Utility of 18-kHz acoustic data for abundance estimation of Atlantic herring (Clupea harengus)

Current acoustic survey protocols for Atlantic herring (Clupea harengus) abundance estimation are principally dependent upon 38-kHz backscatter data. This can constitute a substantial problem for robust stock assessment when 38-kHz data are compromised. Research vessels now typically collect multifrequency data during acoustic surveys, which could be used to remediate such situations. Here, we investigate the utility of using 18-and 120-kHz data for herring abundance estimation when the standard 38-kHz approach is not possible. Estimates of herring abundance/biomass in the Celtic Sea (2007-2010) were calculated at 18, 38, and 120 kHz using the standard 38-kHz target-strength (TS) model and geometrically equivalent TS models at 18 and 120 kHz. These estimates were compared to assess the level of coherence between the three frequencies, and 18-kHz-derived estimates were subsequently input into standard 38-kHz-based population models to evaluate the impact on the assessment. Results showed that estimates of herring abundance/ biomass from 18 and 38 kHz acoustic integration varied by only 0.3-5.4%, and acoustically derived numbers-at-age estimates were not significantly (p. 0.05) different from 1: 1. Estimates at 120 kHz were also robust. Furthermore, 18-kHz-derived estimates did not significantly change the assessment model output, indicating that 18-kHz data can be used for herring stock assessment purposes.