REGIONAL RESEARCH - GIS - Fishing effort distribution

ACTIVITIES
REGIONAL RESEARCH
Use of GIS

Last update: January 2005

Modeling the spatial distribution of fishing effort

combination of scores of activity assigned to variables
for which the spatial distribution is known and considered as
factors contributing to determine the area actually exploited by fishing fleet

Table of contents

Introduction
Network objectives
Participants
Activities
Results
Documents
Perspectives

Introduction

In December 1998, the FAO COPEMED project organized a workshop focusing the use of Geographic Information System (GIS) as a decision support tool for Fisheries Management. The workshop was designed to favor contacts between fishery scientists and GIS experts. Different propositions of regional collaboration were formulated by the participants. A study focusing the creation of a model to simulate the spatial distribution of the fishing effort was one of the projects that COPEMED decided to support.

Network Objectives

to develop an exploratory methodology of spatial distribution of fishery
effort integrating expert knowledge and fisheries data set,
ability to work on rich or poor data set in terms of effort, catches,
environmental conditions, etc...,
to investigate the usefulness of the methodology with data set available
such as within the COPEMED area,
to identify and implement computer tools in order to facilitate the
application of the methodology.

Participants

Name, Institution and E-Mail	Stage I: Conception	Stage 2: Software development	Stage 3: Model testing
Alexis Bensch FAO COPEMED - Roma (Italy) E-Mail:	Co-author of the global model and technical specifications for the software	Supervision, Analysis, Tests, Upgrade to version 1, Documentation	Assistance to users Requirements for version 2
Fabio Carocci FAO - Rome (Italy) E-mail:	Co-author of one of the source model Co-author of the global model and technical Specifications for the software	Supervision, Analysis, Tests, Upgrade to version 1, Documentation	Assistance to users Requirements for version 2
Fabio Corsi, ITC - Enschede ( The Netherlands) E-mail:	Author of one of the source model Co-author of the global model and technical Specifications for the software	Analysis, Tests, Upgrade to version 1	Assistance to users Requirements for version 2
Laurent Drapeau, IRD E-mail:	Co-author of the global model and technical Specifications for the software		Assistance to users
Gildas Le Corre, IFREMER – Sète (France) E-mail:	Co-author of the global model and technical Specifications for the software		Assistance to users
Jesus Morales, CICEM – Huelva (Spain) E-mail:	Co-author of the global model		Data set owner
Cornelius Mende, Geodimensions® Sarl (France)		Analysis and development of the prototype version

Other scientists from different Mediterranean research institutions have joined the network during the phase III to test the model using real case studies:

Just Bayle (Spain - University of Alicante)
Paola Belcari (Italy - CIBM-DSUA)
Malika Bel Hassen (Tunisia - INSTM - Salammbo)
Jorge Baro (Spain - IEO - Málaga)
Najib El Ouamari (Morocco - INRH - Nador)
Rachida Houssa (Morocco - INRH - Nador)
Julio Portela (Spain - IEO - Vigo)
Bruno Reale (Italy - CIBM-DSUA)
Jose Miguel Serna (Spain - IEO - Málaga)

Activities

This study has been organized as a network, and different experts of the Western Mediterranean region were contacted to collaborate to 3 different stages. A workshop was organized during each stage.

Stage 1: definition of a common approach (Workshop held in Montpellier (France) in April 2000)

Analysis of the different approaches proposed: hypothesis and constraints, limits and domain of application
Identification of the commonalities of the different approaches, overlapping and integration considerations
Common approach feasibility analysis, including evaluation of the technical solutions for the development of a computerized simulation tool

Stage 2: Development of simulation tool (Workshop held in Sète (France) in June 2001)

Formalisation of the global approach and technical specifications for the development of a computerized simulation tool
Development of a prototype version of the Fishing Activity Simulation Tool (FAST)

Stage 3: test of the model using case studies

Call for collaboration
Preparation of the data sets
Workshop held in Barcelona (Spain) in September 2002
Upgrade of the FAST software and publication of the version 1

Results

The model

The expert knowledge on the distribution of fishing effort as exerted by a fleet segment is modeled in order to be represented on maps. The fleet segment is a set of boats with similar characteristics and practicing the same fishing activity or “metier”. The spatial distribution of the fishing activity is explained by the spatial distribution of three components: the abundance zone (i.e. the location of targeted biomass), the accessible zone and an authorized zone (i.e. the jurisdiction enforced in the area). The figure 1 shows the causal relationships between a fleet segment and the spatial distribution of its fishing effort.

Fig.1: Relationships between a fleet segment and the spatial distribution of its fishing activity

The modelling process starts with the selection of variables for which the spatial distribution is available (geo-referenced) and considered as factors contributing to determine the area actually exploited by fishing fleet.

Then the expert defines a distribution function of the fishing effort for each of the selected geo-referenced variables:

A scoring function is constructed by attribution of a score of activity to the different values or value ranges of the geo-referenced variable. Tables 1 and 2 give some examples of attribution of scores of activity to two geo-referenced variables: bottom type (qualitative) and distance from the port (quantitative).

Bottom type	Score of Fishing Activity
Sand	0%
Mud	80%
Rock	20%

Table 1 - Scores of fishing activity function of the bottom type

Distance from the port (in nautical miles)		Score of Fishing Activity
Minimum value	Maximum value	Score of Fishing Activity
10	20	5%
20	30	15%
30	40	30%
40	50	35%
50	60	15%

Table 2 - Scores of fishing activity function of the distance from the port

For the distance from the port, a specific model might be used to calculate scores of activity using a parametric model "friction of distance". The model is based on the following equation:

Fd = e^-((x-v)/s)²

Where:

Fd = Friction of Distance
e = base of natural logarithmic
x = distance from the port
v = parameter defining the distance form the port where the peak of effort occurs
s = parameter defining the amplitude of the areas where most of the effort occurs

Fig.2: the Friction of distance model: scores of activity as a function of the geo-referenced variable Distance from the port.

Computation of the fishing effort map: scores of activity layers are overlaid using additive of multiplicative linear model

The scored geo-referenced layers, named constraints layers, are then overlaid using a weighted additive or multiplicative linear model, to produce a map representing the distribution of the index of fishing activity by the fleet segment. The index is calculated for each surface unit (cell). The size of the surface unit corresponds to the resolution selected by the user.
The final step is the computation of the map representing the distribution of the nominal effort associated to the fleet segment. This is done by multiplying the index of fishing activity of each cell of the map by the nominal effort.

The Software

The Fishing Activity Simulation Tool, called FAST has been developed as an extension for the ESRI ArcView GIS Software (v.3.1) in combination with the Spatial Analyst v. 1.0 extension.

A graphic user interface is used to manage the fleet segments and the simulations. All the parameters and resulting maps associated to a simulation may be stored or reload.

Fig. 3: FAST software - user interface for a simulation

Documents

FAST 1.0 Software (last update 3/12/2004)

Spatial Fishing Effort Modelling Network - Conceptual and Methodological Document - Workshop I : 11-14/04/2000 – Montpellier (2001) - F. Carocci, G. Le Corre, A. Bensch, L. Drapeau, F. Corsi, J. Morales

Spatial Fishing Effort Modelling Network - Workshop II: 5 - 7 /06/2001 – IFREMER Sète – France (2001) - A. Bensch, F. Carocci, F. Corsi, G. Le Corre, C. Mende

Outputs of the 3rd Workshop Barcelona (Spain), 16-19/09/2003

Analysis of effort distribution by bottom trawlers from Motril (Spain) using FAST V.06 (Fishing Activities Simulation Tool; FAO - COPEMED)

List of improvements for FAST version 2 (see perspectives section for more information)

Informes y Estudios COPEMED nº 04: Towards the use of Geographic Information Systems as a decision support tool for the management of Mediterranean fisheries. (2000)
A. Bensch, M. Taconet

Spatial fishing effort modeling network presentation (2002)
A. Bensch, F. Carocci

The use of GIS to analyse the spatial distribution of fishing effort in coastal fishery (2003)
A. Bensch, F. Carocci, F. Corsi

Perspectives

A new version of the FAST Arcview 3.x extension is in development. This new version corrects some problems of the version 1. It also includes a new module to model the fishing activity spatial distribution of fleet segments based in different ports.