Evaluating the role of bycatch reduction device design and fish behavior on Pacific salmon (Oncorhynchus spp.) escapement rates from a pelagic trawl
Introduction
Bycatch of Pacific salmon (Oncorhynchus spp.) is a significant driver in the management of the commercial walleye pollock (Gadus chalcogrammus; hereafter ‘pollock’) pelagic trawl fisheries in the North Pacific (Witherell et al., 2012; Ianelli et al., 2013; Madsen and Haflinger, 2014). Protections afforded to many Chinook salmon (O. tshawytscha) stocks under the U.S. Endangered Species Act, combined with a robust pollock fishery that harvests over 1 million metric tons (t) annually (Ianelli et al., 2013; Witherell and Armstrong, 2015), contributed to the setting of annual Prohibited Species Catch (PSC) limits for Chinook salmon by the North Pacific Fishery Management Council. Annual fishery closures are triggered if these limits are reached (Fissel et al., 2019). To meet performance standards and to avoid exceeding Chinook PSC allowances, fishermen target grounds when and where co-occurrence of Chinook salmon with pollock is comparatively infrequent (Gilman et al., 2006; Ianelli and Stram., 2015). Concern for chum salmon (O. keta) bycatch, while not PSC, also causes fishermen to adjust how they target pollock (Fissel et al., 2019). Spatiotemporal limitations can affect catch quality and lead to increased fuel usage and search time to harvest quota. Since 2002, members of the fishing and conservation engineering communities have worked to develop and improve upon bycatch reduction device (BRD) designs that permit salmon to escape from the trawl before they are landed on the vessel (‘excluders’; Stram and Ianelli, 2015). Given high motivation to reduce salmon bycatch (e.g., low Chinook PSC allowances), fishermen and resource managers benefit from understanding the factors affecting escapement rates and from refining excluder designs based upon this knowledge to ensure reliable performance in situ.
Several salmon excluder designs have been tested in the pollock fishery over the past two decades; however, salmon escapement rates have been highly variable by species, fishery (Bering Sea vs. Gulf of Alaska), among designs, and, by design, among trips and tows (Gauvin and Paine, 2004; Gauvin and Gruver, 2008; Gauvin et al., 2011, 2013, 2015; Gauvin, 2016; Fig. 1). The original excluder design to address salmon bycatch in this fishery included escape portals above a square-mesh tapered tunnel in the intermediate section of the trawl (between the net and codend), requiring salmon to access the escape portals by swimming forward, in the direction of tow, and up. Results from testing this design in the Bering Sea supported the hypothesis, posed by Rose (2004), that behavioral differences and the greater swimming ability of salmon (both Chinook and chum salmon) compared with pollock, despite some morphological similarities, were fundamental to the efficacy of the salmon excluder (Gauvin and Paine, 2004; Gauvin and Gruver, 2008). While improvements to excluder designs have been made, a mechanistic understanding of why escapement rates remain so variable is lacking.
Building on industry-driven research and innovation, we developed a new salmon excluder (called the ‘Rope Tube & Funnel’ [RT&F] excluder) to evaluate the potential to increase escapement rates above that of established designs and to evaluate important influences on escapement. We initiated and evaluated the design of the excluder based on the concept that salmon excluder efficacy previously relied on salmon to perceive and access an escapement area by swimming against the flow of water and in the direction of tow, and then to actively escape. To address the key processes of perception and access, the RT&F excluder was designed to significantly reduce water flow around the escapement area to make it more perceptible by salmon, inducing a rheotropic response (Winger et al., 2010). Manipulation of water flow in and around BRDs is an established practice for trying to increase escapement of non-target animals (Engås et al., 1999; Eayrs, 2007; Gauvin and Gruver, 2008; Cha et al., 2011; Parsons et al., 2012; Prasetyo et al., 2017). It has potential for salmon given that they can react to changes in water velocity less than 3 cm/s and are attracted to or deterred from an area based on water velocity (Lyon, 1904; Arnold, 1974; Banks, 1969; Bell, 1991; Lindmark et al., 2008; Duarte et al., 2012; Lindberg, 2016; Gisen et al., 2017). The RT&F design is also intended to break up the visual pattern of the net to disrupt the optomotor response, the tendency of fish to follow the real or apparent relative motion of their surroundings (Lyon, 1904). Finally, the RT&F excluder features a 360° open area for escape with nearly unobstructed access in the path toward the codend, addressing the question of whether salmon will use an excluder even if nearly all physical barriers to escaping are removed.
The RT&F design was based on the theory that the fast towing speeds used in the pollock fishery relative to salmon cruising speeds (approximately 1.2 m/s for larger salmon; Bell, 1991) are a limiting factor to increased escapement. This is evidenced by higher escapement rates on vessels towing at lower speeds. For example, salmon escapement for the same excluder design towed at speeds of 1.5–2.2 m/s in the Bering Sea ranged from 3 to 18% (mean of 11 %; all tows combined by vessel-year-season combination), compared to trials in the Gulf of Alaska towing at slower speeds (1.3–1.5 m/s) in which escapement ranged from 34 to 54% (mean of 40 %; Gauvin et al., 2015; Gauvin, 2016). For excluders to be effective without requiring a change in fishing practices (namely, tow speed), it is important that the excluder facilitate escapement despite the salmon’s swift passage through the escapement area.
The RT&F excluder was developed using computational fluid dynamics simulations and flume tank testing of a scale model before trials at sea. This study evaluated the design process and how elements of the RT&F excluder affected salmon behavior and escapement rates during pollock fishing in the Bering Sea. The information is presented to help explain mechanisms influencing high variability in salmon excluder escapement rates. Further refinement of this, or other, excluder designs following the template provided here has the potential to save considerably on research and development costs and time frames, operationalizing effective conservation tools in a more expedient and less expensive manner.
Section snippets
Salmon excluder design concept
The broad design concept for the RT&F excluder was to split the net at the last tapered section (at a break in the riblines) and attach a diamond mesh funnel encircled by a rope tube (Fig. 2). At the start of the funnel, a straight section of diamond mesh would be attached to the outside, creating a ‘sleeve’. The sleeve would terminate in jibs attached to the rope tube, which would extend over the length of the funnel and, after a given amount of open space, attach to another jib-terminated,
Fluid dynamics modeling and flume tank testing
As a result of the computational fluid dynamics modeling, it was predicted that the conceptual excluder design would create the intended flow field in general: producing a wake region around the funnel and at the collector entrance, and increasing flow velocity at the funnel exit by 2–5 %. Choosing funnel netting with a solidity of 0.3 prevented reduced water flow at the funnel entrance compared with a solidity of 0.5, which led, in simulations, to an undesirable decrease at the funnel entrance
Fluid dynamics modeling and flume tank testing
The computational fluid dynamics simulations and flume tank testing expedited the development of the RT&F excluder by reasonably predicting performance at full scale and under commercial fishing conditions, including different tow speeds. Given the dynamic and cumbersome testing environment at sea, employing multiple tools to design fishing gear before testing at-sea can greatly benefit development (Nguyen et al., 2015; Queirolo et al., 2009). Previous salmon excluder designs benefited from
CRediT authorship contribution statement
Noëlle Yochum: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review & editing, Visualization, Supervision, Project administration, Funding acquisition. Michael Stone: Conceptualization, Methodology, Validation, Investigation, Writing - review & editing. Karsten Breddermann: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing - review & editing, Visualization.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank the captain (Lorin Perry) and crew (Mario Torgerson, Janek Szymul, Mike Urbanczyk, Lauti Tuipala, Ron Ferro, and Tor Storkersen) of the F/V Pacific Explorer. We also acknowledge Harold DeLouche, Craig Hollett, George Legge, and Mark Santos for assistance at the flume tank; Katherine Hellen Schneider for assistance in the field and with video review; Seamus Melly and Chris Brewer (Swan Nets) for feedback on excluder design; John Gruver for feedback on this project; Paul Winger and Dayv
References (48)
- et al.
Fleet communication to abate fisheries bycatch
Mar. Policy
(2006) - et al.
Optimizing attraction flow for upstream fish passage at a hydropower dam employing 3D detached-eddy simulation
Ecol. Eng.
(2017) - et al.
Studies on the use of visual stimuli to control fish escape from codends. II. The effect of a black tunnel on the reaction behavior of fish in otter trawl codends
Fish. Res.
(1995) - et al.
Studies on the use of visual stimuli to control fish escape from codends. I. Laboratory studies on the effect of a black tunnel on mesh penetration
Fish. Res.
(1995) - et al.
The effect of artificial illumination on Chinook salmon behavior and their escapement out of a midwater trawl bycatch reduction device
Fish. Res.
(2019) - et al.
Selectivity in a trawl codend during haul-back operation- an overlooked phenomenon
Fish. Res.
(2008) - et al.
Computer simulation and flume tank testing of scale engineering models: how well do these techniques predict full-scale at-sea performance of bottom trawls?
Fish. Res.
(2015) - et al.
Differences in orientation and swimming of walleye pollock Theragra chalcogramma in a trawl net under light and dark conditions: concordance between field and laboratory observations
Fish. Res.
(2000) - et al.
Comparison between dynamic simulation and physical model testing of new trawl design for Chilean crustacean fisheries
Fish. Res.
(2009) - et al.
Factors affecting hydrodynamic performance of the Nordmøre Grate System: a bycatch reduction device used in the Gulf of Maine shrimp fishery
Fish. Res.
(1995)