Catch, Effort, and eCOsystem impacts of FAD-fishing (CECOFAD)

Abstract

The objective of the European Research project " Catch, Effort, and eCOsystem impacts of FAD-fishing " (CECOFAD) is to improve our understanding of the use of fish-aggregating devices (FAD) in tropical purse seine tuna fisheries on open-sea ecosystems. Due to the relevance of accurate indices of abundance derived from catch per unit of effort, the project will attempt to define a unit of fishing effort for FAD-fishing and to provide reliable estimates of abundance indices and accurate indicators on the impact of FAD-fishing on juveniles of bigeye and yellowfin tunas and on bycatch species. Because science-industry partnerships can improve the quality and availability of data and knowledge, the project research is fostering collaborative research between operators and scientists, without compromising the independence of the latter. CECOFAD is co-funded by EU-DG Mare, 3 scientific institutes (IRD, IEO and AZTI) and 3 professional tuna owner company associations (ANABAC, OPAGAC and ORTHONGEL).

Full text

SCRS/2014/165
Catch, Effort, and eCOsystem impacts of FAD-fishing (CECOFAD)
Gaertner, D.
1
, Ariz, J.
2
, Bez, N.
1
, Clermidy, S.
1
, Moreno, G.
3
, Murua, H.
3
, Soto, M.
4
Abstract
The objective of the European Research project “Catch, Effort, and eCOsystem impacts of FAD-
fishing” (CECOFAD) is to improve our understanding of the use of fish-aggregating devices (FAD) in
tropical purse seine tuna fisheries on open-sea ecosystems. Due to the relevance of accurate indices of
abundance derived from catch per unit of effort, the project will attempt to define a unit of fishing
effort for FAD-fishing and to provide reliable estimates of abundance indices and accurate indicators
on the impact of FAD-fishing on juveniles of bigeye and yellowfin tunas and on bycatch species.
Because science-industry partnerships can improve the quality and availability of data and knowledge,
the project research is fostering collaborative research between operators and scientists, without
compromising the independence of the latter. CECOFAD is co-funded by EU-DG Mare, 3 scientific
institutes (IRD, IEO and AZTI) and 3 professional tuna owner company associations (ANABAC,
OPAGAC and ORTHONGEL).
Introduction
The relationship between catch per unit effort (CPUE) and abundance is central to stock
assessment models and thus, changes in this relationship will ultimately result in changes in scientific
diagnostic and associated management advice. In the lack of fishery-independent information in tuna
fisheries, commercial data are traditionally used to compute CPUE and to derive spatio-temporal
indices of abundance in stock assessments. Although the general process may seem simple in essence,
it needs the proper quantification of the effective effort exerted on tuna stocks. While Nominal efforts
are usually standardized to account for difference among vessels, areas, seasons, and years, but it was
shown in many situations that final estimates of standardized CPUES remained close to nominal
values. One of the major reasons for this is that increasing fishing efficiency through improvements of
fishing gears can strongly modify the relationship between CPUE and abundance over time (Gaertner
and Pallares, 1998; Fonteneau et al, 1999). In addition, the spatial dimension of fishing activities and
resources has to be accurately accounted for in the standardization process as it may severely bias the
estimates of abundance indices (Walters, 2003).
For tropical tuna: yellowfin (Thunnus albacares), skipjack (Katsuwonus pelamis), and bigeye
(Thunnus obesus), an additional problem is that since the implementation of Fish Aggregating Devices
(FAD) in the early 1990s (Ariz et al, 1999; Hallier and Parajua, 1999), progressively equipped with
electronic devices, a fishing effort unit is difficult to be defined for purse seiners and where
catchability can be affected by many factors. In the absence of suitable standardization of purse-seiner
CPUE indices, most of the stock assessments of tropical tunas worldwide are based on longline CPUE
indices which rarely account for changes in technology in the standardization process and only depict
the biomass of the older fraction of tuna populations. In addition, skipjack stock assessments mainly
depend on the accuracy of abundance indices obtained from purse-seiner CPUEs and do not account
for the major changes in purse-seiner strategies that have taken place over the last two decades through
the development of FAD technology and their increasing use. Also, purse seine fishing using FADs
1
Institut de Recherche pour le Développement (IRD), UMR EME, CRH, BP 171, 911 avenue J. Monnet, 34203
Sète Cedex, France
2
Instituto Español de Oceanografía. Centro Oceanográfico de Canarias. Apdo. de Correos 1373. 38080 Santa
Cruz de Tenerife, Islas Canarias (ESPAÑA).
3
AZTI -Tecnalia Herrera Kaia, Portualde z/g 20100 Pasaia Basque Country, Spain
4
Instituto Español de Oceanografía. Corazón de María 8. 28002 Madrid (ESPAÑA)

may have a high by-catch of yellowfin and bigeye juveniles and a potential negative impact on sharks
and other marine organisms and the ecosystem.
The Common Fisheries Policy reform of the European Union identified a number of priorities aimed
at achieving ecological sustainability as well as orientating governance of fisheries towards a
regionalised implementation of principles, defined at Union level. The call for proposals lauched by
UE DG MARE:“Standardization of tropical tuna catch and effort time series for EU purse seine fleets
using FADs in the Atlantic, Indian and Pacific Ocean and estimation of by catch and ecosystem
impacts” (MARE/2012/24) aims to encourage science-industry partnerships to improve the quality and
availability of data and scientific knowledge underpinning decisions on fisheries management
strategies. The expected outcome of the project is to obtain standardized catch-per-unit-effort series
of tropical tuna for the purse seine EU fleet using FADs and information on catch composition and
estimates of potential negative impacts to the ecosystem.
The research project “Catch, Effort, and eCOsystem impacts of FAD-fishing” (hereatfter termed
CECOFAD), is leadered by IRD (France) and included two other Scientific partners: Instituto Español
de Oceanografía (IEO, Spain), AZTI Tecnalia (AZTI, Spain), and 3 professional partners:
Organisation des producteurs de thon tropical congelé et surgelé (ORTHONGEL (France), Asociación
Nacional de Buques Atuneros Congeladores (ANABAC, Spain), Organización de Productores
Asociados de Grandes Atuneros Congeladores (OPAGAC, Spain). Tuna RFMOs, such as the
International Tuna Commission for the Conservation of Atlantic Tunas (ICCAT), the Indian Ocean
Tuna Commission (IOTC), the Inter-American Tropical Tuna Commission (IATTC), as well as the
International Seafood Sustainability Foundation (ISSF) are associated to the project as observers. The
project, co-funded by UE-DG Mare and the partners, began in January 2014 and has a duration of 18
months. The kick-off meeting was held in Montpellier (France) on 1-3 April 2014.
Further information can be found on the web site: http://www.cecofad.eu/ and at the WIKI pages at
http://www.cecofad.eu/w/index.php?title=Main_Page
Objectives
The overall objective of the CECOFAD project is to provide insights into the fishing effort
units (for both fishing modes: FADs and free schools) to be used in the calculation of purse-seiner
CPUEs in the Atlantic as well as in the Indian and the Pacific Oceans, where European purse-seiners
also operate, to ultimately obtain standardized indices of abundance for juveniles and adults of tropical
tunas. With regards to the Ecosystem Approach of Fisheries, the CECOFAD project will contribute to
improve knowledge on the impact of FAD-fishing on the epipelagic ecosystem.
Bearing in mind the multispecies nature of the tropical tuna purse seine fishery and the regular
requests expressed by tuna RFMOs to European tuna scientists to provide reliable estimates of
abundance indices and accurate indicators on the impact of FAD-fishing on juveniles of bigeye and
yellowfin tunas and on bycatch species, the main objectives of the project are:
1) to define a unit of fishing effort for purse-seiners using FADs that accounts for different
factors influencing catchability
2) to standardize catch-per-unit-effort series of the EU purse seine fleet, for juveniles and adults
of the three tropical tuna species and
3) to provide information on catch composition around FADs and estimate impacts on other
marine organisms (e.g. by-catch of sharks, rays, turtles).
Methodology
Structure of the project

The project is organized into 4 Work Packages (WPs), as follows (Figure 1):
- WP 1- Definition of a unit of fishing effort for purse-seiners using FADs that accounts for different
factors influencing catchability (Objective 1 of the project),
- WP 2- Standardization of catch-per-unit-effort series of the EU purse seine fleet, for juveniles and
adults of the three tropical tuna species and exploration of some FAD-regulations in management
strategies (Objective 2),
- WP 3- Alternatives to catch rates (WP 2 and 3 in conjunction will address (Objective 2),
- WP 4.- Provision of information on catch composition around FADs and estimation of potential
impacts on other marine organisms (e.g. by-catch of sharks and criptic mortality; Objective 3).
In addition to these four WPs, transversal activities will be conducted to ensure the coordination and
technical aspects of the project, the data bases management, the web site development and the
administration and management of the project
Description of Work Packages:
WP1- Definition of a unit of fishing effort for purse-seiners using FADs that accounts for different
factors influencing catchability (leader IRD)
Tropical tuna purse seine fisheries capture different species and size of tunas. Large yellowfin,
and in some strata skipjack, are caught in free-swimming schools, while skipjack and juveniles of
yellowfin and bigeye are mainly associated with natural or artificial floating objects. In the traditional
tuna purse seine fishery (characterized by free-swimming schools and natural floating objects), the
fishing effort was expressed as searching time, i.e., the daylight hours devoted to the detection of tuna
schools minus the setting times (Fonteneau, 1978). While this simple definition may be criticized even
for free-swimming schools sets (e.g., due to the non-random distribution of fishing effort, the increase
in fishing power over the years, etc) the implementation of drifting artificial fish aggregating devices
(FAD) since the early 1990s (Ariz et al, 1999; Hallier and Parajua, 1999), progressively equipped
with electronic devices (Moreno et al, 2007; Lopez et al, 2014) have broken the link between
searching time and effective fishing effort for FAD sets. Remote detection of satellite-tracked FADs
often allows fishers to move directly towards a buoy, sometimes at night, avoiding or significantly
reducing searching time (Figure 2).
In addition, the recent development of satellite-tracked echo-sounder fish finder units attached to
floating objects gives fishers realtime information about fish schools aggregating around FADs and
has resulted in an increasing proportion of successful FADs sets. The use of supply vessels, which can
visit FADs and inform purse-seiners on the fish aggregations around these FADs, also contributes to
the efficiency of some purse-seiners (Arrizabalaga et al, 2001, Pallares et al, 2002; Goujon, 2004,
Moreno et al, 2007). For all of these reasons, the increasingly extensive use by tropical tuna purse
seine vessels of FADs has dramatically changed the nature of tuna fishing over the last two decades
(Figure 3). Obviously, the complexity increases when the fishing strategy is based on transitions
between both fishing modes: free-swimming schools and FAD-fishing, which is the most common
situation (Guillotreau et al, 2011). These changes have major consequences for our ability to calculate
useful CPUE values for these fisheries.
During the kick off meeting of CECOFAD the participants agreed that many data on the fishing
technology introduced on board over time should be useful for the project. The collection of the dates
of quantitative and qualitative changes in FAD design and use, as well as in new fishing devices was
initiated during the EU research project Esther (Gaertner and Pallares, 1998) for the French fleet
(Figure 4). Recently, following the study of Moreno et al (2007), new information on technology
associated with FAD-fishing has been collected by Lopez et al (2014) for the Spanish fleet (Figure 5)
and surveys on fishing practices have been made on the French fleet operating in the Indian Ocean
within the framework of the PhD thesis of A. Maufroy (IRD). All these informations will be updated
in cooperation with expert knowledge for both fleets, e.g., questionnaires filled by tuna fishermen on

how and when they perceive a change in catchability over time for both fishing modes (i.e., FAD sets
and free school sets). The active cooperation with the industry which is partner of the project should
ensure that comprehensive data on floating object sets, especially on FADs, and on fishing operations
will be made available to national scientists. The information on the main periods of introduction of
new technology can be used during the standardization of CPUEs or be integrated in a Bayesian
surplus production modelling approach so as to account for increasing fishing power over time. In the
same order of ideas, collecting information on the numbers of active FADs seeded ovet the years by
the EU purse seine fleets is a fundamental. Unfortunately this information is seldom available, keeping
in mind that in the Atlantic and in the Indian Oceans detailed information on the use of FADs,
including the activities conducted by supplies, has been requested only in 2013 in the Indian Ocean
and in 2014 in the Atlantic Ocean (IOTC Res[13-08], and ICCAT Rec[13-01], respectively).
However, some information has been recently collected for the French fleet operating in the Eastern
Atlantic Ocean (Table 1) by Fonteneau et al. (2014). Based on some assumptions on the total catch
and the catch per set for FAD-fishing for the other purse seiner fleets, these authors provide
preliminary estimates of the change in the number of FADs used in the Eastern Atlantic since 2004
(Figure 6).
.
As mentioned previously, one another important component of FAD-fishing is the contribution of the
assistance provided by supply vessels. Consequently, effects of such assistance on the FAD activities
will be explored by means of performance indicator comparisons between individual purse-seiners.
FAD-based fishery indicators are a way to follow the trend in the exploitation level exerted on tuna
juveniles and some by-catch species, e.g., the size of the area fished is a key factor in the potential
catches of a fishery but owing to the mobile nature of the fishery, the total surface explored must
account for the trajectory of the FADs. Individual trajectory obtained from VMS data are essential to
analyze the fishing behaviour of a purse-seiner over a fishing trip (Walker et al, 2010), with the aim to
identify which proportion of the searching time should be related to a fishing effort associated with
FADs and if the FAD was detected at random or with the aid of radio satellite beacons (Figure. 7). The
production of a nominal effort splitting the FAD activities from the conventional daylight searching
time should be a first challenge in this approach.
WP2- Standardization of catch-per-unit-effort series of the EU purse seine fleet, for juveniles and
adults of the three tropical tuna species and exploration of some FAD-regulations in management
strategies
For reason explained previously in WP1, fishing effort on PS (FAD and free school) cannot be
expressed in its traditional form. An alternative to calculate a catch rate depicting as close as possible
the abundance was the concept of catch per set. Until now the standardization process was done for
each fishing mode separately: free school sets or FAD sets (Soto et al, 2009a; Soto et al, 2009b,
respectively). The main explanatory variables traditionally used in the GLM to fit the nominal CPUEs
were : Year, Quarter, Area, a factor termed CatPais (combining the flag and the carrying capacity of
the vessel), the age of the vessel, a factor reflecting the proportion of the species targeted in the catch
and the interactions between the main factors. To account for the presence of a high amount of zero-
catch per fishing day, the delta-lognormal method (Lo et al, 1992) was commonly used and the
specific index for each year was finally calculated as the product of year average fitted values of
lognormal model (for the positive CPUEs) and binomial model (for the proportion of days with catch).
To account for to unbalanced designs, the Least squares means (LSMEANS procedure) are computed
for each factor as it should be for estimate the marginal means over a balanced population
.
Based on the conclusions of the U.E. Research Project ESTHER (Gaertner and Pallares, 1998), for a
better understanding on the specific impact of the new technologies introduced on board, the CPUE
can be decomposed into several sub-indices, such as the number of sets per fishing days (i.e., depicting
the ability to detect a concentration of tuna schools), the proportion of successful sets (the ability of
catching a school) and finally, the amount of catch per positive set (i.e., combining a proxy of the size
of the school and the ability to maximise a catch during the setting). Such approach has been

developed for standardizing CPUEs of yellowfin in free schools in the Indian Ocean by Chassot et al.
(2012). During the kick-off meeting of CECOFAD it was also suggested to explore additional
explanatory factors as the price of the species targetted (or the diference in prices between large
yellofin and skipjack), the density of FADs by strata and a factor identifying the strategy followed by
each captain in terms of fishing mode (obtained for instance asking to each captain who are the 5
fishermen specialized in each fishing mode).
However, other CPUE indicators or standardization procedures (GLMM, Bayesian methods) should
also be explored: integrating spatial correlation in the standardization (Nishida and Chen, 2004),
checking if there is a correlation between catch and effort and an additional spatial correlation (Pereira
et al, 2009), accounting for the changes in the spatial distribution of the fishing effort over time
(Campbell, 2004, Hoyle et al, 2014), which can be done with different methods to integrate the
predicted CPUEs of unfished areas in the calculation of the annual index of abundance (Zhang and
Holmes, 2010; Cao et al, 2011; McKechnie et al, 2013). Effects of spatial distribution and movements
of tuna resources and fishing fleets on the CPUE standardization (i.e., sensitivity to spatial
stratification, application of Poisson kriging to catches and fishery effort data), extension to temporal
data using co-regionalization modeling and Spatial GLM and kriging methods applied to local vs
global CPUEs should be an important part of this working package. It should be stressed that
depending the time unit used (quarter, month, fornight) in the standardization, environmental factors
and short-term decisions taken by fishermen may cause different levels of spatial variability, which
should be integrated in the definition of the spatio-temporal strata (Saulnier, 2014). Furthermore, using
VMS data, associated with the detection of FAD sets, may be used to perform indices of presence of
tuna schools (Bez et al, 2011). Another important aspect is how to estimate the increase in CPUE due
to the assistance brought by a supply vessel or by echo-sounder buoys (Pallares et al, 2002). The
Group agreed that such analysis should be done in priority on the Spanish fleet operating in the Indian
Ocean.
In parallel to the standardization procedure, the participants to CECOFAD recognized the relevance of
different FAD-indicators related to the abundance indices: e.g., Catch per FAD set, Catch / distance
between FAD, Catch per soaking time, Time-at-sea and distance travelled, Density of FAD sets by
1°square, ratio number of FADs fished / number of FADs visited and other related metrics. Another
aspect, not directly related to the estimate of abundance but with FAD-fishing, and consequently
useful for the tuna RFMOs, should be the standardization of the ratio juveniles of bigeye / total bigeye
catch.
WP3- Alternatives to catch rates
Direct indices of tuna abundance may be obtained through the use of echo-sounder buoys
attached to FADs. Behavioral models, calibrated by electronic tagging data, representing the
continuous process of association and non-association, as well as the residence time under FADs can
be an alternative to commercial data in order to perform an abundance estimate. The main tasks
identified in this Work package, whose several are dependent of the full cooperation of the
professional partners, are:
- Direct local abundance by echosounder and modeling the aggregation process of biomass under
FADs: behavior of tunas in terms of residence time and spatial distribution within an array of FADs.
- Preliminary analysis of alternative indices of abundance collected from different sources of
information (e.g. from echosounders).
- Index of FAD density and aggregated abundance derived from VMS data combined with
multispecies sampling data.
Two important aspects of accessibility of fish to be explored in the WP3 will be how to improve the
usability of acoustic data and refine behavior data. This work package is mainly structured around 2
steps: (1) convert acoustic raw data into biologically relevant measures, and (2) link these measures

(number of individuals or biomass at FADs) with tuna abundance around and off the FADs, to help
assessing the population.
First, It was necessary to review if the existing technology is useful for our scientific goals, and if
fishers will continue using these tools in the future (Lopez et al, 2014). Based on interviews conducted
with more than 60 Spanish fishers, 31% of Spanish fishers said that 50 to 75% of the FADs they use
are fitted with echo-sounder buoys and 42% said that 75%-100%. The trend is moving towards 100%
of FADs with echo-sounder buoys, which is promising for our scientific purposes. Regarding the
technology and current measures by these buoys, outputs for the 4 different brands : Zunibal, Satlink,
Marine Instruments and Thalos used by fishers are heterogeneous and cannot be compared (Figure 8).
For one of the brands, raw data was obtained and new algorithms used to convert acoustic raw data
into biomass of 3 main groups (by-catch/ small tuna/large tuna). There is ongoing work with the other
3 manufacturers to find the way to provide acoustic data in dB, as currently for some of the buoys,
data is processed internally in the buoy to be able to deliver it via satellite and is not possible to have
pre-processed data. Another potential field study that could be conducted is the intercalibration of the
4 buoys with a known target, to obtain a standardized output suitable for common analyses.
Nowadays, the biological relevant measures expected from fishers echo-sounder are biomass or
number of individuals for 3 main groups: by-catch, small tuna and large tuna. There is hope for better
discrimination due to new echo-sounder buoys equipped with more than one frequency. These tools
have the potential to discriminate skipjack (no swim-bladder) from bigeye and yellowfin (both with
swimmbladder). Remote target classification could be done in the future by means of using vertical
fish behaviour from electronic tagging in the same area, and also by using species composition and
sizes in the area. That is why for the latter would be interesting to conduct an exploratory analysis to
see the variability of species composition in a given area between adjacent FADs.
The second step on the way to an abundance index is linking our measurements at FADs with the
population. Based on the approach of Sempo et al (2013), a model has been developed to obtain
abundance indices using the time of residence at FADs and out of FADs as well as the number of
aggregation points (Capello, com. pers.). The simplest model assumes that the probability of join and
leave the FAD is constant. Field studies on tuna behaviour from electronic tags would allow feeding
this model.
WP4- Catch composition around FADs and estimation of potential impacts on other marine
organisms (e.g. by-catch of sharks and criptic mortality)
Within the framework of the ecosystem approach of fisheries, Tuna RFMOs, international
bodies, national administrations and NGOs have raised concern for the status of the non-targetted
species caught incidentally by different type of gears (Figure 9). Even discards and by-catch in the
purse seine fishery are moderate compared to other gears, the increase in FAD-fishing impacts not
only the tuna resource, even if this effect is not simple and should be carefully analyzed (Dagorn et al,
2013), but also other non-target species (e.g., sharks, rays, turtles) as showed by different studies
(Delgado de Molina et al, 2000; Amande et al. 2008; 2010; Capietto et al, 2014). Both fishing modes
produce bycatch to a different extent and have a different species composition. FAD-fishing bycatch is
considerably greater than that obtained from fishing on free-swimming schools (Table 2), and this
might have impacted differently over time the different taxonomic groups composing the epipelagic
fauna associated with the tropical tunas (Figure 10). A recent study based on changes in sample-based
rarefaction curves suggests that the species composition of sharks caught on FADs decreased over
time (Torres-Ireneo, 2014a), even if it should be kept in mind that other fishing gears may also have
affected these species.
A great variety of shark species are found within the Tuna RFMO Convention areas, from coastal to
oceanic species, and specifically associated with FADs. As a consequence of the increase in fishing
pressure supported by these species, the recent 30th Session of the Committee on Fisheries (COFI) of
the FAO of the United Nations, encouraged States and RFMOs to take further actions for shark

conservation and management. As an example, ICCAT Res [12-05] asked to all its CPCs to submit,
previously to the 2013 annual meeting, details of their implementation of and compliance with shark
conservation and management measures [Recs. 04-10, 07-06, 09-07, 10-08, 10-07, 11-08 and 11-15].
The main species of sharks associated with FADs are silky shark (Carcharhinus falciformis) and
whitetip shark (Carcharhinus longimanus). In the Atlantic and in the Mediterranean areas of its
jurisdiction, ICCAT has put in place recommendations that prohibit the retention of shark species
identified as at risk due to the impact of fisheries such as thresher sharks (Alopias spp; 09-07), oceanic
whitetip (10-07), hammerhead (Sphyrna spp.; 10-08), silky sharks (11- 08). Similar resolutions
concerning threther sharks exist in the Indian Ocean (Resolution IOTC-2012-09). In addition to
sharks, by-catch generated by FAD-fishing is also known for other species of bony fish, blue marlin,
as well as some turtles, specifically the Green turtle (Chelonia mydas) and the Loggerhead turtle
(Caretta caretta).
Even if progress on fishery statistics for sharks has been recently evidenced by Tuna RFMOs it is
admitted that it is still insufficient to provide quantitative advice on stock status with sufficient
precision to guide fishery management toward optimal harvest levels. Consequently several fishery
indicators related to sharks will be analyzed at a family taxonomic level and as well as possible at a
species level for the shark species listed as endangered. It should be mentioned that the major part of
information used in this Work Package will be obtained from past observer programs and for this
reason special effort will be done to the production of tools allowing to merge information between
logbook and observer data bases.
The impact of FAD-fishing will be evaluated in different ways: in terms of overall quantification of
catch and by-catch species and size composition aggregated under FAD depending on the location,
trajectories, soaking time of the FAD, whenever possible, or accounting for some characteristics of the
FAD (e.g. the length of underwater appendages or the mesh size of the net under the FAD). target
species and bycatch removed by FAD-fishing. Comparative quantitative analysis of biomass removed
by FAD-fishing and for school fishing of both targeted and bycatch (retained or discarded) species
(commonly identified as the Gerrodette approach; Gerrodette et al, 2012): use of different metrics
(diversity indices and species composition for both fishing modes). Part of the PhD thesis of N.
Lezama-Ochoa (AZTI) is related to biodiversity aspects and tuna fisheries.
The assessment of the ecological effects of FAD-fishing would require a precise taxonomic
identification of the species that is usually not recorded. To overcome this problem, groupings will be
defined taking into account the ecological functions of the different taxa, and case studies will
investigate endangered shark species. Regional differences and rich structures in species diversity
should be highlighted and might be used for future conservation issues. In addition, as it was
evidenced that a moratorium on FAD for protecting juveniles of tuna may have different consequences
on the associated fauna (Gaertner et al, 2002), the resulting impact of such type of regulation measure
on whale-sharks and marine mammals will be tentatively analyzed within the PhD of L. Escalle
(UM2/IRD).
References
Amande, J. M., Ariz, J., Chassot, E., Chavance, P., Delgado de Molina, A., Gaertner, D., Murua, H.,
Pianet, R., and Ruiz, J. (2008) By-catch and discards of the European purse seine tuna fishery
in the Indian Ocean. Estimation and characteristics for the 2003-2007 period. IOTC-2008-
WPEB-12
Amande, J. M., Ariz, J., Chassot, E., Delgado de Molina, A., Gaertner, D., Murua, H., Pianet, R., Ruiz,
J., and Chavance, P. (2010). Bycatch and discards of the European purse seine tuna fishery in
the Atlantic ocean. Estimation and characteristics for the 2003-2007 period. Aquat. Living
Resour, 23, 353–362
Ariz, J., Delgado, A., Fonteneau, A., Gonzalez Costas, F., Pallares, P. (1999) Logs and tunas in the
Eastern Tropical Atlantic. A review of present knowledge and uncertainties. In: Scott MD,
Bayliff WH, Lennert-Cody CE, Schaefer KM (eds) Proceedings of the International

Workshop on Fishing for Tunas Associated with Floating Objects, La Jolla, CA, February
11–13, 1992. Inter-Am Trop Tuna Comm Spec Rep 11:21–65
Arrizabalaga, H., Ariz, J., Mina, X., Delgado de Molina, A., Artetxe, I., Pallares, P., Iriondo, A.
(2001). Analysis of the activities of supply vessels in the Indian Ocean from observers data.
Doc. IOTC WPTT-01-11 : 30 p
Arrizabalaga, H., de Bruyn, P., Diaz, G. A., Murua, H., Chavance, P., Delgado de Molina, A.,
Gaertner, D., Ariz, J., Ruiz, J., and Kell, L. T. (2011) Ecological Risk Assessment for species
caught in Atlantic tuna fisheries. Aquat. Living Resour. 24(1): 1-12
Bez, N., Walker, E., Gaertner, D., Rivoirard, J. and Gaspar, P. (2011). Fishing activity of tuna purse-
seiners estimated from VMS data. Can. J. Fish. Aquat. Sci. 68:(11) 1998-2010
Campbell, R. A. (2004) CPUE standardisation and the construction of indices of stock abundance in a
spatially varying fishery using general linear models. Fisheries Research 70: 209‐227
Cao, J., Chen, X. J., Chen, Y. (2011) Generalized linear Bayesian models for standardizing CPUE of
Ommastrephesbartramii for Chinese squid-jigging fishery in northwest pacific ocean. Scientia
Marina, 75 (4): 679-689.
Capietto, A., Escalle, L., Chavance, P., Dubroca, L., Delgado de Molina, A., Murua, H., Floch, L.,
Damiano, A. (2014) Mortality of marine megafauna induced by fisheries : insights from the
whale shark, the world's largest fish. Biological Conservation, 174: 147-151.
Chassot, E., Dubroca, L., Delgado de Molina, A., Assan, C., Soto, M., Floch, L., Fonteneau, A.,
(2012). Decomposing purse seine CPUEs to estimate an abundance index for yellowfin free-
swimming schools in the Indian Ocean during 1981– 2011. IOTC-2012-WPTT14-33
Dagorn, L., Holland, K. N., Restrepo, V., and Moreno, G., (2012). Is it good or bad to fish with
FADs? What are the real impacts of the use of drifting FADs on pelagic marine ecosystems?
Fish and fisheries, 1-25
Delgado de Molina, A., Ariz, J., Gaertner, D., y Santana, J. C. (2000). Estimacion de la importancia de
las capturas de especies accesorias y de descartes en la pesquería de cerco de tunidos
tropicales en el océano Atlantico oriental. Coll. Vol. Sci. Pap. ICCAT, 51(6): 1859-1874
Fonteneau A., (1978) Analyse de l'effort de pêche des thoniers senneurs franco-ivoiro-sénégalais. Cah.
ORSTOM, sér. océanogr., 16 (3-4): 285-307.
Fonteneau, A., Chassot, E and Gaertner, D. (2014) .Managing tropical tuna purse seine fisheries
through limiting the number of drifting fish aggregating devices in the Atlantic: food for
thought. SCRS/2014/133
Fonteneau A., Gaertner D., Nordström V., (1999) An overview of problems in the CPUE-abundance
relationship for the tropical purse seine fisheries. ICCAT Collect. Sci. Pap. 49, 259–276.
Gaertner, D., Ménard, F., Develter, C., Ariz, J., Delgado, A. (2002). By-catch of billfishes by the
European tuna purse seine fishery in the Atlantic Ocean. Fishery Bulletin 100: 683-689.
Gaertner D., Pallares P., (1998) Efficiency of Tuna Purse-Seiners and Effective Effort/Efficacité des
thoniers senneurs et efforts réels ” (ESTHER). Sci. Rep. EU Programme N◦ 98/061.
Gerrodette, T, Olson, R., Reilly, S., Watters, G., and Perrin, W. (2012). Ecological metric of biomass
remved by three methods of purse-seine fishing for tunas in the eastern tropical Pacific Ocean.
Biol. Cons. 26: 2, 248-256.
Goujon, M. (2004). Informations sur l'activité des thoniers senneurs et du supply français et sur les
observations en mer d'après les données collectées pendant les plans de protection des
thonidés de l'Atlantique 2000-2001, 2001-2002 et 2002-2003. Col. Vol. Sci. Pap. ICCAT
56(2): 432-442
Guillotreau, P., Salladarré, F., Dewals, P., Dagorn, L. (2011). Fishing tuna around Fish Aggregating
Devices (FADs) vs free swimming schools: Skipper decision and other determining factors.
Fisheries Research 109:234-242
Hallier, J.P., Parajua, J.I. (1999). Review of tuna fisheries on floating objects in the Indian Ocean. In:
Scott MD, Bayliff WH, Lennert-Cody CE, Schaefer KM (eds) Proceedings of the
International Workshop on Fishing for Tunas Associated with Floating Objects, La Jolla,
CA, February 11–13, 1992. Inter-Am Trop Tuna Comm Spec Rep 11:195–221
Hoyle, S., Shono, H., Okamoto, H., Langley, A.D. (2010). Analyses of Japanese longline operational
catch and effort for bigeye tuna in the WCPO. WCPFC-SC6-2010/SA-WP-02

Lo, N.C, Jacobson, L.D, and Squire, J. L., (1992). Indices of relative abundance from fish spotter data
based on delta-lognormal models. Can. J. Fish. Aquat. Sci. 49:2515-2526
Lopez, J., Moreno, G., Sancristobal, I., Murua, J. (2014) Evolution and current state of the technology
of echo-sounder buoysused by Spanish tropical tuna purse seiners in the Atlantic, Indian and
Pacific Oceans. Fisheries Research 155 : 127–137
Maufroy, A. (2012). Caractérisation de la dynamique spatio-temporelle des Dispositifs de
Concentration de Poissons (DCP) utilisés par les pêcheries de senneurs français dans les
océans Atlantique et Indien. Rapport de stage de Mémoire de fin d’études Diplôme
d’Ingénieur, Spécialisation halieutique d’Agrocampus Ouest
McKechnie, S, Hoyle, S., and Harley, S. (2013) Longline CPUE series that account for changes in the
spatial extent of fisheries. WCPFC-SC9-2013/SA-IP-05
Moreno, G., Dagorn, L., Sancho, G., García, D., and Itano, D. (2007) Using local ecological
knowledge (LEK) to provide insight on the tuna purse seine fleets of the Indian Ocean useful
for management. Aquat. Living Resour. 20, 367–376
Nishida, T., Chen, D-G. (2004) Incorporating spatial autocorrelation into the general linear model with
an application to the yellowfin tuna (Thunnus albacares) longline CPUE data. Fisheries
Research 70 (2004) 265–274
Pallares, P., Mina, X., Delgado de Molina, A., Artetxe, I., Ariz, J.,Arrizabalaga, H. (2002) Analisis de
la actividad de los buques auxiliares en el océano Atlantico y su efecto sobre el rendimiento de
atuneros congeladores. Col. Vol. Sc. Pap. ICCAT, 54(1): 139-156
Pereira, J. C., Leandroa, R. A., Petrere Jr. M., Nishida, T. (2009). Comparing three indices of catch per
unit effort using Bayesian geostatistics. Fisheries Research, 100: 200-209
Saulnier, E. (2014) Dynamique spatio-temporelle de l’effort de pêche sur DCP des thoniers senneurs
français dans l’océan indien.
Rapport de stage de Mémoire de fin d’étude Diplôme
d’Ingénieur de l’Institut Supérieur des Sciences Agronomiques, Agroalimentaires,
Horticoles et du Paysage
Sempo, G., Dagorn, L., Robert, M. and Deneubourg, J-L. (2013) Impact of increasing deployment of
artificial floating objects on the spatial distribution of social fish species. Journal of Applied
Ecology. 50 (5): 1081-1092
Soto, M., Pallarés, P., Delgado de Molina, A., and Gaertner, D. (2009a). Actualizacion de la CPUE
estandardizada de rabil adulto (Thunnus albacares) de la flota de cerco tropical Europea en el
Atlantico de 1980 a 2006. Col. Vol. Sc. Pap. ICCAT, 64(4):1032 ‐1043
Soto, M., Pallarés, P., Delgado de Molina, A., and Gaertner, D. (2009b). Stardardized CPUE for
juvenile Atlantic yellowfin and bigeye and skipjack tunas caught by the purse seine fleet
fishing with FADs. Col. Vol. Sc. Pap. ICCAT, 64(4): 1044-1053
Torres-Irineo, E., Amande, M. J., Gaertner, D., Delgado de Molina, A., Murua, H., Chavance, P.,
Ariz, J., Ruiz, J., Lezama-Ochoa, N. (2014a) Bycatch species composition over time by tuna
purse-seine fishery in the eastern tropical Atlantic Ocean. Biodivers Conserv. 23: 1157–1173
Torres-Irineo, E., Gaertner, D., Chassot, E., Dreyfus-León, M. (2014b) Changes in fishing power and
fishing strategies driven by new technologies: The case of tropical tuna purse seiners in the
easternAtlantic Ocean.
Fisheries Research, 155: 10-19
Walker, E., Gaertner, D., Gaspar, P. and Bez, N. (2010). Fishing activity of tuna purse-seiners
estimated from VMS data and validated by observers' data. Col. Vol. Sc. Pap. ICCAT, 65(6):
2376-2391
Walters, C., (2003). Folly and fantasy in the analysis of spatial catch rate. Can. J. Fish. Aquat. Sci.
60: 1433.1436
Zhang, Z., and Holmes, J. (2010).Use of generalized linear Bayesian model to mitigate the impact of
spatial contraction in fishing pattern on the estimation of relative abundance. Fisheries
Research, 106:413-419

Table 1
. Numbers of FADs used by French purse seiners: seeded yearly in the Atlantic ocean and
active ones on a quarterly basis, number of PS and total catches on FADs. (From Orthongel, in
Fonteneau et al 2014)
Nb of
Active
buoys/PS
Nb buoys
seeded
yearly/PS
Ratio Nbs
FAD Seeded
& active
Nb PS
Average
Nb of
active
FADs
Total Nb of
seeded
buoys
yearly
FAD
catches
Average
Catches on
FAD by each
PS
Average
catch per
buoy
seeded
2004
41
13
533
19 818
1524
37,2
2005
41
9
369
13 521
1502
36,6
2006
47
7
329
4 896
699
14,9
2007
42
5
210
4 452
890
21,2
2008
54
7
378
3 051
436
8,1
2009
60
10
600
7 311
731
12,2
2010
68
72
1,05
10
683
720
16125
1613
22,4
2011
71
82
1,16
9
635
738
13195
1466
17,9
2012
96
118
1,23
9
861
1062
16956
1884
16,0
2013
90
156
1,74
9
808
1404
16749
1861
11,9
2014
200
9
1800
Average 2004-2013
81
71
1,30
9
747
634
11607
1261
19,8
Table 2 Ratio of bycatch (FADs vs Free school) in weight for the UE purse-seiner fleets
operating in the Atlantic and Indian oceans (from Amande el al, 2008; 2010).
Ratio FADs / Free school Atlantic Ocean Indian Ocean
Total by-catch 5.3 3.9
Tunas 6.2 3.6
Bony fishes 21.3 5.8
Billfishes 0.5 2.0
Sharks 6.0 3.6
Rays 0.1 0.5

Figure 1 Structure of the CECOFAD project depicting the links between the Work Packages and the
different partners, the Tuna RFMOs, the fisheries managers, stakeholders and public.
Figure 2- Before FADs fishing, Free schools and natural logs were randomly detected, while after
FADs fishing the purse-seiner detects electronically its own FADs and may run directly in the good
direction. Free school, logs and foreign FADs continue to be randomly searched during daily hours
(from Fonteneau et al, 1999).

Figure 3. Example of spatial distribution of satellite-tracked FADs for the period 2007-2011
(from Maufroy, 2012).
Figure 4
Year (or time period), as indicated by rectangles, in which a new technology was introduced
to the French tropical tuna purse seine fleet operating in the eastern Atlantic Ocean Right side
identifies the broad technology type whilst the left side distinguishes the different model
specificationswithin the technology type. Some devices are specific-fishing mode (e.g., automatic
radar plotting aid (ARPA), radio equipped buoys, supply vessels for FAD fishing), othersare affecting
the fishing efficiency at a whole (from Torres-Irineo et al, 2014b).

Figure 5 Timeline of the most important events that occurred in the development of buoys technology
and the Spanish tropical purse seine DFAD fishery in the Indian Oceanfor the last 30 years (from
Lopez et al, 2014).
Figure 6 Estimate of total number of FADs seeded yearly in the Eastern Atlantic ocean, based on
French data (Orthongel) and estimated for the other PS fleets (from Fonteneau et al, 2014)

Figure 7. Example of an individual trajectory of a tuna purse-seiner during a fishing trip from VMS
data (Maufroy, comm.. pers.).
Figure 8 Use of the different brands of echo-sounder buoys by the Sp anish fleet in the Indian Ocean
showing the heterogeneity in the information available for scientific studies (G. Moreno, comm.. pers.)

Figure 9. Summary plots of the ICCAT bycatch list. Number of species reported to have interacted
with each species group, by fishing gear. An occurrence is defined as a species reported to have
interacted at least once with a given fishing gear. The presence of a species in the list does not imply
that it is caught in significant quantities, or that individuals that are caught necessarily died as a result
of the interaction (from Arrizabalaga et al, 2011).
Figure 10 Quarterly occurrence probability per set by species group for each fishing mode (Free-
swimming school and FAD-associated school) during time period 1 (1997–1999), and 2 (2005–2008).
Quarter: 1 January–March, 2 April–June, 3 July–September, 4 October–December (From Torres-
Ireneo et al 2014a).