Matt K. Broadhurst

a NSW Department of Primary Industries, Fisheries Conservation Technology Unit, National Marine Science Centre, Southern Cross University, 2 Bay Drive, Coffs Harbour, NSW, 2450, Australia

b Marine and Estuarine Ecology Unit, School of Biological Sciences, University of Queensland, Brisbane, QLD, 4072, Australia

Keywords:

ABSTRACT

1. Introduction

Penaeid trawling occurs throughout the tropic-temperate zone and is recognised as one of the worlds’ least selective fishing methods-traditionally characterised by disproportionate ratios of non-target(‘bycatch’) to target catches (Andrew & Pepperell, 1992; Gillett, 2008;Kelleher, 2005, p. 131). The poor selectivity of penaeid trawls is attributed in part to their small mesh sizes (＜50 mm stretched mesh opening; SMO; Vendeville, 1990, p. 75), but the main problem is their spatial-fishing mechanisms. Penaeid trawls require contact with the substrate (typically sand or mud, and via chain or weighted ground ropes) at towing speeds usually ＞1.2 msto stimulate the benthic-orientated penaeids into the tapered netting panels, and are deployed across inshore/estuarine areas inhabited by diverse assemblages of small fish (i.e. ＜20 cm total length; TL). Most small fish have insufficient swimming capabilities to avoid entering penaeid trawls and many are caught and discarded, often dead (Andrew & Pepperell, 1992;Gillett, 2008).

Recognition of the negative ecological consequences associated with excessive collateral mortalities of discarded penaeid-trawl bycatch has precipitated global resolution efforts (reviewed by Broadhurst, 2000;McHugh, Broadhurst, & Sterling, 2017). Most focus has been directed towards developing and installing physical bycatch reduction devices(BRDs) in codends designed to exclude unwanted species (typically by 30-70%); via differences in either behaviour or size (Broadhurst, 2000).While rarely validated, organisms escaping codends are assumed to incur few negative impacts, and so BRDs are thought to alleviate unaccounted fishing mortality. More recently, attempts at eliminating mortalities have been made via modifications to the anterior trawl that entirely preclude the capture of unwanted species, and mostly by exploiting behavioural differences (McHugh et al., 2017).

In addition to their poor selectivity, penaeid trawls are substantially＜100% effective for catching all penaeids in front of the trawl opening(Broadhurst, Sterling, & Millar, 2015a; Eayrs, 2002; Goeden, Coles, &Mellors, 1990). Specifically, while there are few in situ studies, during research in aquaria (using tow tanks), Goeden et al. (1990) estimated nearly 50% of tiger prawns, Penaeus semisulcatus that contacted the ground chain of a trawl were not caught, with many presumed to be damaged as they passed under, or collided with various components.Such poor efficiency appears typical in many penaeid fisheries, whereby trawlers often maintain similar catches of penaeids as those that have immediately trawled the same area. There are concerns that physical impacts to escaping penaeids could result in unaccounted fishing mortalities, representing a wasted resource (Broadhurst, Suuronen, &Hulme, 2006).

A simple method for improving catching efficiency for the targeted penaeids involves modifying the ground gear (e.g. Broadhurst et al.,2015a; Deshpande & Sivan, 1962; Goeden et al., 1990). In particular,some penaeid trawls have so-called ‘tickler chains’ attached between the wing ends or otter boards and typically comprising light chain that contacts the bottom anterior to the ground chain, but still beneath the overhang of the upper panel (termed ‘lead-ahead’) (Dellapenna, Allison,Gill, Lehman, & Warnken, 2006; Deshpande & Kartha, 1967; Deshpande& Sivan, 1962; Watson, Workman, Taylor, & Serra, 1984). Tickler chains function by stimulating some penaeids to contract their abdomen and propel themselves upwards into the upper panel and eventually the trawl (Goeden et al., 1990). The posteriorly located ground gear then similarly stimulates some of the remaining penaeids.

Tickler chains are also used in some other crustacean and teleost otter- and beam-trawl fisheries where they clearly improve catches of various benthic teleosts (Chittenden & van Engel, 1972; Rijnsdorp et al.,2008; van Marlen, Wiegerinck, van Os-Koomen, & van Barneveld,2014). Considering benthic teleosts are a large component of penaeid-trawl bycatches, there are concerns tickler chains could negatively affect species selectivity (and increase bycatches, e.g. Chittenden& van Engel, 1972), but there are very few published accounts in the primary literature, with most studies limited to national reports(Deshpande & Kartha, 1967; Deshpande & Sivan, 1962; Goeden et al.,1990).

Irrespective of tickler chains, there are potential benefits associated with altering conventional ground chains; both to improve efficiency for penaeids and reduce bycatch (Broadhurst et al., 2015a) As one example,Broadhurst et al. (2015a) showed that, although non-significant,compared to a 6-mm diameter (Ø) stainless-steel ground chain, those made from 8- or 10-mm caught slightly more school prawns, Metapenaeus macleayi, and the latter caught nearly twice as many unwanted forktail catfish, Arius graeffei. These results were attributed to the greater surface area of the ground chain creating greater stimuli for both benthic species; although in the absence of additional studies such conclusions are speculative.

Considering the dearth of manipulative experiments, additional research is required to address hypotheses concerning the effectiveness of subtle changes to ground gears and/or the use of tickler chains for affecting the species-specific efficiencies of penaeid trawls. This study contributes towards addressing the shortfall in studies by testing the hypothesis of no changes in the catching efficiencies for targeted penaeids and bycatches of a typical Australian penaeid trawl fished with two sizes of ground chain and with and without a tickler chain. The work was done in one fishery, but the results have broader national and international application.

2. Materials and methods

The experiment was completed during December 2019 in the Clarence River/Lake Wooloweyah estuary (2926 S 15322 E) NSW,Australia using a double-rigged trawler (10 m with an 89-kW engine)(Fig.1a). The trawler was fitted with two hydraulic winches spooled with 8-mm diameter (Ø) stainless warps and 40-m bridles (6-mm Ø stainless wire) attached to paired, stainless-steel otter boards (53.00 kg;1.08 ×0.73 m) (Fig.1). Onboard electronic equipment included a global positioning system (GPS; Lowrance) to measure speed over the ground(SOG; ms), Notus trawl-monitoring system to monitor wing-end spreads (m) and an ecosounder to measure depths (m).

Fig.1. Schematic diagram of (a) double-rigged (b) trawl used in the experiment.

2.1. Trawls and ground-gear configurations

Two identical trawls were constructed and fished; both comprising two panels and all made from the same bundle of netting (nominal 38-mm stretched mesh opening; SMO and 1.10-mm Ø braided polyethylene-PE twine) orientated with the knots providing positive lift(Broadhurst, Sterling, & Millar, 2016). The trawls had headline and footrope lengths of 7.53 and 8.46 m, respectively creating 12 meshes of lead-ahead, and were attached to sweeps (2.89 m) via snap clips at the wing ends (Figs. 1 and 2). Both trawls had extension sections made from nominal 41-mm SMO PE mesh and measuring 100 meshes in the transverse direction and 30 meshes in the normal direction. Each extension had a Nordmøre-grid BRD (comprising 20-mm bar spaces;Broadhurst, 2000) installed and was attached to a T45 codend made from nominal 27-mm mesh and measuring 120 ×75 bars (Fig.2).

Two replicates of two different ground gears (all 8.46 m long) and one tickler chain (7.99 m) were constructed for use with either trawl(Figs. 1 and 2). The ground gears comprised 13 drops (114 mm long)made from four links of 6-mm Ø chain in 32-mm Ø plastic tubing (spaced 550/600 mm apart) and were attached to either 8-mm Ø stainless-steel chain (termed the ‘8-mm ground chain’ and with a weight of 13.00 kg; n= 2) or 10-mm Ø stainless-steel chain (termed the ‘10-mm ground chain’; 17.90 kg; n = 2) (Fig.2). The ground gears were secured (and removed) to the footropes of either trawl using plastic cable ties. The tickler chain was made from 4-mm Ø chain (2.30 kg) and could be clipped to the last links of the ground chains, which meant that during fishing it orientated on the substrate, midway between the headline and footrope (～250 mm anterior to the centre of the ground chain; Fig.1b).

2.2. Experimental design and data collected

Fifteen replicate meshes were measured for SMO (to the nearest 1 mm) in each of the two trawls, extensions and codends using a local,purpose-built gauge with constant tension. At the start of each fishing day, either one of the replicate 8-mm and one of the 10-mm ground chains, both 8-mm ground chains or both 10-mm ground chains were secured to each trawl, which were then were clipped to the sweeps on each side of the vessel. The Notus paired wing-end sensors were secured to the trawl wing ends and the tickler chain was randomly assigned to either trawl. The double configuration was then deployed between 08:00 and 14:00 h over up to nine replicate 35-min deployments at one of two separate locations in the fishery. The trawls were swapped from side-to-side after each day, and one of the three possible combinations of ground gears (with and without the tickler chain on each side) was reassigned. Over six days of fishing, two replicate days of all three possible ground-gear configurations (8- vs 10-mm, 10- vs 10 mm and 8-vs 8-mm) with and without the tickler chain were completed.

Technical data collected during each deployment included the: (1)depth; (2) SOG; (3) distance trawled (otter boards on and off the bottom-obtained from the plotter and Notus trawl-monitoring system);and (4) the wing-end spreads (in m recorded at 1-min intervals during each deployment). The latter two variables were used to calculate ha trawled. All catches were removed from the codends in each deployment, separated onto a partitioned tray and recorded. Biological data collected from each trawl included the total weights of penaeids,bycatch and jellyfish, Catostylus mosaicus (whole individuals and any pieces were combined), the numbers of all other bycatch species, and the total lengths (TL in mm) of virtually all teleosts. Random samples of～200 g (～100 individuals) of penaeids were collected from each trawl catch, placed into plastic bags and transferred to the laboratory where they were separated into species, weighed and measured for carapace length CL (using Vernier callipers to the nearest 1.0 mm). These data were used to calculate species-specific weights and mean CLs in each deployment.

Fig.2. Plan of the (a) trawls used, and their ground gear, with (b) 4-mm Ø (diameter) chain used for the tickler, (b) 6-mm Ø chain for the drops, and (d) 8- and (e)10-mm Ø chains for the ground gears. T, transversals; N, normal; B, bars; SMO, stretched mesh opening; PE, polyethylene.

2.3. Statistical analyses

The hypothesis of no differences in the nominal SMOs of each trawl,extension and codend was tested using separate linear models (LM).Data describing key engineering and biological variables were analysed in linear mixed models (LMMs). Catch variables with sufficient replication across deployments were log-transformed so differences between treatments were modelled to act multiplicatively, rather than additively.Technical data and CLs were analysed raw.

All LMMs all included ‘ground chain’ (8- v 10-mm), ‘tickler chain’(with or without) and their interaction as fixed, while ‘fishing locations’,‘sides of the vessel’, ‘trawls’, ‘days’ and ‘deployments’ (within days and locations) were random. For the LMM assessing wing-end spread, SOG and depth were included as additional fixed effects, and the most appropriate model chosen via forward selection. The LMMs were fitted using the ASReml package of the R statistical language and the significance of trawl configuration was determined using a Wald F-test (Butler,Cullis, Gilmour, & Thompson, 2018). Predicted mean numbers and weights (deployment) were obtained by back-transforming log-predictions from the LMMs and are presented along with raw means and SEs.

3. Results

3.1. Technical data

The two trawls, extensions and codends had the same SMOs, with overall means (±SE) of 37.9 (±0.1), 41.4 (±0.2) and 27.3 (±0.1) mm,respectively (LM, p ＞0.05). In total, the trawls were used in 65 deployments (all 35 min and at SOGs of 1.1-1.4 ms) comprising 14-18 replicates each of the 8- and 10-mm ground chains with and without the tickler. There were no significant effects of ground chain, tickler chain,their interaction or SOG on wing-end spread (overall mean ±SE of 4.1± 0.1 m), with the final model reduced to depth only (positive coefficient) (Table 1). Similarly, the area trawled was not affected by any of the fixed categorical effects (LMM, p ＞0.05; Table 1).

Table 1 Summaries of Wald F-statistics from linear mixed models (LMM) assessing the importance of the fixed effects of ground chain (8 vs 10 mm), tickler chain (with or without) and their interaction for explaining variability among technical and biological responses. Excluding the mean carapace lengths (CL) of school prawns, Metapenaeus macleayi and eastern king prawns, Penaeus plebejus, all numbers and weights (kg) were log-transformed. Random effects included ‘locations’, ‘days’, ‘trawl sides’, and ‘deployments within days and locations’ for all LMMs. -, not appropriate for model.

3.2. Biological data

Total catches included 297 kg of penaeids (encompassing 247 kg of school prawns, 50 kg of eastern king prawns, and 0.2 kg of greasyback prawns, Metapenaeus bennettae) and 109 kg of bycatch (32 species),mostly comprising fish ＜10-15 cm TL (Table 2). Eight species dominated bycatch (＞80%, and formed the basis of analyses), including jellyfish (comprising mostly whole small individuals, but also some pieces of tentacles), yellowfin bream, Acanthopagrus australis, southern herring,Herklotsichthys castelnaui, pink-breasted siphonfish, Siphamia roseigaster,silver biddy, Gerres subfasciatus, squid, Uroteuthis sp., blue swimmer crabs, Portunus armatus, and Ramsey’s perchlet, Ambassis marianus(Tables 1 and 2).

There were significant main effects of ground chain and/or tickler chain on the weights of total penaeids and jellyfish, school prawns and total bycatch and the numbers of southern herring, blue swimmer crabs and squid (LMM, p ＜0.05; Table 1; Figs. 3 and 4), and with no interactions, although the p-value was 0.05 for the latter two variables(Table 1; Figs. 3 and 4). In terms of ground-chain effects, compared to trawls rigged with the 8-mm ground chain, those with the heavier 10-mm ground chain caught significantly less total penaeids (predicted means reduced by 14%), total bycatch (by 39%) and jellyfish (by 39%)(LMM, p ＜0.05; Table 1, Fig.3a, d and e). Most of the effects on total penaeids appeared to be caused by school prawns, but there were no significant differences in mean CLs of this species (16.4 ± 0.9 mm and 16.3 ±0.9 mm) or eastern king prawns (15.3 ±0.9 mm and 15.2 ±0.9 mm) between ground gears (LMM, p ＞0.05; Table 1, Fig.5a and b).

Table 2 Common and scientific names, quantities (numbers, unless stated otherwise)and sizes of organisms (in alphabetical order of common name) caught during the experiment. Sizes were carapace length in mm for penaeids and total length in cm for fish. Na means that these species were not measured. ^economically important in other fisheries.

Fig.3. Differences in raw (+SE; grey histograms) and predicted (white) mean catches 35-min deployment-1 between trawls rigged with 8- or 10-mm ground chains for the weights of (a) total penaeids, (b) school prawns, Metapenaeus macleayi, (c) eastern king prawns, Penaeus plebejus, (d) total bycatch and (e) jellyfish, Catostylus mosaicus and the numbers of (f) yellowfin bream, Acanthopagrus australis, (g) southern herring, Herklotsichthys castelnaui, (h) pink-breasted siphonfish, Siphamia roseigaster, (i) silver biddy, Gerres subfasciatus, (j) squid, Uroteuthis sp., (k) blue swimmer crabs, Portunus armatus, and (l) Ramsey’s perchlet, Ambassis marianus. ＞and= indicate significant differences (or otherwise) detected in linear mixed models (p ＜0.05).

Irrespective of the ground-chain diameter, trawls rigged with a tickler chain caught significantly more total penaeids (increase in predicted means by 1.13×) and school prawns (1.14×) (p ＜0.05), but not eastern king prawns (LMM, p ＞0.05; Table 1, Fig.4a-c). There were no significant differences in mean CLs due to the presence or absence of a tickler chain for either species (16.4 ± 0.9 mm vs 16.3 ± 0.9 mm for school prawns, and 15.5 ±0.9 mm and 15.0 ±0.9 mm for eastern king prawns) (LMM, p ＞0.05; Table 1, Fig.5a and b). Installing a tickler chain also significantly increased the weight of jellyfish (by 1.23×) and blue swimmer crabs (by 1.52×), but significantly reduced the numbers of southern herring (by 38%) and squid (by 37%) (LMM, p ＜0.01;Table 1; Fig.4e, g, j and k). There were no other significant effects on any other variables (Figs. 3 and 4).

4. Discussion

The data support the utility of simple changes to penaeid-trawl ground gears for improving efficiency that are often anecdotally implied, but rarely quantified (Broadhurst et al., 2015a; Chittenden &van Engel, 1972; Deshpande & Kartha, 1967; Deshpande & Sivan, 1962;Goeden et al., 1990; Watson et al., 1984). It is also clear the effects were species-specific and encompassed not only differences among taxa that dominated catches (decopods, teleosts, cephalopods and Scyphozoa),but between genera for penaeids. Attempts at deciphering the mechanisms supporting the observations are best made by considering known taxon-specific behavioural responses to the two technical main effects of interest (ground chain size and presence/absence of a tickler chain).This information can then be used to recommend future changes/modifications to the studied fishery and other similar fisheries.

In terms of penaeid behaviour, it is well established that while there are species-specific diel patterns, individuals typically reside on or are buried in the substrata, especially during daylight, and respond to external mechanical stimuli such as the ground gear of a trawl by ventrally flexing their abdomen, and often multiple times (Coles, 1979;Eayrs, 2002; Ruello, 1973; Watson, 1989). This flection propels individuals backwards in varying directions depending on their orientation, but typically upwards or sideways and, because many are orientated on the bottom, into the trawl mouth where some eventually become impinged against netting panels and are directed to the codend.Vision appears to have minimal impact on the behavioural responses of penaeids to trawls (Eayrs, 2002; Watson, 1989).

Fig.4. Differences in raw (+SE; grey histograms) and predicted (white) mean catches 35-min deployment-1 between trawls rigged without or with a tickler chain for the weights of (a) total penaeids, (b) school prawns, Metapenaeus macleayi, (c) eastern king prawns, Penaeus plebejus, (d) total bycatch and (e) jellyfish, Catostylus mosaicus and the numbers of (f) yellowfin bream, Acanthopagrus australis, (g) southern herring, Herklotsichthys castelnaui, (h) pink-breasted siphonfish, Siphamia roseigaster, (i) silver biddy, Gerres subfasciatus, (j) squid, Uroteuthis sp., (k) blue swimmer crabs, Portunus armatus, and (l) Ramsey’s perchlet, Ambassis marianus. ＞and= indicate significant differences (or otherwise) detected in linear mixed models (p ＜0.05).

By comparison, teleosts and especially those that school, have more complex reactions to trawls (Bayse, Pol, & He, 2016; Ryer, 2008; Watson, 1989). Many teleosts can visually detect the approaching trawl components and attempt escape, which usually involves orientating away and swimming at speeds dictated by their physiology, and mostly their size (Videler & He, 2010; Watson, 1989). Some teleosts can escape over the headline or under the footrope, while others are directed back into the trawl by the various anterior components (e.g. otter boards and sweeps) and herded by the netting panels to the codend (Watson, 1989).

Comparatively few studies have assessed the behaviour of cephalopods (and more specifically, Decapodiformes) in trawls and there is some conjecture over responses within (e.g. Doryteuthis pealei; Glass,Sarno, Milliken, Morris, & Carr, 1999; Bayse et al., 2016) and between species (e.g. D. pealeii; Bayse, He, Pol, & Chosid, 2014 and Photololigo etheridgei; Scandol, Underwood, & Broadhurst, 2006). Nevertheless, visual cues appear to be important and depending on their mantle orientation when they detect a trawl, like for teleosts, some attempt escape in response to stimuli (Bayse et al., 2014; Glass et al., 1999). In contrast to all other taxa, the response of Scyphozoa would be passive and, owing to their very poor swimming ability (Neil & Askew, 2018), their entry into trawls likely depends on their orientation in the water column at the entrance.

Considering the behaviours above, it is possible to postulate causes for the observed differences among treatments. The 10-mm chain had a slightly greater weight and surface area, which although not sufficient to affect wing-end spread, intuitively might be expected to create more stimuli for penaeids. But, irrespective of the presence of a tickler chain,trawls with the 8-mm Ø chain caught significantly more total penaeids(mostly school prawns) than trawls with the 10-mm Ø chain. Broadhurst et al. (2015a) did not observe the same effect for these ground chains used on beam trawls, which might support one hypothesis concerning bottom contact. By being smaller, and comprising more links, the 8-mm chain may have better followed the bottom contours, and especially in an otter trawl where the movement of one component of the spreading mechanism (otter board) was independent of the other. In contrast, any variations in bottom conditions that affect the sled on a rigid beam will transfer to the other sled and the entire ground chain. This possible effect may also explain the significantly greater weights of jellyfish in the trawls with the 8-mm Ø chain and because this was the dominant species by weight, the total weight of bycatch. At least some jellyfish may have been on the bottom and were more effectively directed into the trawl by the lighter chain.

Although speculative, a possible factor contributing to reduced catches of penaeids and jellyfish by the heavier chain is that because the same length drops were used (to preclude confounding the distance between the chains and the footrope), the 10-mm chain increased the footrope height from the substrate by 4.2 mm, which although small was～4% of the total. This increase may have been sufficient to allow some school prawns and or jellyfish to escape between the footrope and the ground chain. Obviously, such results are hypothetical in the absence of video (which would nevertheless be difficult owing to the poor water visibility).

Fig.5. Carapace-length (measured) frequencies of (a) school prawns, Metapenaeus macleayi and (b) eastern king prawns, Penaeus plebejus retained in trawls with either 8- or 10-mm ground chains, with or without tickler chains. n, number measured.

Unlike the ground chains, the effects of tickler chains on the key abundant species were more predictable. Increasing the mechanical stimulus on the substrate in front of the trawl increased total catches of penaeids by 1.13×, but this was entirely due to school prawns. Catches of eastern king prawns were not affected. Such species specificity probably reflects emergence behaviours (Coles, 1979). Eastern king prawns are primarily nocturnal and burrow during the day, while school prawns are characterised by relatively greater diurnal activity. The tickler chain was sufficient to stimulate school prawns at or near the substrate, but the heavier posterior ground chain may have been required for eastern king prawns. The species-specific variations in catches were independent of size.

The tickler chain also increased catches of the economically important blue swimmer crabs (the only other species caught in sufficient numbers), and probably following the same mechanisms as for school prawns. Chittenden and van Engel (1972) observed comparable results for another portunid, the blue crab, Callinetetcs sapidus, in a USA penaeid-trawl fishery, whereby catches were nearly tripled in a trawl with a tickler chain. The only other species affected by tickler chains here was jellyfish. Following the logic above, potentially relatively more individuals on the bottom were directed upwards into the trawl, rather than passing through the space between the footrope and the ground chain.

Unlike school prawns, blue swimmer crabs and jellyfish, installing a tickler chain reduced the numbers of one teleost, southern herring and the only squid species encountered (both economically important).These results may reflect behavioural responses to the tickler chain being dragged across the bottom and, more specifically, some visual stimulus (perhaps created by sand or mud clouds) that might have facilitated detection and escape: either over the headline or between the footrope and ground chain.

Such effects have been suggested as the general mechanisms supporting so-called ‘counter herding devices’ proposed by several authors(e.g. Ryer, 2008) and then tested in penaeid (Broadhurst, Sterling, &Millar, 2015b; McHugh, Broadhurst, Sterling, & Millar, 2015; Melli,Broadhurst, & Kennelly, 2019) and other crustacean-trawl fisheries(Melli, Karlsen, Feekings, Herrmann, & Krag, 2018). Similar to tickler chains, counter-herding devices are rigged anterior to the trawl, either suspended in the water column (Broadhurst et al., 2015b; McHugh et al.,2015) or on the bottom (Melli et al., 2018) and are designed to generate visual stimuli for key species to react. Of note, in the same fishery assessed here, Melli et al. (2019) showed that counter-herding devices comprising canvas banners located midway between the headline and footrope significantly reduced catches of southern herring by 44-59%-reductions only slightly larger than observed here for the tickler chain(38%).

Irrespective of the various mechanisms supporting species-specific responses to the ground-gear modifications, the data clearly show tickler chains can not only increase the catches of some targeted penaeids, but also positively and negatively affect the bycatches of some other key species. In some fisheries, there are concerns about tickler chains increasing total bycatch (Chittenden & van Engel, 1972; Geoden et al., 1990), while in other fisheries these have been refuted (Deshpande & Kartha, 1967; Deshpande & Sivan, 1962). Such variability probably reflects not only the different species, including their orientations in the water column (benthic vs semi pelagic) and abundances in bycatches and, but also the different trawls and/or gear-specific rigging arrangements.

The ecological impacts of any increases or decreases among the key bycatch species discussed above require careful consideration. Here,significant increases in catches of jellyfish and blue swimmer crabs were observed, but these are unlikely to have negative ecological impacts because of the small numbers, and also because most blue swimmer crabs probably survive discarding (Wassenberg & Hill, 1989). Nevertheless, further research is warranted. By comparison, southern herring and cephalopods in general have discard mortalities approaching 100%(Broadhurst et al., 2006; 2008), and so any modifications that support their bycatch reduction should be promoted-assuming escapees survive.

Notwithstanding the need for ongoing research, in the interim tickler chains represent a simple, inexpensive mechanism by which regional fishers can consistently increase (across different ground gears and various fishing conditions) catches of school prawns without significantly impacting total bycatches. Practical modifications like tickler chains should be encouraged both locally and more broadly, because improved fishing efficiencies through maximised economic return with minimal environmental impact will contribute towards more effective harvesting.

CRediT authorship contribution statement

Matt K. Broadhurst: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing.

Declaration of competing interest

I declare I have no conflict of interest.

Acknowledgements

This study was funded by the New South Wales (NSW) Department of Primary Industries. Thanks are extended to the (NSW) Professional Fishermen’s Association, and especially Steve Everson and Don Johnson. Sean Blake is especially thanked for his field and technical assistance.