Elsevier

Fisheries Research

Volume 164, April 2015, Pages 286-292
Fisheries Research

Defining small-scale fisheries in the EU on the basis of their operational range of activity The Swedish fleet as a case study

https://doi.org/10.1016/j.fishres.2014.12.013Get rights and content
Under a Creative Commons license
open access

Abstract

Extending the definition of small scale fisheries is a recurrent issue in policy and research debates. A broader definition of small scale fisheries would need to encompass, in addition to vessel size attributes such as vessel length, variables relating to their local operational range, their social role in coastal communities and the economics of the enterprise. In this study, data mining and geospatial analysis techniques were used to explore the relationship between vessel characteristics and local operational range. The process relies heavily on the availability of detailed logbook data and involves two main steps: (1) clustering vessels on the basis of operational range attributes and (2) finding vessel characteristics that best match the operational range classes through machine learning algorithms. The analysis was carried out using the Swedish fishing fleet as a case study and considers the fishing activity of the entire fleet over the period 2007–2013. Swedish logbook data offers the advantage of providing precise spatial information on the location of the catch. Results clearly identified three operational range clusters: local, medium and long range. When considering engine power and vessel tonnage as explanatory variables, the classification algorithms were able to represent the operational range classes with a success rate of 94%. However, the fact that medium size vessels operate and compete in the same operational range class of small size vessels limits, in practice, the possibility of using vessel characteristics to represent univocally the local operational range characteristics of small scale fisheries, unless very high thresholds for power and tonnage are used.

Keywords

Spatial
GIS
Data mining
Logbook data
Classification
Clustering

Cited by (0)