A least squares method of estimating length to target strength relationships from in situ target strength distributions and length frequencies

Abstract

A least squares method is presented for estimating length to target strength relationships for a target species and associated species using a series of target strength distributions and associated trawl catches. A by-product of the estimation procedure is an objective determination of the correspondence between modal lengths in the trawl catches and the modal lengths in the associated target strength distributions. The method is illustrated by applying it to a data set collected to determine the length to target strength relationship for hoki (Macruronus novaezelandiae).

Improvements in estimating an acoustic target strength–length relationship for hoki (Macruronus novaezeladiae)

“…Over the length range of hoki typically encountered, the new relationship gave much higher estimates of hoki TS than data from the current New Zealand TS-length relationship of Macaulay (2006), TS = 12.2log 10 (TL) − 63.9, which was based on in situ ensemble measurements on hoki aggregations. The TS-length relationship for New Zealand in situ data was derived by a least-squares fitting procedure which matches modes in the observed TS distribution with modal lengths in associated trawl catches (Cordue et al, 2001). As discussed by Kloser et al (2011), the mean of a TS distribution is likely to be much higher than the mode, so an estimate of the TS-length relationship based on fitting TS modes is likely to be biased low.…”

“…There are currently several contradictory TS-length relationships for hoki obtained from in situ measurements and scattering models (Coombs and Cordue, 1995;Cordue et al, 2001;Kloser et al, 2011), and large differences in the TS used to estimate hoki abundance in Australia and New Zealand (Kloser et al, 2011;O'Driscoll, 2012). For a recent New Zealand survey, use of the Australian TS-length relationship (Kloser et al, 2011) changed hoki abundance estimates by a factor of four (O'Driscoll, 2012).…”

“…The current New Zealand TS-length relationship (Macaulay, 2006), was derived using a least-squares fitting procedure which matches modes in the observed TS distribution with modal lengths in associated trawl catches (Cordue et al, 2001). Kloser et al (2011) have noted a number of difficulties with this 'mode-matching' approach and have suggested that this is one of the major sources of difference between the various TS relationships for hoki.…”

Acoustic ontogeny of a teleost

“…Backscatter and fish length data can be measured directly in controlled conditions (McCartney & Stubbs, 1971; Nakken & Olsen, 1977) or indirectly by combining field backscatter measurements from individual fish ( i.e. in situ ) with coincident length samples from directed net hauls (Foote & Traynor, 1988; Cordue et al , 2001).…”

“…: þ12062216890; fax: þ12062216939; email: jhorne@u.washington.edu Journal of Fish Biology (2008) 73, 1444-1463doi:10.1111/j.1095-8649.2008 conditions (McCartney & Stubbs, 1971;Nakken & Olsen, 1977) or indirectly by combining field backscatter measurements from individual fish (i.e. in situ) with coincident length samples from directed net hauls (Foote & Traynor, 1988;Cordue et al, 2001).…”