Defining fishing spatial strategies from VMS data: Insights from the world's largest monospecific fishery
Introduction
Understanding fishermen's spatial behavior and fishing effort is essential to the design of fisheries management systems (Salas and Gaertner, 2004, Wilen, 2004, Babcock et al., 2005, Garcia and Cochrane, 2005). Fishermen's spatial behavior results from external stresses (e.g., biotic and abiotic conditions, management rules, economic incentives) and ‘internal’ factors (e.g., skippers’ skill and personality, characteristics of the vessels), which are reflected to some extent in the geometry of the fishing trip track. The characterization of fishermen's movement patterns at the scale of a fishing trip has been made easier by the implementation of vessel monitoring systems (VMS). However, to date this technology has only been used for discriminating métiers1 (Russo et al., 2011).
The Peruvian anchovy (Engraulis ringens) fishery provides an opportunity to compare fishing strategies and tactics, as defined in Salas and Gaertner (2004), which for simplicity will be called strategies. This very large fishery (4–9 million tons landed each year over the last decade, Fréon et al., 2008, SOFIA, 2014) involves a single purse seine métier, subject to the intense environmental variability of the Northern Humboldt Current System (NHCS) over multidecadal to intraseasonal timescales (Chavez et al., 2008). This environmental variability determines the extent of the tridimensional anchovy habitat (Bertrand et al., 2004, Bertrand et al., 2011), which in turn affects fish availability for fishermen (Bertrand et al., 2008, Joo et al., 2014) and leads to adaptations in fishing strategies.
In addition, neo-liberal economic policies during 1990–2000 and the recovery of the stock after the 1982–1983 El Niño provided an opportunity for a rapid expansion of fishing capacity: reduced tariffs and import restrictions, relaxed domestic price controls and fewer constraints on foreign investment encouraged investment in vessels and plant modernization and construction (Aguilar Ibarra et al., 2000, Aranda, 2009). Moreover, in 1998, the Peruvian government promulgated a law authorizing owners of wooden vessels larger than 30 m3 of fish-hold capacity to join the pelagic industrial fleet. Since then, the industrial fleet has been composed of two segments: a steel fleet (vessels made of steel and with at least 120 m3 of fish-hold capacity; Aranda, 2009) and a wooden fleet (wooden-hulled and between 30 m3 and 119 m3 of fish-hold capacity). During the last decade, the number of vessels operating in the fishery increased substantially, reaching a peak of ∼1200 active vessels by day in 2006 (Fréon et al., 2008). Overcapacity, together with an open-access management regime, made the race for fish more intense, increasing the pressure on anchovy and leading to a reduction in the length of the fishing season (the total allowable catch was attained in fewer days each year; Fréon et al., 2008). Starting in 2009, individual vessel quota allocations (IVQs) were implemented in the anchovy fishery, with the aim of stopping the race for fish. They had an immediate effect, lengthening the annual fishing season and reducing the total number of active vessels (Tveteras et al., 2011).
The management of the anchovy fishery is adaptive on short time scales to cope with environmental, biological and fishing effort variation. For instance, when juveniles account for more than 10% of the catches in a given port, landings may be prohibited in that area within two days (Arias Schreiber et al., 2011). Decisions on opening and closure dates for fishing seasons (usually there are two fishing seasons per year) are also made on short time scales. Moreover, there are distinct management policies for the north-center (from 3° S to 16° S) and south regions (from 16° S to the frontier with Chile). Catch shares are established independently for each region; the fishing season is longer in the south than in the north-center; and, during the last decade, the ban on fishing within the first few nautical miles from the coastline was set to 5 nm in the north-center region, while it varied between 1.5 nm and 3 nm in the south.
The spatial behavior of Peruvian anchovy fishermen might be shaped by all these elements. Patterns of fishermen's trajectories can be investigated from VMS data. From time series of vessel positions, it is possible to derive fishing trip descriptors such as duration, distance traveled, maximum distance to the coast and time spent fishing, among others. In Peru, VMS has been mandatory for the industrial pelagic fleet since 2000. In practice, while the steel fleet was almost entirely covered with VMS by 2000, the coverage of the wooden fleet was much more gradual. Here we analyzed 352,711 fishing trips monitored by VMS during the period from 2000 to 2009, from which a set of descriptors were computed. By means of a hierarchical cluster analysis, we study how the trips associate into groups without establishing a priori the number of clusters. We then examine which factors determined those clusters of contrasted spatial strategies.
Section snippets
Data and pre-processing
We used VMS positioning records from the Peruvian anchovy industrial fleet corresponding to the decade 2000–2009 (∼100 m of accuracy; ∼1 record per hour; Table 1 and Fig. 1). Pre-processing of raw VMS data was performed based on the criteria and algorithms described in Bertrand et al., 2005, Bertrand et al., 2007 and Joo et al. (2011). For each fishing trip, we first computed the following seven global metrics: duration (Dur), total distance traveled (Dist), maximum distance from the coast
Results
For the PCA, not all variables were used as active variables. Dist, Lat. Max and Lon. Min were only used as supplementary variables since they were highly correlated to other active variables: Dur (r = 0.89), Bef.Fishing (r = 0.74) and Time.Cruising (r = 0.73), Lat.Min (r = 0.97) and Lon.Max (r = 0.96), respectively. Moreover, when comparing absolute times and proportions in each activity as active variables, absolute times were more closely correlated to the principal components, so we used absolute
Discussion
From the classification of the 352,711 fishing trip tracks, four clusters emerged (Fig. 5). They were mostly explained by fleet characteristics, skipper behavior and regional management rules.
The first and largest cluster, labeled ‘typical’, presented the most common features of the Peruvian anchovy fishing industry: the trips were made by vessels with steel hulls in the north-center region, without outstanding duration or distances.
The second cluster comprised more than 75% of the trips
Acknowledgements
We would like to express our gratitude to Marilu Bouchon and Erich Diaz for the valuable information and discussions about fisheries management in Peru. We would also like to thank Carlos Requena, Federico Iriarte, ‘cabezon Kili’ and his crew for valuable insights on fishermen's tactics, strategies and behavior. We are also grateful to the UPRSIG from the Instituto del Mar del Perú (IMARPE) for making VMS data available for this study. This work is a contribution to the cooperative agreement
References (46)
Developments on fisheries management in Peru: the new individual vessel quota system for the anchoveta fishery
Fish. Res.
(2009)- et al.
Impacts of Kelvin wave forcing in the Peru Humboldt Current system: scenarios of spatial reorganizations from physics to fishers
Prog. Oceanogr.
(2008) - et al.
The northern Humboldt Current System: brief history, present status and a view towards the future
Prog. Oceanogr.
(2008) - et al.
Interdecadal variability of anchoveta abundance and overcapacity of the fishery in Peru
Prog. Oceanogr.
(2008) - et al.
Ecosystem scenarios shape fishermen spatial behavior. The case of the Peruvian anchovy fishery in the Northern Humboldt Current System
Prog. Oceanogr.
(2014) - et al.
Optimization of an artificial neural network for identifying fishing set positions from VMS data: An example from the Peruvian anchovy purse seine fishery
Ecol. Model.
(2011) - et al.
Standardizing catch and effort data: a review of recent approaches
Fish. Res.
(2004) - et al.
In situ measurements of the speed of Peruvian anchovy schools
Fish. Res.
(2014) - et al.
When behaviour reveals activity: assigning fishing effort to métiers based on VMS data using artificial neural networks
Fish. Res.
(2011) - et al.
The relationship of anchovy and sardine to water masses in the Peruvian Humboldt Current System from 1983 to 2005
Prog. Oceanogr.
(2008)