Factors affecting the availability of walleye pollock to acoustic and bottom trawl survey gear

ICES Journal of Marine Science ICES Journal of Marine Science (2015), 72(5), 1425– 1439. doi:10.1093/icesjms/fsv011 Original Article Factors affecting the availability of walleye pollock to acoustic and bottom trawl survey gear Stan Kotwicki 1,2 *, John K. Horne1,2, André E. Punt 2, and James N. Ianelli 1 National Marine Fisheries Service, Alaska Fisheries Science Center, National Oceanic and Atmospheric Administration, 7600 Sand Point Way NE, Seattle, WA 98115, USA 2 School of Fishery and Aquatic Sciences, University of Washington, Box 355020, Seattle, WA 98195, USA *Corresponding author: tel: +1 206 526 6614; fax: +1 206 526 6723; e-mail: Kotwicki, S., Horne, J. K., Punt, A. E., and Ianelli, J. N. Factors affecting the availability of walleye pollock to acoustic and bottom trawl survey gear. – ICES Journal of Marine Science, 72: 1425 – 1439. Received 2 September 2014; revised 6 January 2015; accepted 8 January 2015; advance access publication 7 February 2015. Abundances of semi-pelagic fish are often estimated using acoustic or bottom trawl surveys, both of which sample only a fraction of the water column. Acoustic instruments are effective at sampling the majority of the water column, but they have a near-surface blind zone and a nearbottom acoustic dead zone (ADZ), where fish remain undetected. Bottom trawls are effective near the seabed, but miss fish that are located above the effective fishing height of the trawl. Quantification of the extent of overlap between these gears is needed, particularly in cases where environmental factors play a role. We developed logistic regression models to predict the availability (qa) of walleye pollock (Gadus chalcogrammus) to both acoustic and bottom trawl gears using factors shown to affect qa (depth, light intensity, fish length) and introducing additional factors (tidal currents, surface and bottom temperature, sediment size). Results build on earlier studies and quantify the uncertainty associated with the estimation of the ADZ correction using Bayesian methods. Our findings indicate that on average during the day, walleye pollock are more available to the bottom trawl than to the acoustics. Availability to both gears depends mostly on bottom depth, light conditions, and fish size, and to a lesser extent sediment size. Availability to the acoustic gear is also related on surface temperature. Variability in availability to both gears also depends on environmental factors. Keywords: catchability, acoustics, bottom trawl, availability, vertical distribution, abundance estimates, survey, walleye pollock. Introduction Semi-pelagic fish species occupy both demersal and pelagic habitats. The semi-pelagic gadoid walleye pollock (Gadus chalcogrammus; Coulson et al., 2006) in the eastern Bering Sea (EBS) is surveyed using both bottom trawl (BT) and acoustic sampling gears (Karp and Walters, 1994). Bottom trawls catch fish from near bottom up to the effective fishing height of the trawl (EFH; Aglen, 1996; Hjellvik et al., 2003). The EFH can be higher than the trawl headrope height because fish in the pelagic zone may swim towards the bottom in response to vessel noise, towing wire, and doors (Ona and Godø, 1990; Nunnallee, 1991). The bottom trawl will thus catch fish that originally were situated higher in the water column (Aglen, 1996). Acoustic instruments detect fish through the majority of the water column except near boundaries that include a nearsurface blind zone and a near-bottom acoustic dead zone (ADZ, Hjellvik et al., 2003). The near-surface blind zone is not a concern for pollock because they are rarely found in surface waters (Honkalehto et al., 2011). Although pollock BT and acoustic surveys have been performed in the EBS for decades (Honkalehto et al., 2011, Lauth and Nichol, 2013), understanding of the relationship between abundance indices derived from these two methods is still lacking, and consequently two separate indices of abundance are currently used in the pollock stock assessment (Ianelli et al., 2012). Knowledge of catchability of each survey is necessary to understand if and how BT and acoustic abundance indices are related. Catchability is defined as the product of availability (qa) to the sampling gear and the efficiency (qe) of that gear (e.g. Godø, 1994). The availability of pollock to the BT gear (qa,BT) is defined as the proportion of fish in the water column that are present in the bottom trawl zone (BTZ, i.e. between the substrate and the EFH), and availability Published by Oxford University Press on behalf of International Council for the Exploration of the Sea 2015. This work is written by (a) US Government employee(s) and is in the public domain in the US. 1 1426 using ADZ correction estimates and BT EFH parameter estimates from acoustic and BT data (Kotwicki et al., 2013). An evaluation of the relationship between qa and environmental covariates improved our understanding of pollock vertical availability to both the BT and acoustic surveys, and contributed to general understanding of their vertical behaviour. It also has the potential to improve how survey results inform pollock stock assessment models. Methods The acoustic backscatter and bottom trawl catch data used in this investigation were collected during the annual EBS Alaska Fisheries Science Center (AFSC) BT surveys conducted in June and July between 2005 and 2009 during daylight hours—30 min after sunrise and 30 min before sunset (Lauth and Nichol, 2013). The acoustic data analysed here do not come from the biennial acoustictrawl survey of walleye pollock (Honkalehto et al., 2011). However, methods for collecting data are the same as during these EBS acoustic surveys. They were collected using acoustic transducers mounted on the hulls of the BT survey vessels. BT surveys were conducted using the chartered fishing vessels FV Aldebaran, FV Arcturus, and FV Northwest Explorer to sample fixed stations at the centres of a 20 × 20 nautical mile grid cells (Figure 1) using a standard 83 – 112 eastern otter trawl (Stauffer, 2004). The corners of the grid cells were also sampled in areas surrounding St Matthew Island and the Pribilof Islands. Surveys started in the southeastern corner of Bristol Bay and proceeded westward. The standard tow duration and speed was 30 min at 1.54 m s21 (3 knots). Processing of acoustic data is described in Kotwicki et al. (2013). In summary, acoustic backscatter at 38 kHz collected while the vessel was trawling was processed using a semi-automated procedure (Kotwicki et al., 2009) to produce the nautical area scattering coefficient, sA (m2 nautical mile22; a linear measure of backscatter Figure 1. Bottom trawl (points) and acoustic (lines) survey locations. Acoustic data have been also collected on all bottom trawl stations. to the acoustic gear (qa,A) as the proportion of pollock present above the ADZ. The efficiency of the BT (qe,BT) is defined as a proportion of fish within the BTZ that are caught and (...truncated)