168 
Fishery Bulletin 1 14(2) 
A 
B 
(A) Beam width [or angle] (a), depth range (h), and diameter (d) at 115 m of the Sat- 
link echosounder buoy. (B) An example of the echogram display for the 10 depth lay- 
ers (ranging from 3 m to 115 m): raw acoustic backscatter (numbers between paren- 
theses) and their corresponding biomass estimates in metric tons (numbers in black) 
for skipjack tuna ( Katsuwonus pelamis) based on the manufacturer’s algorithms. The 
acoustic information used in this study was collected by a commercial Spanish purse 
seiner operating in the central and eastern Atlantic Ocean between 2009 and 2011. 
tail about this process should contact the manufacturer 
directly). 
The model 
Because the manufacturer’s method is not designed to 
provide abundance estimates for the different species 
and sizes of fish that usually compose fish aggregations 
at DFADs (Fonteneau et al., 2013), we proposed a new 
method consisting of developing specific algorithms by 
ocean or regional sites. The specificities of the regional 
algorithms would be based on 1) existing knowledge 
of the vertical behavior of fish at FADs, 2) appropriate 
TS and weight values for different species and sizes to 
perform a multispecies and size echo-integration, and 
3) further parameter optimization against a set of fish- 
ing operations on DFADs of each region. In the present 
study, and as an example of application of the method, 
we developed a specific algorithm for DFADs in the At- 
lantic Ocean using a set of 21 fishing sets. 
Assigning a species group to each depth layer Knowing 
the vertical distribution of species within the observa- 
tional range of the buoy is essential for accurately con- 
verting acoustic backscatter into biomass of different 
fish groups. For that purpose, we reviewed all avail- 
able scientific studies focused on the investigation of 
the behavior of tuna and nontuna species when associ- 
ated with FADs (Brill et al., 1999; Musyl et al., 2003; 
Schaefer and Fuller, 2005; Matsumoto et al. 4 ; Dagorn 
et al., 2007b; Dagorn et al., 2007c; Taquet et al., 2007a, 
2007b; Babaran et al., 2009; Leroy et ah, 2009; Gov- 
inden et al. 5 ; Filmalter et al., 2011; Mitsunaga et al., 
2012; Muir et al. 6 ; Govinden et al., 2013; Schaefer and 
Fuller, 2013; Weng et al., 2013; Matsumoto et al., 2014; 
Forget et al., 2015), as well as on the spatial distribu- 
tion of the echo-traces recorded around DFADs during 
scientific acoustic surveys (Moreno et al., 2007). Addi- 
tional information was obtained from 25 echosounder 
buoys deployed in the Indian Ocean between 2009 and 
2012 (a total of about 2000 acoustic samples). The plot 
4 Matsumoto, T., H. Okamoto, and M. Toyonaga. 2006. Be- 
havioral study of small bigeye, yellowfin and skipjack tunas 
associated with drifting FADs using ultrasonic coded trans- 
mitter in the central Pacific Ocean. Scientific commit- 
tee, second regular session. Western and Central Pacific 
Fisheries Commission Inf. Pap. WCPFC-SC2-2006/FT IP-7, 
25 p. [Available at website.] 
5 Govinden, R., L. Dagorn, M. Soria, and J. Filmalter. 
2010. Behaviour of tuna associated with drifting fish ag- 
gregating devices (FADs) in the Mozambique Channel. In- 
dian Ocean Tuna Commission (IOTC), Working Party Tropi- 
cal Tuna ( WPTT ) IOTC-2010-WPTT-25, 22 p. [Available at 
website.] 
6 Muir, J., D. Itano, M. Hutchinson, B. Leroy, and K. Hol- 
land. 2012. Behavior of target and non-target species on 
drifting FADs and when encircled by purse seine gear. Sci- 
entific committee, eigth regular session. Western and Cen- 
tral Pacific Fisheries Commission Inf. Pap. WCPFC-SC8-2012/ 
EB-WP-13, 7 p. [Available at website.] 
