable importance scores are shown in parentheses. Rug plots indicate the distribution of observations in relation to the predictor variable. CoM-EN: Co-governance and El Niño; RovBan1: Sea cucumber re-opening phase; RovBan2: Sea cucumber expansion phase; MarZon: Sea cucumber overexploitation phase and marine zoning; Crisis: Sea cucumber collapse phase and global financial crisis; Recovery: Spiny lobster recovery.

Fig 9 — The response variable (diver-hours) has been centered by subtracting its mean. Variable importance scores are shown in parentheses. Rug plots indicate the distribution of observations in relation to the predictor variable. CoM-EN: Co-governance and El Niño; RovBan1: Sea cucumber re-opening phase; RovBan2: Sea cucumber expansion phase; MarZon: Sea cucumber overexploitation phase and marine zoning; Crisis: Sea cucumber collapse phase and global financial crisis; Recovery: Spiny lobster recovery.

Fishing effort showed a non-monotonic relationship with the previous spiny lobster catch in all BRT models. At regional level, fishing effort increased with previous catch until the latter reached 60 kg tail/fishing trip, then decreased until previous catch was approximately 105 kg tail/fishing trip, increasing again to a second peak at 110 kg tail/fishing trip, and leveling off subsequently. A similar pattern was observed at port level. In Puerto Villamil and Baquerizo Moreno, fishing effort also increased with the previous spiny lobster catch, to 30 and 55 kg tail/fishing trip, respectively, decreasing afterwards (Figs 8 and 10). In both cases, fishing effort increased again up to a previous spiny lobster catch at 105 and 60 kg tail/fishing trip, respectively, and leveled off subsequently. In contrast, fishing effort in Puerto Ayora increased abruptly up to a previous spiny lobster catch between 40 and 60 kg tail/fishing trip, decreasing afterwards (Fig 9).

Fishing effort showed different patterns according to the time period. At regional level, fishing effort increased dramatically during the reopening of the sea cucumber fishery (RovBN1), then declined gradually until reaching a minimum value during the sea cucumber collapse phase and global financial crisis 2007–09, increasing afterwards until reaching a maximum during the spiny lobster recovery period (Fig 7). A similar pattern was observed in Puerto Villamil, although in this case fishing effort decreased to a minimum during the marine zoning (MarZon) period, increasing afterwards (Fig 8). In Puerto Ayora, fishing effort also showed a maximum peak during RovBN1 period. However, unlike Puerto Villamil and Baquerizo Moreno, fishing effort decreased gradually until reaching a minimum value during the spiny lobster recovery period (Fig 9). In Baquerizo Moreno, fishing effort showed maximum values between CoM-EN and RovBN2 periods, then decreased until reaching a minimum value during the sea cucumber collapse phase and global financial crisis 2007–09. Afterwards, fishing effort increased slightly (Fig 10).

Spatially, fishing effort showed different distribution patterns across the archipelago, according to each fishing fleets’ homeport. At regional level, fishing effort reached maximum values between 1.0° N and 1.5° N; i.e., around Darwin and Wolf islands, in the northernmost part of the Galapagos (Fig 7). In relation to longitude, fishing effort showed a decreasing trend from 91.3° W to 90.2° W; i.e., from the Eastern side of Fernandina toward Santiago Island; this was then followed by a steep increase from 90.0° W to 89.5° W; i.e., from the Western part of Santa Cruz Island toward Española and the Southwestern part of San Cristobal. Partial dependence plots for homeports showed that Puerto Ayora’s fishing fleet was the only one that fished around these two islands between 1997 and 2011 (Fig 9). In this case, fishing effort showed a decreasing trend latitudinally from 90.0° W to 92.0° W; i.e., from the Western of Santa Fe Island toward Santa Cruz, Isabela and Fernandina Islands. In contrast, fishing effort for Puerto Villamil showed two peaks at latitudes 1.1 S and 0.5 S, while longitude showed an increasing trend in fishing effort from 90.8° W to 91.2° W (Fig 8). These results suggest that fishing effort increased from Puerto Villamil toward the Southwestern part of Isabela Island, reaching a peak probably in the hotspot located in the South part of this island (Figs 5 and 6). Then fishing effort increased toward the North, reaching a peak probably in the hotspot located in the Western side of Isabela Island (Figs 5 and 6). Finally, fishing effort for Baquerizo Moreno showed a peak at -1.5°S, probably around Española Island, then increasing very slightly from South to North (Fig 10). These results suggest that fishing effort is largely influenced by the location of fishing ports, as it was also demonstrated by the strong performance of the DistHP variable as a fishing effort predictor (Table 4) and by the location of hotspots and fishing fleets’ core areas in fishing grounds adjacents to Puerto Villamil, Puerto Ayora and Baquerizo Moreno (Figs 3–6).

Most BRT models showed that fishing effort increased when sea cucumber revenues, obtained the fishing season before the beginning of lobster season (SeaCucRev), ranged between US$2000 and US$6,400/fishing season. At regional level, fishing effort decreased with higher sea cucumber revenues, including a steep decrease when revenues exceeded US$11000/fishing season (Fig 7). In Puerto Villamil, fishing effort increased with sea cucumber revenues up to the latter reaching US$2500/fishing season, showing a second and highest peak when revenues were higher than US$6400/fishing season (Fig 8). In Baquerizo Moreno, fishing effort showed a positive relationship with sea cucumber revenues only when they reached values between US$4000 and US$5000/fishing season (Fig 10). Before this threshold no relationship was found. In Puerto Ayora, this explanatory variable was not relevant (Fig 9).

Finally, at regional level, fishing effort increased when the oceanographic variable ONI ranged between -1.0 and 2.0. Similar patterns were observed among ports (Figs 8–10). In Puerto Villamil and Puerto Ayora, fishing effort showed an increasing trend from -1.0 to 1.5 (Figs 8 and 9), while in Baquerizo Moreno fishing effort increased gradually after 0, showing a peak around 2.0 (Fig 10). ONI values equal to or higher than +0.5 indicate El Niño conditions, meaning that the East-central tropical Pacific is significantly warmer than usual. In contrast, ONI values equal to or lower than -0.5 indicate La Niña conditions, meaning that the region is cooler than usual. Therefore, our results suggest that fishing effort increased during El Niño conditions.

Discussion

To our knowledge, this paper represents the first empirical study that illustrates how GIS techniques can be used in combination with BRT models to evaluate and predict the spatial distribution of fishing effort, and its response to human and climatic drivers of change, in this case, inside a multiple-use MPA. Our results showed that substantial changes in the spatio-temporal distribution of fishing effort occurred in the Galapagos spiny lobster fishery due to interaction among various climatic and human drivers, acting at multiple temporal and spatial scales. In this section, we first reviewed how these drivers of change influenced the large-scale dynamics of fishing effort in the Galapagos spiny lobster fishery. That is followed with an examination of micro-scale dynamics of fishing patterns in the small-scale fleet. Finally, we examined the impact of no-take zones and their management implications, leading to recommendations to improve the design and effectiveness of Galapagos marine zoning to reconcile conservation and fishery management objectives.

Drivers of change: ‘Macro’ adaptive responses

Our results showed that the spiny lobster fishery of the Galapagos, and the spatio-temporal distribution of its fishing effort, were especially affected by two major drivers, the boom-and-bust exploitation of the sea cucumber fishery and the global financial crisis 2007–09. Both drivers of change triggered substantial macro-scale changes in fishing effort dynamics within the GMR, notably reflected in a remarkable reduction of fishing capacity, and shifts in post-harvest arrangements.

The sequence of events examined here began with the re-opening of the sea cucumber fishery, one year after the creation of the GMR. This event caused severe overcapitalization of the entire Galapagos small-scale fishing sector, with fishing capacity increasing not only in the sea cucumber fishery but also in the lobster fishery [2,24,52]. Only a moratorium on new entrants in 2002 stopped the exponential growth in the number of fishers and vessels registered in Galapagos, that had occurred between 1997 and 2000 [51]. Subsequently, the collapse of the sea cucumber fishery in 2006 caused a severe economic perturbation, which was intensified a few years later by the global financial crisis 2007–09 [38,52]. The latter led to a sharp contraction in the consumption of lobsters in the United States, the main foreign market for Galapagos lobsters [53], and a price drop of 32% between 2008 and 2009 [53]. In response to this economic perturbation, a significant number of fishers abandoned not only the sea cucumber fishery, but also the spiny lobster fishery, leading to a 56% reduction in fishing effort between 2005 and 2008 [2]. This resulted in declines of total catch, and exports to mainland Ecuador, by 23% and 45%, respectively [2,38].

Our results suggest that those fishers who decided to remain in the spiny lobster fishery after 2006 responded to the crisis either by re-expanding their distribution ranges and core areas (in the cases of Puerto Villamil and Baquerizo Moreno) or by diversifying their products and markets (for Puerto Ayora). In the latter case, some fishers reacted to the global financial crisis 2007–09 in two ways [38]: (1) diversifying their product by trading whole fresh lobsters instead of lobster tails, as had been done since the 1960s; and (2) diversifying their market by selling their product directly to the local hospitality sector and general public instead of middlemen. The restructuring of the value chain improved fishers’ revenues by increasing local consumption of whole lobsters and increasing ex-vessel prices [54]. Diversification of products and markets was enabled by the fact that tourist and land-based infrastructure (hotels and restaurants) is more extensive in Puerto Ayora than in Baquerizo Moreno or Puerto Villamil. This socioeconomic feature made fishers from Puerto Ayora less vulnerable to the economic perturbations caused by total closure of the sea cucumber fishery and the global financial crisis 2007–09. Puerto Ayora’s fishers faced the crisis by adding value to their catches, rather than expanding the spatial range of fishing, as did fishers in other ports. Indeed the Puerto Ayora fishers shifted their fishing effort to nearer their homeport, which likely increased their profits by reducing variable costs (e.g., diesel fuel).

Different adaptive responses to the global financial crisis 2007–09 were reported by Castrejón and Defeo [38] for two Mexican spiny lobster fisheries, one in Punta Allen, Quintana Roo, and the other in Baja California. In Punta Allen, fishers from Vigía Chico’s fishing cooperative stopped lobster fishing for three months until market conditions improved and, since then, they have acquired the infrastructure, technology and expertise needed to export live lobsters to Asia and Europe. In Baja California, fishers from the Federation of Cooperative Societies of the Fishing Industry of Baja California (FEDECOOP) also adapted their harvesting and trading strategies according to global market conditions. In this case, a 10-day early closure of the spiny lobster fishing season was agreed upon, and implemented in a coordinated way, by the 10 cooperatives that made up the FEDECOOP. Thanks to this, and other harvest and trading strategies, such as the agreement of spiny lobster unit price and harvesting levels before the beginning of each fishing season, and the diversification of markets and products, FEDECOOP was able to reach maximum historic prices after the conclusion of the global financial crisis 2007–09. These two case studies, together with the Galapagos spiny lobster fishery, reinforce the notion that crises represent opportunities for learning, adapting, and entering onto more sustainable pathways [55]. Such crises triggered adaptive responses, either individual or collective, which were shaped by the social and geographic attributes of the fishing communities in which fishing cooperatives are embedded, and the capacity and willingness of individuals and fishing cooperatives to take actions to re-organize themselves, change harvesting and trading strategies, and implement self-regulatory mechanisms to face the economic perturbations caused by external drivers of change [38].

Finally, it should be noted that while the global financial crisis 2007–09 was detrimental for Galapagos fishers, it was beneficial for spiny lobster stocks. Two years after the official end of the recession, lobster CPUE and catch increased 91% and 102%, respectively, whereas fishing effort only increased 6% between 2009 and 2011 [2]. According to Defeo et al. [56], the recovery of spiny lobster stocks could be attributed to the substantial reduction in fishing effort, together with the combined effect of market forces and favorable environmental conditions. Our results support this hypothesis.

Drivers of change: ‘Micro’ adaptive responses

The macro-scale changes in the spiny lobster fishery, described above, in turn altered the micro-scale dynamics of fishing patterns in the small-scale fleet, reflected in spatio-temporal variations in the fishing fleets’ core areas and fishing effort distribution. According to the BRT models, these changes in fishing effort were shaped by six main predictor variables: the distance from homeport (DistHP), the latitude and the longitude, the particular time period considered (Period), the Oceanic Niño Index (ONI), and the lobster catch obtained in previous fishing trips (PrevCatch). The average sea cucumber revenue obtained the fishing season before the beginning of lobster season (SeaCucRev) also showed a good predictive performance, but not for Puerto Ayora, for the reasons explained above.

The coastal nature of the spiny lobster fishery, and the geographic and socioeconomic features of each homeport, help to explain why the six explanatory variables were the most relevant as fishing effort predictors. Each port showed different adaptive responses to these drivers due to differences regarding number of fishers, composition of the fishing fleet, and available land-based tourism infrastructure. For example, Baquerizo Moreno has historically had the largest concentration of fishers and mother boat vessels [51]. Such features probably have forced fishers to fish farther away from their homeport to reduce competition with their peers, thereby reducing the influence of static variables on fishing effort distribution. In contrast, the reduced number of mother boats in Puerto Ayora and Villamil increased the influence of static variables, which may explain why these fishers catch spiny lobsters near their homeports.

A special feature of Galapagos is the limited number of landing sites (Fig 1). This, together with the limited range of the Galapagos small-scale fishing fleet and the close proximity of homeports to the most productive fishing grounds [36,53], explains why static rather than dynamic explanatory variables were more relevant as fishing effort predictors. Thus, distance to homeport, latitude and longitude are the most important variables explaining why fishers from the same homeport tend to use similar fishing grounds. This leads to exclusive core areas, which usually do not show overlaps. Similar results were found by Bucaram et al. [20] whose short-term analysis of factors affecting fishing behaviour in the Galapagos spiny lobster fishery identified travel distance from vessels’ home ports to fishing grounds and expected revenues as the most important factors affecting spatial allocation of fishing effort. They also found that fishing behaviour is sensitive to changes in sea conditions and sea surface temperature, but not to precipitation or moon visibility.

In contrast, dynamic variables explain why fishing patterns changed during certain periods of time. In this sense, our results suggest that two variables, the revenues produced by the sea cucumber fishery and the previous lobster catch, are responsible for the alternate expansion and contraction of fishing fleets’ core areas and distribution ranges during the boom-and-bust exploitation of the sea cucumber fishery, marine zoning implementation and the global financial crisis 2007–09. Higher revenues produced by the sea cucumber fishery and higher previous lobster catches were associated with increasing trends in fishing effort in the spiny lobster fishery. Our results suggest that higher revenues produced during the reopening and expansion period of the sea cucumber fishery probably acted as subsidies that allowed spiny lobster fishers to extend their fishing trips for longer times and farther away from their homeports. In contrast, lower revenues caused by the overexploitation of the sea cucumber and spiny lobster fishery, and the global financial crisis 2007–09, led to decreasing trends in fishing effort, which were reflected in the contraction of fishing fleets’ core areas and distribution ranges. These hypotheses are supported by Bucaram and Hearn [57], who found that the decision to participate in the spiny lobster fishery is significantly influenced by the average catch per trip of spiny lobsters and sea cucumbers during the previous fishing season. In other words, the higher the average catch per trip of both species obtained by a vessel in the previous year, the more likely that the vessel will decide to participate in the next spiny lobster fishing season, and the greater the extent of participation.

The oceanographic variable ONI was also identified in our analysis as a relevant fishing effort predictor. From partial dependence plots, fishing effort increased during El Niño conditions; this could be caused by the redistribution of spiny lobster stocks from inshore to deeper waters, making them inaccessible to fishing by hooka diving (cf. [58]). Such reproductive migrations are influenced by temperature. According to Vega [59], warmer temperatures during El Niño periods accelerate the time of breeding of Panulirus interruptus significantly, while the converse occurred under colder temperatures caused by La Niña. Based on these studies, fishing effort probably was higher in the Galapagos spiny lobster fishery during El Niño events, with reproductive migration of spiny lobsters to deeper waters making the lobster less accessible to fishing, leading to increased search times and increased diving hours per fishing trip.

Impact of no-take zones and management implications

The above results of our integrated analysis showed how Galapagos fishing communities coped with the interactions of human and climatic drivers of change, both temporally and spatially. Our results showed that ‘macro’ and the ‘micro’ fishers’ adaptive responses varied according to the magnitude, extent and intensity of the social-ecological perturbations caused by the drivers of change analyzed, and were shaped by geographic, economic and oceanographic factors, including the socioeconomic attributes of the three fishing communities analyzed.

Furthermore, our results indicated that among the possible effects producing a recovery of the spiny lobster stock, the implementation of no-take zones within the GMR was not a significant factor. Indeed, there is no scientific evidence (see also [2]) that adoption of no-take zones contributed directly to the sustainability of Galapagos shellfish fisheries [60,61]. The lack of a ‘fishing the line’ effect around no-take zones and the poor performance of the NearNTZ variable as a fishing effort predictor suggest that marine zoning, after its implementation, had little impact on the spatio-temporal distribution of fishing effort, particularly in Puerto Ayora and Baquerizo Moreno. This result may well be due to the manner in which locations of no-take zones were chosen across Galapagos. According to Edgar et al. [21], fishers sought to minimize perceived impacts on their livelihood by advocating the location of no-take zones in areas with low densities of the most valuable commercial species (sea cucumber and spiny lobster), while tourism operators and sport divers promoted the protection of areas containing high densities of species important for tourism (e.g., sharks). As a result, sea cucumber and spiny lobster baseline densities were much more abundant (3 and 2.7 times higher, respectively) in fishing zones compared to no-take zones, although differences between both zone types were not significant for spiny lobsters [21]. The location of no-take zones in areas with relatively low abundance of the most lucrative species explains why fishers have shown a lack of interest in fishing near these areas, thereby explaining the lack of a ‘fishing the line’ effect around no-take zone boundaries after marine zoning implementation.

The above results have implications for fisheries policy and management in the GMR. First, there is a need to re-evaluate the distribution of no-take zones across the GMR, to promote the sustainability of the spiny lobster fishery and conserve key biodiversity areas. Effective no-take zones should be implemented in areas that ensure the protection of a proportion of the breeding stock and critical reproduction and nursery habitats. However, the geographic location of these areas across the Galapagos archipelago is still uncertain. It may be useful to consider fishing effort hotspots, which probably overlap with the location of spawning and nursery areas, a hypothesis that should be evaluated by future studies. Unfortunately, declaring fishing effort hotspots as no-take zones will represent a challenge, considering the lack of evidence of the ecological and economic benefits provided by no-take zones and the high opportunity and transaction costs associated with their implementation and enforcement. That reality has contributed to reduce the acceptability and legitimacy of what could be potentially a valuable tool to manage Galapagos shellfish fisheries [22]. In consequence, additional research and management efforts are required to create the conditions for the effective planning, implementation, monitoring and enforcement of the GMR’s marine zoning.

Second, since no-take zones represent only one of multiple management tools available for successful implementation of spatial EBM [22], a more effective approach could be a combination of a coastal network of no-take zones with co-managed harvested areas that allocate exclusive spatial fishing rights to local communities, potentially a more robust approach to address the roots of fisheries management failures that led to overexploitation of fisheries [2,17,18,62]. This could produce a set of strategically-placed Territorial Use Rights in Fishing (TURF) areas. Based on the distribution of core areas and fishing effort hotspots across the archipelago for the three fishing fleets, the most strategic places for the experimental implementation of TURFs in the GMR are those located in the southern part of Isabela Island, the western part of Santa Cruz Island and the southeastern part of San Cristobal Island. These places are strategic because (1) each is used exclusively by one fishing fleet, which reduces the likelihood of potential conflicts among different fishing fleets arising over competition for the same ocean space, and (2) each is located near the corresponding homeport, which facilitates surveillance, control and monitoring activities and creates an economic incentive for TURF co-management. The active involvement of local communities in the co-management of strategically placed TURFs could contribute, under certain enabling conditions, to generate a sense of stewardship among fishers [4,63–65]. This management approach could promote the implementation, by fishers themselves, of effective monitoring, control and surveillance procedures, and the accomplishment of objectives for management and conservation [66], as has been observed in spiny lobster fisheries of Baja California, México and Chile [38,67,68], all currently certified by the Marine Stewardship Council as sustainable [69].

Third, the geographic definition of new management areas is needed, based on core areas and distribution ranges of the three fishing fleets. This could include area-based co-management that could enhance the acceptability and legitimacy of GMR’s marine zoning and help to mitigate the potential conflict associated with the redistribution of no-take zones. Within the area-based co-management system, we suggest creating specific co-management councils for each management area to promote the involvement and participation of local stakeholders in their planning, implementation, monitoring and enforcement. Each co-management council should be made up exclusively of those fishers, and other relevant stakeholders, who would be most affected by implementation of no-take zones, and/or the experimental allocation of TURFs, inside their core area and distribution ranges. An area-based co-management approach could be useful to ensure a strategic distribution of no-take zones across the archipelago and to minimize the impact of zoning on fishing communities’ livelihoods, helping to improve the acceptance, legitimacy and compliance of the new marine zoning scheme.

Conclusions and lessons learned

The first and most important conclusion and lesson learned lies in the reality that fishery systems face social-ecological impacts produced by a diverse range of human and climatic external drivers of change, acting at different spatial and temporal scales, usually simultaneously. These include not only the implementation of new regulations, such as multiple-use MPAs and marine zoning schemes, and extreme climatic events associated with global climate change (e.g., El Niño), but also socioeconomic perturbations caused by the globalization of markets and the development and/or collapse of alternative fisheries.

Second, our results demonstrated the need for a broad-based and integrated social-ecological approach to fishery management and marine conservation, whether in planning MPAs or in any fisheries management, or indeed, any natural resource management. In this context, MPAs must be designed and implemented taking into consideration not only the spatial-temporal dynamics of key biodiversity areas and fishery resources, but also the dynamics of fishing fleets, and fishers’ adaptive responses to human and climatic drivers of change. Only in this way will management and conservation measures, such as no-take zones, be useful tools for rebuilding depleted fish stocks, conserving marine ecosystems and improving fishing communities’ livelihoods.

Third, assessments of the effectiveness of MPAs should not assume that any change in fishing patterns is caused exclusively by the implementation of an MPA. Instead, a comprehensive understanding of how local fishing communities cope with relevant human and climatic drivers is fundamental to properly assess the socio-ecological outcomes generated by an MPA. This knowledge will reduce the risk of errors in planning, implementing and assessing the effectiveness of MPAs and marine zoning more broadly.

These conclusions and lessons learned apply broadly, not only to situations involving MPAs, but to any social-ecological system in which ecosystem-based management, marine zoning and other management approaches are being considered to improve the governance and sustainability of fisheries and the conservation of key biodiversity areas.

Supporting information

S1 Table. Summary of the fishery monitoring data gathered for the spiny lobster fishery at the three main ports of the Galapagos Marine Reserve from 1997 to 2011.

(DOCX)

Click here for additional data file.^{(17.1KB, docx)}

S2 Table. Fishing fleets estimated mean core areas and distribution ranges (in km²), according to port interviews and observer onboard data collected in the Galapagos Marine Reserve from 1997 to 2011.

(DOCX)

Click here for additional data file.^{(13.6KB, docx)}

S3 Table. Fishing fleets estimated site fidelity (IOR95) to similar core areas and distribution ranges, according to port interviews and observer onboard data collected in the Galapagos Marine Reserve from 1997 to 2011.

(DOCX)

Click here for additional data file.^{(21.2KB, docx)}

Acknowledgments

We thank Peter Tyedmers, Boris Worm, Jeffrey Hutchings, Robert Steneck, as well as the editor and referees, for their helpful suggestions and comments to improve this research article. We acknowledge the Galapagos National Park Service and the Charles Darwin Foundation for sharing the fishery-related data required to conduct this study.

Data Availability

The Galapagos National Park Service is the owner of the data used for this paper. Therefore, there are legal restrictions on sharing a de-identified data set. However, the data underlying the results presented in the study are available on request at investigacion@galapagos.gob.ec. We confirm that other researchers would be able to access the data set in the same manner as we did, and we did not have any special access privileges that others would not have.

Funding Statement

MC is grateful for the financial support provided by the Consejo Nacional de Ciencia y Tecnología (CONACYT-Mexico), and the World Wildlife Fund’s Russell E. Train Education for Nature Program. AC acknowledges funding support from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Social Sciences and Humanities Research Council of Canada (SSHRC), through the Community Conservation Research Network (www.communityconservation.net). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Worm B, Hilborn R, Baum JK, Branch TA, Collie JS, Costello C, et al. Rebuilding Global Fisheries. Science. 2009;325: 578–585. 10.1126/science.1173146 [DOI] [PubMed] [Google Scholar]
2.Defeo O, Castrejón M, Pérez-Castañeda R, Castilla JC, Gutiérrez NL, Essington TE, et al. Co-management in Latin American small-scale shellfisheries: assessment from long-term case studies. Fish Fish. 2016;17: 176–192. 10.1111/faf.12101 [DOI] [Google Scholar]
3.Micheli F, Saenz-Arroyo A, Greenley A, Vazquez L, Espinoza Montes JA, Rossetto M, et al. Evidence that marine reserves enhance resilience to climatic impacts. PLoS One. 2012;7: e40832 10.1371/journal.pone.0040832 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Alcala A, Russ G. No-take marine reserves and reef fisheries management in the Philippines: a new people power revolution. Ambio. 2006;35: 245–254. 10.1579/05-a-054r1.1 [DOI] [PubMed] [Google Scholar]
5.OECD. Marine protected areas: economics, management and effective policy mixes Paris: OECD Publishing; 2017. [Google Scholar]
6.Maestro M, Pérez-Cayeiro M, Chica Ruiza J, Reyez H. Marine protected areas in the 21st century: current situation and trends. Ocean Coast Manag. 2019;171: 28–36. [Google Scholar]
7.Guarderas AP, Hacker SD, Lubchenco J. Current status of marine protected areas in Latin America and the Caribbean. Conserv Biol. 2008;22: 1630–1640. 10.1111/j.1523-1739.2008.01023.x [DOI] [PubMed] [Google Scholar]
8.Guarderas AP, Hacker SD, Lubchenco J. Ecological effects of marine reserves in Latin America and the Caribbean. Mar Ecol Prog Ser. 2011;429: 219–225. 10.3354/meps09103 [DOI] [Google Scholar]
9.Charles A, Wilson L. Human dimensions of marine protected areas. ICES J Mar Sci. 2009;66: 6–15. [Google Scholar]
10.Charles AT. Human dimensions in marine ecosystem-based management In: Fogarty MJ, McCarthy J, editors. The Sea, Volume 16: Marine Ecosystem-Based Management. Cambridge: Harvard University Press; 2014. [Google Scholar]
11.Christie P, Bennett N, Gray N, Wilhelm T, Lewis N, Parks J, et al. Why people matter in ocean governance: incorporating human dimensions into large-scale marine protected areas. Mar Policy. 2017;84: 273–284. [Google Scholar]
12.Salas S, Gaertner D. The behavioural dynamics of fishers: management implications. Fish Fish. 2004;5: 153–167. 10.1111/j.1467-2979.2004.00146.x [DOI] [Google Scholar]
13.Stelzenmüller V, Maynou F, Bernard G, Cadiou G, Camilleri M, Crec’hriou R, et al. Spatial assessment of fishing effort around European marine reserves: implications for successful fisheries management. Mar Pollut Bull. 2008;56: 2018–2026. 10.1016/j.marpolbul.2008.08.006 [DOI] [PubMed] [Google Scholar]
14.Horta e Costa B, Batista MI, Gonçalves L, Erzini K, Caselle JE, Cabral HN, et al. Fishers’ behaviour in response to the implementation of a marine protected area. PLoS One. 2013;8: e65057 10.1371/journal.pone.0065057 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Halpern BS, Lester S, Kellner J. Spillover from marine reserves and the replenishment of fished stocks. Env Conserv. 2010;36: 268–276. 10.1017/S0376892910000032 [DOI] [Google Scholar]
16.Kerwath SE, Winker H. Marine protected area improves yield without disadvantaging fishers. Nat Commun. 2013;4: 2347 10.1038/ncomms3347 [DOI] [PubMed] [Google Scholar]
17.Charles A, Westlund L, Bartley DM, Fletcher WJ, Garcia S, Govan H, et al. Fishing livelihoods as key to marine protected areas: insights from the World Parks Congress. Aquat Conserv Mar Freshw Ecosyst. 2016;26: 165–184. 10.1002/aqc.2648 [DOI] [Google Scholar]
18.Westlund L, Charles A, Garcia SM, Sanders J. Marine protected areas: interactions with fishery livelihoods and food security Rome: FAO Fisheries Aquaculture Technical Paper; 603; 2017. [Google Scholar]
19.Soykan CU, Eguchi T, Kohin S, Dewar H. Prediction of fishing effort distributions using boosted regression trees. Ecol Appl. 2014;24: 71–83. 10.1890/12-0826.1 [DOI] [PubMed] [Google Scholar]
20.Bucaram SJ, White JW, Sanchirico JN, Wilen JE. Behavior of the Galapagos fishing fleet and its consequences for the design of spatial management alternatives for the red spiny lobster fishery. Ocean Coast Manag. 2013;78: 88–100. 10.1016/j.ocecoaman.2013.03.001 [DOI] [Google Scholar]
21.Edgar GJ, Bustamante RH, Fariña JM, Calvopiña M, Martínez C, Toral-Granda M V. Bias in evaluating the effects of marine protected areas: the importance of baseline data for the Galapagos Marine Reserve. Environ Conserv. 2004;31: 212–218. 10.1017/S0376892904001584 [DOI] [Google Scholar]
22.Castrejón M, Charles A. Improving fisheries co-management through ecosystem-based spatial management: the Galapagos Marine Reserve. Mar Policy. 2013;38: 235–245. 10.1016/j.marpol.2012.05.040 [DOI] [Google Scholar]
23.Viteri C, Chávez C. Legitimacy, local participation, and compliance in the Galápagos Marine Reserve. Ocean Coast Manag. 2007;50: 253–274. 10.1016/j.ocecoaman.2006.05.002 [DOI] [Google Scholar]
24.Szuwalski CS, Castrejon M, Ovando D, Chasco B. An integrated stock assessment for red spiny lobster (Panulirus penicillatus) from the Galapagos Marine Reserve. Fish Res. 2016;177: 82–94. 10.1016/j.fishres.2016.01.002 [DOI] [Google Scholar]
25.Defeo O, Castrejón M, Ortega L, Kuhn AM, Gutiérrez NL, Castilla JC. Impacts of climate variability on Latin American small-scale fisheries. Ecol Soc. 2013;18: 30 10.5751/ES-05971-180430 [DOI] [Google Scholar]
26.DPNG. Plan de manejo de las áreas protegidas de Galápagos para el buen vivir. Izurieta A, Tapia W, Mosquera G, Chamorro S, editors. Galapagos, Ecuador: Dirección del Parque Nacional Galápagos; 2014.
27.Edgar GJ, Banks S, Fariña JM, Calvopiña M, Martínez C. Regional biogeography of shallow reef fish and macroinvertebrate communities in the Galapagos Archipelago. J Biogeogr. 2004;31: 1107–1124. [Google Scholar]
28.Wolff M, Ruiz D, Taylor M. El Niño induced changes to the Bolivar Channel ecosystem (Galapagos): comparing model simulations with historical biomass time series. Mar Ecol Prog Ser. 2012;448: 7–22. 10.3354/meps09542 [DOI] [Google Scholar]
29.Bustamante RH, Reck G, Ruttenberg BI, Polovina J. The Galápagos spiny lobster fishery In: Phillips BF, Kittaka J, editors. Spiny lobsters: Fisheries and culture. Fishing News Books; 2000. [Google Scholar]
30.PNG. Plan de Manejo del Parque Nacional Galápagos: un pacto por la conservación y desarrollo sustentable del archipiélago Galapagos, Ecuador: Parque Nacional Galápagos; 2005. [Google Scholar]
31.Epler B. Tourism, economy, population growth, and conservation in Galapagos Galapagos: Charles Darwin Foundation; 2007. [Google Scholar]
32.Reyes H, Ramírez J. Informe técnico de la pesquería de langosta espinosa (Panulirus penicillatus y Panulirus gracilis) 2012 en la Reserva Marina de Galápagos Galapagos: Dirección del Parque Nacional Galápagos; 2012. [Google Scholar]
33.Heylings P, Bravo M. Evaluating governance: a process for understanding how co-management is functioning, and why, in the Galapagos Marine Reserve. Ocean Coast Manag. 2007;50: 174–208. [Google Scholar]
34.Heylings P, Bensted-Smith R, Altamirano M. Zonificación e historia de la Reserva Marina de Galápagos In: Danulat E, Edgar GJ, editors. Reserva Marina de Galápagos, línea base de la biodiversidad. Quito, Ecuador: Fundación Charles Darwin and Servicio Parque Nacional Galápagos; 2002. pp. 10–21. [Google Scholar]
35.Hall SJ. Climate change an other external drivers in small-scale fisheries: practical steps for responding In: Pomeroy RS, Andrew N, editors. Small-scale fisheries management: Frameworks and approaches for the developing world. London, UK: CABI; 2011. pp. 132–159. [Google Scholar]
36.Reck G. The coastal fisheries in the Galapagos Islands, Ecuador. Descriptions and consequences for management in the context of marine environmental protection and regional development. PhD thesis, Christian-Albrechts-Universität zu Kiel, Germany. 1983.
37.Chasiluisa C, Banks S. Sitios de pesca en la costa de la Reserva Marina de Galápagos In: Danulat E, editor. Evaluación de las pesquerías de la Reserva Marina de Galápagos Informe compendio 2003: análisis comparativo con los años 1997–2002. Galapagos: Charles Darwin Foundation; 2004. pp. 130–140. [Google Scholar]
38.Castrejón M, Defeo O. Co-governance of small-scale shellfisheries in Latin America: institutional adaptability to external drivers of change In: Chuenpagdee R, Jentoft S, editors. Interactive governance for small-scale fisheries: Global reflections. MARE Publications Series 13. Springer International Publishing; 2015. pp. 605–625. 10.1007/978-3-319-17034-3 [DOI] [Google Scholar]
39.Moreno J, Peñaherrera C, Hearn A. Evaluación de la pesquería de langosta espinosa (Panulirus penicillatus y P. gracilis) en la Reserva Marina de Galápagos 2006. Galapagos: Charles Darwin Foundation; 2007. [Google Scholar]
40.Fortin M, Dale M. Spatial analysis: a guide for ecologists. Cambridge: Cambridge University Press; 2005. [Google Scholar]
41.Morrissey J, Gruber S. Home range of juvenile Lemon sharks Negaprion brevirostris. Copeia. 1993;2: 425–434. [Google Scholar]
42.Horta e Costa B, Gonçalves L, Gonçalves EJ. Vessels’ site fidelity and spatio-temporal distribution of artisanal fisheries before the implementation of a temperate multiple-use marine protected area. Fish Res. 2013;148: 27–37. [Google Scholar]
43.Ord J, Getis A. Local spatial autocorrelation statistics: distributional issues and an application. Geogr Anal. 1995;27: 286–306. [Google Scholar]
44.Ord J, Getis A. Testing for local spatial autocorrelation in the presence of global autocorrelation. J Reg Sci. 2001;41: 411–432. [Google Scholar]
45.Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction New York: Springer-Verlag; 2001. [Google Scholar]
46.Elith J, Leathwick J, Hastie T. A working guide to boosted regression trees. J Anim Ecol. 2008;77: 802–813. 10.1111/j.1365-2656.2008.01390.x [DOI] [PubMed] [Google Scholar]
47.Martínez-Rincón R, Ortega-García S, Vaca-Rodríguez J. Comparative performance of generalized additive models and boosted regression trees for statistical modeling of incidental catch of wahoo (Acanthocybium solandri) in the Mexican tuna purse-seine fishery. Ecol Modell. 2012;233: 20–25. [Google Scholar]
48.Friedman J. Greedy function approximation: A gradient boosting machine. Ann Stat. 2001;29: 1189–1232. [Google Scholar]
49.Friedman J, Meulman J. Multiple additive regression trees with application in epidemiology. Stat Med. 2003;22: 1365–1381. 10.1002/sim.1501 [DOI] [PubMed] [Google Scholar]
50.Toral-Granda V. Galapagos Islands: a hotspot of sea cucumber fisheries in Latin America and the Caribbean In: Toral-Granda V, Lovatelli A, Vasconcellos M, editors. Sea cucumbers: a global review of fisheries and trade. Rome: FAO Fisheries and Aquaculture Technical Paper; No. 516; 2008. pp. 231–256. [Google Scholar]
51.Castrejón M. Co-manejo pesquero en la Reserva Marina de Galápagos: tendencias, retos y perspectivas de cambio Mexico: Fundación Charles Darwin, Kanankil, and Plaza-Valdés; 2011. [Google Scholar]
52.Shepherd SA, Martinez P, Toral-Granda MV, Edgar GJ. The Galapagos sea cucumber fishery: Management improves as stocks decline. Env Conserv. 2004;31: 102–110. 10.1017/S0376892903001188 [DOI] [Google Scholar]
53.Ramírez J, Castrejón M, Toral-Granda MV. Mejorando la pesquería de langosta en la Reserva Marina de Galápagos. Ramírez J, Castrejón M, Toral-Granda MV, editors. Quito, Ecuador: World Wildlife Fund; 2012.
54.Viteri C, Moreno J. Estudio de impactos socio-económicos de la comercialización de langosta entera en Galápagos Galapagos: Conservation International; 2014. [Google Scholar]
55.Folke C, Colding J, Berkes F. Synthesis: Building resilience and adaptive capacity in socio-ecological systems In: Berkes F, Colding J, Folke C, editors. Navigating the dynamics of social–ecological systems: Building resilience for complexity and change. Cambridge: Cambridge University Press; 2003. pp. 325–383. [Google Scholar]
56.Defeo O, Castrejón M, Ortega L, Kuhn AM, Gutiérrez NL, Castilla JC. Impacts of climate variability on Latin American small-scale fisheries. Ecol Soc. 2013;18 10.5751/ES-05971-180430 [DOI] [Google Scholar]
57.Bucaram SJ, Hearn A. Factors that influence the entry-exit decision and intensity of participation of fishing fleet for the Galapagos lobster fishery. Mar Policy. 2014;43: 80–88. 10.1016/j.marpol.2013.05.005 [DOI] [Google Scholar]
58.Bertelsen R, Hornbeck J. Using acoustic tagging to determine adult spiny lobster (Panulirus argus) movement patterns in the Western Sambo Ecological Reserve (Florida, United States), New Zealand: J Mar Freshw Res; 2009;43: 35–46. [Google Scholar]
59.Vega A. Reproductive strategies of the spiny lobster Panulirus interruptus related to the marine environmental variability off central Baja California, Mexico: Management implications. Fish Res; 2003;65: 123–135. 10.1016/j.fishres.2003.09.011 [DOI] [Google Scholar]
60.Jones PJS. A governance analysis of the Galápagos Marine Reserve. Mar Policy. 2013;41: 65–71. 10.1016/j.marpol.2012.12.019 [DOI] [Google Scholar]
61.Hearn A. The rocky path to sustainable fisheries management and conservation in the Galápagos Marine Reserve. Ocean Coast Manag. 2008;51: 567–574. 10.1016/j.ocecoaman.2008.06.009 [DOI] [Google Scholar]
62.Salas S, Chuenpagdee R, Charles A, Seijo J. Coastal Fisheries of Latin America and the Caribbean Rome: FAO Fisheries and Aquaculture Technical Paper 544; 2011. [Google Scholar]
63.White C, Costello C. Matching spatial property rights fisheries with scales of fish dispersal. Ecol Appl. 2011;21: 350–362. 10.1890/09-1188.1 [DOI] [PubMed] [Google Scholar]
64.Aburto J, Stotz W. Learning about TURFs and natural variability: failure of surf clam management in Chile. Ocean Coast Manag. 2013;71: 88–98. [Google Scholar]
65.Charles A. Rights-based fisheries management: the role of use rights in managing access and harvesting In: Cochrane K, Garcia SM, editors. A fishery manager’s guidebook. Oxford, UK: Wiley-Blackwell; 2009. pp. 253–282. [Google Scholar]
66.Charles A. People, oceans and scale: governance, livelihoods and climate change adaptation in marine social-ecological systems. Curr Opin Environ Sustain. 2012;4: 351–357. 10.1016/j.cosust.2012.05.011 [DOI] [Google Scholar]
67.Gelcich S, Hughes TP, Olsson P, Folke C, Defeo O, Fernandez M, et al. Navigating transformations in governance of Chilean marine coastal resources. Proc Natl Acad Sci U S A. 2010;107: 16794–16799. 10.1073/pnas.1012021107 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.McCay BJ, Micheli F, Ponce-Díaz G, Murray G, Shester G, Ramirez-Sanchez S, et al. Cooperatives, concessions, and co-management on the Pacific coast of Mexico. Mar Policy. 2014;44: 49–59. 10.1016/j.marpol.2013.08.001 [DOI] [Google Scholar]
69.Pérez-Ramírez M, Castrejón M, Gutiérrez N, Defeo O. The Marine Stewardship Council certification in Latin America and the Caribbean: a review of experiences, potentials and pitfalls. Fish Res. 2016;182: 50–58. 10.1016/j.fishres.2015.11.007 [DOI] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0228094.r001

Decision Letter 0

Heather M Patterson

1 Nov 2019

PONE-D-19-26580

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

PLOS ONE

Dear Dr. Castrejón,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I found this to be a really interesting study. The inclusion of social drivers was especially interesting. However, both reviewers have flagged a number of issues that need to be addressed before the manuscript can be published. The main issue with this manuscript is that it is just far too long and wordy. It would be much more accessible for the reader if it was written in a more concise manner. Instead, it is very repetitive and often includes details that are not germane to the actual study. The Discussion is especially repetitive and in need of some serious editing. Both reviewers have provided very detailed comments to improve the clarity of the paper and assist the authors in their revisions. I have also provided extensive editorial comments (PLoS editorial comment file) to assist the authors. I strongly encourage the authors to consider all the comments provided when making your revisions.

We would appreciate receiving your revised manuscript by Dec 16 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Heather M. Patterson, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Review of PONE-D-19-26580. “Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve”. Castrejon, M. and A. Charles

This study evaluated how the spatiotemporal allocation of fishing effort for lobsters in the Galapagos multiple-use Marine Protected Area was affected by the interaction of diverse climatic and human drivers, before and after implementation of no-take zones. The study used GIS data on fishing effort and BRTs to attempt to identify how these drivers affected spatial fishing patterns. The paper concludes that the boom-and-bust exploitation of the sea cucumber fishery and the global financial crisis (2007-2009), rather than no-take zone implementation, were the most important drivers affecting the distribution of fishing effort for lobsters across the archipelago. The study is spatially and temporally extensive (most of the Galapagos Islands, 1997-2011), the data are fairly well-analyzed and interpreted, and the manuscript is well-written. I have no major disagreements with the conclusions. I also have some sympathy with the suggestions that the MPA network placement could be revisited or even revised, and outside the network TURFS encouraged. My comments are mostly to assist the authors with publication.

Major Comments.

1. This is a paper that, in effect, quantifies spatial and temporal trends in fishing effort of a lobster fishery in a developing country. Yet the emphasis chosen is how this data informs effects of MPA implementation. It is highly commendable that the study includes before and after implementation data. In fact, this is such an important aspect of the study, I would recommend that the authors stress this point more in the paper. However, this MPA network is also well-known as a “classic” case where fishers ensured that no-take zones were NOT placed where fishers fished (Edgar et al 2004 Ref. 22 in this manuscript). That is, it is a case where you might NOT expect much change in spatial effort in the lobster fishery pre- and post-implementation of the MPA network (which is what they found). This very important point is not even mentioned until Lines 999-1010 in the Discussion. I recommend that you mention this much earlier in the paper, probably in the Abstract and Introduction.

2. You place a substantial amount of faith in the “explanatory” powers of your BRTs. This needs to be tempered a fraction. Table 4 indicates that the deviance explained by the BRTs is 29.47% (Regional), 35.73% (PV), 32.66% (PA) and 15.74% (BM). If I understand Figures 7-10 and Table 4 correctly, this amount of deviance explained is then partitioned among 14 potential explanatory predictor variables. Thus, Distance from Port, your strongest driver in the Regional analysis, explains 22.4% of 29.47% (i.e. 6.6%) of the variance. For BM, your strongest driver, Longitude, explains about 17% of 15.74% (i.e. 2.7%). Clearly all of the weaker drivers “explain” very small percentages of the spatial trends. Thus, describing small peaks and troughs in the trends shown in individual panels in Figs. 7-10 is almost describing details unnecessarily. That said, I agree that the major 6 drivers in the BRTs are as you indicate at Lines 1045-47.

3. The spatial scale at which you measure effort (2.25 km2) may be rather coarse to be making confident statements about the lack of evidence for “fishing the line”. Many of the studies of spillover (see references cited below at Line 86) often report this effect at much smaller spatial scales than this. You should at least acknowledge this point.

4. The Discussion is far too long and repetitive (17 pages, with a Summary of almost 7 pages). This should be condensed considerably.

5. Lines 93-97 (Introduction) and 952-954 (Discussion) “…to our knowledge, no study has examined yet how fishers respond to those situations in which they have to cope simultaneously with implementation of an MPA, and with the interaction of external drivers…”. A relevant, similar, example is the perceived effect of the rezoning of Australia’s Great Barrier Reef Marine Park in 2004 on local fisheries described by Fletcher WJ et al (2015) Large-scale expansion of no-take closures within the Great Barrier Reef has not enhanced fishery production. Ecol. Appl. 25: 1187-1196 and critiqued by Hughes TP et al (2016) A critique of claims for negative impacts of marine protected areas on fisheries. Ecol. Appl. 26: 637-641. I would recommend that you cite these two papers.

Minor Comments.

Abstract.

Line 30. MP Area (omit s).

Line 31. Note change in font size of text at full stop.

Line 37. Unfeasible (not infeasible).

Introduction

Line 73. “…pay greater attention to the human dimensions of MPAs [10,11]…” In addition references 10 and 11 cited, both by the authors of the current paper, a very relevant example possibly worth citing here would be: Alcala A.C. and G.R. Russ (2006). No-take marine reserves and reef fisheries management in the Philippines: A new people power revolution. Ambio 35(5): 245-254.

Line 86 (and 198). In addition to the Kellner reference (15) on spillover and fishing the line, which is a modelling paper, and Ref. 26 (line 198) and Ref. 70 (line 1264) regarding spillover, three excellent empirical papers on spillover that could be cited are the review by Halpern BS et al (2010) Spillover from marine reserves and the replenishment of fished stocks. Env. Cons. 36: 268-276; Goni R et al (2010) Net contribution of spillover from a marine reserve to fishery catches. Mar. Ecol. Prog. Ser. 400:233-243 (on lobsters in the Mediterranean); and Kerwath SE et al (2013) Marine protected area improves yield without disadvantaging fishers. Nature Communications 4:2347. You should also acknowledge the possibility of larval (as opposed to adult) export from reserves to fished areas, for example: Harrison HB et al (2012) Larval export from marine reserves and the recruitment benefit for fish and fisheries. Current Biology 22:1023-1028.

Lines 93-97. Note major point 5 above.

Lines 116-117. Indicate here the year when the MPAs were implemented (2000).

Materials and Methods.

Line 198. “….and spillover to fishing grounds may occur ([26]”

Lines 208, 209. Tourist or tourism (not touristic).

Lines 273-274. Why calculate effort by dividing catch by catch-per-unit-effort (CPUE)? Surely you measured catch and effort directly to calculate CPUE?

Line 302. ..affected by the potential drivers (add potential).

Lines 310-312. You make it clear that the re-zoning was confounded by the sea-cucumber over-exploitation phase (see also Table 3). Thus, when you talk of changes to effort associated with the zoning (e.g. Lines: 600-605, 989-990, 1035-1036) you must acknowledge this confounding. At lines 1096-1099 you DO acknowledge the confounding, and should in other places in the manuscript.

Lines 367, 401, 407. Insert “the” before: normality assumption, input field, z score.

Results.

Line 505. (Fig. 2a, b, c) should read (Fig. 2d, e, f).

Line 520. (Fig. 2d, e, f) should read (Fig. 2a, b, c).

Line 601. Acknowledge confounding of zoning and sea-cucumber over-exploitation phase.

Line 644. ..the eastern part, ..the southeastern part (insert the).

Line 678. Fishers (add s).

Line 702. These types of fishers..

Line 718. Suggest (not suggests).

Line 888. Western side of

Discussion.

Lines 952-954. Note comment re Fletcher et al (2015) and Hughes et al (2016) above.

Lines1174-1176. Good point. The lobster recovery may not be related to the implementation of the MPAs.

Lines 1188-1190 and 1218-1220. When suggesting a re-evaluation of the MPA zoning, you must be clear about why the MPAs were established: conservation, fisheries management, or both.

Lines 1194. The TURFS suggestion outside the MPAs is a good one.

Lines 1199-1201. Alcala and Russ (2006) could be cited here also.

Lines 1224-1227. Why would an MPA network placed in a biased manner help the fishery if it was set up to avoid the fishery?

Line 1238. ..replicates.

Line 1241. Thirdly (not Fourthly).

Lines 1249-1262. In addition to the Kay example in the Channel Islands, which is a good one, you could also mention the Goni et al (2010) lobster example from the Mediterranean.

Line 1264. Ref. 70 in support of the idea of spillover is inadequate. See references to cite on spillover suggested above.

Line 1291. To support (not the support).

Fig. 2. What do the dark grey and light grey shaded areas of time represent? El Nino/La Nina? Specify in caption.

Figs. 3 and 4. What are the units here? Effort (diver hours)? Specify in caption.

Figs. 3-6. I find it difficult to differentiate Fig. 3 from 4, or Fig. 5 from Fig. 6, simply by eye.

Figures 7-10. Specify acronyms for all of the predictor variables in the caption of Fig. 7, then refer to this in the captions of Figs. 8-10. Reader must be reminded what these variables are in the caption.

Fig. 7. I agree, NearNTZ has no pattern.

Table 1. Caption. Sampling method (not smapling).

Table 3. Caption Line 2: occurring (not occurred).

Table 4. Perhaps call the variables “Predictor Variables” in the caption?

Reviewer #2: General comments:

In this study, the authors aim to investigate the effects of management, biophysical data and socioeconomic factors on the distribution of fishing effort. They use a variety of analytical tools to detect global and local drivers, from the Global Financial Crisis and climatic drivers to the distribution of MPAs. Given the need to better understand drivers of social and ecological dynamics, it will be good to see this paper published. There are two primary concerns that need to be addressed, however.

The first (and most serious) is that there is no mention of overfishing as a possible driver. This may be hard to measure, but in any boom-and-bust dynamic this must be one of the factors investigated. By reading this manuscript, the reader has no idea what kind of fishing effort the spiny lobster and sea cucumber populations in this area can sustainably endure. Ideally, the authors need to weave this consideration into the whole manuscript, and if there is no way of adding actual data on this, they need to make a substantial effort to include information from other studies.

The second is that as it stands, this paper is extremely long and gets way too bogged down in the detail. This whole manuscript needs to be clearly structured and significantly tightened. The introduction neglects to adequately develop the relevant background, and can be much improved with examples and references. The most important points are often lost in the detail, and there is a lot of unnecessary repetition, both between sections and within sections. The authors need to go through the manuscript carefully and re-develop it around the main points they are trying to make.

Further detailed comments are listed below.

Introduction

L55: MPAs more than just a topic of discussion - it would be a stronger opening for your introduction to acknowledge their widespread and increasing implementation.

L65: Change "spatial management and integrated management" to "spatial and integrated management".

You could also briefly mention where Marine Spatial Planning (a term widely used in the Western Pacific) comes in.

L68-69: Please provide one or two examples of this, with references.

L69-72: Please provide one or two examples of this, with references.

L78-79: Who is discussing this? Please provide references. A discussion implies some weighing up of pros and cons; please give examples.

L80: Remove the comma after “grounds”.

L88: Change "on" to "for".

L94: Change the phrase to "...no study has yet examined..."

L96: Remove the "s" from "markets".

L102: Change "on" to "to".

L104: Recommendations cannot be mislead. Perhaps you mean something like "misleading management agencies into making inadequate decisions"?

L110: Remove the comma after “drivers”.

L111: Remove the comma after “MPA”.

L113: The Introduction needs to make a case for why this is a good place for this study. You can use some of the information already in the Methods section, to avoid repetition. I have indicated below with section would fit better here than in the Methods.

L120: The management implications of what / who?

L124: A clearer way to frame the goals of this study, which then can also streamline the structure of the paper, is to pose a list of questions. Then the methods, results and discussion sections can be structured accordingly.

Materials & Methods

L133: Who created this division? Please provide a reference.

L141: Change "in" to "across".

L143: Do you mean that it's officially protected as National Parks, or it's just uninhabited?

L178-232: All this could go in the Introduction. It also needs tightening and streamlining; as it is, it's much too long.

L189: Briefly say what this means for environmental conditions around Galapagos.

L198: Change "spillover" to "spill over". In this context, it's being used as a verb.

L243: This is incorrect - it needs to be expressed as "data points" or "records".

L245: Change "daily-basis" to "daily basis".

L286: This is rather hard to follow. It would be much improved by a table of what data were collected when, and with what method. A little of that is contained within Table 1, but this could be moved to an expanded data collection table.

L316: Does this mean you can't tell which one - stocks collapse or global financial crisis - actually drove the profitability of fishing?

L335: What statistic was used, and what software was used?

L379: These are all examples of goals of your analysis that could be framed as questions and added to the end of the Introduction.

L409: Explain the difference between a hotspot and a cold spot.

L417: Should this be diver hours per unit area?

L452: change "being" to "are".

Results

L524: I don't see these illustrated in the figure. One way to show this in the figure itself would be to add arrows for when these events occurred.

L527: The Results section is not the place to try and find reasons for the results - move all these kinds of inferences to the Discussion section. The Results section is simply for describing results.

L621: These key patterns would be more useful if they were moved to the beginning of each section. The authors could begin with the key patterns and the describe some of the detail.

L625: The best place for this next paragraph would be in the Discussion, where it could then be followed by more detailed discussion about these patterns and their reasons and implications.

L684: The best place for this next paragraph would be in the Discussion, where it could then be followed by more detailed discussion about these patterns and their reasons and implications.

L718-720: This is a very awkward way to start - clearly state your main result. This whole following section is way too long. Please tighten it and clearly highlight the key results that you will discuss in the Discussion sections.

Discussion

L952: To make the reader want to read more, highlight your most important and interesting results at the beginning of the Discussion. You only need one sentence to "sell" the novelty of the methods used.

L976-985: This is what you could start the Discussion with.

L989-997: There's no need to re-iterate detailed results. Stick to discussing them in the context of current knowledge, and the implications of your findings.

L999: Insert "The" at the start of this sentence.

L1017-1029: This seems out of place here. Stick to discussing your results.

L1028: This has already been said. This repetition is not helpful and makes the Discussion hard to read.

L1099: This is a little confusing - the discussion about fishing the line further above suggests that marine zoning was implemented to not affect areas preferred for fishing - but here there's a suggestion that zoning did have a significant effect on fisheries.

L1130: This is good - please develop this further by setting it in context of other studies that may have found similar patterns, and then discuss the implications for management.

Summary, Implications, Conclusions – this back end is far too long, please condense.

Tables and Figures

Table 2: This table really belongs with the paragraphs describing the different factors that have influenced these fisheries.

Figure 2: Throughout the caption, please change "relation" to "relationship". Could you add p-values to the regression figures?

Figure 4 caption: Change "three-time" to "three time".

Figures 8-10: There’s no reason to use the abbreviations in the axis titles. Please write them out in full.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Attachment

Submitted filename: Review of PONE_D_19_26580_Castrejon_Charles.pdf

Click here for additional data file.^{(50.1KB, pdf)}

Attachment

Submitted filename: PLoS editorial comment_Spatial fishing patterns in the GMR.docx

Click here for additional data file.^{(25.1KB, docx)}

PLoS One. 2020 Jan 23;15(1):e0228094. doi: 10.1371/journal.pone.0228094.r002

Author response to Decision Letter 0

19 Dec 2019

December 15th, 2019

Heather M. Patterson, Ph.D.

Academic Editor

PlosOne

Dear Dr. Patterson,

Please find enclosed the revised version of the manuscript entitled “Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve” by Mauricio Castrejón (corresponding author) and Anthony Charles.

As required, a revised version of our manuscript that addresses the points raised during the review is submitted as a single Word document file. Given the novel nature of the topic and lack of previous long-term quantitative studies on Galapagos marine zoning, the paper required a longer treatment that the normal article length for PlosOne. Nevertheless, the whole manuscript was edited and reduced, as recommended by reviewers.

Also please find attached a rebuttal letter that contains our responses to each point raised by the academic editor and two reviewers. We are grateful for the insightful and constructive comments made by all of you. We accepted most of your suggestions. We have written the entire manuscript in a more concise manner, particularly the discussion section. We have also modified some figures and tables as suggested by the referees. We feel that the revision process has greatly enhanced the clarity and quality of the paper.

We hope you will be pleased with this new version of the manuscript. Please do not hesitate to contact me with any questions or concerns. Thank you very much for your consideration. We look forward to receiving the acknowledgment of the manuscript and your decision.

Sincerely yours,

Dr. Mauricio Castrejón

Interdisciplinary PhD Program

Dalhousie University

Halifax, Canada

mauricio.castrejon@dal.ca

PONE-D-19-26580

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

Rebuttal letter

We acknowledge useful comments provided by the academic editor and two referees. Most of their suggestions have been included in the revised version of the manuscript. In this letter, we explain how and where the corrections have been incorporated into the text. We follow the editorial order and have numbered referees' suggestions and comments.

Editorial comments

The issue with this manuscript is that it is just far too long and wordy. It would be much more accessible for the reader if it was written in a more concise manner. Instead, it is very repetitive and often includes details that are not germane to the actual study. There are also too many figures and some should be moved to the Supporting Information, too many sub-headings in the methods, results and discussion (every paragraph does not need its own sub-heading) and too many acronyms, that are often not used or are redefined over and over which is unnecessary. Please only introduce an acronym if it is used several times again in the manuscript and then use it consistently; it only has to be defined once. The Discussion is a particular problem and is 17 pages on its own, which is just far too long and there are a number of paragraphs that add little and should be deleted. The text overall needs work and contains quite a few grammatical errors.

R: We have written the entire manuscript in a more concise manner, particularly the discussion section. We have also modified some figures and tables as suggested by the referees. The number of sub-headings in the Methods, Results and Discussion sections was reduced, as well as the number of acronyms. Grammatical errors were corrected.

Line 30: Do not capitalize ‘marine protected areas’

R: Suggestion taken.

Line 31: Looks like the font size at the end of this line changes

R: Font size changed.

Line 35: Write as ‘interpretation of assessments of’

R: Suggestion taken.

Line 36: delete ‘assessments’

R: Suggestion taken.

Line 39: ‘long-term, spatially-explicit’

R: Suggestion taken.

Line 43: delete ‘(GIS)’ here as this is not used again in the abstract

R: Suggestion taken.

Line 55: Do not capitalise ‘protected areas’

R: Suggestion taken.

Line 64: ‘spatially-demarcated areas’

R: Suggestion taken.

Lines 70-71: don’t understand this text offset by en dashes. Doesn’t follow the previous text so either delete or rewrite.

R: Text was edited.

Line 72: Should the ‘and’ before ‘policy’ be deleted? Doesn’t make sense as written.

R: Text was edited.

Line 73: replace the hyphen with a comma and delete the comma after ‘countries’

R: Suggestion taken.

Line 80: delete the comma

R: Suggestion taken.

Line 94: no study has yet examined’

R: Suggestion taken.

Line 95: delete the comma after ‘MPA’

R: Suggestion taken.

Lines 104-105: Not sure what this means ‘potentially misleading….’ How can you mislead something that has been adopted? Please clarify or delete.

R: Text was edited.

Line 110: Delete the comma

R: Suggestion taken.

Line 11: Delete the comma

R: Suggestion taken.

Lines 112-113: Move this to the beginning of the sentence ‘The Galapagos Marine Reserve…..’

R: Suggestion taken.

Line 135: Add ‘the’ before ‘El Nino’

R: Suggestion taken.

Lines 137: Move the ‘according to Edgar’ out of the main title of the figure. Either put it in the notes or just include it in the text.

R: Suggestion taken. Cite included in the text.

Line 141: Write as 25,144 and ‘distributed on’

R: Suggestion taken.

Line 150: Delete ‘to’ after ‘provides’

R: Suggestion taken.

Line 153: Write as ‘large, wooden boats’

R: Suggestion taken.

Line 169: Write as ‘fishing season since 1999’

R: Suggestion taken.

Line 170: delete ‘since 1999’

R: Suggestion taken.

Lines 178-232: This is very long and wordy and much of it does not seem directly relevant to the study. I would cut this back to only the information that is applicable to the study.

R: Suggestion taken. Paragraph was reduced and edited.

Line 179: The abbreviation GMR has already been established and does not need to be established here again so just use ‘GMR’

R: Suggestion taken.

Line 184: The colon should be a comma

R: Text was edited.

Line 184: Do not introduce the abbreviation PMB

R: Text was edited.

Line 185: Do not introduce the abbreviation IMA

R: Text was edited.

Line 189: 1997-98

R: Text was edited.

Line 201: Delete the apostrophe in miles

R: Text was edited.

Line 208: Should be ‘tourist activities’ and would be good to have an example of what that means.

R: Text was edited.

Line 209: tourist activities

R: Text was edited.

Line 216: Need to define CPUE here and define the abbreviation if you are going to use it

R: Text was edited.

Lines 224-232: Most of this is a repeat from the introduction on why the study is being done so delete

R: This paragraph was deleted.

Line 238: ‘spatially-explicit’

R: Suggestion taken.

Line 242: ‘Fisheries-related data’

R: Suggestion taken.

Line 245: Table 1

R: Suggestion taken.

Line 258: delete ‘up’

R: Suggestion taken.

Line 263: 17,723

R: Suggestion taken.

Line 272: ‘from 2009’

R: Suggestion taken.

Line 274: Need to define CPUE at line 216, not here, so just use ‘CPUE’

R: Text edited.

Line 283: delete the comma at the end of the line

R: Suggestion taken.

Line 290: Too many sub-headings, should delete some

R: Suggestion taken. The number of sub-headings was reduced.

Line 292: I would say ‘The most potentially relevant’ because there is no evidence these are relevant.

R: Suggestion taken. Text edited.

Line 296: Again ‘that potentially affected’

R: Suggestion taken. Text edited.

Line 307: ‘boom-and-bust’

R: Suggestion taken.

Line 312: delete ‘the Galapagos’ as this is not necessary

R: Suggestion taken.

Line 313: delete the comma

R: Suggestion taken.

Line 317: Again, delete ‘in Galapagos’

R: Suggestion taken.

Line 343: Don’t need a part 1 and 2 for the sub-heading

R: Suggestion taken. Sub-headings were edited.

Line 346: abbreviation GIS has already been established and does not need to be established here again so just use ‘GIS’

R: Suggestion taken.

Line 357: Delete the apostrophe

R: Suggestion taken.

Lines 370-371: Don’t need a new sub-heading

R: Suggestion taken. Sub-heading deleted.

Line 374: Should be ‘is concentrated’

R: Suggestion taken.

Line 388: ‘fine-scale distribution’

R: Suggestion taken.

Line 407: Why are there vertical lines around 1.96?

R: Vertical lines deleted.

Lines 447-457: Is all this background on BRT required?

R: As BRT model are relatively a new analytical tool, BRT background is needed. Nevertheless, text was reduced and edited.

Lines 450-451: Do not capitalise ‘general linear models’ or ‘general additive models’ and if the abbreviations are not used again delete them

R: Suggestion taken.

Line 489: results should be in the past tense so please check the text

R: Suggestion taken. Text was revised and edited.

Line 507: Long-term variation

R: Suggestion taken.

Line 524: This is confusing as it is not clear to me how the authors have decided that the GFC contributed to this. Is this just due to the time overlap? Not sure how Fig 2 demonstrates this either.

R: Figure 2 and text were edited to explain how fishing capacity varied during the five periods analyzed, including GFC.

Line 531: Use the abbreviation GMR

R: Suggestion taken.

Line 534: Don’t need a part 1 and 2

R: Suggestion taken.

Line 561: The authors introduced an abbreviation for core area and distribution ranges on line 539 so they should use them throughout the manuscript or do not introduce the abbreviation

R: Suggestion taken. Abbreviations were eliminated.

Line 564: Should be ‘were shared’

R: Suggestion taken.

Line 566: Should be ‘showed’

R: Suggestion taken.

Line 571: Should be ‘trends of the’

R: Suggestion taken.

Lines 588-589: This is very awkward and the results section is not the place for this. Just write is as a simple statement of the results (not in italics) ‘Based on these results there is a pattern where…..’

R: Text was edited.

Lines 619-620: Same comment as above.

R: Text was edited.

Line 626: ‘boom-and-bust’

R: Suggestion taken.

Line 630: Replace ‘In consequence’ with ‘Therefore’

R: Suggestion taken.

Line 634: Don’t need a part 2

R: Suggestion taken.

Line 686: ‘fishing effort was concentrated’ and delete the apostrophe at the end of the line.

R: Suggestion taken.

Line 688: boom-and-bust

R: Suggestion taken.

Line 698: replace the dash with a comma

R: Suggestion taken.

Line 702: ‘This type of fisher’

R: Suggestion taken.

Line 721: pearson’s

R: Suggestion taken.

Line 724: 35.73%, 32.66%

R: Suggestion taken.

Line 725: Pearson’s

R: Suggestion taken.

Line 728: delete ‘VI’

R: Suggestion taken.

Line 731: Use the abbreviation BRT

R: Suggestion taken.

Line 745: Just use the abbreviation RN

R: Suggestion taken.

Line 768: Need a comma after ‘islands’

R: Suggestion taken.

Lines 781-782: This is confusing as written. Notes there was a decline but the revenue goes up to $11,000? Also, write as US$2000 etc. Please change throughout manuscript.

R: Text was edited to improve understanding.

Line 793: delete ‘VI’

R: Suggestion taken.

Line 811: replace the dash with a comma

R: Suggestion taken.

Line 819: delete ‘the random number variable’

R: Suggestion taken.

Line 829: Replace the dashes with commas

R: Suggestion taken.

Line 845: Delete ‘a value of’

R: Suggestion taken.

Line 854: delete ‘VI’

R: Suggestion taken.

Line 860: delete ‘VI’

R: Suggestion taken.

Line 866: delete ‘VI’

R: Suggestion taken.

Line 870: Write as US$2500

R: Suggestion taken.

Line 871: same as above

R: Suggestion taken.

Line 873: same as above

R: Suggestion taken.

Line 886: Should be ‘islands’

R: Suggestion taken.

Line 942: delete the apostrophe and write as ‘fishing fleet core areas’

R: Suggestion taken.

Line 944: ‘boom-and-bust’

R: Suggestion taken.

Line 950: Too many sub-headings

R: Suggestion taken. Sub-headings were reduced.

Lines 956-962: This text is repetitive and unnecessary so delete and move the next paragraph up to follow the opening sentence of the paragraph.

R: Suggestion taken. Text was deleted and Discussion was edited.

Lie 970: ‘boom-and-bust’

R: Suggestion taken.

Line 981: GMR

R: Suggestion taken.

Line 985: ‘fishing fleet core areas’

R: Suggestion taken.

Line 994: Don’t redefine the abbreviation nearNTZ, just use it.

R: Suggestion taken.

Line 1000: GMR

R: Suggestion taken.

Line 1008: Move ‘historically’ to line 1009 after ‘areas’

R: Suggestion taken.

Line 1045: ‘these adaptive responses’? what responses?

R: Text edited.

Line 1048: ‘although this was not the case for Puerto Ayora’

R: Suggestion taken.

Line 1053: delete ‘on the water’

R: Suggestion taken.

Line 1065: Reference not formatted correctly, needs a reference number

R: References were formatted.

Line 1066: Do not capitalise ‘discrete choice models’ and delete the acronym as it is not used again.

R: Suggestion taken.

Line 1074: Replace the dash with a comma

R: Suggestion taken.

Line 1075: replace the dash with a comma

R: Suggestion taken.

Line 1082: Delete ‘fishing fleets’

R: Suggestion taken.

Line 1086: Reference not formatted correctly, needs a reference number

R: References were formatted.

Line 1111: delete the comma and ‘on the water’

R: Suggestion taken.

Line 1133: CPUE has already been defined and does not need to be defined again here so just use CPUE

R: Suggestion taken.

Line 1145: Delete ‘has’

R: Suggestion taken.

Line 1149: delete ‘probably’

R: Suggestion taken.

Line 1150: add ‘likely’ before ‘irrelevant’

R: Suggestion taken.

Lines 1153-1165: This is very confusing. I have no idea why the authors are providing a summary half way through the Discussion. This is unnecessary and repetitive so please delete.

R: Suggestion taken. This paragraph was deleted.

Line 1166: Use ‘GMR’

R: Suggestion taken.

Line 1173: Use ‘GMR’

R: Suggestion taken.

Lines 1178-1183: This is a sentence, not a paragraph, so please move it up to join the paragraph above.

R: Suggestion taken.

Line 1178: Delete both commas

R: Suggestion taken.

Line 1180: Use ‘GMR’

R: Suggestion taken.

Line 1185: replace the dash with a comma

R: Suggestion taken.

Line 1187: replace the dash with a comma

R: Suggestion taken.

Line 1189: Use ‘GMR’

R: Suggestion taken.

Line 1192: Use ‘GMR’

R: Suggestion taken.

Line 1197: Use ‘GMR’

R: Suggestion taken.

Line 1205: Do not introduce the abbreviation MSC

R: Suggestion taken.

Lines 1208-1216: Move this up to join the previous paragraph (which is actually a sentence).

R: Suggestion taken.

Line 1210: Use ‘GMR’

R: Suggestion taken.

Line 1222: Replace ‘usefulness’ with ‘utility’

R: Suggestion taken.

Line 1223: EBM has already been defined and does not need to be defined again here so just use EBM

R: Text was edited.

Line 1226: Use ‘GMR’

R: Suggestion taken.

Line 1231: replace ‘over’ with ‘on’ and do not need to provide the scientific name as this has already been established

R: Suggestion taken.

Line 1235: Delete ‘Using non-parametric statistics’

R: Suggestion taken.

Line 1253: Do not introduce the abbreviation SBCI

R: Suggestion taken.

Line 1254: replace ‘over’ with ‘on’ and abbreviate as P. interruptus

R: Suggestion taken.

Line 1257: Should be vs.

R: Suggestion taken.

Line 1259: Don’t need z=0.59

R: Text was edited.

Line 1260: Spell out SBCI

R: Text was edited

Lines 1263-1270: This is very repetitive and adds nothing to the discussion in my view so I would delete.

R: Suggestion taken. These lines were deleted.

Line 1283: Delete ‘enabling’

R: Suggestion taken.

Line 1290: Delete ‘enabling’

R: Suggestion taken.

Line 1293: ‘based on the core areas and distribution ranges of fishing fleets’

R: Suggestion taken.

Line 1303: Delete ‘fishing fleets’

R: Suggestion taken.

Lines 1305-1315: Again, I find this text unnecessary and it could be deleted.

R: Text was deleted.

Lines 1319-1323: This is a sentence, not a paragraph

R: Text was edited.

Line 1321: EBM has already been defined and does not need to be defined again here so just use EBM

R: Text was edited.

Line 1328: ‘should not be taken for granted; MPAs are not a panacea. (delete the rest ‘i.e. a one-size…..’

R: Suggestion taken.

Line 1332: ‘with global climate change’

R: Suggestion taken.

Line 1336: delete ‘learned’ and write as ‘is that assessments of MPA effectiveness’

R: Suggestion taken.

Reviewer 1

Major Comments.

R: Suggestion taken. Both points suggested are mentioned in the Introduction.

R: The deviance explained is not partitioned among the explanatory predictor variables. Table 4 and Figures 7-10 show the variable importance (VI), which was estimated by averaging the number of times a variable is selected for splitting and the squared improvement resulting from these splits. VI scores provide a measure of the relative influence of predictor variables used to build the model. Values are scaled so that the sum adds to 100, with higher numbers indicating a stronger influence on the response variable. This explanation is provided in Lines 730-735.

R: We agree that a finer scale probably would be needed to evaluate a spiny lobster spillover effect, but this is not the objective of this study. We evaluated finer and coarser spatial scales to conduct the hotspot analysis and the 2.25km2 scale was the most proper scale to visualize the results and to evaluate the presence of a fishing the line effect around no-take zones. On the other hand, a fishing the line effect around the Galapagos Marine Reserve was detected by Bucaram at al. 2018 using a coarser scale of analysis (see “Assessing fishing effects inside and outside an MPA: The impact of the Galapagos Marine Reserve on the Industrial pelagic tuna fisheries during the first decade of operation”). Therefore, we think that a 2.25km2 is an appropriate spatial scale of analysis for our case study.

4. The Discussion is far too long and repetitive (17 pages, with a Summary of almost 7 pages). This should be condensed considerably.

R: Suggestion taken. Discussion was considerably reduced and edited.

R: We reviewed Fletcher et al (2015) and, even though they recognize that MPAs can be affected by diverse drivers of change, they do not evaluate their impact in a quantitative way as we did. Therefore, we edited the text in the following way: “…to our knowledge, no study has examined yet, in a quantitative way, how fishers respond to those situations in which they have to cope simultaneously with implementation of an MPA, and with the interaction of external drivers…”.

Minor Comments. Abstract.

Line 30. MP Area (omit s).

R: Suggestion rejected. MPAs is a term commonly used in the scientific literature.

Line 31. Note change in font size of text at full stop.

R: Suggestion taken.

Line 37. Unfeasible (not infeasible).

R: Suggestion taken.

Introduction

R: Suggestion taken. Cite added.

R: Suggestion taken. We replace Kellner reference by Halpern et al (2010) and Kerwarth et al (2013). Line 198 was eliminated.

Lines 93-97. Note major point 5 above.

R: Same response as in point 5 above.

Lines 116-117. Indicate here the year when the MPAs were implemented (2000).

R: Suggestion taken. Text was edited.

Materials and Methods.

Line 198. “….and spillover to fishing grounds may occur ([26]” Lines 208, 209. Tourist or tourism (not touristic).

R: Suggestion taken.

Lines 273-274. Why calculate effort by dividing catch by catch-per-unit-effort (CPUE)? Surely you measured catch and effort directly to calculate CPUE?

R: We estimated total fishing effort by dividing total catch by average CPUE per fishing season. We conducted this analysis to estimate the amount of sampling effort per fishing season. For all analysis presented in the paper, we used CPUE data estimated directly from catch and effort data per fishing trip.

Line 302. ..affected by the potential drivers (add potential).

R: Suggestion taken.

Lines 310-312. You make it clear that the re-zoning was confounded by the sea- cucumber over-exploitation phase (see also Table 3). Thus, when you talk of changes to effort associated with the zoning (e.g. Lines: 600-605, 989-990, 1035- 1036) you must acknowledge this confounding. At lines 1096-1099 you DO acknowledge the confounding, and should in other places in the manuscript.

R: Suggestion taken. Text was edited.

Lines 367, 401, 407. Insert “the” before: normality assumption, input field, z score.

R: Suggestion taken.

Results.

Line 505. (Fig. 2a, b, c) should read (Fig. 2d, e, f). Line 520. (Fig. 2d, e, f) should read (Fig. 2a, b, c).

R: Suggestion taken.

Line 601. Acknowledge confounding of zoning and sea-cucumber over-exploitation phase.

R: Suggestion taken.

Line 644. ..the eastern part, ..the southeastern part (insert the). Line 678. Fishers (add s).

R: Suggestions taken.

Line 702. These types of fishers. Line 718. Suggest (not suggests). Line 888. Western side of

R: Suggestion taken.

Discussion.

Lines 952-954. Note comment re Fletcher et al (2015) and Hughes et al (2016) above.

R: Same response as in point 5 above.

Lines1174-1176. Good point. The lobster recovery may not be related to the implementation of the MPAs.

R: No response needed.

Lines 1188-1190 and 1218-1220. When suggesting a re-evaluation of the MPA zoning, you must be clear about why the MPAs were established: conservation, fisheries management, or both.

R: Suggestions taken. Text was edited as: “…we suggest re-evaluating the distribution of no-take zones across the GMR to promote the sustainability of the spiny lobster fishery and conserve key biodiversity areas”.

Lines 1194. The TURFS suggestion outside the MPAs is a good one. Lines 1199-1201. Alcala and Russ (2006) could be cited here also.

R: Suggestions taken. Cite added.

Lines 1224-1227. Why would an MPA network placed in a biased manner help the fishery if it was set up to avoid the fishery?

R: We agree. However, there was no scientific evidence about the long-term impact of no-take zones on the fishing effort dynamic for the spiny lobster fishery before this study. Our study highlights the need to redistribute no-take zones to accomplish conservation and fishery management objectives.

Line 1238. ..replicates.

R: Suggestion taken.

Line 1241. Thirdly (not Fourthly).

R: Suggestion taken.

Lines 1249-1262. In addition to the Kay example in the Channel Islands, which is a good one, you could also mention the Goni et al (2010) lobster example from the Mediterranean.

R: During the edition of the Discussion, we decided to eliminate Lines 1249-1262. We took this decision to put more emphasis in the discussion of other results directly associated to the objectives of our study.

Line 1264. Ref. 70 in support of the idea of spillover is inadequate. See references to cite on spillover suggested above.

R: The academic editor suggested to eliminate Lines 1263-1270. We accepted this suggestion as we decide to put more emphasis in other results of the paper.

Line 1291. To support (not the support).

R: Suggestion taken.

Fig. 2. What do the dark grey and light grey shaded areas of time represent? El Nino/La Nina? Specify in caption.

R: Dark grey and light grey areas represents the periods analyzed. Figure 2 has been edited. The name of the periods was added. Caption was edited.

Figs. 3 and 4. What are the units here? Effort (diver hours)? Specify in caption.

R: Standard deviation ellipses (SDE) polygons represent graphical summaries of the central tendency, dispersion and directional trends of fishing fleets. Core areas and distribution ranges refer to those areas covering 68% (1 SDE) and 95% (2 SDE) of the full spatial extent of fishing fleet distribution, respectively. Explanation is provided in Lines 324-327 of the revised manuscript.

Figs. 3-6. I find it difficult to differentiate Fig. 3 from 4, or Fig. 5 from Fig. 6, simply by eye.

R: Text was edited to improve explanation of spatial patterns. References to Figures 3-6 was added to the text.

R: Full name of acronyms was added to caption of Figures 7-10.

Fig. 7. I agree, NearNTZ has no pattern.

R: No response needed.

Table 1. Caption. Sampling method (not smapling). Table 3. Caption Line 2: occurring (not occurred).

R: Text edited.

Table 4. Perhaps call the variables “Predictor Variables” in the caption?

R: Suggestion taken.

Reviewer 2

General comments

R: As the objective of this study was to predict fishing effort distribution rather than catch or catch-per-unit-effort (CPUE), we focused our analysis on the human element (effort), rather than the interaction between humans and the target species themselves (catch or CPUE). This approach helped us to simplify the interpretation of the results and more accurately predict fishing effort. On the other hand, as overfishing of the spiny lobster fishery is the consequence of the external drivers of change analyzed in this study, particularly of the boom-and-bust exploitation of the sea cucumber fishery, we did not consider overfishing explicitly as a driver of change. Nevertheless, we analyzed the factors influencing fishing effort, including the previous lobster catch and sea cucumber revenues, which were relevant as fishing effort predictors. These two predictors were affected by the overexploitation of the sea cucumber and spiny lobster fisheries. We edited the text to highlight this fact in the manuscript (Lines 1988-1991 of the revised manuscript). In addition, we explained in the Discussion that the spiny lobster fishery was overexploited due to the overcapitalization caused by the expansion of the sea cucumber fishery and explained the consequences of overexploitation on fishing capacity (Lines 1858-1853).

R: We have written the entire manuscript in a more concise manner, particularly the discussion section. Introduction was improved by including relevant background about the Galapagos Marine Reserve. The Methods, Results and Discussion sections were restructured, reduced and edited to avoid unnecessary repetition.

Further detailed comments are listed below.

Introduction

L55: MPAs more than just a topic of discussion - it would be a stronger opening for your introduction to acknowledge their widespread and increasing implementation.

R: Suggestion taken. Introduction was edited following reviewer’s suggestions.

L65: Change "spatial management and integrated management" to "spatial and integrated management".You could also briefly mention where Marine Spatial Planning (a term widely used in the Western Pacific) comes in.

R: Text edited. We did not mention the term Marine Spatial Planning. Instead, we edited the introduction to put more emphasis on the widespread and increasing implementation of MPAs and to explain in a better way the justification for this study.

L68-69: Please provide one or two examples of this, with references.

R: These lines were eliminated, but additional references were added at the beginning of the introduction.

L69-72: Please provide one or two examples of this, with references.

R: These lines were eliminated, but additional references were added at the beginning of the introduction.

L78-79: Who is discussing this? Please provide references. A discussion implies some weighing up of pros and cons; please give examples.

R: These lines were eliminated and Introduction was edited.

L80: Remove the comma after “grounds”.

R: Suggestion taken.

L88: Change "on" to "for".

R: Suggestion taken.

L94: Change the phrase to "...no study has yet examined..."

R: Suggestion taken.

L96: Remove the "s" from "markets".

R: Suggestion taken.

L102: Change "on" to "to".

R: Suggestion taken.

L104: Recommendations cannot be mislead. Perhaps you mean something like "misleading management agencies into making inadequate decisions"?

R: Text was edited.

L110: Remove the comma after “drivers”.

R: Suggestion taken.

L111: Remove the comma after “MPA”.

R: Suggestion taken.

R: Suggestion taken. We moved some information from the Method section to the Introduction section to make stronger our case to conduct this study in the Galapagos Marine Reserve.

L120: The management implications of what / who?

R: Suggestion taken. Management implications for fisheries management.

R: Suggestion taken. We framed in a better way the goals of this study and restructured the Results and Discussion sections.

Materials & Methods

L133: Who created this division? Please provide a reference.

R: Suggestion taken. A reference was provided.

L141: Change "in" to "across".

R: Suggestion taken.

L143: Do you mean that it's officially protected as National Parks, or it's just uninhabited?

R: Text edited. It is officially protected as a National Park.

L178-232: All this could go in the Introduction. It also needs tightening and streamlining; as it is, it's much too long.

R: Suggestion taken. The Method section was reduced.

L189: Briefly say what this means for environmental conditions around Galapagos.

R: We think that this is not necessary as we mentioned at the beginning of the Method section that El Niño has a strong influence in the abundance and distribution of fish and macro-invertebrate assemblages.

L198: Change "spillover" to "spill over". In this context, it's being used as a verb.

R: Suggestion taken.

L243: This is incorrect - it needs to be expressed as "data points" or "records".

R: Suggestion taken.

L245: Change "daily-basis" to "daily basis".

R: Suggestion taken.

R: Text was reduced and edited for better understanding. In addition, Table S1 provides a summary of the fishery monitoring data gathered for the spiny lobster fishery at the three main ports of the Galapagos Marine Reserve from 1997 to 2011.

L316: Does this mean you can't tell which one - stocks collapse or global financial crisis - actually drove the profitability of fishing?

R: Both drivers caused economic perturbations that affected the profitability of the small-scale fishing sector in Galapagos. For this reason they were grouped together in the same period.

L335: What statistic was used, and what software was used?

R: Suggestion taken. This information is now provided in the text.

L379: These are all examples of goals of your analysis that could be framed as questions and added to the end of the Introduction.

R: Suggestion taken. We framed in a better way the goals of this study in the Introduction and restructure the Results and Discussion sections.

L409: Explain the difference between a hotspot and a cold spot.

R: An explanation is provided in Lines 623-627 of the revised paper.

L417: Should this be diver hours per unit area?

R: No, just diver-hours.

L452: change "being" to "are".

R: Suggestion taken.

Results

L524: I don't see these illustrated in the figure. One way to show this in the figure itself would be to add arrows for when these events occurred.

R: Suggestion taken. Figure 2 has been edited. The name of the periods was added.

R: Suggestion taken. Inferences were moved to the Discussion section.

L621: These key patterns would be more useful if they were moved to the beginning of each section. The authors could begin with the key patterns and the describe some of the detail.

R: Suggestion taken. Results section was edited following reviewer’s comments.

L625: The best place for this next paragraph would be in the Discussion, where it could then be followed by more detailed discussion about these patterns and their reasons and implications.

R: Suggestion taken. Paragraph eliminated and integrated in the Discussion.

L684: The best place for this next paragraph would be in the Discussion, where it could then be followed by more detailed discussion about these patterns and their reasons and implications.

R: Suggestion taken. Paragraph eliminated and integrated in the Discussion.

R: Suggestion taken. Paragraph was reduced and edited.

Discussion

R: Suggestion taken. Discussion was edited to highlight our most important and interesting results.

L976-985: This is what you could start the Discussion with.

R: Suggestion taken. Discussion was edited.

L989-997: There's no need to re-iterate detailed results. Stick to discussing them in the context of current knowledge, and the implications of your findings.

R: Suggestion taken. Discussion was edited.

L999: Insert "The" at the start of this sentence.

R: Suggestion taken.

L1017-1029: This seems out of place here. Stick to discussing your results.

R: Suggestion taken. These lines were edited.

L1028: This has already been said. This repetition is not helpful and makes the Discussion hard to read.

R: Suggestion taken. Text was edited.

R: Suggestion taken. Text was edited to improve understanding.

L1130: This is good - please develop this further by setting it in context of other studies that may have found similar patterns, and then discuss the implications for management.

R: Suggestion taken. We have added a paragraph (Lines 910-930) that address the points raised by the reviewer.

Summary, Implications, Conclusions – this back end is far too long, please condense.

R: Suggestion taken. Text was reduced and edited.

Tables and Figures

Table 2: This table really belongs with the paragraphs describing the different factors that have influenced these fisheries.

R: Suggestion taken. Table 2 was moved to Study area section.

Figure 2: Throughout the caption, please change "relation" to "relationship". Could you add p-values to the regression figures?

R: Suggestion taken. P-values were added to regression figures.

Figure 4 caption: Change "three-time" to "three time".

R: Suggestion taken.

Figures 8-10: There’s no reason to use the abbreviations in the axis titles. Please write them out in full.

R: Full names for some predictor variables, such as NearNTZ, are too long to be added to the axis titles. Instead, full name of abbreviations was added to legends of Figures 7-10.

Attachment

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.^{(304.1KB, pdf)}

PLoS One. doi: 10.1371/journal.pone.0228094.r003

Decision Letter 1

Heather M Patterson

31 Dec 2019

PONE-D-19-26580R1

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

PLOS ONE

Dear Dr. Castrejón,

The manuscript is much improved so I thank the authors for their efforts and Reviewer 2 is satisfied that all the suggestions have been address (Reviewer 1 was not available). That said, Reviewer 2 has noted that the Discussion still requires some work and I agree. The authors should emphasis the important findings first, rather than leading with the 'fishing the line' story. I still think the Discussion could be trimmed down to be more concise as well so I encourage the authors to make their revisions with that in mind. I have some minor editorial corrections as well.

That said, I think these changes can be made relatively quickly and that a more focused and concise Discussion will improve the paper. Once these changes have been made the paper can be accepted. I look forward to seeing the final version of this interesting and timely paper in the near future.

We would appreciate receiving your revised manuscript by Feb 14 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Heather M. Patterson, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #2: I commend the authors for the considerable effort they have put into their revisions. Points where they opted not to make changes were adequately explained.

However, the Discussion is still a bit of a mess. Somehow now the discussion starts with the "fishing the line" story, which is not really one of the main points this paper is making. The following paragraphs then jump around various topics, and the main points are mostly still lost in the detail. My suggestion is: 1) Present all the broad topics and key points in the opening paragraph. 2) Before writing the text, list all the key points and sub-points in order of importance (or sequentially as they appear in the results section. 3) Organize the text accordingly. The content is all there, it just needs to be organized and edited for flow. It's also still very long. Consider what sentences you could lose without affecting the content or the message.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Attachment

Submitted filename: PLoS editorial comment_GMR.docx

Click here for additional data file.^{(12KB, docx)}

PLoS One. 2020 Jan 23;15(1):e0228094. doi: 10.1371/journal.pone.0228094.r004

Author response to Decision Letter 1

6 Jan 2020

January 5th, 2020

Heather M. Patterson, Ph.D.

Academic Editor

PlosOne

Dear Dr. Patterson,

Also please find attached a rebuttal letter that contains our responses to each point raised by the academic editor and reviewers. We are grateful for the insightful and constructive comments made by all of you. We accepted all your suggestions. We have written the discussion in a more focused and concise manner, as requested.

Sincerely yours,

Dr. Mauricio Castrejón

Interdisciplinary PhD Program

Dalhousie University

Halifax, Canada

mauricio.castrejon@dal.ca

PONE-D-19-26580R1

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

Rebuttal letter

We acknowledge useful comments provided by the academic editor and one referee. All their suggestions have been included in the revised version of the manuscript. In this letter, we explain how and where the corrections have been incorporated into the text. We follow the editorial order and have numbered referees' suggestions and comments.

Editorial comments

The manuscript is much improved, so I thank the authors for their efforts and Reviewer 2 is satisfied that all the suggestions have been address (Reviewer 1 was not available). That said, Reviewer 2 has noted that the Discussion still requires some work and I agree. The authors should emphasis the important findings first, rather than leading with the 'fishing the line' story. I still think the Discussion could be trimmed down to be more concise as well so I encourage the authors to make their revisions with that in mind. I have some minor editorial corrections as well.

R: We have written the Discussion section in a more focused and concise manner, emphasizing the most important findings first. We shifted the material around and consolidated the discussion into four main subsections. The first of these is about the 'big picture' of change in the Galapagos, in response to the drivers of change analyzed. The second subsection focuses on the detailed assessment of what caused dynamics of fishing effort. The third brings together all the no-take zones analysis. The fourth summarizes the conclusions and lessons learned. The whole Discussion section was edited and reduced.

Minor Comments

Line 30: Rewrite as ‘Assessments of the effectiveness of marine protected areas (MPAs)’

R: Suggestion taken.

Line 65: Write as ‘particularly those designed for multiple use’ (no hyphen in multiple use here’

R: Suggestion taken.

Line 67: I think this should be ‘in which as assessment of the performance of MPAs’

R: Suggestion taken.

Line77: Should be ‘processes’

R: Suggestion taken.

Line 146: Do not introduce the abbreviation ‘ENSO’ as it is only used once again in the manuscript. It is PLoS formatting that abbreviations must be used at least 3 times if they are introduced.

R: Suggestion taken.

Line 185: Just abbreviate the scientific name here as P. penicillatus as the full name has already been provided

R: Suggestion taken.

Line 194: I would use MPA rather than marine reserve here to be consistent in the terminology.

R: Suggestion taken.

Line 234: Do not introduce the abbreviation ‘GPS’ as it is not used again in the manuscript.

R: Suggestion taken.

Line 309: What is ‘GNP’? This has not been defined yet. I would just spell it out.

R: Suggestion taken.

Line 403: Do not introduce the abbreviation ‘ENSO, just spell it out.

R: Suggestion taken.

Line 1057: Should be ‘shellfish fisheries’

R: Suggestion taken.

Attachment

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.^{(218.1KB, pdf)}

PLoS One. doi: 10.1371/journal.pone.0228094.r005

Decision Letter 2

Heather M Patterson

8 Jan 2020

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

PONE-D-19-26580R2

Dear Dr. Castrejón,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Heather M. Patterson, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0228094.r006

Acceptance letter

Heather M Patterson

14 Jan 2020

PONE-D-19-26580R2

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: the Galapagos Marine Reserve

Dear Dr. Castrejón:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Heather M. Patterson

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Summary of the fishery monitoring data gathered for the spiny lobster fishery at the three main ports of the Galapagos Marine Reserve from 1997 to 2011.

(DOCX)

Click here for additional data file.^{(17.1KB, docx)}

(DOCX)

Click here for additional data file.^{(13.6KB, docx)}

(DOCX)

Click here for additional data file.^{(21.2KB, docx)}

Attachment

Submitted filename: Review of PONE_D_19_26580_Castrejon_Charles.pdf

Click here for additional data file.^{(50.1KB, pdf)}

Attachment

Submitted filename: PLoS editorial comment_Spatial fishing patterns in the GMR.docx

Click here for additional data file.^{(25.1KB, docx)}

Attachment

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.^{(304.1KB, pdf)}

Attachment

Submitted filename: PLoS editorial comment_GMR.docx

Click here for additional data file.^{(12KB, docx)}

Attachment

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.^{(218.1KB, pdf)}

Data Availability Statement

[pone.0228094.ref001] 1.Worm B, Hilborn R, Baum JK, Branch TA, Collie JS, Costello C, et al. Rebuilding Global Fisheries. Science. 2009;325: 578–585. 10.1126/science.1173146 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref002] 2.Defeo O, Castrejón M, Pérez-Castañeda R, Castilla JC, Gutiérrez NL, Essington TE, et al. Co-management in Latin American small-scale shellfisheries: assessment from long-term case studies. Fish Fish. 2016;17: 176–192. 10.1111/faf.12101 [DOI] [Google Scholar]

[pone.0228094.ref003] 3.Micheli F, Saenz-Arroyo A, Greenley A, Vazquez L, Espinoza Montes JA, Rossetto M, et al. Evidence that marine reserves enhance resilience to climatic impacts. PLoS One. 2012;7: e40832 10.1371/journal.pone.0040832 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228094.ref004] 4.Alcala A, Russ G. No-take marine reserves and reef fisheries management in the Philippines: a new people power revolution. Ambio. 2006;35: 245–254. 10.1579/05-a-054r1.1 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref005] 5.OECD. Marine protected areas: economics, management and effective policy mixes Paris: OECD Publishing; 2017. [Google Scholar]

[pone.0228094.ref006] 6.Maestro M, Pérez-Cayeiro M, Chica Ruiza J, Reyez H. Marine protected areas in the 21st century: current situation and trends. Ocean Coast Manag. 2019;171: 28–36. [Google Scholar]

[pone.0228094.ref007] 7.Guarderas AP, Hacker SD, Lubchenco J. Current status of marine protected areas in Latin America and the Caribbean. Conserv Biol. 2008;22: 1630–1640. 10.1111/j.1523-1739.2008.01023.x [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref008] 8.Guarderas AP, Hacker SD, Lubchenco J. Ecological effects of marine reserves in Latin America and the Caribbean. Mar Ecol Prog Ser. 2011;429: 219–225. 10.3354/meps09103 [DOI] [Google Scholar]

[pone.0228094.ref009] 9.Charles A, Wilson L. Human dimensions of marine protected areas. ICES J Mar Sci. 2009;66: 6–15. [Google Scholar]

[pone.0228094.ref010] 10.Charles AT. Human dimensions in marine ecosystem-based management In: Fogarty MJ, McCarthy J, editors. The Sea, Volume 16: Marine Ecosystem-Based Management. Cambridge: Harvard University Press; 2014. [Google Scholar]

[pone.0228094.ref011] 11.Christie P, Bennett N, Gray N, Wilhelm T, Lewis N, Parks J, et al. Why people matter in ocean governance: incorporating human dimensions into large-scale marine protected areas. Mar Policy. 2017;84: 273–284. [Google Scholar]

[pone.0228094.ref012] 12.Salas S, Gaertner D. The behavioural dynamics of fishers: management implications. Fish Fish. 2004;5: 153–167. 10.1111/j.1467-2979.2004.00146.x [DOI] [Google Scholar]

[pone.0228094.ref013] 13.Stelzenmüller V, Maynou F, Bernard G, Cadiou G, Camilleri M, Crec’hriou R, et al. Spatial assessment of fishing effort around European marine reserves: implications for successful fisheries management. Mar Pollut Bull. 2008;56: 2018–2026. 10.1016/j.marpolbul.2008.08.006 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref014] 14.Horta e Costa B, Batista MI, Gonçalves L, Erzini K, Caselle JE, Cabral HN, et al. Fishers’ behaviour in response to the implementation of a marine protected area. PLoS One. 2013;8: e65057 10.1371/journal.pone.0065057 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228094.ref015] 15.Halpern BS, Lester S, Kellner J. Spillover from marine reserves and the replenishment of fished stocks. Env Conserv. 2010;36: 268–276. 10.1017/S0376892910000032 [DOI] [Google Scholar]

[pone.0228094.ref016] 16.Kerwath SE, Winker H. Marine protected area improves yield without disadvantaging fishers. Nat Commun. 2013;4: 2347 10.1038/ncomms3347 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref017] 17.Charles A, Westlund L, Bartley DM, Fletcher WJ, Garcia S, Govan H, et al. Fishing livelihoods as key to marine protected areas: insights from the World Parks Congress. Aquat Conserv Mar Freshw Ecosyst. 2016;26: 165–184. 10.1002/aqc.2648 [DOI] [Google Scholar]

[pone.0228094.ref018] 18.Westlund L, Charles A, Garcia SM, Sanders J. Marine protected areas: interactions with fishery livelihoods and food security Rome: FAO Fisheries Aquaculture Technical Paper; 603; 2017. [Google Scholar]

[pone.0228094.ref019] 19.Soykan CU, Eguchi T, Kohin S, Dewar H. Prediction of fishing effort distributions using boosted regression trees. Ecol Appl. 2014;24: 71–83. 10.1890/12-0826.1 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref020] 20.Bucaram SJ, White JW, Sanchirico JN, Wilen JE. Behavior of the Galapagos fishing fleet and its consequences for the design of spatial management alternatives for the red spiny lobster fishery. Ocean Coast Manag. 2013;78: 88–100. 10.1016/j.ocecoaman.2013.03.001 [DOI] [Google Scholar]

[pone.0228094.ref021] 21.Edgar GJ, Bustamante RH, Fariña JM, Calvopiña M, Martínez C, Toral-Granda M V. Bias in evaluating the effects of marine protected areas: the importance of baseline data for the Galapagos Marine Reserve. Environ Conserv. 2004;31: 212–218. 10.1017/S0376892904001584 [DOI] [Google Scholar]

[pone.0228094.ref022] 22.Castrejón M, Charles A. Improving fisheries co-management through ecosystem-based spatial management: the Galapagos Marine Reserve. Mar Policy. 2013;38: 235–245. 10.1016/j.marpol.2012.05.040 [DOI] [Google Scholar]

[pone.0228094.ref023] 23.Viteri C, Chávez C. Legitimacy, local participation, and compliance in the Galápagos Marine Reserve. Ocean Coast Manag. 2007;50: 253–274. 10.1016/j.ocecoaman.2006.05.002 [DOI] [Google Scholar]

[pone.0228094.ref024] 24.Szuwalski CS, Castrejon M, Ovando D, Chasco B. An integrated stock assessment for red spiny lobster (Panulirus penicillatus) from the Galapagos Marine Reserve. Fish Res. 2016;177: 82–94. 10.1016/j.fishres.2016.01.002 [DOI] [Google Scholar]

[pone.0228094.ref025] 25.Defeo O, Castrejón M, Ortega L, Kuhn AM, Gutiérrez NL, Castilla JC. Impacts of climate variability on Latin American small-scale fisheries. Ecol Soc. 2013;18: 30 10.5751/ES-05971-180430 [DOI] [Google Scholar]

[pone.0228094.ref026] 26.DPNG. Plan de manejo de las áreas protegidas de Galápagos para el buen vivir. Izurieta A, Tapia W, Mosquera G, Chamorro S, editors. Galapagos, Ecuador: Dirección del Parque Nacional Galápagos; 2014.

[pone.0228094.ref027] 27.Edgar GJ, Banks S, Fariña JM, Calvopiña M, Martínez C. Regional biogeography of shallow reef fish and macroinvertebrate communities in the Galapagos Archipelago. J Biogeogr. 2004;31: 1107–1124. [Google Scholar]

[pone.0228094.ref028] 28.Wolff M, Ruiz D, Taylor M. El Niño induced changes to the Bolivar Channel ecosystem (Galapagos): comparing model simulations with historical biomass time series. Mar Ecol Prog Ser. 2012;448: 7–22. 10.3354/meps09542 [DOI] [Google Scholar]

[pone.0228094.ref029] 29.Bustamante RH, Reck G, Ruttenberg BI, Polovina J. The Galápagos spiny lobster fishery In: Phillips BF, Kittaka J, editors. Spiny lobsters: Fisheries and culture. Fishing News Books; 2000. [Google Scholar]

[pone.0228094.ref030] 30.PNG. Plan de Manejo del Parque Nacional Galápagos: un pacto por la conservación y desarrollo sustentable del archipiélago Galapagos, Ecuador: Parque Nacional Galápagos; 2005. [Google Scholar]

[pone.0228094.ref031] 31.Epler B. Tourism, economy, population growth, and conservation in Galapagos Galapagos: Charles Darwin Foundation; 2007. [Google Scholar]

[pone.0228094.ref032] 32.Reyes H, Ramírez J. Informe técnico de la pesquería de langosta espinosa (Panulirus penicillatus y Panulirus gracilis) 2012 en la Reserva Marina de Galápagos Galapagos: Dirección del Parque Nacional Galápagos; 2012. [Google Scholar]

[pone.0228094.ref033] 33.Heylings P, Bravo M. Evaluating governance: a process for understanding how co-management is functioning, and why, in the Galapagos Marine Reserve. Ocean Coast Manag. 2007;50: 174–208. [Google Scholar]

[pone.0228094.ref034] 34.Heylings P, Bensted-Smith R, Altamirano M. Zonificación e historia de la Reserva Marina de Galápagos In: Danulat E, Edgar GJ, editors. Reserva Marina de Galápagos, línea base de la biodiversidad. Quito, Ecuador: Fundación Charles Darwin and Servicio Parque Nacional Galápagos; 2002. pp. 10–21. [Google Scholar]

[pone.0228094.ref035] 35.Hall SJ. Climate change an other external drivers in small-scale fisheries: practical steps for responding In: Pomeroy RS, Andrew N, editors. Small-scale fisheries management: Frameworks and approaches for the developing world. London, UK: CABI; 2011. pp. 132–159. [Google Scholar]

[pone.0228094.ref036] 36.Reck G. The coastal fisheries in the Galapagos Islands, Ecuador. Descriptions and consequences for management in the context of marine environmental protection and regional development. PhD thesis, Christian-Albrechts-Universität zu Kiel, Germany. 1983.

[pone.0228094.ref037] 37.Chasiluisa C, Banks S. Sitios de pesca en la costa de la Reserva Marina de Galápagos In: Danulat E, editor. Evaluación de las pesquerías de la Reserva Marina de Galápagos Informe compendio 2003: análisis comparativo con los años 1997–2002. Galapagos: Charles Darwin Foundation; 2004. pp. 130–140. [Google Scholar]

[pone.0228094.ref038] 38.Castrejón M, Defeo O. Co-governance of small-scale shellfisheries in Latin America: institutional adaptability to external drivers of change In: Chuenpagdee R, Jentoft S, editors. Interactive governance for small-scale fisheries: Global reflections. MARE Publications Series 13. Springer International Publishing; 2015. pp. 605–625. 10.1007/978-3-319-17034-3 [DOI] [Google Scholar]

[pone.0228094.ref039] 39.Moreno J, Peñaherrera C, Hearn A. Evaluación de la pesquería de langosta espinosa (Panulirus penicillatus y P. gracilis) en la Reserva Marina de Galápagos 2006. Galapagos: Charles Darwin Foundation; 2007. [Google Scholar]

[pone.0228094.ref040] 40.Fortin M, Dale M. Spatial analysis: a guide for ecologists. Cambridge: Cambridge University Press; 2005. [Google Scholar]

[pone.0228094.ref041] 41.Morrissey J, Gruber S. Home range of juvenile Lemon sharks Negaprion brevirostris. Copeia. 1993;2: 425–434. [Google Scholar]

[pone.0228094.ref042] 42.Horta e Costa B, Gonçalves L, Gonçalves EJ. Vessels’ site fidelity and spatio-temporal distribution of artisanal fisheries before the implementation of a temperate multiple-use marine protected area. Fish Res. 2013;148: 27–37. [Google Scholar]

[pone.0228094.ref043] 43.Ord J, Getis A. Local spatial autocorrelation statistics: distributional issues and an application. Geogr Anal. 1995;27: 286–306. [Google Scholar]

[pone.0228094.ref044] 44.Ord J, Getis A. Testing for local spatial autocorrelation in the presence of global autocorrelation. J Reg Sci. 2001;41: 411–432. [Google Scholar]

[pone.0228094.ref045] 45.Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction New York: Springer-Verlag; 2001. [Google Scholar]

[pone.0228094.ref046] 46.Elith J, Leathwick J, Hastie T. A working guide to boosted regression trees. J Anim Ecol. 2008;77: 802–813. 10.1111/j.1365-2656.2008.01390.x [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref047] 47.Martínez-Rincón R, Ortega-García S, Vaca-Rodríguez J. Comparative performance of generalized additive models and boosted regression trees for statistical modeling of incidental catch of wahoo (Acanthocybium solandri) in the Mexican tuna purse-seine fishery. Ecol Modell. 2012;233: 20–25. [Google Scholar]

[pone.0228094.ref048] 48.Friedman J. Greedy function approximation: A gradient boosting machine. Ann Stat. 2001;29: 1189–1232. [Google Scholar]

[pone.0228094.ref049] 49.Friedman J, Meulman J. Multiple additive regression trees with application in epidemiology. Stat Med. 2003;22: 1365–1381. 10.1002/sim.1501 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref050] 50.Toral-Granda V. Galapagos Islands: a hotspot of sea cucumber fisheries in Latin America and the Caribbean In: Toral-Granda V, Lovatelli A, Vasconcellos M, editors. Sea cucumbers: a global review of fisheries and trade. Rome: FAO Fisheries and Aquaculture Technical Paper; No. 516; 2008. pp. 231–256. [Google Scholar]

[pone.0228094.ref051] 51.Castrejón M. Co-manejo pesquero en la Reserva Marina de Galápagos: tendencias, retos y perspectivas de cambio Mexico: Fundación Charles Darwin, Kanankil, and Plaza-Valdés; 2011. [Google Scholar]

[pone.0228094.ref052] 52.Shepherd SA, Martinez P, Toral-Granda MV, Edgar GJ. The Galapagos sea cucumber fishery: Management improves as stocks decline. Env Conserv. 2004;31: 102–110. 10.1017/S0376892903001188 [DOI] [Google Scholar]

[pone.0228094.ref053] 53.Ramírez J, Castrejón M, Toral-Granda MV. Mejorando la pesquería de langosta en la Reserva Marina de Galápagos. Ramírez J, Castrejón M, Toral-Granda MV, editors. Quito, Ecuador: World Wildlife Fund; 2012.

[pone.0228094.ref054] 54.Viteri C, Moreno J. Estudio de impactos socio-económicos de la comercialización de langosta entera en Galápagos Galapagos: Conservation International; 2014. [Google Scholar]

[pone.0228094.ref055] 55.Folke C, Colding J, Berkes F. Synthesis: Building resilience and adaptive capacity in socio-ecological systems In: Berkes F, Colding J, Folke C, editors. Navigating the dynamics of social–ecological systems: Building resilience for complexity and change. Cambridge: Cambridge University Press; 2003. pp. 325–383. [Google Scholar]

[pone.0228094.ref056] 56.Defeo O, Castrejón M, Ortega L, Kuhn AM, Gutiérrez NL, Castilla JC. Impacts of climate variability on Latin American small-scale fisheries. Ecol Soc. 2013;18 10.5751/ES-05971-180430 [DOI] [Google Scholar]

[pone.0228094.ref057] 57.Bucaram SJ, Hearn A. Factors that influence the entry-exit decision and intensity of participation of fishing fleet for the Galapagos lobster fishery. Mar Policy. 2014;43: 80–88. 10.1016/j.marpol.2013.05.005 [DOI] [Google Scholar]

[pone.0228094.ref058] 58.Bertelsen R, Hornbeck J. Using acoustic tagging to determine adult spiny lobster (Panulirus argus) movement patterns in the Western Sambo Ecological Reserve (Florida, United States), New Zealand: J Mar Freshw Res; 2009;43: 35–46. [Google Scholar]

[pone.0228094.ref059] 59.Vega A. Reproductive strategies of the spiny lobster Panulirus interruptus related to the marine environmental variability off central Baja California, Mexico: Management implications. Fish Res; 2003;65: 123–135. 10.1016/j.fishres.2003.09.011 [DOI] [Google Scholar]

[pone.0228094.ref060] 60.Jones PJS. A governance analysis of the Galápagos Marine Reserve. Mar Policy. 2013;41: 65–71. 10.1016/j.marpol.2012.12.019 [DOI] [Google Scholar]

[pone.0228094.ref061] 61.Hearn A. The rocky path to sustainable fisheries management and conservation in the Galápagos Marine Reserve. Ocean Coast Manag. 2008;51: 567–574. 10.1016/j.ocecoaman.2008.06.009 [DOI] [Google Scholar]

[pone.0228094.ref062] 62.Salas S, Chuenpagdee R, Charles A, Seijo J. Coastal Fisheries of Latin America and the Caribbean Rome: FAO Fisheries and Aquaculture Technical Paper 544; 2011. [Google Scholar]

[pone.0228094.ref063] 63.White C, Costello C. Matching spatial property rights fisheries with scales of fish dispersal. Ecol Appl. 2011;21: 350–362. 10.1890/09-1188.1 [DOI] [PubMed] [Google Scholar]

[pone.0228094.ref064] 64.Aburto J, Stotz W. Learning about TURFs and natural variability: failure of surf clam management in Chile. Ocean Coast Manag. 2013;71: 88–98. [Google Scholar]

[pone.0228094.ref065] 65.Charles A. Rights-based fisheries management: the role of use rights in managing access and harvesting In: Cochrane K, Garcia SM, editors. A fishery manager’s guidebook. Oxford, UK: Wiley-Blackwell; 2009. pp. 253–282. [Google Scholar]

[pone.0228094.ref066] 66.Charles A. People, oceans and scale: governance, livelihoods and climate change adaptation in marine social-ecological systems. Curr Opin Environ Sustain. 2012;4: 351–357. 10.1016/j.cosust.2012.05.011 [DOI] [Google Scholar]

[pone.0228094.ref067] 67.Gelcich S, Hughes TP, Olsson P, Folke C, Defeo O, Fernandez M, et al. Navigating transformations in governance of Chilean marine coastal resources. Proc Natl Acad Sci U S A. 2010;107: 16794–16799. 10.1073/pnas.1012021107 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228094.ref068] 68.McCay BJ, Micheli F, Ponce-Díaz G, Murray G, Shester G, Ramirez-Sanchez S, et al. Cooperatives, concessions, and co-management on the Pacific coast of Mexico. Mar Policy. 2014;44: 49–59. 10.1016/j.marpol.2013.08.001 [DOI] [Google Scholar]

[pone.0228094.ref069] 69.Pérez-Ramírez M, Castrejón M, Gutiérrez N, Defeo O. The Marine Stewardship Council certification in Latin America and the Caribbean: a review of experiences, potentials and pitfalls. Fish Res. 2016;182: 50–58. 10.1016/j.fishres.2015.11.007 [DOI] [Google Scholar]

PERMALINK

Human and climatic drivers affect spatial fishing patterns in a multiple-use marine protected area: The Galapagos Marine Reserve

Mauricio Castrejón

Anthony Charles

Roles

Abstract

Introduction

Materials and methods

Study area

Fig 1. Marine biogeographical regions of the Galapagos Islands.

Table 1. General features of the three main fishing ports of the Galapagos Islands, including a summary of the fishery information analyzed in this study, for each sampling method used.

Table 2. Main climatic and human drivers that potentially affected the spiny lobster fishery from the Galapagos Islands between 1997 and 2011.

Fishing effort data

Data analysis

Table 3. Periods defined to evaluate the spatio-temporal dynamic of fishing patterns in the spiny lobster fishery, based on the most relevant climatic and human drivers occurring between 1997 and 2011.

Interaction between sea cucumber and spiny lobster fisheries

Spatiotemporal analysis of fishing patterns

Climatic and human drivers affecting spatial fishing effort allocation

Results

Interaction between sea cucumber and spiny lobster fisheries

Fig 2. Long-term variation in fishing capacity in the spiny lobster and sea cucumber fishery from the Galapagos Marine Reserve.

Spatiotemporal analysis of fishing patterns

Fig 3. Core areas (filled ellipses) and distribution ranges (unfilled ellipses) of the fishing fleets from Puerto Ayora, Baquerizo Moreno and Puerto Villamil in the Galapagos Marine Reserve between 1997 and 2011, based on port interview data, for each of the six time periods.

Fig 4. Core areas (filled ellipses) and distribution ranges (unfilled ellipses) of the fishing fleets from Puerto Ayora, Baquerizo Moreno and Puerto Villamil in the Galapagos Marine Reserve between 2001 and 2008, based on observer onboard data.

Fig 5. Fishing effort hotspots in the Galapagos Marine Reserve for the spiny lobster fishery between 1997 and 2011, based on port interview data.

Fig 6. Fishing effort hotspots in the Galapagos Marine Reserve for the spiny lobster fishery between 2001 and 2008, based on observer onboard data.

Climatic and human drivers affecting spatial fishing effort allocation

Table 4. For each predictor variable, the variable importance score (summing to 100) and the ranking is shown, for regional results and for each port.

Fig 7. Variation of fishing effort (in diver-hours) in relation to predictor variables for the spiny lobster fishery of the Galapagos Marine Reserve, according to the regional BRT model.

Fig 10. Variation of fishing effort (in diver-hours) in relation to predictor variables for the spiny lobster fishery of the Galapagos Marine Reserve, according to the BRT model for Baquerizo Moreno.

Fig 8. Variation of fishing effort (in diver-hours) in relation to predictor variables for the spiny lobster fishery of the Galapagos Marine Reserve, according to the BRT model for Puerto Villamil.

Fig 9. Variation of fishing effort (in diver-hours) in relation to predictor variables for the spiny lobster fishery of the Galapagos Marine Reserve, according to the BRT model for Puerto Ayora.

Discussion

Drivers of change: ‘Macro’ adaptive responses

Drivers of change: ‘Micro’ adaptive responses

Impact of no-take zones and management implications

Conclusions and lessons learned

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Heather M Patterson

Roles

Author response to Decision Letter 0

Decision Letter 1

Heather M Patterson

Roles

Author response to Decision Letter 1

Decision Letter 2

Heather M Patterson

Roles

Acceptance letter

Heather M Patterson

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases