Spatial Dynamics Modeling of Small-Scale Fishing Fleets With a Random Walk Approach

• Main Authors: Daniel R. Quijano Quiñones, Jorge A. López-Rocha, Isis Hernández-Herrera, Edgar Torres-Irineo
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-05-01
• Series: Frontiers in Marine Science
• Subjects: small scale fisheries; small scale fleet; spatial dynamics; random walk; modeling
• Online Access: https://www.frontiersin.org/articles/10.3389/fmars.2021.669112/full

The understanding of fishing effort allocation has been recognized as a key feature within the spatial management planning of fisheries. In small-scale fisheries this assessment is challenging because of their multi-specific context. This has resulted in management regulations that do not fulfill their objectives. Therefore, the analysis of vessels movement is needed for the better assessment and management of small-scale fisheries. Thus, we used a Random Walk (RW) modeling approach to characterize the fishing trips of small-scale multi-gear vessels targeting multiple species. We analyzed activities of three different gears (hand net, handline, longline) and two fishing methods (diving, “gareteo”) recorded with a GPS from 156 fishing trips along three fishing seasons (2018–2020) by the small-scale fleet of Yucatan, Mexico. We fitted seven RW models for the movement data and compared their relative and absolute fit through a delta Akaike test and G-Tests respectively. A total of 143 trajectories showed conclusive evidence to establish a CCRW-type movement; the remaining 13 trajectories fit three models (CCRW, TRW and CRW). The Random Walk models were useful for modeling a multi-gear, small-scale fleet operating in southeast Mexico. According to our results, the Composite Correlated Model was the most suitable RW for this fleet. For future studies, we suggest increasing the number of fishing trips and conducting specific studies by gear and fishing method. Also, the application of relative and absolute fit tests to compare the performance of RW models is highly recommended.