Diurnal variation in acoustic densities: why do we see less in the dark?

Abstract

Diurnal fluctuations in total integrated echo abundance and in vertical density profiles were examined using data from the Norwegian combined acoustic and bottom-trawl survey for demersal fish during winter in the Barents Sea. The total echo abundance was about 40%-50% higher at day than at night. An unknown amount of fish was lost close to the seabed in the acoustic dead zone, but the systematic changes in the near-bottom vertical density profiles did not indicate that migration in and out of the dead zone was the major reason for the large diurnal differences in echo abundance. A more plausible explanation could be that diurnal changes in fish behaviour affect the mean acoustic target strength. Based on the present study, we recommend that the time series of acoustic surveys should be reanalysed, taking the diurnal bias into account. Any comparison of the fish densities indicated by trawl and acoustic surveys will suffer if this bias is not corrected. We believe that model development utilizing this type of information is crucial for future ecosystem-based monitoring.Résumé : Nous avons examiné les fluctuations journalières des abondances totales intégrées basées sur l'écho et des profils verticaux de densité; nous avons utilisé une combinaison de données norvégiennes provenant d'inventaires acoustiques et d'inventaires au chalut de fond de poissons démersaux durant l'hiver dans la mer de Barents. L'abondance totale déterminée à l'écho est d'environ 40-50 % plus élevée le jour que la nuit. Une quantité inconnue de poissons est perdue près du fond dans la zone acoustique morte; cependant, les changements systématiques qui se produisent dans les profils verticaux de densité près du fond n'indiquent pas que l'immigration ou l'émigration dans la zone morte est la raison majeure des importantes différences journalières dans les abondances déterminées à l'écho. Une explication plus plausible est que les changements diurnes de comportement des poissons affecte la force moyenne des cibles acoustiques. Sur la base de notre étude, nous recommandons une nouvelle analyse des séries chronologiques des inventaires acoustiques qui tienne compte des sources d'erreur journalières. Toute comparaison entre les densités de poissons déterminées par les inventaires au chalut et celles données par les inventaires acoustiques sera affectée si cette erreur n'est pas corrigée. Nous croyons que la mise au point de modèles qui utilisent ce genre d'information est essentielle pour la surveillance future à l'échelle de l'écosystème.[Traduit par la Rédaction] Hjellvik et al. 2254

Seasonal abundance of two important forage species in the North Pacific Ocean, Pacific herring and walleye pollock

“…All surveys began after daybreak and concluded before dusk. Some fish species, particularly herring, undergo diurnal migrations through the water column that can affect acoustic density estimates (Huse and Korneliussen, 2000;Hjellvik et al, 2004). We chose to sample only during daylight because of the diel variation in acoustic density estimates and to facilitate observations of foraging sea lions while collecting acoustic data.…”

Vertical migration in adult Atlantic cod (Gadus morhua)

“…The vertical positioning of cod therefore potentially has a great impact on measurements of cod. Hjellvik et al (2004) found that the acoustic abundance of fish in the Barents Sea was two times higher during the day than at night and attributed this difference to diurnal differences in vertical distribution and tilt of the fish. Although their study was based on aggregated data and not only cod, it illustrates the importance of considering fish behaviour in stock assessment.…”

Schooling behaviour of small pelagic fish: phenotypic expression of independent stimuli

“…Describing such aggregative patterns is essential to modelling of the in situ schooling behaviour of fish (Azzali et al 1985, Viscido et al 2004. Schooling behaviour must be considered in the management of fisheries of pelagic species because of its impact on the estimation of catchability coefficients (Marchal & Petitgas 1993, Hjellvik et al 2004. Nevertheless, in situ measurements of basic behaviour, such as schooling and diel aggregative patterns, are complex to obtain and to interpret.…”