Irruptive prey dynamics following the groundfish collapse in the Northwest Atlantic: an illusion?

ICES Journal of Marine Science ICES Journal of Marine Science (2013), 70(7), 1299– 1307. doi:10.1093/icesjms/fst111 Irruptive prey dynamics following the groundfish collapse in the Northwest Atlantic: an illusion? Kenneth T. Frank 1*, William C. Leggett 2, Brian D. Petrie 1, Jonathan A. D. Fisher 3, Nancy L. Shackell1, and Christopher T. Taggart 4 1 *Corresponding author: tel: 1-902-426-3498; fax: 1-902-426-6927; e-mail: Frank, K. T., Leggett, W. C., Petrie, B., Fisher, J. A. D., Shackell, N. L., and Taggart, C. T. 2013. Pelagic fish outbreak in the Northwest Atlantic - reality or illusion? – ICES Journal of Marine Science, 70: 1299 – 1307. Received 5 April 2013; accepted 8 June 2013; advance access publication 24 August 2013. The collapse of Northwest Atlantic groundfish in the early 1990s yielded a “natural experiment” within which to explore responses of ecosystems to a major perturbation. The “Pelagic Outburst” hypothesis was developed to explain an up to 900% increase in the abundance of small-bodied forage fishes and macroinvertebrates following this collapse and a subsequent trophic cascade extending across four trophic levels. Recently, this theory has been challenged and an alternative “Suprabenthic Habitat Occupation” (SHO) hypothesis has been advanced; it proposes the prey outburst associated with the forage fish component was an illusion created by changes in the vertical distribution of small pelagic fishes after the cod collapse in favour of a more bottom-oriented distribution that increased their vulnerability to bottom trawls. We evaluated the SHO hypothesis as it applied to the relationship between changes in the biomass of cod and the vertical distribution of herring and sand lance, the major small pelagic species of the Scotian Shelf ecosystem off eastern Nova Scotia. Contrary to predictions of the SHO hypothesis our initial conclusion that a pelagic outburst occurred in that ecosystem was confirmed and we found no evidence of a predator effect on vertical distributions of these species. We also explored the acoustic survey design and execution that generated the data that form the cornerstone of the SHO hypothesis, and the coherence between the behaviour depicted in these data and catch rates in the surface-oriented purse-seine fishery for herring operating at the time of these surveys. In combination, the results of our re-analysis of the population dynamics and behaviour of herring on the eastern Scotian Shelf, lead us to conclude that the SHO hypothesis, at least as it relates to the post-cod collapse dynamics of the affected Northwest Atlantic ecosystems, is not supported. Keywords: ecosystem dynamics, predator– prey, small pelagics. Introduction The growing acceptance that the traditional single species approach to the management of commercial marine resources is inadequate has led to enhanced interest in the applicability of ecosystem-based management models (Link, 2010). Central to this approach is a more complete knowledge of the interdependency of the numerous trophic levels and species interactions that govern the dynamics of large marine ecosystems (Cury et al., 2011; Hunsicker et al., 2011). The well-known collapse of cod (Gadus morhua) and other groundfish populations in the western North Atlantic in the early 1990s produced something akin to a natural experiment (Jensen et al., 2012) # Crown copyright 2013. that has been exploited by several researchers to develop, refine and evaluate ecological theory as it relates to the response of large marine ecosystems to natural and anthropogenic perturbations (Bundy, 2005; Frank et al., 2005, 2011; Savenkoff et al., 2007; Shackell et al., 2010). Frank et al. (2006) documented a suite of post-collapse trophic cascades that involved up to four trophic levels (large-bodied benthic predators, small pelagic fishes, zooplankton, and phytoplankton) in several Northwest Atlantic ecosystems. Small pelagic planktivores, once the primary prey of the collapsed groundfish species on the eastern Scotian Shelf (Figure 1), exhibited a 900% Department of Fisheries and Oceans, Bedford Institute of Oceanography, PO Box 1006, B2Y 4A2, Dartmouth, Nova Scotia, Canada Department of Biology, Queen’s University, K7L 3N6, Kingston, Ontario, Canada 3 Centre for Fisheries Ecosystem Research, Fisheries and Marine Institute of Memorial University of Newfoundland, A1C 5R3, St John’s, Newfoundland, Canada 4 Department of Oceanography, Dalhousie University, B3H 4J1, Halifax, Nova Scotia, Canada 2 1300 increase in biomass (Frank et al., 2005). Zooplankton species, their principal prey, declined and phytoplankton concentrations increased. Petrie et al. (2009) subsequently demonstrated that the differential susceptibility of large marine ecosystems to such top-down alterations was linked in a predictable way to water temperature and species richness—colder and less species rich ecosystems being more vulnerable to top-down forcing. Detailed studies of the dynamics of the altered eastern Scotian Shelf ecosystem led Frank et al. (2011) to inquire whether these lower trophic level effects, and the failure of cod and other collapsed groundfish species to recover in spite of a . 20-year moratorium on exploitation, resulted from the post-collapse increase in the abundance and biomass of small pelagic fishes, notably herring (Clupea harengus harengus), sand lance (Ammodytes dubius) and capelin (Mallotus villous). The same authors hypothesized that this produced a predator–prey role reversal in which the once pelagic fish prey became important predators of the now diminished egg and larval stages of the collapsed demersal species, thereby inhibiting their recovery. McQuinn (2009) developed an alternate view of the postcollapse dynamics of these small pelagic species by focusing on herring. McQuinn’s analysis of the post-collapse vertical distribution of commercially exploited herring in the waters off western Newfoundland (Figure 1), as reflected in acoustic surveys conducted at night between 1991 and 2002, led to the conclusion that the collapse of cod in the region resulted in a rapid (,2 year), dramatic, and persistent (decadal) behavioural change in which the classic night-time, near-surface distribution of herring (e.g. Horwood and Cushing, 1978; Blaxter and Hunter, 1982; Cardinale et al., 2003) was no longer evident. The acoustic survey data indicated that in the years following the collapse of cod approximately 80% of the population was distributed within 6 m of the bottom at night at depths in excess of 80 m compared with 10% estimated from one acoustic survey conducted in 1989 before cod collapsed. McQuinn (2009) attributed this apparent behavioural change to reduced predation pressure resulting from the collapse of cod, a behaviour characterized as Suprabenthic Habitat Occupation (SHO). This hypothesis was then generalized to apply to other small-bodied forage fishes known to be prey of cod (e.g. sand lance), and to all (...truncated)