Content uploaded by Paul Brown
Author content
All content in this area was uploaded by Paul Brown on Nov 08, 2015
Content may be subject to copyright.
Effects of gear type, entrance size and soak time
on trap efficiency for freshwater crayfish
Cherax destructor and C. albidus
Paul Brown
A
,
C
,
D
,Taylor L. Hunt
A
and Khageswor Giri
A
,
B
A
Fisheries Victoria, Department of Environment and Primary Industries, Queenscliff,
Vic. 3225, Australia.
B
Biometrics Unit, Agricultural Research Division, Department of Environment and
Primary Industries, Parkville, Vic. 3053, Australia.
C
Present affiliation: The Murray–Darling Freshwater Research Centre and La Trobe University,
PO Box 3428, Mildura, Vic. 3501, Australia.
D
Corresponding author. Email: paul.brown@latrobe.edu.au
Abstract. Freshwater crayfish support significant commercial and recreational fisheries worldwide. The genus Cherax
is fished in Australia with a variety of fishing gears, yet little is known of the relative efficiency of the different fishing
gears and methods. Additionally, freshwater-crayfish traps can pose a risk to air breathing by-catch such as aquatic
mammals, reptiles and birds, so by-catch mitigation is important. We sought to understand whether freshwater-crayfish
fishing can be undertaken efficiently, using passive traps and nets, without undue risk to air-breathing by-catch species.
In field-experiments, we compared the efficiency of six gear types and tested the effect of five exclusion rings on catch
performance over three soak times. The efficiency of gear types varied significantly by soak times. In productive locations,
catch can be maximised by repeatedly deploying open-topped gear for short soak times. Opera-house traps fitted with fixed
entrance rings (45–85-mm diameter) were not size-selective for yabbies. Encouragingly, open-topped gear and opera-
house traps fitted with fixed ring entrances much smaller than many commercially available (45-mm diameter) still fish
effectively for yabbies. We believe that smaller fixed ring-entrance size is likely to be correlated with a reduced risk of
by-catch for air-breathing fauna.
Additional keywords: by-catch exclusion, crawfish, fishing-gear efficiency, yabby trap.
Received 15 September 2014, accepted 30 November 2014, published online 7 April 2015
Introduction
Fisheries for freshwater crayfish are economically important in
several states in the USA and Australia, and in Europe, for human
consumption and also as fishing bait (FAO 2012). USA indus-
tries rely on Procambarus spp., whereas in Australasia, it is
several species within the endemic genus Cherax, across much of
the Australian states of Western Australia, South Australia,
Victoria, New South Wales and Queensland that support sig-
nificant commercial and recreational fisheries. An Australian
national recreational and indigenous fish survey estimated an
annual catch in 2000–2001 of almost 7.5 million freshwater
crayfish (Henry and Lyle 2003). The genus Cherax (Family:
Parastacidae) comprises 26 species of freshwater crayfish, with
four others under investigation (McCormack 2012). Commercial
and recreational catches of freshwater crayfish vary strongly, for
example between 1974 and 1984, commercial harvest in New
South Wales varied from 30 to 170 t, with population abundance
strongly linked to hydrologic conditions (Rankin 2000). A boom
in freshwater-crayfish stocks after drought-breaking rains
(van Dijk et al. 2013) across Victoria and the rest of southern
Australia brought an increase in the popularity and productivity
of recreational fishing for freshwater crayfish, both for bait and
for the table (ABC 2011). The Victorian fishery is mainly for two
similar species, Cherax destructor and C. albidus; both were
encountered in the present study. Genetic and morphological
studies indicated that C. albidus may merit subspecific classifi-
cation (Campbell et al. 1994).
A range of innovative new ‘pyramid’ trap designs are now
commercially available and have become widespread in the
community alongside lift nets and opera-house traps. Although
the legality of use of these designs varies across different state
jurisdictions and waterways, most state recreational fishing
regulations allow for the more traditional approach of using lift
nets (in public waters) and opera-house traps (see Fig. 1; at least
in private waterways) (DEPI 2013). Lift nets require an active
approach to fishing, with fishers regularly attending the nets,
whereas traps enable a more passive approach so that freshwater
crayfish can be harvested over long soak times.
CSIRO PUBLISHING
Marine and Freshwater Research, 2015, 66, 989–998
http://dx.doi.org/10.1071/MF14284
Journal compilation ÓCSIRO 2015 www.publish.csiro.au/journals/mfr
Some recreational and commercial freshwater-crayfish fish-
ers are concerned about the potentially high efficiency of new
trap designs on the market compared with lift nets (Charas 2010;
Vic DPI 2011). Little scientific data were available on the
relative efficiency of the different fishing gears and methods
(Campbell and Whisson 2001), and published data were insuf-
ficient to support the substantial interstate variation in permitted
types and quantities of recreational fishing gear.
Compared with lift nets, traps pose a risk to air-breathing
by-catch such as aquatic birds, mammals and reptiles (Limpus
et al. 2006;Serena and Williams 2010). Some of the available
traps have entrance holes lined with a metal ring to exclude
large by-catch such as turtles and to hold the entrance open to
enable air-breathing fauna to escape more easily (NSW DPI
2011). Entrance rings are of benefit because they prevent the
entry of by-catch above a threshold size-range. As a general
principle, the smaller entrance the rings have, the less by-catch is
expected. Small-entrance rings may also reduce the freshwater-
crayfish catch rate (M. Allanson and S. Thurstan, NSW DPI,
pers. comm.). However, more information is required to
clarify the relationship between exclusion-ring diameter and
freshwater-crayfish trap performance and whether longer soak
times can compensate for this.
Exclusion of by-catch from freshwater-crayfish traps by
restricting entry is a more appropriate by-catch mitigation
practice than is provision of escape vents. Freshwater-crayfish
traps are passive capture devices that pose a risk to species such
as platypus (Ornithorhynchus anatinus), water rats (Hydromys
Fig. 1. Clockwise from top left: lift net (LN), pyramid trap long side (PTLS), pyramid trap short side (PTSS), pyramid
trap ringed funnel (PTRF), opera-house net ringed funnel (OHRF) and opera-house net collapsible funnel (OHCF).
990 Marine and Freshwater Research P. Brown et al.
chrysogaster) and turtles (Chelodina spp. and Emydura spp.).
Once inside a trap, air-breathing animals undergo significant
cumulative stress in the process of escaping traps, possibly
resulting in death (Davis 2002;Broadhurst 2008). Exclusion of
by-catch fauna avoids these risks. As freshwater-crayfish traps
are inexpensive to purchase and in wide use, the development
and implementation of other effective by-catch reduction
methods, such as ‘escape vents’ for all potential by-catch
species in freshwater-crayfish waters, may be unfeasible
(Grant et al. 2004).
Using a field-experimental approach we sought to compare
the efficiency of a range of freshwater-crayfish traps and to test
the effect of exclusion rings on freshwater-crayfish catching
performance. How can recreational fishing for freshwater cray-
fish be undertaken efficiently and safely, using passive gear such
as traps, while limiting the risk to air-breathing by-catch species?
Materials and methods
Experimental sites
Field trials were undertaken where reasonable catch-rates of
freshwater crayfish were anticipated, as determined after con-
sultation with fisheries compliance professionals and drawing
on local knowledge about recent recreational fishing activity
and success rates.
For Experiment 1, the experimental sites were Charam
Swamp (3685304500S, 14182703200 E) and Lake Charlegrark
(3684600100S, 14181401700 E), in the Wimmera River catchment,
and Reedy Lake (3684300100S, 14580600700 E) in the Goulburn
River catchment. Charam Swamp and Lake Charlegrark are
shallow swamps (IWC 2012), with an average depth of ,1–3 m
during March 2012 and areas of 40 and 56 ha respectively.
Reedy Lake is categorised as a 300-ha deep swamp (.5m)
(IWC 2012); however, in April 2012, the average depth was
1–3 m.
For Experiment 2, the experimental sites were Miga Lake
(3685504100S, 14183701900 E) and Clear Lake (3685504800S,
14185105800E) in the Wimmera River Catchment. Miga Lake
and Clear Lake are both open-water shallow swamps ,50 ha in
extent, with maximum depths in September 2012 of ,2m.
Water quality
At each sampling site during Experiment 1, water-quality
parameters were measured (YSI meter). At the beginning and
end of each day, water temperature, dissolved oxygen and pH
were measured offshore and at ,0.5-m depth. During Experi-
ment 2, water temperature was measured at midday with a spirit
thermometer.
Water temperatures at Charam Swamp and Lake Charlegrark
in March were between 17 and 218C, and for Reedy Lake in
May, water temperature had dropped to between 12 and 148C.
Sampling sites were well oxygenated. Dissolved oxygen in
March was 7–8 mg L
1
and in May at Reedy Lake it varied from
7to10mgL
1
. All sites were close to neutral pH, with the range
of measurements between 6.8 and 7.5 pH units.
Experiment 1: evaluating different gear types
A63 factorial randomised-block design was used to
answer our research objectives. Experimental fishing was
undertaken comparing six types of popular freshwater-cray-
fish gear at three different soak times. The six different types
of freshwater-crayfish gear were lift nets (LN), opera-house
traps (2 collapsible funnels) (OHCF), opera-house traps
(75-mm-diameter ring in funnels 2) (OHRF), pyramid
traps (open top) long sides (PTLS), pyramid traps (open top)
short sides (PTSS), pyramid trap (closed top with 90-mm ring
in funnels 4) (PTRF) (Fig. 1).
Each gear was used for the following three different soak
times: a short soak (1 h), to simulate active fishing methods
typical of widely used lift nets; a medium soak (6 h), to simulate
the strategy of deploying and retrieving gear at the beginning
and end of a day trip; and a long soak overnight typically ,12 h),
to simulate fishing gear deployed on multi-day trips.
There were three replicates for each gear type at each soak
time. At each day, 6 33 (54) traps were deployed for
fishing. This experimental fishing was undertaken at three
different sites, and at each site, 3 consecutive days (including
three overnight soak times) of fishing was carried out.
Experiment 2: evaluating effects of by-catch exclusion rings
on trap performance
A53 factorial randomised-block design was used with five
experimental ring diameters fitted to opera-house trap entrance
funnels. Three soak times were used consistent with Experiment
1 (i.e. 1 h, 6 h and 12 h). The rings used were nominally 45, 55,
65, 75 and 85 mm in diameter, with each trap being fitted with
two equivalent-sized rings.
There were three replicates for each ring size at each soak
time. At each day, 5 33 (45) traps were deployed for
fishing. This experimental fishing was undertaken at two differ-
ent sites. At Miga Lake, 2 consecutive days (including two
overnight soak times) of fishing were carried out; at Clear Lake,
1 day, including one overnight soak time, was completed.
Experimental fishing
Baits used in freshwater-crayfish gear were pieces of fish. Bait
was uniform in size (,150 g) and type and held in bait bags
constructed of ‘mussel-mesh’ (polypropylene, 10-mm mesh
size). Bait bags were secured to the base of each trap or net near
the centre in a uniform manner.
At all locations on each day, gear was deployed in a randomly
allocated manner to remove the risk of subjective bias from
operators choosing which gear went where. Each day, gear was
fished in a different randomly chosen portion of the available
habitat to ensure individual freshwater crayfish were encoun-
tered only once. For each soak-time treatment, gears of six
different types, or the five different exclusion-ring sizes, were
deployed in a randomly allocated sequence.
Three different float types were used to mark each soak-
time treatment. This enabled operators to see easily what gear
to haul, and what to leave, when multiple soak-time treatments
overlapped.
Freshwater crayfish captured in each trap or net were all
counted and the total catch from each trap or net was weighed
(liveweight, 1 g).
For Experiment 1, a subsample of up to five individuals per trap
or net was randomly chosen and occipital carapace length (OCL)
Freshwater crayfish gear efficiency and by-catch exclusion Marine and Freshwater Research 991
was measured using vernier callipers (1 mm). During Experi-
ment 2, all individuals were measured (OCL, mm) from each trap.
All freshwater crayfish were returned to the water immedi-
ately after counting, weighing and measuring were completed.
Statistical analyses
For Experiments 1 and 2, catch by number and catch by weight
were analysed after standardising to nominal soak-times of 1, 6
and 12 h. The duration required for setting and hauling gear and
processing the catch meant that actual soak times varied for
individual experimental units (i.e. traps and nets). To enable
standard comparisons, the catch rate for each individual
experimental unit iwas calculated as
cpuei¼ci
ðthaul;itset;iÞ
where cpue is catch per unit effort (catch per hour), c
i
is catch in
gear unit i(total number of individual freshwater crayfish,
number of large freshwater crayfish or total weight of freshwater
crayfish caught per haul) and t
haul
and t
set
are the times that unit i
is hauled or set.
For each experimental unit, nominal soak times, S, are
multiplied by individual-trap cpue, to estimate standardised
catch, as follows:
Catch ¼cpueiS
where Sis 1-, 6- or 12-h soak-time treatment.
For Experiments 1 and 2, the standardised catch and weight
data from all sites were analysed using general analysis of
variance, as follows:
For Experiment 1,
Treatment structure ¼Gear type Soak time Sites
Blocking structure ¼Sites þSites Dates
For Experiment 2,
Treatment structure ¼Ring diameter Soak time Sites
Blocking structure ¼Sites þSites Dates
To reduce the skewness and stabilise the variance of residuals
of the standardised data, mathematical transformations of the
data were necessary for both experiments as follows:
For the catch data,
Catch !logðCatch þ5Þ
For the standardised weight data,
Weight !ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Weight
p
For Experiment 2, the standardised catch of large freshwater
crayfish and the proportion of large freshwater crayfish in the
catch,
Large catch !logðLarge catch þ0:5Þ
Large catch%!logðLarge catch%þ0:005Þ
One trap in Lake Charlegrark, that caught a single large
freshwater crayfish, was deleted from the analysis of standar-
dised weight in Experiment 1 as an extreme statistical outlier.
Two extreme outlier traps were deleted from the analysis of
standardised catch (both from Miga Lake), and the analysis
of standardised weight (one from Miga Lake and one from
Clear Lake).
The difference between predicted means is judged signifi-
cant if its magnitude is greater than the least significant
difference.
Size analysis
The proportion of large freshwater crayfish in the measured
sample was compared for each gear type. There is no minimum
legal size for freshwater crayfish (Cherax spp.) in Victoria
(DEPI 2013). Therefore, to determine an appropriate benchmark
size for ‘large’ freshwater crayfish for human consumption, we
conducted a brief survey of the commercial retail market.
A review of market size among commercial freshwater-
crayfish growers suggested that individual freshwater crayfish
larger than 60 g would be an acceptable ‘benchmark’ for harvest
for human consumption. Length data from sampled freshwater
crayfish allowed us to classify them as ‘large’ or ‘small’ on the
basis of whether they were larger than 45-mm carapace length
equating to a liveweight of ,60 g.
To compare the distribution of the catch of large freshwater
crayfish across sites and gear types the proportion and standard
error of large freshwater crayfish ($45-mm OCL) in the catch at
each site, and for each gear type, were calculated and the
confidence limits on this proportion were determined as follows:
sp¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pð1pÞ
n1
r
where s
p
is the standard error of the sample proportion pfrom the
sample n, and the 95% confidence interval of the proportion p
with the normal approximation is [p–s
p
1.96, pþs
p
1.96 ].
Proportions with non-overlapping confidence intervals were
considered significantly different.
Results
Experiment 1: evaluating different gear types
The total catch was 3826 freshwater crayfish, weighing 112 kg.
Analyses of the length subsample indicated that ,26% of the
catch was of generally acceptable harvest size (i.e. 45 mm, OCL
and ,60 g).
General analysis of variance of the freshwater-crayfish catch
data showed that catch efficiency among locations depended on
soak times (F¼38.8, P,0.001). Under the assumption that
catch rate broadly indicated population density, the optimum
soak time to maximise the catch depended on the density
of freshwater crayfish at the fishing location. With lower-
density populations such as Lake Charlegrark and Reedy Lake,
the catch continued to accumulate over longer soak times.
Long overnight soak times yielded the highest catches using
all gears. With higher-density lakes such as Charam Swamp, the
catches generally peaked after 6 h and declined for longer
overnight soaks.
992 Marine and Freshwater Research P. Brown et al.
The catch efficiency among gear types depended on soak
times (F¼9.1, P,0.001) (Fig. 2,Table 1).
The catch using the PTSS was similar to that using PTLS.
However, the catch using PTSS was significantly higher than
those with the other gears over the short (1 h) soak time (Fig. 2).
Generally, gear types with constrained entrances, for example,
opera-house traps and pyramid trap with funnel entrances, had the
lowest catch efficiency over short soak times. In the simulation of
active fishing with 1-h soak times, with the exception of the open-
topped PTSS, all gears had similar catches (Table 1).
Over medium soak times, traditional LNs had the lowest
catch efficiency and catches were significantly lower than those
obtained with all other gear types (Fig. 2). All pyramid-trap
variants (PTSS, PTLS and PTRF) had similar catches and
performed similarly to opera-house variants with rings in the
funnels (OHRF) or collapsible funnels (OHCF). The OHCF was
clearly the most efficient during medium soak times at almost
twice the catch of LNs. Lift nets were no more effective than
during short soak treatments, whereas all other gear types were
more effective than during short soak treatments.
After long overnight soak times, the OHCF traps were the
most effective, catching significantly more than the OHRF,
which in turn caught significantly more than all other gears
(Fig. 2). The slight increased catches in LN were not signifi-
cantly different from LN catches at medium or at short soak
treatments.
OHCF traps had significantly increased catches after longer
soak times (Fig. 2). There was no significant change in catch
between medium and long soak times for LN and OHRF.
Catches significantly declined in all pyramid-trap variants
fished for the long soak times.
General analysis of variance of the freshwater-crayfish catch
(weight) data showed that results varied by site and gear type
broadly mirror the previously reported results for numbers of
freshwater crayfish caught. At short soak times, both open-
topped pyramid traps (PTSS and PTLS) had significantly
heavier catches of freshwater crayfish than did other gears. At
medium and long soak times, results were equivalent to catch-
by-numbers and will not be reported further.
The size distribution of the catch varied depending on the site
sampled (Fig. 3). The catch from Charam Swamp and Reedy
Lake had a similar size distribution with a single identifiable size
class of 25–55-mm OCL. The catch from Lake Charlegrark had
two distinct size classes, including freshwater crayfish larger
than 60-mm OCL (Fig. 4).
The mean size of freshwater crayfish caught varied among
location by gear type (F¼5.0, P,0.001) and by soak time
(F¼2.8, P¼0.026). Whereas mean size of freshwater crayfish
did not differ by gear type at Charam Swamp and Reedy Lake, at
Lake Charlegrark, where the size range of freshwater crayfish
included larger individuals, different gear types caught different
sizes of freshwater crayfish. At Lake Charlegrark, mean length
of all freshwater crayfish caught by pyramid traps with short
sides was the highest at 54-mm OCL. The mean length of all
freshwater crayfish caught by the LNs and pyramid traps with
rings and funnels was the lowest at 41-mm OCL.
Although the proportion of large freshwater crayfish
($45-mm OCL) in the catch varied among sites (24–32%),
there was no indication that any gear type was better at catching
large freshwater crayfish overall (Fig. 5). The differences in
mean size of freshwater crayfish caught among gear types were
largely driven by the catches at Lake Charlegrark. At Lake
Charlegrark, the greatest proportion of large freshwater crayfish
was caught by open-topped pyramid traps (PTSS and PTLS);
and the lowest proportion of large freshwater crayfish was
caught by the LN; however, sample size was small (n,20)
and confidence intervals overlapped for all gear types, indicat-
ing no statistically significant difference.
Experiment 2: evaluating effects of by-catch exclusion rings
on trap performance
Opera-house traps with exclusion rings in the entrance funnels
(OHRF) of five different nominal diameters were deployed in
two waterways over 3 days and three different soak times for 135
individual trap lifts. The total catch was 2699 freshwater
crayfish, weighing 89 kg. The whole catch from each trap lift
was counted, weighed (g) and sampled for individual lengths.
For most trap lifts, all the freshwater crayfish caught were
10
4
2
0
0
Soak time (h)
Yabbies caught
8
6
PTLS
PTRF
PTSS
OHCF
OHRF
LN
126
Fig. 2. Mean (back-transformed) standardised catch per trap (number of
freshwater crayfish) for each of six types of gear fished over soak times of 1,
6 and 12 h at Charam Swamp, Lake Charlegrark and Reedy Lake. LN, lift
net; OHCF, opera-house trap with collapsible funnel; OHRF, opera-house
trap with ring funnel; PTRF, pyramid trap with closed top and ring funnels;
PTLS, pyramid trap with open top and original long sides; PTSS, pyramid
trap with open top and modified short sides.
Table 1. Mean (back-transformed) standardised catch of freshwater
crayfish per trap for each of the six types of gear (see Fig. 1 for gear type
key) at the three soak times trialled
Soak time Gear type
LN OHCF OHRF PTLS PTRF PTSS
1 2.7 2.0 2.7 3.0 2.1 3.7
6 3.5 7.0 5.7 5.9 5.6 6.2
12 3.7 8.8 5.4 4.0 3.4 3.5
61h
A
16.3 12.2 16.3 18.1 12.4 22.0
12 1h
A
32.6 24.4 32.5 36.2 24.8 44.0
A
Simulated catches for active fishing in 1-h soaks over 6- and 12-h periods
included for comparison.
Freshwater crayfish gear efficiency and by-catch exclusion Marine and Freshwater Research 993
individually measured. Analyses of the lengths indicated
that ,3% of the catch was of generally acceptable harvest size
(i.e. $45-mm OCL and ,60 g). Actual ring diameters varied,
although average actual size was within 3 mm of nominal size
in each case (Table 2).
General analysis of variance of the freshwater-crayfish catch
data showed that catch efficiency for opera-house traps was not
dependent on the exclusion-ring diameter during Experiment 2.
None of the two-way interactions or the three-way interaction
was statistically significant. Consistent with the results of
Experiment 1, the numbers of freshwater crayfish caught varied
only by soak times (F¼101, P,0.001), not by ring diameters.
The mean numbers of freshwater crayfish caught at 6- and 12-h
soak times were approximately five times higher than that at the
1-h soak time (Fig. 6).
The catch data for large freshwater crayfish ($45 mm, OCL)
showed that catch efficiency for opera-house traps was not
dependent on the exclusion-ring diameter during Experiment
2. None of the two-way interactions or the three-way interaction
was statistically significant. The number of large freshwater
crayfish caught varied only by soak times (F¼8.6, P,0.001),
not by ring diameter. As with the total number of freshwater
crayfish, the mean numbers of large freshwater crayfish caught
at 6- and 12-h soak times were approximately five times higher
than that caught at the 1-h soak time (Fig. 6).
The data for weight of freshwater crayfish (g) caught indi-
cated that catch efficiency for opera-house traps was not
dependent on the exclusion-ring diameter during Experiment
2. None of the two-way interactions or the three-way interaction
was statistically significant. The weight of freshwater crayfish
caught varied only by soak times not by ring diameters.
The weights of freshwater crayfish caught at 6- and 12-h soak
times were approximately six times higher than that caught at
the 1-h soak time.
0%
10%
20%
30%
40%
50%
60%
70%
Reedy Lake, n 298
Charam Swamp, n 579 Lake Charlegrark, n 60
Percentage of large yabbies
(45 mm OCL)
Fig. 4. The proportion (95% CI) of large freshwater crayfish in the combined catch from all gears at
three sites in Victoria. (OCL, occipital carapace length.)
0%
10%
20%
30%
15 20 25 30 35 40 45 50 55 60 65 70
Frequency
Charam Swamp, n 579
0%
10%
20%
30%
15 20 25 30 35 40 45 50 55 60 65 70
Lake Charlegrark, n 60
0%
10%
20%
30%
15 20 25 30 35 40 45 50 55 60 65 70
Carapace len
g
th class (mm)
Reedy Lake, n 298
Fig. 3. Percentage frequency distribution of freshwater crayfish size (carapace length) caught from three
locations in Victoria during March and May 2012, using multiple gear types in Experiment 1. End points of
5-mm length classes are shown as x-axis labels.
994 Marine and Freshwater Research P. Brown et al.
Mean length (mm, OCL) of freshwater crayfish caught did
not vary according to the exclusion-ring diameter or soak-time
during Experiment 2. Miga Lake generally had bigger freshwa-
ter crayfish; however, mean size was not significantly different
from that in Clear Lake.
The proportion of large freshwater crayfish ($45-mm OCL)
caught indicated that this proportion varied only by soak times
and did not vary by exclusion-ring diameter during Experiment
2. The proportion of large freshwater crayfish caught at 6- and
12-h soak times were approximately four times higher than that
caught at the 1-h soak time.
Discussion
A review of marine crustacean trap fisheries considered many
factors important in regulating the catch of crabs and lobsters,
including trap size, bait quantity and quality, soak time and
prevention of escape through the entrance; however, the largest
potential for increasing trap catches was by increasing ease of
entry and reducing the effect of gear saturation (animals inside
traps preventing those outside from entering; Miller 1990). Our
results suggest that this is also the case for freshwater-crayfish
trap catches, with the effectiveness of new open-topped pyramid
traps enhanced by modifying the traps to cut down the length of
the sides, making it easier for freshwater crayfish to enter.
Although lift nets lying flat on the substrate should also be easy
to enter, they are equally easy for freshwater crayfish to leave,
and this may be the difference between their effectiveness and
that of the open-topped pyramid designs. Short soak times are
one way to minimise gear saturation. In our experiments, an
active approach yielded greater catches of freshwater crayfish
than did a passive one (see Table 1). An active fisher repeatedly
sets and resets each trap while moving to new locations in
between each set. At medium and long soak treatments, none of
the gears achieved the catch equivalent to those at short soaks
over the same period. Studies of freshwater crawfish (Pro-
cambarus clarkii) in the USA have shown that the highest catch
rates are achieved with empty traps, as opposed to traps with live
crawfish already inside (Hardee 2009). The presence of live
crawfish in traps inhibited the entry of additional crayfish
because of their aggressive, agonistic behaviour. Active fishing
for freshwater crayfish may capitalise on similar behaviour, and
improvements in catch size can be 200–300% higher than those
achieved by passive approach.
The present findings are that passive rather than active
fishing produces best results, which is opposite to current
regulation trends in Victoria. Indeed, a global trend in recrea-
tional fisheries management is to regulate fishing activity
towards active methods rather than passive ones such as reduc-
ing the number of fishing rods used (FAO 2012). Examples
include prohibition of recreational mesh nets, and more recently
of baited set lines in Victorian inland waters (Victorian Govern-
ment 2014). The common view of this trend is that it provides a
more conservative approach and limited harvesting for recrea-
tion rather than subsistence; however, paradoxically, the present
study suggests that active fishing and trends for regulating
against gears that promote passive fishing are likely to maximise
catches of freshwater crayfish in many circumstances. There is
presently no assessment of possible impacts of such trends on
the sustainability of freshwater crayfish stocks. The driving
need for recreational freshwater crayfish fishing regulations in
Table 2. Summary of actual measurements of exclusion-ring
treatments for Experiment 2
Parameter Nominal inside-diameter of exclusion rings (mm)
‘45’ ‘55’ ‘65’ ‘75’ ‘85’
Minimum 45 53 65 71 85
Mean 46 54 66 73 88
Maximum 48 55 67 75 90
All sites
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
LN OH OH PT PT PT
CF RF RF LS SS
Gear type
Percentage of large yabbies
(45 mm OCL)
Fig. 5. The proportion (95% CI) of large freshwater crayfish in the catch from each gear type. LN, lift net;
OH, opera-house trap with closed funnel (CF), or fixed ring in funnel (RF); PT, pyramid trap with fixed ring
funnel (RF), open top and original long sides (LS) or open top and modified short sides (SS). (OCL, occipital
carapace length.)
Freshwater crayfish gear efficiency and by-catch exclusion Marine and Freshwater Research 995
Australia has been to minimise by-catch mortality of non-target,
air-breathing aquatic fauna (Serena and Williams 2010).
We examined the role of fixed-diameter entrance rings as by-
catch exclusion devices on the freshwater-crayfish catch perfor-
mance and size-structure of the catch. Our assumption was that
the most effective by-catch exclusion-ring size in the present
study was the smallest-diameter ring studied (45-mm diameter).
There was no statistical difference between catches using this
diameter entrance and catches from traps of any of the larger
entrances. The only previous study of catches in opera-house
traps modified with entrance rings of various sizes was incon-
clusive; however, it indicated that catch variability was high and
no statistical difference in numbers of freshwater crayfish
caught was found among traps with variable entrance-ring size
(M. Allanson and S. Thurstan, NSW DPI, pers. comm.).
Recreational fishers generally catch freshwater crayfish
either to eat (i.e. large freshwater crayfish) or to use as bait for
finfish (i.e. small freshwater crayfish). The popularity of a by-
catch reduction device will likely depend on its ability to catch
both sizes. The size distribution of freshwater-crayfish popula-
tions sampled in our Experiment 2 showed no evidence of
relative size-selectivity in catches of traps with any particular
entrance size. Large and small freshwater crayfish, typical of
those harvested by recreational fishers, were adequately cap-
tured in traps with small (45-mm diameter) entrances. A study of
eight trap designs used to catch red-swamp crayfish (P. clarkii)
in France was carried out in a population with a size distribution
comparable to that of the present study (Paillisson et al. 2011).
These authors also showed that traps with small (40-mm
diameter) entrances caught good numbers of large crayfish
(.45-mm OCL) in comparison with other trap types with larger
entrances. Also, the standard inside diameter of entrances to
traps preferred in the Louisiana commercial crawfish fishery for
crawfish (Procambarus sp.) (McClain et al. 2007) of a preferred
size similar to that of Cherax sp. in the present study is
44–51 mm, on the basis of published length–weight relation-
ships (Do¨rr et al. 2006).
The optimum gear for maximising a recreational fisher’s
catch depends on freshwater-crayfish abundance and choice of
soak time. If abundance is low, catches are maximised by using
either open-topped pyramid traps (PTSS or PTLS) fished in
an active manner (i.e. short soaks), or opera-house nets with
long, overnight soak times. If abundance is high, maximum
catches are obtained by using modified open-topped pyramid
traps (PTSS) in an active manner (i.e. short soaks). Where
freshwater-crayfish abundance was low, such as Lake Charle-
grark during the present study, greater catches were achieved by
fishing in a more passive manner. Under these conditions, the
use of opera-house nets with collapsible funnels (OHCF) was
most effective. Campbell and Whisson (2001) found similar
results using OHCF for Marron (C. tenuimanus) in eight water-
ways in Western Australia. Surprisingly, in our low-abundance
populations, open-topped pyramid traps still gave greater yields
when fished actively rather than passively, although the consid-
erable effort required to achieve the small catch may make this
unlikely to be a popular behaviour among recreational fishers.
When considering fishing using long soak times, such as
overnight, the catch does not increase with soak time for all but
one gear type (OHCF). In abundant populations, this may be
because freshwater crayfish escape the gear after the bait
becomes depleted, freshwater crayfish become satiated or the
trap becomes overcrowded. The lowest overnight catches came
from gear with open tops (LN, PTLS, PTSS) or multiple short
funnels (PTRF) that may be easier to escape, whereas both
opera-house net designs retained higher catches. The opera-
house net with collapsible funnels (OHCF) was more effective
45 (1 h)
45 (6 h)
45 (12 h)
55 (1 h)
55 (6 h)
55 (12 h)
65 (1 h)
65 (6 h)
65 (12 h)
75 (1 h)
75 (6 h)
75 (12 h)
85 (1 h)
85 (6 h)
85 (12 h)
45 (1 h)
45 (6 h)
45 (12 h)
55 (1 h)
55 (6 h)
55 (12 h)
65 (1 h)
65 (6 h)
65 (12 h)
75 (1 h)
75 (6 h)
75 (12 h)
85 (1 h)
85 (6 h)
85 (12 h)
60
50
40
30
20
10
0
4
3
2
1
0
Fig. 6. Box plots showing a summary of catch data from Experiment 2. The total catch (left) and the catch of large freshwater
crayfish ($45-mm occipital carapace length)(right) in opera-house traps with five different diameter entrance rings (45 mm, 55 mm,
65 mm, 75 mm and 85 mm) standardised to three different soak times of 1 h (dotted), 6 h (dashed) and 12 h (solid). Boxes describe
inter-quartile range (Q3–Q1), with the median indicated withinthe box. Bars indicate range of data, with points indicating outliers.
996 Marine and Freshwater Research P. Brown et al.
the longer it was deployed, indicating very low, possibly zero,
rates of escape.
Reduced catches at long soak times could be caused by bait
depletion, satiation of freshwater crayfish or reduced bait
attractiveness. Jones et al. (1996) showed that food consumption
in C. destructor was 2–5% of bodyweight per day, declining as
freshwater crayfish grew. At these consumption rates, it would
take between 3 and 7.5 kg of freshwater crayfish to consume
150 g of bait on 1 day. In the present study, catches rarely
exceeded 600 g per trap. It is therefore unlikely that baits were
entirely depleted by consumption during experimental fishing.
Satiation of freshwater crayfish attracted to the trap seems more
likely. Bait was always present as the traps were hauled,
although the amount remaining was not measured.
There was a reduced rate of increase in catch for OHCF
between 6-h and 12-h soak times. Either because fish pieces
used for bait became less attractive after 12 h than they were at
6 h, or because entry rates declined as a result of inhibition from
freshwater crayfish already in the trap (Hardee 2009). Future
studies to determine optimum soak times could consider weigh-
ing bait before deployment and after retrieval, and looking for
statistical correlations between bait weight and catch at a range
of soak times.
Two other issues were not well defined and require further
study. At one location (Lake Charlegrark), we measured a
significant difference in mean size caught by some gears.
Neither entrance-ring size nor trap mesh size explain this as
traps with larger entrance rings (PTRF) caught smaller freshwa-
ter crayfish, and gear with the larger mesh sizes did not
consistently catch larger freshwater crayfish. In some trap
fisheries for crustaceans, size selectivity occurs as dominant
animals inhibit smaller ones from entering the trap (Miller 1979;
Frusher and Hoenig 2001). Our study found evidence of size
selectivity only at the site where large freshwater crayfish were
abundant. Larger sample sizes are necessary to confirm which
types of gear catch the largest freshwater crayfish.
No by-catch was actually encountered. Whereas non-target
by-catch was not encountered during the present study, at least
partially because of careful site choice, by-catch mortality is
only one possible end point of interactions between fishing gear
and non-target organisms. Other possibilities are encounter and
avoidance, or encounter and capture followed by escape.
Destructive sampling is practically flawed as a field-based
evaluation method of the risk of fishing gear to aquatic fauna
without parallel studies of avoidance and escape behaviour
(Grant et al. 2004). An alternative approach is required to
monitor the whole interaction of by-catch fauna with the gear
(i.e. encounter, entrance and exit). Further work is required to
evaluate relative by-catch risk of these gears, either in water-
ways with higher levels of non-target fauna, or under more
controlled laboratory conditions.
The present study was carried out in freshwater crayfish
stocks in five different freshwater lakes in Victoria. The findings
are likely to be broadly applicable in fisheries for freshwater
crayfish across similar types of habitat. However, differences in
catch efficiency among traps types are likely to occur across
different types of habitat (such as, for example, vegetated
swamps and flowing streams). Paillisson et al. (2011) noted
between-trap variations in standardised catch rate for different
habitats and suggested that environmental factors such as
vegetation cover and water level may influence the efficiency
of traps. Furthermore, the likelihood of different behavioural
and density-dependent factors associated with different species
means that caution is advised before applying these findings to
other species or genera of freshwater crayfish.
In conclusion, freshwater-crayfish catch varies by gear type
and, in productive locations, could be maximised by fishing
actively, repeatedly deploying open-topped gear (LN, PTSS or
PTLS) for short soak times and relocating between soaks.
Opera-house traps fitted with small fixed ring entrances
(45-mm diameter) fish effectively for freshwater crayfish and
are not size-selective. This suggests that opera-house traps could
be either retro-fitted with, or legislated to have, a maximum
entrance-ring diameter of 45 mm to limit the risk of by-catch,
without compromising their effectiveness as freshwater-cray-
fish traps.
Acknowledgements
Thanks go to Travis Dowling, Murray Burns and David Trickey, Department
of Environment and Primary Industries (DEPI); and to Melody Serena,
Australian Platypus Conservancy, for helpful suggestions during the project
planning. Thanks go to Murray Burns, Ash Hastie, Nathan O’Mahony,
Corey Green and Sean Blake, DEPI, for assistance with the fieldwork. Also,
thanks go to James Andrews, DEPI, and two anonymous referees, for helpful
comments on earlier drafts of this paper.
References
ABC (2011). Yabbies spike sparks net shortage. ABC News Online, 28 April
2011. Available at http://www.abc.net.au/news/2011-04-28/yabbies-
spike-sparks-net-shortage/2696030 [Verified September 2014].
Broadhurst, M. (2008). Maximising the survival of bycatch discarded from
commercial estuarine fishing gears in NSW. Final report to the Fisheries
Research & Development Corporation for Project 2005/056. NSW
Department of Primary Industries Fisheries Final Report Series number
107, Sydney.
Campbell, L., and Whisson, G. J. (2001). Catch efficiency of five freshwater
crayfish traps in south-west Western Australia. Journal of the Interna-
tional Association of Astacology 13, 58–66.
Campbell, N., Geddes, M., and Adams, M. (1994). Genetic-variation in
yabbies, Cherax destructor and C. albidus (Crustacea, Decapoda, Para-
stacidae), indicates the presence of a single, highly sub-structured
species. Australian Journal of Zoology 42, 745–760. doi:10.1071/
ZO9940745
Charas, S. (2010). We need your support for the protection of tabby and
shrimp against over fishing before September 10 2010. Available at
http://www.myenvironment.net.au/index.php/me/Work/Biodiversity/
Biodiversity-Resources/We-need-your-support-for-the-protection-of-
Yabby-and-Shrimp-against-over-fishing-before-September-10-2010
[Verified September 2014].
Davis, M. W. (2002). Key principles for understanding fish bycatch discard
mortality. Canadian Journal of Fisheries and Aquatic Sciences 59,
1834–1843. doi:10.1139/F02-139
DEPI (2013).‘Victorian Recreational Fishing Guide 2013.’ (Victorian
Government: Melbourne.)
Do¨ rr, A. J. M., La Porta, G., Pedicillo, G., and Lorenzoni, M. (2006). Biology
of Procambarus clarkii (Girard, 1852) in Lake Trasimeno. Bulletin
Francais de la Peche et de la Pisciculture 380–381, 1155–1168.
doi:10.1051/KMAE:2006018
FAO (2012). ‘Recreational Fisheries.’ (Food and Agriculture Organisation:
Rome.)
Freshwater crayfish gear efficiency and by-catch exclusion Marine and Freshwater Research 997
Frusher, S. D., and Hoenig, J. M. (2001). Impact of lobster size on selectivity
of traps for southern rock lobster (Jasus edwardsii). Canadian Journal
of Fisheries and Aquatic Sciences 58, 2482–2489. doi:10.1139/F01-181
Grant,T.R.,Lowry,M.B.,Pease,B.,Walford,T.R.,andGraham,K.
(2004). Reducing the by-catch of platypuses (Ornithorhynchus ana-
tinus) in commercial and recreational fishing gear in New South
Wales. Proceedings of the Linnean Society of New South Wales 125,
259–272.
Hardee, D. C. (2009). The effects of bait type, trap-soak duration and trap
modification on harvest of red swamp crawfish. M.Sc. Thesis, Louisiana
State University, Baton Rouge, LA.
Henry, G. W., and Lyle, J. M. (Eds) (2003). ‘The National Recreational and
Indigenous Fishing Survey.’ (NSW Fisheries: Sydney.)
IWC (2012). Index of Wetland Condition. Available at http://mapshare2.
dse.vic.gov.au/MapShare2EXT/imf.jsp?site=iwc [Verified 3 November
2014].
Jones, P. L., Silva, S. S., and Mitchell, B. D. (1996). The effect of dietary
protein source on growth and carcass composition in juvenile Australian
freshwater crayfish. Aquaculture International 4, 361–376. doi:10.1007/
BF00120951
Limpus, D., Hodge, W., and Limpus, C. (2006). ‘Impacts of Dams and Weirs
on Freshwater Turtles: Fairbairn Dam, March 2006.’ (Queensland
Environmental Protection Agency: Brisbane, Qld.)
McClain, W. R., Romaire, R. P., Lutz, C. G., and Shirley, M. G. (2007).
‘Louisiana Crawfish Production Manual.’ (Louisiana State University
Agriculture Centre: Baton Rouge, LA.)
McCormack, R. B. (2012). ‘A Guide to Australia’s Spiny Freshwater
Crayfish.’ (CSIRO Publishing: Melbourne.)
Miller, R. J. (1979). Saturation of crab traps: reduced entry and escapement.
Journal du Conseil 38, 338–345. doi:10.1093/ICESJMS/38.3.338
Miller, R. J. (1990). Effectiveness of crab and lobster traps. Canadian
Journal of Fisheries and Aquatic Sciences 47, 1228–1251. doi:10.1139/
F90-143
NSW DPI (2011). Fishing for yabbies in New South Wales. (New South
Wales Department of Primary Industries.) Available at http://www.dpi.
nsw.gov.au/__data/assets/pdf_file/0015/54204/Fishing-for-yabbies-in-
NSW.pdf [Verified September 2014].
Paillisson, J.-M., Soudieux, A., and Damien, J.-P. (2011). Capture efficiency
and size selectivity of sampling gears targeting red-swamp crayfish in
several freshwater habitat. Knowledge and Management of Aquatic
Ecosystems 401, 06. doi:10.1051/KMAE/2011015
Rankin, T. A. (2000). Status of the NSW commercial yabby fishery. NSW
Fisheries Resource Assessment Series Report 9., Sydney.
Serena, M., and Williams, G. (2010). Factors contributing to platypus
mortality in Victoria. The Victorian Naturalist 125, 178–183.
van Dijk, A. I. J. M., Beck, H. E., Crosbie, R. S., Jeu, R. A. M., Liu, Y. Y.,
Podger, G. M., Timbal, B., and Viney, N. R. (2013). The Millennium
drought in southeast Australia (2001–2009): natural and human causes
and implications for water resources, ecosystems, economy, and society.
Water Resources Research 49, 1040–1057. doi:10.1002/WRCR.20123
Vic DPI (2011). Recreational Forums: List of Angler Suggestions 2011.
Available at http://www.dpi.vic.gov.au/fisheries/environment-and-
community/regional-roundtables/list-of-suggestions-2011 [Verified
January 2012].
Victorian Government (2014). ‘Fisheries Regulations 2009.’ (Victorian
State Government: Melbourne.)
www.publish.csiro.au/journals/mfr
998 Marine and Freshwater Research P. Brown et al.
A preview of this full-text is provided by CSIRO Publishing.
Content available from Marine and Freshwater Research
This content is subject to copyright.
