Data were collected continuously throughout the survey with two calibrated echosounders (Foote et al., 1987) SIMRAD EK80 connected to two 7° splitbeam transducers centred on the frequencies of 70 kHz and 200 kHz, operated simultaneously in narrow band (CW) transmission. Transducers were attached with a stainless-steel pole on the port side of a sport fishing boat (10 m long). The acquisition was performed with transmitted powers of 600 W and 90 W for 70 and 200 kHz, respectively. With the aim of measuring target strength (TS) and track individual detections, the ping rate was set to “maximum” for a maximum acquisition range of 100 m (on the continental shelf) and to 1 ping per sec, off the shelf, where the maximum acquisition range was set to 400 m. Due to the rather windy condition around Fernando de Noronha pulse duration had to be long enough to improve the signal-to-noise ratio and the maximum range for the two transducers. A pulse duration of 1.024 ms for both frequencies was a good compromise between a pulse long enough allowing low noise at the maximum range of 400 m at 70 kHz, an identical pulse duration at both frequencies, easier for further processing and short enough to separate individual fish by about 40 cm. Continuous measurements were made at a typical vessel speed of 3.5 knots. During FAROFA 3 survey, one day was dedicated to a magic square operation during twelve hours from 2 pm to 2 am.

Acoustic data were scrutinized, corrected and analysed using the “Movies3D” software developed at the “Institut Français de Recherche pour l'Exploitation de la Mer” (Ifremer; Trenkel et al., 2009) combined with the French “Institut de Recherche pour le Développement” (IRD) open source tool “Matecho” (Perrot et al., 2018), developed in MATLAB.

The near-field noise was removed (3 m deep from the transducer, that is 9 m deep from the surface) from the analysis; phantom bottom echoes were excluded and the bottom line was corrected manually. Echograms were cleaned by eliminating unwanted signals such as one impulsive noise, transient noise due to the interferences with other electronic devices or other sources. Background noise was estimated and subtracted using methods described by De Robertis and Higginbottom (2007). The weather was very quiet during these surveys, thus no further cleaning was necessary. The acoustic nautical area scattering coefficient (sA in m2 nmi-2 or NASC), an indicator of marine organisms biomass, and the acoustic volume backscattering strength (Sv in dB re 1m-1), an indicator of marine organisms density, were calculated. Data were echo-integrated onto 1 m high layers over 3 pings ESDU (elementary sampling distance unit) with a -90 dB threshold. The data provided are the results of the echointegration processing.

Sensors metadata and variables are described in the report "ICES. 2016. A metadata convention for processed acoustic data from active acoustic systems. Series of ICES Survey Protocols SISP 4-TG-AcMeta. 48 pp".