153 



Measurements of total scattering spectra 

 from bocaccio iSebastes paucispinis) 



Stephane G. Conti (contact author)^ 

 Benjamin D. Maurer^ 

 Mark A. Drawbridge^ 

 David A. Demer' 



Email address for S. G. Conti: sconti@ucsd.edu 



' Southwest Fishenes Science Center 

 8604 La Jolla Shores Drive 

 La Jolla, California 92037 



Present address for S. G. Conti; Marine Physical Laboratory 



Scripps Institute of Oceanography 

 University of California San Diego 

 9500 Gilman Drive 

 La Jolla, California 92093-0238 



^ Hubbs-SeaWorld Research Institute 

 2595 Ingraham St. 

 San Diego, California 92109 



Marine sportfishing in southern 

 California is a huge industry with 

 annual revenues totaling many bil- 

 lions of dollars. However, the stocks 

 of lingcod and six rockfish species 

 have been declared overfished by the 

 Pacific Fisheries Management Coun- 

 cil. As part of a multifaceted fisheries 

 management plan, marine conser- 

 vation areas, covering many million 

 square nautical miles, have been 

 mandated. To monitor the recovery 

 of the rockfish stocks in these areas, 

 scientists are faced with the follow- 

 ing challenges: 1) multiple species of 

 rockfish exist in these areas; 2) the 

 species reside near or on the bottom 

 at depths of 80 to 300 m; and 3) they 

 are low in numerical density. To meet 

 these challenges, multifrequency 

 echosounders, multibeam sonar, and 

 cameras mounted on remotely oper- 

 ated vehicles are frequently used 

 (Reynolds et al., 2001). The accuracy 

 and precision of these echosounder 

 results are largely dependent upon 

 the accuracy of the species classifica- 

 tion and target strength estimation 

 (MacLennan and Simmonds, 1992). 



Broad bandwidth characteriza- 

 tion of sound scatter from marine 

 organisms has some potential for re- 

 motely classifying fish species (Conti 

 and Demer, 2003), shapes and sizes 



(Conti et al., 2005), behaviors (Conti 

 et al., 2006b), and to validate models 

 for target strength estimation (De- 

 mer and Conti, 2003). All of these 

 studies have employed variants of a 

 new method for measuring the broad 

 bandwidth total scattering cross sec- 

 tion ia-p) of animals moving in a re- 

 verberant tank. 



With the new method, the total 

 scattering cross section (oj) of live 

 animals in tanks is obtained from a 

 comparison of the coherent and inco- 

 herent acoustical intensities reverber- 

 ated in a tank (de Rosny and Roux, 

 2001, 2003). The accuracy of this 

 measurement technique was shown 

 by using standard metal spheres (De- 

 mer et al., 2003). This technique was 

 successfully used on krill (Demer 

 and Conti, 2003; Conti et al., 2006a), 

 fish (Conti and Demer, 2003), and 

 humans (Conti et al., 2004). In our 

 study, we explored the potential and 

 limitations of the method to char- 

 acterize the broad bandwidth sound 

 scattering from bocaccio (Sebastes 

 paucispinis). 



Materials and methods 



The total scattering cross section, o-p, 

 of bocaccio was measured over acous- 



tic frequencies ranging from 10 to 

 150 kHz with a group of fish {n=20) 

 swimming freely in a large, insulated 

 fiberglass tank at Hubbs-SeaWorld 

 Research Institute, San Diego, CA, on 

 1 and 2 July 2004. The tank had 5.1 

 cm of foam insulation on the exterior, 

 measured 2.44 m in diameter, and was 

 filled with seawater to a depth of 1.37 

 m (V [volume] = 6.4 m^). The pool was 

 thermostated at approximately 12°C. 

 The acoustic measurement technique 

 and a variety of its applications have 

 been well documented (de Rosny and 

 Roux, 2001; Conti and Demer, 2003; 

 Demer and Conti, 2003; Demer et al., 

 2003; Conti et al., 2004). However, 

 the general procedure and details of 

 these experiments are presented here 

 for convenience and clarity. 



Each of the 20 fish was handled 

 one time, a week prior to the experi- 

 ment, to measure their weight (W) 

 and total length (L). These data were 

 summarized and plotted in graphs 

 (Table 1, Fig. 1). 



An emitter transmitted M acousti- 

 cal pulses into the tank every other 

 second {dT=2s). The corresponding 

 reverberation time-series hf,{t) were 

 simultaneously recorded on multiple 

 receivers while the fish were swim- 

 ming between consecutive shots. The 

 boundaries, volume, as well as the 

 positions of the emitter and the re- 

 ceivers in the tank remained identi- 

 cal during the measurements. 



The time series hi,(t) were composed 

 of echoes from the boundaries of the 

 tank and the fish. For two consecu- 

 tive time series h/^it) and h^^j(t), the 

 contributions from the boundaries of 

 the tank were identical, whereas the 

 contributions from the fish were not. 

 The coherent 



'" k=i 



and incoherent 



1 M 



Manuscript submitted 19 July 2005 

 to the Scientific Editor. 



Manuscript approved for publication 

 7 July 2006. 



Fish. Bull. 105:153-157 (2007). 



