An investigation of the habitat use of juvenile Astacopsis gouldi in the Emu River, Tasmania.
Water Assessment Branch Water Resources Division, DPIW 13 St Johns Avenue, New Town, TAS 7008
Report Series; WA 06/01 December 2006
Author: Christopher Bobbi Copyright Notice: Material contained in the report provided is subject to Australian copyright law. Other than in accordance with the Copyright Act 1968 of the Commonwealth Parliament, no part of this report may, in any form or by any means, be reproduced, transmitted or used. This report cannot be redistributed for any commercial purpose whatsoever, or distributed to a third party for such purpose, without prior written permission being sought from the Department of Primary Industries and Water, on behalf of the Crown in Right of the State of Tasmania.
Disclaimer: Whilst DPIW has made every attempt to ensure the accuracy and reliability of the information and data provided, it is the responsibility of the data user to make their own decisions about the accuracy, currency, reliability and correctness of information provided. The Department of Primary Industries and Water, its employees and agents, and the Crown in the Right of the State of Tasmania do not accept any liability for any damage caused by, or economic loss arising from, reliance on this information.
Preferred Citation: DPIW (2006) An investigation of the habitat use of juvenile Astacopsis gouldi in the Emu River, Tasmania . Water Assessment Branch, Department of Primary Industries and Water, Hobart. Technical Report WAP 06/01
ISSN: 1449-5996
Cover Photo: Radio-tagged juvenile Astacopsis franklinii (CPL 45mm) being released into New Town Rivulet, southern Tasmania.
The Department of Primary Industries and Water The Department of Primary Industries and Water provides leadership in the sustainable management and development of Tasmania’s resources. The Mission of the Department is to advance Tasmania’s prosperity through the sustainable development of our natural resources and the conservation of our natural and cultural heritage for the future. The Water Resources Division provides a focus for water management and water development in Tasmania through a diverse range of functions including the design of policy and regulatory frameworks to ensure sustainable use of the surface water and groundwater resources; monitoring, assessment and reporting on the condition of the State’s freshwater resources; facilitation of infrastructure development projects to ensure the efficient and sustainable supply of water; and implementation of the Water Management Act 1999 , related legislation and the State Water Development Plan.
ii Executive Summary The Giant Freshwater Lobster ( Astacopsis gouldi - Parastacida e) is a long-lived freshwater crayfish species that is endemic to many of the rivers draining the north coast Tasmania. In the past, this species was subject to significant recreational fishing pressure and together with a number of other factors, the species is now listed as ‘vulnerable’ under Tasmanian and Commonwealth species protection legislation. Adults of the species are known to prefer still or slow-flowing habitat in rivers, where there is an abundance of decaying wood, plenty of riparian shade and undercut river banks. Less has been documented about the habitat preference of juvenile crayfish, although recent work in smaller, more elevated streams has shown that they favour habitat with higher levels of boulder and logs as substrate and low levels of silt (Davies 2004). In larger rivers juveniles are often mentioned as occurring mostly in shallow riffle zones.
To date no environmental flows studies undertaken for water management planning have specifically examined the habitat needs of A. gouldi , and this was identified as a knowledge gap during the development of the Great Forester Catchment Water Management Plan (DPIWE 2003). Through the use of radio telemetry this study aimed to better understand habitat use of juvenile crayfish in larger river systems.
During initial reconnaissance, the study found that because of excessive sedimentation in the middle and lower Great Forester River, little suitable riffle habitat remains for juvenile crayfish. As a result, the study was undertaken in the lower reaches of the Emu River at Burnie in northwest Tasmania. Through a combination of manual sampling and surveillance of juveniles fitted with radio transmitters, it was found that juvenile crayfish preferred riffle habitat dominated by larger substrate elements (boulder and cobbles) where finer sediments (gravel, sand and silt) were absent. Juvenile crayfish were also found to prefer habitat with shallow water depth (0.1-0.25 m) and low to moderate water velocity (0.1-0.7 m.s -1 ), depending on the nature of the substrate.
Radio-tracking showed that while the majority of juvenile crayfish movement was restricted to small-scale ‘foraging’ behaviour within the riffle in which they were initially captured and then released, they are also capable of traversing distances in excess of 300 m in a 24-hour period. In doing this, individuals crossed deep pool habitat inhabited by large adult crayfish as well as adult brown trout and platypus, known predators of juvenile crayfish. The stimuli for this larger-scale ‘nomadic’ behaviour is unknown, though it may be a response either to disturbance or tagging of individuals that were already in the process of relocating to new habitat.
Natural and artificial changes to flow conditions were not found to prompt a marked response from juveniles, in particular the abandonment of riffle habitat for deeper pools. It appears more likely that reduction in flow and seasonal elevation in water temperature will elicit a response whereby juveniles burrow into the substrate, where environmental conditions are less harsh.
The study has gathered valuable data regarding the use of riffle habitat by juvenile A. gouldi , and this will assist with the future development of appropriate environmental flow allocations for rivers where this species occurs.
iii Acknowledgments I would like to thank members of the Water Assessment Branch of DPIW for their efforts in supporting this study; Dave Horner, Justine Latton and Martin Read for valuable assistance with field work, and Donald Hine, Mic Yemm and Danielle Warfe for her help and advice on technical issues and data analysis.
I would also like to thank Todd Walsh for help with site selection and sampling of crayfish for the study.
A number of other people were helpful in the initial design and in some cases providing equipment for the study; Peter Davies, Sarah Munks, Jean Jackson, Matt Webb and Alistair Richardson.
This study was undertaken entirely with funds from the Water Assessment Branch budget and formed part of its commitment to gathering more information to support the Water Management Planning process in the Great Forester River and elsewhere in the Tasmania.
iv Table of Contents
EXECUTIVE SUMMARY III
1. INTRODUCTION 1
2. STUDY AIM 2
2.1 Feasibility issues 2
3. MATERIALS AND METHODS 4
3.1 Study location 4
3.2 Specimen collection and radio-tagging 6
3.3 Tracking and other monitoring 7
3.4 Habitat mapping 8
3.5 Water level manipulation 9
3.6 Data Analysis 9
4. RESULTS 10
4.1 Crayfish tagging and site of capture data 10
4.2 Environmental monitoring 11
4.3 Crayfish movement 11
4.4 Habitat use 15
4.5 Changes in habitat availability with flow 20
5. DISCUSSION 23
6. REFERENCES 26
APPENDIX A: Environmental condition monitoring 28
APPENDIX B: Accuracy of spatial locations 31
APPENDIX C: Habitat use data for juvenile A. gouldi 33
v 1. Introduction During the formulation of the Great Forester Catchment Water Management Plan (DPIWE 2003), a number of information gaps were identified and studies were recognised as being required to fill these gaps prior to any future review of the plan. A commitment was made by DPIWE (now DPIW) to report back to the catchment Consultative Group on work to fill some of these information gaps, one of which is to examine the habitat requirements of the Giant Freshwater Lobster ( Astacopsis gouldi ) in the Great Forester River. During earlier studies to derive an ‘environmental flow’ for the river (McKenny 1999), the requirements of this species was not included, despite it being a listed species under State and Commonwealth threatened species legislation.
Astacopsis gouldi is a long-lived species that is thought to reach ages in excess of 30 years (Hamr 1990) and can grow up to one metre long and weigh as much as 6.5 kg (recently captured individual). The species is generally found in still or slow- flowing rivers and streams along Tasmania’s north coast, at elevations up to about 400 m above sea level (Horwitz 1994). Most research to date has focussed on adults of the species, which are known to inhabit deeper pool habitat within rivers, preferring sites with an abundance of decaying timber and with plenty of riparian shade and undercut riverbanks (Lynch 1997; Lyall 2000; Webb 2001). Juvenile crayfish have been found to occur in higher densities in shallow, faster-flowing water where boulders are present (Forteath 1987). In a study focussing on headwater streams, Davies and Cook (2004) found that densities of juvenile crayfish were significantly higher in larger rivers (third order and above), being replaced by larger crayfish in smaller fourth order streams. In smaller streams, juveniles have been found to favour habitat with low levels of silt, moderate to high percentage substrate as boulders and logs, and with bed slope less than 15% (Davies 2004). In larger rivers, this habitat is utilised by adult crayfish, and juveniles are restricted to shallower riffle habitat containing larger substrate (Inland Fisheries Service pers. comm.). However it is not known whether riffles in larger rivers are the habitat of choice for juvenile A. gouldi , or a consequence of competitive or predator-prey interactions with adults, which are known to be cannibalistic.
Riffles habitat is highly productive and tends to support higher densities of benthic invertebrates (Gordon 2004). During low flows this habitat can become exposed, and is the instream habitat that is most at risk from over-extraction of water during the summer. High water temperature often coincides with periods of low-flow, posing proportionally greater risk to animals that inhabit this part of the river. It is therefore logical to presume that juvenile crayfish may experience physiological stress if they remain in riffles during extreme low-flows, and that if animals are forced to move out of this habitat, they may be at risk of predation by adult crayfish or other predators within deeper, pool habitat.
Given that there is already a good body of information available on the habitat requirements of adult crayfish, the main aim of this study was to improve our understanding of the habitat used by juveniles, specifically in larger rivers where existing water use may impact on low-flows. Although the impetus for the project arose from the water assessment and planning work carried out in the Great Forester catchment, future water management planning activities elsewhere along the north
1 coast of Tasmania are also likely to require this knowledge. With this wider applicability in mind, the following study was devised.
2. Study Aim The main aim of the study was to determine if juvenile A. gouldi are restricted to shallow riffle habitat during low flow periods. If very low flows are found to influence movement and habitat use by juveniles, this information will assist in developing appropriate flow management protocols for rivers containing this species.
The study sought to answer two simple questions: 1. If juvenile A. gouldi inhabit riffle zones in larger rivers, are individuals forced to leave this habitat during periods of low flow? 2. Is water quality different in pools and riffles during warm periods, and are environmental conditions likely to cause stress to adult and juvenile crayfish?
To answer these questions, a radio-tracking study was proposed. The proposal included capturing and tagging juveniles of less than 70 mm carapace length (CPL) from two small river reaches. One reach was to act as a control, while the other was to undergo flow manipulation in an attempt to prompt movement by tagged individuals. The study was to be conducted over a 3-4 week period in late summer- autumn, when river flow was likely to be at its lowest, and water temperature at its highest. To assist in characterising water quality within the study reaches, instream probes were to be deployed, along with water level monitoring equipment. Following completion of the radio-tracking, animals were then to be removed and habitat within both reaches intensively mapped for substrate type, water depth and water velocity, so that habitat use during tracking could be examined.
2.1 Feasibility issues During the inception of the study a number of potential technical issues were identified. These are briefly outlined below.
Size of transmitting device Given that the main interest for the study was on juvenile A. gouldi of <70 mm CPL, the practicality of attaching transmitters to animals was a potentially limiting factor. If the units were too large they might interfere with movement of animals and their capacity to hide beneath cobbles and boulders. Some pilot testing was therefore undertaken in a small river near to Hobart, using a related species A. franklinii . This confirmed the practicality of attaching transmitters to small crayfish and also provided an opportunity for training and technical refinement of the methodology for radio- tracking.
Potential for large movements by individuals From the radio-tracking study of Webb (2001) it appears that adult A. gouldi (in particular adult males) can move as much as 700 m in a single 24-hour period. There is little information regarding the patterns of movement of juveniles, however it is reasonable to assume that the movement of juveniles is more restricted due to their smaller size and the potential for conflict with larger adults if juveniles move out of riffles.
2 Moulting Another significant factor that might impact on the study is the moulting frequency of juveniles. Information on moulting in A. gouldi that is available from (Hamr 1990) indicates that moulting generally occurs between October and March. Although not well document for wild crayfish, juveniles of A. gouldi (classified by Hamr 1990 as being those having a carapace length of <70 mm) may moult 2-3 times during the summer. This was seen as having obvious implications for the length of time over which monitoring of tagged individuals could be carried out.
Availability of suitable study sites A preliminary survey of the main stem of the Great Forester River was conducted in December 2004 to look for potential sites to conduct the study. The survey found that the reach of river used for the original environmental flows assessment (McKenny 1999) was heavily impacted by the deposition of gravel and fine sediment. This was particularly the case for riffle habitat, where almost all of the interstitial spaces that would have been present during the original assessment were now in-filled by sediment.
Active searching with a fine hand-net within the shallow sections of the reach failed to locate any juvenile crayfish. Downstream of this reach the river is dominated by sand and gravel, and contains little or no riffle habitat containing cobble or pebble sized material. Reconnaissance of the river upstream did not locate a suitable reach where the river was unaffected by riparian or catchment disturbance, and is of sufficient size for the study. While active searching did locate juvenile A. gouldi within McKenzie Rivulet (a tributary where prior sampling by the Inland Fisheries Service had found an abundance of juvenile crayfish), this stream was too small to allow study results to be justifiably translated to larger rivers within the catchment or elsewhere in northern Tasmania. As a result, it was decided that the field study would be conducted in the lower Emu River near Burnie. This river is known to contain an abundance of both adult and juvenile crayfish, and a visual inspection found habitat that is reasonably representative of other rivers where A. gouldi occurs.
It is important to note that the cursory survey of the Great Forester River suggests there is limited habitat for juveniles in the main stem of the river, particularly downstream of Prosperity Road. The preliminary results imply that juvenile crayfish in this catchment may rely more heavily on habitat contained in tributary streams (such as McKenzie Rivulet), and in this catchment it is the tributaries that contain the better sites for large water storages. Therefore in the interest of conserving A. gouldi in the Great Forester catchment it is recommended that a catchment-wide survey be undertaken to identify parts of the drainage system where crayfish are most abundant, and develop catchment-scale management provisions to ensure high priority areas are protected from negative developments. The Conservation of Freshwater Ecosystem Values (CFEV) database would be very helpful in such an undertaking.
3 3. Materials and methods
3.1 Study location The study was conducted on the lowest section of the Emu River, which is a 58 km long unregulated, first order river ( sensu Strahler) flowing northwards into Bass Strait at Burnie on Tasmania’s northwest coast (41 o4’12”S, 145 o 55’12”E). The river at the location of the study lies just above sea level, ranges in width from 5-25 m and contains pools up to about 2 m deep. The 254 km 2 catchment of the Emu River is narrow and deeply incised, and receives annual rainfall of 900-1500 mm. Because of its narrow geometry, the majority of active land-use is restricted to the hilltops, and as a result the river throughout most of its length is fringed by a thick and relatively undisturbed wet schlerophyll forest (Figure 1). The lower section of the river is heavily confined by bedrock, and is composed predominantly of boulder, cobble and pebble material, reflecting the high-energy characteristics of the river.
Burnie
Figure 1. Satellite image of the City of Burnie and the heavily forester margins of the Emu River (outlined in white), northwest Tasmania. Star indicates location of crayfish radio- tagging study reaches.
Two small reaches of the lower river within the Fern Glade Reserve were selected for study. The first is located at about 5 mASL (Figure 2) and this was selected as the ‘control’ for the study. The second reach is located approximately 700 m upstream at an elevation of about 9 mASL (Figure 3), where the river is slightly less confined and contains more cobble and pebble substrate rather than boulder. This section of river was chosen as the ‘test’ site, where flow manipulation was carried out to examine whether juvenile crayfish respond to alterations in flow conditions.
4 Elevation (mASL)
6
5.75
5.5
5.25
5
4.75
4.5
4.25
Figure 2. Topographic map and photograph of the ‘control’ reach on the lower Emu River. The two diagonal lines indicate the section of river within which animals where radio-tracked and physical habitat data (also used to produce the topographic map) were collected. Zero on the axis for the map is located on the upstream left-hand bank when looking upstream (view in photo).
Elevation 20 (mASL)
18
11
16 10.5
14 10
9.5 12
9 10
8.5 8
8
6 7.5
4
2
0 0 2 4 6 8 10 Figure 3. Topographic map and photograph of the ‘test’ reach on the lower Emu River. The two diagonal lines indicate the section of river within which animals where radio-tracked and physical habitat data (also used to produce the topographic map) were collected. Zero on the axis for the map is located on the upstream right hand bank when looking downstream (view in photo).
5 Because of the largely forested nature of the catchment, water quality in the river is good, with low conductivity (generally < 100 µS.cm -1 ) and healthy levels of dissolved oxygen (9-12 mg.L -1 ). Like many rivers along the northwest coast of Tasmania, the pH of the water is slightly acidic (6.0 – 7.0) and has a brown hue from the high level of tannins present in the water. In summer following prolonged periods of low flows, those parts of the river that are less confined by the valley walls and experience higher levels of incident sunlight experience blooms of filamentous algae.
3.2 Specimen collection and radio-tagging Crayfish were collected from the ‘test’ reach on the afternoon of the 9 th March 2006, while the ‘control’ reach was sampled on the morning of the 10 th March 2006. At the ‘test’ site, very young crayfish were also collected from a riffle approximately 100 m upstream for additional information about habitat use. Crayfish were captured using an active hand-searching method as described in (Robinson 2000) and used by Davies and Cook (2004) in Tasmanian headwater streams. Stones on the bed of the river were slowly removed by hand to look underneath for crayfish. Those that could not be captured directly by hand were often carried by the current into a fine mesh hand-net purposely placed directly downstream. Captured animals were placed in a bucket and the stone replaced and marked with a small painted rock.
When crayfish were found, the water depth and substrate composition at the location of capture was recorded, and the carapace length (CPL) of the crayfish measured. In cases where crayfish were too small for use in radio-tracking, animals were simply measured and released back into the river at their capture location.
For the radio-tracking study, two-stage radio transmitters (type BD-2T Holohil Systems, Canada), weighing 0.81 grams and with a battery life of approximately 28 days, were used to track crayfish. The transmitter and battery were encapsulated in an inert, waterproof epoxy with an overall dimension of 13 x 6.5 x 2.5 mm, to which a trailing stainless steel whip antenna (160 mm in length) was attached (Figure 4a & b). Individual transmitters were distinguished by having different frequencies between 150.00 and 150.40 MHz.
Figure 4a & b. Juvenile crayfish with radio-transmitters attached. Crayfish (a) is an A. franklinii (CPL 42 mm) captured in New Town Rivulet for testing of the technique, and crayfish (b) is an A. gouldi (CPL 32 mm) captured in the Emu River and tracked during the study.
6 Animals used for the study ranged between 9 g and 105 g (wet weight), which meant that the radio-transmitters attached to crayfish ranged between 9% and <1% of body weight. This is a similar range of tag mass: body mass ratio to what has been reported for other crayfish radio-tracking studies (Robinson 2000; Bubb 2002; Aquiloni 2005). None of these studies reported that radio-tags interfered with crayfish mobility or behaviour.
Eleven suitable juvenile A. gouldi were captured and fitted with radio-transmitters. Transmitters were attached directly to the upper lateral area of the carapace using rapid-set epoxy resin, after Webb (2001). In each case, crayfish were held in a shallow tray with sufficient depth of water to submerge the gills, but not wet the upper half of the carapace. Upon contact, each transmitter was held in place by hand for about 5 minutes to allow the epoxy to set, then each animal was retained in the tray for another 30 minutes before being released back into the river at the site of capture. To prevent immediate movement of animals away from release locations in a ‘fright’ response similar to that mentioned as a cause of initial movement by Robinson et al. (2000), a small-mesh plastic cage measuring about 400 x 400 mm was dropped over the site of release and left for several hours.
Upon completion of the study, most of the animals were recaptured and the radio- transmitters removed. This was done by carefully using a scalpel to make an incision into the edges of the epoxy holding the transmitter to the carapace. Once this was done, the transmitter could then easily be peeled away from the carapace without damage to the animal. Following removal, each animal was once again held in a bucket of water for observation before being released back into the river at the site of recapture.
3.3 Tracking and other monitoring Crayfish were tracked using an Australis 26K tracking receiver (Titley Electronics) in combination with two different styles of antennae. To locate animals from set benchmark pegs on the river-bank, a collapsible 3-element Yagi antenna was used. Tagged animals could be detected up to 500 m away by holding this antenna 2 m or more above the level of the river. To locate animals within each of the study reaches, bearings were taken from 2 or 3 benchmark pegs situated on the bank of the river. Using this method, animals could be pin-pointed to within about a 2 m radius (see Appendix B: Accuracy of spatial locations).
Radio-tracking was carried out over 20 days between 10 th - 30 th March, 2006. Crayfish were located every day, with the exception of two 2-day periods when monitoring was not undertaken.
Once animals had been roughly located and their position in the river noted, a ‘stripped coaxial cable antenna’ (Beeman 2004) was attached to the radio receiver. This antenna was fixed to a 1.2 m wooden pole so that it could be used as an underwater antenna. Whilst carefully wading in the river (taking care to avoid other animals being tracked), the end of this antenna could be poked in and around boulders and rocks to pin-point animal locations to within a radius of 0.25 m (Figure 5). This was particularly successful when the gain on the receiver was turned down low and the antennae bent or twisted. Once animals were pin-pointed using this method,
7 bearings of the location were once again taken from the benchmark pegs on the river- bank.
Figure 5. Using a ‘stripped’ coaxial cable antenna to pin-point the location of juvenile A. gouldi in the ‘control’ reach of the study site in the Emu River. Inset showing the twisted end of the coaxial antenna.
On each occasion, in addition to locating animals, the pulse rate of the transmitted signal was recorded using ‘Period / PPM’ meter (Titley Electronics). This signal was then converted to a temperature value using the calibration curve for each transmitter (mentioned above). Water temperature data was also collected at 2 to 3 locations within each study reach using small environmental temperature loggers (HOBO Water Temp Pro, Onset Computer Corporation, USA) attached to painted rocks. These loggers recorded water temperature every 15 minutes and were retrieved and downloaded following completion of the tracking experiment on March 30 th . Additional spot measurements of water temperature and dissolved oxygen were taken throughout the study using hand-held probes (YSI 550A, YSI Incorporated).
Water level, conductivity and temperature were also monitored at the deepest section of the upstream ‘test’ reach using fully submersible environmental loggers (YSI 690 Sonde and ‘Diver’ Van Essen Instruments). These loggers were attached to a metal picket driven into the bed of the river and recorded water level every hour and water quality every 15 minutes throughout the period of the study. To understand how changes in water level at the site related to flow, several flow gaugings were taken in the river at the upstream end of the ‘test’ reach using a wading rod and flow-meter (FlowMate 2000, Marsh-McBirney Incorporated).
3.4 Habitat mapping Upon completion of the radio-tracking component of the study, the instream habitat at both ‘control’ and ‘test’ reaches was assessed. This was done using the protocol detailed in Bovee (1982) to collect data on bed elevation, water depth, water velocity, substrate composition, and algal, silt and detritus cover at 10 transects perpendicular to the river channel. Transects were placed between 1 m and 3 m apart, with transects being more closely spaced within the area used most by crayfish. Habitat information was collected at 1 m intervals across each transect, and a dumpy level was used to tie
8 transects together and link them with the benchmark pegs used during the radio- tracking phase of the study.
3.5 Water level manipulation The experimental component of the study involved manipulating water level at the upstream ‘test’ reach. This was done by constructing a temporary weir part-way across the river to divert flow to one side of the river, consequently lowering water level in a part of the river containing radio-tagged crayfish. The weir was installed on March 20 th , about half-way through the study, and resulted in water level immediately downstream being lowered by up to 80 mm (in an area where water depth previously ranged between 100 – 250 mm). The modification to water depth was noted at 12 fixed locations within the river downstream from the weir. Figures 6 shows the ‘test’ reach after installation of the temporary weir.
Figure 6. The temporary sandbag weir installed part way across the Emu River at the ‘test’ reach on March 20 th , 2006.
3.6 Data Analysis The habitat data collected at the end of the study was used in two ways. In the first instance, the data was entered into Microsoft Excel® (Microsoft Corporation, USA) and then imported into Surfer ® (Golden Software Incorporated, USA) to create elevation and habitat distribution maps for each of the study reaches. To create these maps the transect data was gridded using the ‘Point Kriging’ interpolation function in Surfer ® and the resulting data was then used to create contour maps. Crayfish location data was then plotted onto the contour maps as point data by transforming the bearing data into X-Y co-ordinate data within the river transect matrix and then importing into Surfer ® files for plotting. Habitat use and movement by crayfish was then visually analysed.
To examine the habitat preference of juvenile crayfish, and how habitat availability changes with flow, the hydraulic river analysis packages HEC-RAS (HEC 2005) and RAP (Stewardson M. and Marsh 2004) were used. The HEC-RAS software package was used to convert the data from the cross-sectional transects into a hydraulic model of the ‘control’ reach. Data from the spot gaugings that were recorded during the study were used to check the accuracy of the model. The geometry and flow output
9 files from this model were then used as inputs to the hydraulic analysis module of RAP (River Analysis Package). Within RAP, changes in the availability of juvenile crayfish habitat during different flow conditions was examined. 4. Results
4.1 Crayfish tagging and site of capture data During the 2 half-days over which active searching for animals was carried out, a total of 21 juvenile A. gouldi were caught (Table 1), ranging from 9 mm CPL (new recruits) up to 70 mm CPL (sub-adult). Radio transmitters were attached to 11 of the larger individuals, and the rest were weighed and measured and then released back to the river at the site of capture. Crayfish below 28 mm CPL could not be reliably weighed and therefore were not tagged.
For crayfish with a CPL less than about 25 mm, the site-of-capture data shows that there was a distinct preference for shallow water depths, with all animals in this size category being found in less than 0.15 m of water. While animals larger than this were also captured in shallower habitat, it is clear that once crayfish grow beyond about 30 mm CPL they are able to live in deeper water where higher flow velocity often occurs. The largest crayfish that was captured during the study was found in the deepest section of river sampled.
Table 1. Juvenile A. gouldi sampled by active searching in the lower Emu River within the Fern Glade Reserve on March 9 th - 10 th , 2006. Also included are the characteristics of the habitat at the site of capture, presented as relative proportions of substrate size category (see below). Reach CPL Weight Water depth Water Habitat Transmitter (mm) (g) (m) velocity characteristics number (m.s -1 ) (% B,C,P,G)# ‘Control’ 22 0.07 0.17 50:35:12:3 No tag ‘Control’ 35 11 0.34 0.04 75:20:3:2 110101 ‘Control’ 42 30 0.24 0.47 70:25:5:0 110100 ‘Control’ 70 105 0.46 0.28 85:5:8:2 110092 ‘Control’ 30 10 0.05 0.01 40:20:30:10 110091 ‘Control’ 34 12 0.18 0.07 80:10:7:3 110102 ‘Control’ 28 10 0.34 0.37 65:25:5:5 110098 ‘Control’ 59 75 0.22 0.02 75:20:3:2 No tag ‘Control’* 31 10 0.11 0.15 35:10:40:15 110099 ‘Control’ 22 0.09 0.12 25:25:35:15 No tag ‘Test’ 30 9 0.14 0.45 25:15:60:0 110095 ‘Test’ 29 11 0.15 0.49 10:55:30:5 110094 ‘Test’ 36 15 0.14 0.42 5:70:20:5 110093 ‘Test’ 39 30 0.13 0.31 25:60:10:5 110097 ‘Test’ 20 0.04 0.27 40:30:25:5 No tag ‘Test’ 21 0.10 0.18 40:30:25:5 No tag ‘Test’ 10 0.09 0.38 45:40:10:5 No tag ‘Test’ 11 0.08 0.53 60:35:3:2 No tag ‘Test’ 9 0.10 0.30 60:35:3:2 No tag ‘Test’ 9 0.12 0.28 55:30:10:5 No tag ‘Test’ 13 0.08 0.32 45:45:5:5 No tag # ‘B’ = boulder (>256 mm), ‘C’ = cobble (64 - 256 mm), ‘P’ = pebble (8 – 64 mm), ‘G’ = gravel (2 – 8 mm). * Juvenile that was captured in the ‘Control’ reach and translocated to the ‘Test’ reach to boost test animal numbers.
10 4.2 Environmental monitoring Water quality (temperature and dissolved oxygen) monitoring undertaken at both the ‘control’ and ‘test’ reaches showed that there was little difference between conditions within riffles and in deeper pools, either within reaches or between reaches. Maximum water temperature reached 18.5 oC during the course of the study, and dissolved oxygen varied in saturation between 90% and 113%. The environmental condition data collected during the study is presented and discussed more fully in Appendix A. However, in summary it was concluded that under low flows and elevated seasonal temperatures, there is unlikely to be any ecologically significant difference in environmental conditions between riffle and pool habitat, unless there is some localised impact that might cause such a situation.
4.3 Crayfish movement This section presents data on the range and pattern of movement by individuals and an assessment of some of the potential factors that might trigger movement including the impact of artificial flow modification on crayfish movement at the ‘test’ reach.
Prior to using the data obtained from the directional Yagi antenna to assess crayfish position and movement, an analysis was made of the spatial accuracy of these data. To do this, data from the Yagi antenna was compared to the much more accurate and precise data obtained from the stripped coaxial cable antenna (accurate to +/- 0.25 m). It showed that on average, locations obtained using the directional antenna differed from those obtained by the coaxial antenna by about 1.4 m, but could be as high as 3.2 m. This needs to be kept in mind when viewing the following text, which presents data on crayfish movement that was obtained using the Yagi directional antenna. Smaller scale movements (<2 m) may be due to error in locating animals, rather than actual movement by crayfish. A more detailed presentation of the accuracy of directional location data is made in Appendix B.
The pattern of movement by juvenile A. gouldi can generally be described as ‘small- scale ranging’, where individuals tended to move around intermittently within an area of about 20 m radius. There was substantial variation in the level and manner in which individual crayfish changed position during the study, with one individual remaining almost stationary throughout the entire 20-day tracking period and another moving more than 460 m in a single night. To aid in describing the movement of crayfish, individual movements were characterised as either primarily lateral (where positional change over a single period was >45 o to long axis of the river) or longitudinal (positional change <45 o to long axis of the river). These data are shown in Tables 2a and 2b along with summaries of the distance moved by individual crayfish. Although the data in the tables uses all estimated distances moved, as stated above, it must be kept in mind that individual movements of less than 2 m may be due to errors associated with radio-tracking.
When the movement data is viewed in this manner, it is evident that the majority of all crayfish tagged tended to move more in a lateral direction than longitudinally and the total distance moved during the study was greater along the lateral axis. The notable exceptions to this general pattern of movement were crayfish #110099 & #110093 (both from the ‘test’ reach), which appear to have been tagged either while in a nomadic, wandering phase (Gherardi 2000), or moved in response to capture and tagging. Crayfish #110099, which moved more than 80 m upstream immediately
11 following release, also happened to be the individual that was translocated from the ‘control’ to the ‘test’ reach to boost the number of individuals. Crayfish #110093, having moved more than 450 m upstream on the night following release at the ‘test’ reach, was later recaptured and returned, only to immediately move more than 300 m upstream the following night.
Table 2a & b: Descriptive statistics of movements by crayfish during the study period. Single movements are categorised as either ‘lateral’ or ‘longitudinal’. In some cases, animals did not appear to move for extended periods, and therefore have a lower total number of recorded movements.
‘Control’ reach # 110101 #110100 #110092 #110098 #110102 #110091 (CPL 35mm) (CPL 42mm) (CPL 70mm) (CPL 28mm) (CPL 34mm) (CPL 30mm) No. of lateral movements (n) 7 7 6 9 4 1 Max distance moved (m) 4.0 7.0 3.3 7.3 3.3 1.8 Total distance moved (m) 12.4 27.9 13.8 29.6 9.3 1.8 No. longitudinal movements (n) 4 5 4 2 7 1 Max distance moved (m) 2.1 4.75 1.2 4 4.3 1.8 Total distance moved (m) 5.1 15.1 3.4 5.8 18.5 1.8
‘Test’ reach # 110094 #110095 #110097 #110099 #110093 (CPL 29mm) (CPL 30mm) (CPL 39mm) (CPL 31mm) (CPL 36mm) No. of lateral movements (n) 10 7 5 5 0 Max distance moved (m) 5.8 9.5 1.8 9.2 - Total distance moved (m) 18.0 21.2 5.7 15.7 - No. longitudinal movements (n) 2 6 3 3 3 Max distance moved (m) 4.5 6.7 1.8 61 462 Total distance moved (m) 6.9 23.6 3.3 82.8 770
The movement data for crayfish from both reaches is displayed graphically in Figures 7 and 8. These plots show crayfish movement as ‘cumulative distance moved’ over the entire period of the study. Overlayed on both graphs is a blue vertical column indicating the period during days 7-8 when the small ‘flow event’ in the river occurred. In addition, the graph for crayfish movement at the ‘test’ reach (Figure 8) also contains a brown column indicating when the temporary weir structure was installed part-way across the river channel to modify flow conditions in part of the reach.
The data from the ‘control’ reach shows that for some crayfish, movement was fairly consistent through time (e.g #110098 & #110101) while for others (e.g. #110092 & #110100) there appears to be more definite periods of movement in between periods of inactivity. The graph also suggests that there is no major movement in response to the rise in river level that occurred during days 7-8. In contrast, 2 of the 3 crayfish that were still present in the ‘test’ reach when the event occurred (Figure 8), appear to have responded to the flow event with relatively substantial movements. From this it might be concluded that while juvenile crayfish may respond to natural changes in flow, response is highly variable between individuals.
12 (crayfish moved beyond study reach) 50 50
110101 110094 110100 110095 110092 110097 40 110098 40 110102 110099 110091 110093
30 30
20 20 Distance moved (m) moved Distance Distance moved (m) moved Distance
10 10
0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Study days Study days
Figure 8. Cumulative distance moved by juvenile A. gouldi in the ‘test’ reach of Figure 7. Cumulative distance moved by juvenile A. gouldi in the ‘control’ reach th th of the lower Emu River between 10 th – 28 th March, 2006. The vertical blue the lower Emu River between 9 – 29 March, 2006. The vertical blue column column indicates the occurrence of a small ‘freshet’ in the river during days 7-8 of indicates the occurrence of a small ‘freshet’ in the river during days 7-8 of the the study. study, and the vertical brown column indicates when the temporary weir was installed.
13 The impact of artificial flow modification (installation of the weir) at the ‘test’ reach is less clear. Because only 3 crayfish remained within the confines of the ‘test’ reach following radio-tag attachment the results from the flow manipulation cannot be examined in a statistically rigorous manner. Visually, the data suggests that there might have been some response by crayfish #110094 and #110095 to installation of the weir, however Figure 7 indicates that the majority of crayfish at the ‘control’ reach also moved at this time. This suggests that some environmental cue other than those that were recorded during this study may be responsible for initiating crayfish movement.
In an effort to further examine crayfish movement, the position of individual crayfish were plotted on spatial maps of water velocity at the ‘test’ reach prior to and following weir installation (Figures 9 and 10). While there was also some change in the spatial distribution of water depth throughout the reach following installation of the weir, the greatest change occurred in the velocity variable, therefore this is the one that has been used. Location data from both the ‘directional antenna’ and ‘coaxial antenna’ are shown on the maps for comparison.
Water Velocity Water Velocity (m/s) (m/s) 18 18 1 1
0.9 0.9 16 16 0.8 0.8
0.7 0.7
0.6 0.6 14 14 0.5 0.5
0.4 0.4 12 12 0.3 0.3
0.2 0.2
0.1 0.1 10 10 0 0
110094 110094 8 110095 8 110095 110097 110097
6 6
4 4
2 2
0 2 4 6 8 10 0 2 4 6 8 10 Pre-weir Post-weir Figure 9a & b. Positions maintained by crayfish #110097 prior to (pre-weir) and following flow manipulation (post-weir) at the ‘test’ reach. Maps show crayfish positions (as triangles) in relation to water velocity. The circle enclosing the triangle on the pre-weir map is the capture and release location, and triangles connected by lines are instances where the crayfish was located using both the directional and coaxial antennas. The weir is shown as heavy black lines on the post-weir map.
When individual crayfish location data are viewed from this perspective, the nature of positional change clearly differed between animals. The crayfish that had the smallest range of movement prior to flow manipulation (#110097) did not significantly alter its ranging behaviour and continued to move around within the same area following weir installation. This was despite a significant change in water velocity within the range
14 of this crayfish (Figure 9a & b). This crayfish appeared to heavily favour maintaining a position near to or under the largest boulder in the reach, irrespective of a substantial decline in water velocity due to installation of the weir immediately upstream.
In contrast, both crayfish #110094 and #110095 showed greater movement prior to flow manipulation, ranging over a variety of water depths and velocities (demonstrated by the recorded positions of #11095 in Figure 10a). Following installation of the weir, crayfish #11095 moved away from deeper water with higher velocity, finally taking up residence directly behind the apex of the weir structure (Figure 10b), whilst crayfish #11094 (data not shown) remained in slightly deeper water in an area where water velocity was more variable.
Water Velocity Water Velocity (m/s) (m/s) 18 18 1 1
0.9 0.9 16 16 0.8 0.8
0.7 0.7
0.6 0.6 14 14 0.5 0.5
0.4 0.4 12 12 0.3 0.3
0.2 0.2
0.1 0.1 10 10 0 0
110094 110094 8 110095 8 110095 110097 110097
6 6
4 4
2 2
0 2 4 6 8 10 0 2 4 6 8 10 Pre-weir Post-weir Figure 10a & b. Positions maintained by crayfish #110095 prior to (pre-weir) and following flow manipulation (post-weir) at the ‘test’ reach. Maps show crayfish positions (as black dots) in relation to water velocity. The circle enclosing the dot on the pre-weir map is the capture and release location, and dots connected by lines are instances where the crayfish was located using both the directional and coaxial antennas. The weir is shown as heavy black lines on the post-weir map.
When the location data is examined in a spatial manner such as this, it provides further support to the conclusion that while juvenile crayfish probably react to changes in flow conditions, there is no strong trend in the response of individuals.
4.4 Habitat use During the initial capture of crayfish a record was made of the characteristics of the habitat from which animals were taken. These data were presented earlier in Table 1. Whilst these ‘site of capture’ data suggests that there is some preference by juveniles for habitat with shallower water depth and larger substrate (predominantly boulder and cobble material), this may be an artefact of the sampling technique that was used.
15 Areas of shallower depth are easier to sample and the capture of animals is more likely, therefore these data cannot be used alone to confidently characterise habitat use/preference by juvenile crayfish. To better understand this, data from the tracking component has been overlayed on habitat distribution maps for each of the study reaches to indicate habitat used by tracked crayfish during the course of the study.
In the ‘control’ reach, the river bed is dominated by boulder-sized (>256 mm) and cobble-sized (64 – 256 mm) material, while the upstream ‘test’ reach contains significantly fewer boulders and is dominated more by cobble and pebble (8 – 64 mm) material. Maps showing the distribution of the main habitat types for each of the reaches are shown in Figures 11(a-d) and 12(a-f) below. In each case, the crayfish locations that were recorded during the radio-tracking component of the study are overlayed to indicate in a visual manner the type of habitat that was utilised. The habitat utilisation data were subsequently read off the maps and analysed using frequency histograms.
When the data from both reaches is combined and plotted in the form of habitat utilisation histograms (Figure 13a-g), the results clearly show that radio-tracked juvenile crayfish tended to utilise areas where water depth was between 0.1-0.25 m and flow velocity varied between 0.1-0.5 m.s -1 . The substrate data shows that crayfish tended to utilise areas where cobble comprised 20-60% of the bed substrate and areas where there was larger, more stable boulder material. Crayfish were much less frequently located in habitat with substantial levels of pebble material (>50%), and there appears to be a clear preference for areas with less than 20% gravel. It is also interesting to note that the radio-tracked crayfish preferred habitat with only low levels of algal cover (10-20%).
When the results from the two reaches are analysed separately, the differences in river morphology between them becomes more evident in the location data (data from each of the reaches are included in Appendix C). Within the ‘control’ reach, where the benthic morphology is dominated by boulders, it can be seen that habitat use is fairly evenly spread across areas where boulder comprise anywhere between 10% and 70% of the substrate. In this reach, where there is plenty of boulder substrate, the presence of cobble substrate may have more of an influence on habitat use. At the ‘test’ reach, few boulders were present, and the data shows that as a consequence of this, the habitat utilisation is skewed more towards habitat with higher proportions of cobble (30-60%) and pebble (~50%) material.
The other feature of the location data from the ‘test’ reach is that there appears to be more of a bi-modal distribution with respect to the water velocity used by crayfish. The data suggests that crayfish utilised areas of both very slow-moving water (~0.1 m.s -1 ) and very fast-moving water (0.5-0.6 m.s -1 ). However upon closer inspection of the data it emerges that this is driven by the installation of the temporary weir and the movement of 2 individuals to the slack water directly behind it. For this reason, and the fact that only three crayfish were tracked in the ‘test’ reach, data from both reaches were combined to gain a more general view of habitat use.
16 Units (m) Units (m/s) 20 20 0.5 1
0.45 0.9 18 18 0.4 0.8
0.35 0.7
16 0.3 16 0.6
0.25 0.5
14 0.2 14 0.4
0.15 0.3
0.2 12 0.1 12 0.05 0.1
0 0 10 10
110091 110091 8 110092 8 110092 110098 110098 110100 110100 110101 110101 6 110102 6 110102
4 4
2 2
0 0 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 (a) Water depth (b)Velocity
Units (%) Units (%) 20 20 100 90
90 80 18 18 80 70 70 60 16 60 16 50 50 40 14 40 14 30 30 20 12 20 12 10 10
0 0 10 10
110091 110091 8 110092 8 110092 110098 110098 110100 110100 110101 110101 6 110102 6 110102
4 4
2 2
0 0 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 (c) %Boulder (d) %Cobble
Figure 11(a-d). Contour maps showing the distribution of major habitat descriptors for the ‘control’ reach in the lower Emu River, overlayed with location data for juvenile A. gouldi radio-tracked between 10 th – 29 th March 2006. Maps for %Boulder and %Cobble show spatial variation in the percentage of each substrate type within the reach. The direction of flow in all cases is from left to right.
17 Units (m) Units (m) 18 18
0.45 0.45
0.4 0.4 16 16 0.35 0.35
0.3 0.3 14 14 0.25 0.25
0.2 0.2
0.15 12 12 0.15
0.1 0.1
0.05 0.05 10 10 0 0
110094 110094 8 110095 8 110095 110097 110097
6 6
4 4
2 2
0 2 4 6 8 10 0 2 4 6 8 10 (a) Water Depth: Pre-weir (b) Water Depth: Post-weir
Units (m/s) Units (m/s) 18 18 1 1 0.9 0.9 16 16 0.8 0.8
0.7 0.7 0.6 14 14 0.6 0.5 0.5
0.4 0.4 12 0.3 12 0.3
0.2 0.2
0.1 0.1 10 10 0 0
110094 110094 8 110095 8 110095 110097 110097
6 6
4 4
2 2
0 2 4 6 8 10 0 2 4 6 8 10 (c) Velocity: Pre-weir (d) Velocity: Post-weir
Units (%) Units (%) 18 18
90 90
80 80 16 16 70 70
60 60 14 14 50 50
40 40
12 30 12 30
20 20
10 10 10 10 0 0
110094 110094 8 110095 8 110095 110097 110097
6 6
4 4
2 2
0 2 4 6 8 10 0 2 4 6 8 10 (e) %Cobble (f) %Pebble Figure 12(a-f). Contour maps showing the distribution of major habitat descriptors for the ‘test’ reach in the lower Emu River, overlayed with location data for juvenile A. gouldi tracked between 10 th – 29 th March 2006. Maps have been included for depth and velocity for periods prior to and following installation of the temporary weir. Maps for %Cobble and %Pebble show spatial variation in the percentage of each substrate type within the reach. The direction of flow in all cases is from left to right. 18 60 60
50 (a) 50 (b)
40 40
30 30
20 20 Number of observations Numberof observations 10 10
0 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Depth (m) Velocity (m/s) 60 60
50 (c) 50 (d)
40 40
30 30
20 20 Number of observations of Number 10 observations of Number 10
0 0 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 %Boulder %Cobble 60 60
50 (e) 50 (f)
40 40
30 30
20 20
Number of observations of Number 10 Number of observations 10
0 0 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 %Pebble %Gravel 60
50 (g) Histograms showing the habitat utilisation of crayfish radio-tracked at both the ‘control’ 40 and ‘test’ reaches of the Emu River during
30 10-29 March, 2000. Size Categories: 20 Boulder = >256 mm Cobble = 64-256 mm Number of observations of Number 10 Pebble = 8-64 mm 0 Gravel = 2-8 mm 0 10 20 30 40 50 60 70 80 90 100 %Algae Figure 13. Frequency histograms of combined location data from juvenile crayfish radio- tracking in the lower Emu River between 10 th – 29 th March 2000.
19 4.5 Changes in habitat availability with flow This section examines the changes in the amount of preferred habitat that is available to juvenile A. gouldi under different river flows. To do this, the water depth and velocity data that was presented in Figure 13 was used along with the additional data from Table 1, as input to the RAP software. Using the hydraulic modelling module in RAP these data were combined to construct a ‘habitat use curve’, and this curve was then used to develop a ‘rating curve’ that predicts the amount of habitat that is available between flows of 0-175 Ml.day -1 for the ‘control’ reach (Figure 14). The graph shows that the availability of preferred habitat within the river (expressed as m2.m -1 of river length) increases in a log-normal fashion with increasing flow, approaching a maximum of about 11.5 m 2.m -1 when flows reach 110 Ml.day -1 .
14
12
10
8
6
4
2 Available (m2 area per m of river)
0 0 50 100 150 Flow (Ml/day)
Figure 14. Rating curve for preferred habitat for juvenile A. gouldi in the ‘control’ reach of the lower Emu River. The curve was derived using water depth and velocity data from both the ‘control’ and ‘test’ reaches.
Although the rating has not been developed for flows above 175 Ml.day -1 , it is likely that as flows exceed this, habitat availability for juvenile crayfish once again declines as water depth and velocity exceeds the levels preferred by juvenile crayfish. However, for the purposes of this study, we are interested only in how habitat availability is affected by a reduction in flow. The curve for these data indicates that once flow drops below about 70 Ml.day -1 , the availability of habitat preferred by juvenile crayfish declines rapidly. If the maximum habitat available for juvenile crayfish in the ‘control’ reach is taken as the area that is available at 110 Ml.day -1 , then from the curve it is possible to classify potentially important flow horizons in terms of percentage reduction in available habitat. The result of this analysis is presented in Table 3. In brief, the data shows that a 43% reduction in flow (62 Ml.day -1 ) reduces available habitat by 20%, and that even if flow declines by 75% (down to 27 Ml.day -1 ) there is still 50% of preferred habitat available within the river.
20 Table 3. The predicted availability of preferred habitat within the ‘control’ reach of the lower Emu River under a range of different river flows. Percentage of preferred Flow (Ml.day -1 ) habitat available 100% 108 90% 78 80% 62 75% 50 50% 27 25% 11
These data form a good base from which the long-term flow record can be examined. The following example has been included simply to illustrate how the habitat data that has been collected during this study can be used to conduct a hydrological analysis for the river that has some ecological meaning in terms of juvenile A. gouldi . While the focus in this instance is on the Emu River, the results also have application to other rivers on the north coast where A. gouldi is present, and river morphology and instream habitat is similar.
Since flow monitoring is not currently carried out in the Emu River, the record from the nearby Leven River has been used. Flow in the Leven River has been monitored at Bannons Bridge (station #14207) since the early 1990’s. Making corrections for the difference in catchment size, and using the spot gaugings taken in the Emu River during the study to calibrate the transposed flow record from the Leven River, the following flow statistics have been calculated (Table 4). Alongside the lower flow percentiles are conclusions as to the predicted availability of preferred habitat for juvenile A. gouldi .
Table 4. Predicted habitat available to juvenile crayfish under different flow categories. Percentage of time Flow Value Predicted habitat available flow value is exceeded (Ml.day -1 ) 100 29.4 Maintains >50% available habitat 95 57.6 Maintains >75% available habitat 90 69.5 Maintains >85% available habitat 85 73.6 Maintains >90% available habitat 80 100.6 Maintains >95% available habitat 75 124.6 Maintains 100% available habitat 50 349.5 Not known 25 792.8 Not known 0 13377 Not known
The results show that at the absolute minimum flow that occurs in the river (the 100% exceedance flow), more than 50% of preferred habitat for juvenile crayfish remains available. This highlights the preference of juvenile crayfish for low velocity, shallow habitat, and the fact that despite the low level of discharge there is still a large area of the river channel that remains wet.
21 Habitat availability increases significantly with only a small increase in flow, so that at the 95% exceedance flow (in this case a flow of 57 Ml.day -1 ), more than 75% of instream habitat preferred by juvenile crayfish is available. Within the water management and planning arena, these data can provide a firm basis against which ‘risks’ to juvenile crayfish from water extraction can be assessed.
This analysis is made solely for the purpose of illustrating how habitat availability changes with flow. In using the flow record from the Leven River it is recognised that there are differences in the level of water use between the two catchments, and the transposed flow data is only likely to be indicative of the flow regime for the Emu River.
22 5. Discussion This study aimed to provide answers to two relatively simple questions: 1. If juvenile A. gouldi inhabit riffle zones in larger rivers, are they forced to abandon this habitat during low flows? and 2. Is water quality different in pools and riffles during warm periods, and are environmental conditions likely to cause stress to adult and juvenile crayfish? Both of these questions were framed in light of recent ecological studies on A. gouldi and are targeted at contributing knowledge that will help in making informed decisions in the area of water management planning and environmental flow allocations.
Radio telemetry proved to be a useful approach for studying the habitat use and movements of juvenile crayfish in the Emu River. While prior work on A. gouldi in Tasmania indicated that juveniles appear to favour riffle zones of larger rivers, little has been documented about the ecology of juvenile crayfish in larger rivers, and this technique has provided finer scale information about their habitat use and movements.
The habitat preference data that was collected during the capture phase of the study, and later substantiated by habitat use data collected using radio telemetry, indicates that juveniles tend to prefer habitat within the river that is dominated by boulder and cobble sized material, where water depth is reasonably shallow (50-250 mm) and where water velocity is moderate to low (0.1-0.7 m.s -1 ). Very small crayfish (<25 mm CPL) were found to be more abundant in sections of riffle where water depth was very shallow and velocity very low. Where boulder substrate is not available, juvenile crayfish show a clear preference for riffle habitat that contains a greater proportion of cobble and pebble material.
These data were collected during a period when flow in the river was at its lowest (early autumn), and therefore are only applicable during periods of low flow. During higher flows this habitat within the river is subject to much greater water depth and velocity. While juvenile crayfish may need to move out of riffles to avoid such unfavourable conditions this is unlikely to occur, as crayfish are more than able to burrow into the bed of the river, taking refuge in the protection of the shallow hyporheic zone. The ability to excavate cavities in substrate was observed during the initial survey work for the study and emphasises the need for suitable interstitial space within riffle zones if crayfish are to employ this tactic. This may also be the main reason why no juveniles could be located at the original environmental flow study site in the Great Forester River, where fine gravels have largely filled these spaces. It is recognised that sediment deposition in Tasmania rivers is a significant threat to this species (Growns 1995; Lynch 1997; Bryant 1999; Walsh 2002) and observations made during the course of this study support this.
The results of the radio-tracking showed that although juveniles generally tend to make small movements whilst remaining within the confines of a riffle, they are capable of moving very large distances, traversing habitat that is known to be preferred by adults (Hamr 1990; Webb 2001). This provides evidence that suggests juvenile crayfish are fully able to travel through habitat that is known to be favoured by adults and where they may be exposed to increased risk of predation (particularly by trout and platypus, but also by larger adult crayfish).
23 Furthermore, the study showed that movement of juveniles in response to either natural or artificial changes in flow conditions appears to vary between individuals, and there was no clear indication that minor reductions in flow stimulated a uniform or substantial response by this sector of the crayfish population. While the results were not totally conclusive, the evidence from the flow manipulation does suggest that reducing flow and velocity within the riffle zone does not cause larger-scale movement by juvenile crayfish into deeper water.
Although the study period was quite short, the temporal pattern of movement shown by animals in this study was similar to that described in other crayfish movement studies. In European studies of introduced freshwater crayfish, movement has been described as “periodic movements intercalated with longer periods of slow or null movement” (Aquiloni 2005) or “stationary phases interposed with nomadic bursts of movement” (Gherardi 2000). The results of this study support the application of these descriptions to the movement of juvenile A. gouldi . In this study, some juveniles showed very stationary behaviour (crayfish #110091 & #110092), while others (crayfish #110099 & #110093) were markedly nomadic. Why this might vary between individuals is not clear, but may be due to differences in growth stage or a result of competitive interactions for suitable habitat (Gherardi 2000).
While most of the individuals that were tagged remained within the riffle from which they were initially captured and released, the tracking data showed that the majority of movement during the 20-day study period was perpendicular to the river channel. Few published studies have specifically examined crayfish movement from a directional viewpoint. However in a radio-tracking study of an introduced freshwater crayfish (Pacifastacus leniusculus ) in England, Bubb et al. (2002) found that lateral and longitudinal movement of crayfish was about equal during autumn and winter, and greater longitudinal movement was more likely when temperatures were warmer. The main reasons given by these authors for a more localised pattern of activity during autumn and winter are cooler water temperatures. They also suggest that movement during the colder months is much more limited to foraging around suitable refuge habitat. Although this study was conducted in a temperate region with a comparable temperate climate, radio-tracking was confined to 20-day period in autumn, when water temperature in northern Tasmanian rivers is at or near its highest. It therefore appears that the localised pattern of activity displayed by juvenile A. gouldi has less to do with water temperature and more to do with a preference at this developmental stage for riffle habitat.
During the study, selected water quality parameters (water temperature and dissolved oxygen) were monitored, both within shallow riffle habitat and in deeper pools. This clearly demonstrated that there was no substantial difference between environmental conditions in riffle and pool habitat during low flows and elevated seasonal temperatures. Although the maximum water temperature recorded during this study was only 18.5 oC, and the temperature in larger rivers along the northwest coast of Tasmania commonly reaches 22 oC at the height of summer (WIST database), it can safely be concluded that except under extreme or unusual conditions, elevated water temperature is unlikely to cause juvenile crayfish to abandon riffle habitat in favour of deeper pool habitat. If climatic conditions do cause stress to crayfish through elevated water temperature, the most likely response is likely to be one whereby crayfish burrow into substrate seeking less harsh conditions. Studies examining surface water-
24 groundwater interactions have shown that the amplitude of surface water temperature signals is significantly dampened with depth into the hyporheic zone, and daytime peaks in temperature of surface water can be as much as 9 oC lower in the hyporheic zone (Malcolm 2004). Given the burrowing ability of A. gouldi , this behavioural response to temperature stress is more likely.
In conclusion, it is clear that since juvenile A. gouldi are mobile and readily move through deeper water (where the presumed risk of predation is higher), the occurrence of low flows does not appear to pose a significant threat to this component of the population. The radio-tracking data suggests that while juvenile crayfish are willing to leave riffle zones, they are unlikely to be forced to move because of a lack of suitable habitat or environmental stress during periods of low flow. In terms of water management, the study has provided valuable habitat use information that can contribute to the development of holistic, environmental flows. The study shows that juvenile A. gouldi are likely to cope well with periods of low flows in rivers, and that as a part of the invertebrate community that inhabit riffle zones, are unlikely to require special attention during future environmental flows studies.
25 6. References Aquiloni, L., Ilheu, M. and Gherardi, F. (2005). 'Habitat use and dispersal of the invasive crayfish Procambarus clarkii in ephemeral water bodies of Portugal.' Marine and Freshwater Behaviour and Physiology 38(4): 225-236 pp. Beeman, J. W., Grant, C. and Haner, P.V. (2004). 'Comparison of three underwater antennas for use in radiotelemetry.' Nth. Amer. J. of Fish. Manag. 24: 275-281 pp. Bovee, K. D. (1982). A guide to stream habitat analysis using the instream flow incremental methodology. Washington, US Fish and Wildlife Service : 249 p. Bryant, S. L. a. J., J. (1999). Tasmania's Threatened Fanu Handbook: what, where and how to protect Tasmania's threatened animals . Hobart, Parks and Wildlife Service. Bubb, D. H., Lucas, M.C. and Thom, T.J. (2002). 'Winter movement and activity of signal crayfish Pacificus leniusculus in an upland river, determined by radio telemetry.' Hydrobiologia 483: 111-119 pp. Davies, P. E. and Cook, L.S.J. (2004). Juvenile Astacopsis gouldi in headwater streams - relative abundance and habitat. Hobart, Forest Practices Board : 40 p. DPIWE (2003). Great Forester Catchment Water Management Plan. New Town, Tasmania, Dept. Primary Industries, Water and Environment. Forteath, N. (1987). The aquaculture potential of the giant freshwater crayfish Astacopsis gouldi. Launceston, School of Applied Science, Tasmanian State Institute of Technology : 22 p. Gherardi, F., Barbaresi, S. and Salvi, G. (2000). 'Spatial and temporal patterns in the movement of Procambarus clarkii , and invasive crayfish.' Aquatic Sciences 62: 179-193 pp. Gordon, N. D., McMahon, T.A., Finlayson, B.L., Gippel, C.J. and Nathan, R.J. (2004). Stream hydrology: An introduction for ecologists . Chichester, England, Wiley. Growns, I. O. (1995). ' Astacopsis gouldi Clark in streams of the Gog Range, northern Tasmania: effects of catchment disturbance.' Papers and Proc. of the Royal Soc. Tasmania 29: 1-6 pp. Hamr, P. (1990). Comparative reproductive biology of the Tasmanian freshwater crayfishes Astacopsis gouldi Clark, Astacopsis franklinii Gray and Parastacoides tasmanicus Clark (Decapoda: Parastacidae). Department of Zoology. Hobart, University of Tasmania. HEC (2005). HEC-RAS. River Analysis System. Davis, California., Hydrologic Engineering Centre, US Army Corps of Engineers. Horwitz, P. H. J. (1994). 'Distribution and conservation status of the Tasmanian giant freshwater lobster Astacopsis gouldi (Decapoda: Parastacidae).' Biological Conservation 69: 199-206. Lyall, J. (2000). Biotic and abiotic factors affecting populations of Astacopsis gouldi in Northwest Tasmania. School of Applied Science. Burnie, University of Tasmania. Lynch, T. P. a. B., D.R. (1997). Reservation assessment and habitat requirements of the giant Tasmanian freshwater lobster, Astacopsis gouldi . Hobart, Inland Fisheries Commission : 74 p.
26 Malcolm, I. A., Soulsby, C., Youngson, A.F., Hannah, D.M., McLaren, I.S. and Thorne, A. (2004). 'Hydrological influences on hyporheic water quality: implications for salmon egg survival.' Hydrol. Process. 18: 1543-1560. McKenny, C. and Read, M. (1999). Ecological flow requirements for the Great Forester River. Hobart, Department of Primary Industries, Water and Environment. WRA 99/15 39 p. Robinson, C. A., Thom, T.J. and Lukas, M.C. (2000). 'Ranging behaviour of a large invertebrate, the white-clawed crayfish Austropotamobious pallipes .' Freshwater Biology 44(3): 509-521. Stewardson M. and Marsh, N. (2004). Hydraulic Analysis Module: River Analysis Package., Cooperative Research Centre for Catchment Hydrology, Monash University, Melbourne, Australia. Walsh, T. and Nash, W. (2002). Factors influencing the health of the giant freshwater lobster in Tasmanian rivers. Moonah, Tasmania, Inland Fisheries Service : 30 p. Webb, M. (2001). Movement patterns and habitat use of adult Astacopsis gouldi in the Dip River, north-west Tasmania. School of Zoology. Hobart, University of Tasmania : 99 p.
27 APPENDIX A: Environmental condition monitoring During the course of the radio-tracking, selected water quality monitoring was undertaken at both the ‘control’ and ‘test’ reaches to determine if changes in environmental conditions during the study might help to explain movements by juvenile crayfish. In undertaking this, the aim was also to identify whether conditions in riffle habitat differed markedly from that occurring in deeper pool habitat, where shading and greater water depth might provide a less stressful environment for crayfish. Miniature temperature loggers were deployed in the riffles at both study reaches, and a multi-sensor logger deployed at the deepest part of the pool situated at the bottom of the ‘test’ reach.
The time series data from this logger is displayed in Figure A1, and shows how water temperature, electrical conductivity and water level varied during the entire period of the study. As the data show, water temperature was highest during the early stages of the study, as fine, warm weather conditions prevailed on the north coast of Tasmania. This ended abruptly with the arrival of a cold front on day 4, resulting in rainfall throughout the catchment, and causing a moderate rise in water level in the river on day 8. During this event, gauged flow in the river increased from about 50 Ml.day -1 to 136 Ml.day -1 . Following passage of the cold front and the subsequent rainfall, water temperature and conductivity was markedly lower, rising only gradually over the second half of the study period.
2.00
1.80
1.60
1.40
1.20
1.00 Units
0.80 Water Level (m)
0.60 Water Temperature (Factored by 0.1)
Conductivity (Factored by 0.01) 0.40
0.20
0.00 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Study days Figure A1. Time series of changes in water level (recorded in metres), water temperature ( oC) and electrical conductivity ( µS.cm -1 ) at the ‘test’ reach of the Emu River during the course of the study. To allow all three time series to be displayed on a single plot, water temperate and conductivity data have both been factored to fit on a y-axis scale of 0 to 2.
Comparison of the daily change in water temperature recorded in the riffles with that recorded in pool habitat (Figure A2) during the early days of the study shows that there is little difference between temperature in the two habitats. Both clearly show the same pattern and scale of diurnal change, with peak temperature occurring at all
28 locations late in the day (at about 1900 hours). At the time of peak temperature, it does appear that water temperature in the pool is always slightly lower than in the riffles, however the difference is always less than 0.1 oC. It is also interesting to note that there is virtually no difference in water temperature between the two reaches, despite the difference in their morphology and more particularly in the amount of incident light they receive. It was visually noted during the study that the riffle in the ‘test’ reach received considerably more incident sunlight than the ‘control’ reach, where the more confined topography shaded the river more.
19.0 Mid-riffle SN931630 Control Mid-riffle SN931631 18.5 Mid-riffle SN931629 Test Mid-riffle SN931632
18.0 Test - Pool
17.5
17.0
16.5 Watertemperature (Degrees C)
16.0
15.5
15.0 10-Mar 11-Mar 12-Mar 13-Mar 14-Mar 15-Mar Figure A2. Comparison of changes in water temperature ( oC) recorded during the earlier part of the study by mid-riffle loggers located at the ‘control’ and ‘test’ reaches with that recorded in the deeper pool habitat situated below the ‘test’ reach on the Emu River. The vertical arrows indicate times when spot readings of water temperature and dissolved oxygen were recorded using hand-held equipment.
In addition to providing spot checks on the water temperature data being collected by the instream loggers, measurements of water quality by the hand-held instrument also provided an indication of the level of variation in dissolved oxygen both within and between sites. The data from these spot checks is provided in Table 2, and clearly shows that there is no substantial or ecologically meaningful difference in water temperature or dissolved oxygen level between habitats. The difference between reaches is simply a function of readings being taken at a different time of day. It can therefore be concluded that under low flows and elevated seasonal temperatures, there is unlikely to be an ecologically significant difference in environmental conditions between riffle and pool habitat, unless there is some localised impact that might cause such a situation.
29 Table A1. Spot water quality measurements taken at locations of loggers situated within both ‘control’ and ‘test’ reaches of the Emu River. Date Location Wat Temp DO [DO] (oC) %Saturation (mg.l -1 ) 10/03/3006 8:45 Test 1 - pool logger 15.4 91 8.9 10/03/3006 8:45 Test 1: Upper Hobo - 931629 15.7 90.8 8.9 10/03/3006 8:45 Test 1: Lower Hobo - 931632 15.7 92.7 9.1 10/03/3006 11:45 Control: Upper Hobo - 931630 15.9 91.7 9.07 10/03/3006 11:45 Control: Lower Hobo - 931631 15.9 92.3 9.12 11/03/2006 12:15 Test 1 - pool logger 17.6 108.2 10.33 11/03/2006 12:15 Test 1: Upper Hobo - 931629 17.2 106 10.2 11/03/2006 12:15 Test 1: Lower Hobo - 931632 17.3 107.5 10.3 11/03/2006 11:00 Control: Upper Hobo - 931630 16.6 91.5 9.00 11/03/2006 11:00 Control: Lower Hobo - 931631 16.6 93 9.06 12/03/2006 8:25 Test 1: Upper Hobo - 931629 18.5 113 10.6 12/03/2006 8:25 Test 1: Lower Hobo - 931632 18.5 111.5 10.5 13/03/2006 11:00 Test 1: Upper Hobo - 931629 18 100.4 9.55 13/03/2006 11:00 Test 1: Lower Hobo - 931632 18 101.5 9.58
30 Appendix B: Accuracy of spatial locations In an effort to assess the accuracy of the location data that was collected using the hand-held directional antenna at fixed benchmark points (a standard approach to radio-tracking), a second technique to locate individuals was used. This involved replacing the hand-held Yagi antenna with a ‘stripped coaxial cable antenna’ (Beeman 2004) and wading into the stream as described in Section 2. During the pilot testing conducted prior to the study, and also at the end of the study when animals were recaptured, it was estimated that the use of the ‘coaxial antenna’ allowed the position of radio-tagged crayfish to be located to within +/- 0.25 m of their actual position.
As outlined in Section 2, the accuracy of the standard ‘directional antenna’ approach was assessed by firstly locating crayfish from fixed benchmarks on the river bank using the Yagi antenna and a line-of-sight compass to record bearings. Once individual crayfish had been located in this manner, the ‘coaxial antenna’ was then used to locate crayfish more precisely, and bearings were once again taken from the river bank benchmarks. To allow data from the two approaches to be compared, locations were plotted on a gridded map of the reach and the straight-line difference between the two locations measured. This process was carried out on 34 occasions during the study, and these data combined for use in assessing the accuracy of the ‘directional antenna’ approach.
The results of the assessment are presented below in the form of a statistical summary (Table B1) and a frequency histogram displaying the degree of departure of the ‘directional antenna’ data from the more accurate ‘coaxial antenna’ data (Figure B1). The results show that on average the ‘directional antenna’ locations differed from the ‘coaxial antenna’ locations by 1.4 m, but could be as high as 3.2 m. When the raw data is examined more closely, the largest differences appear to have occurred when bearings were recorded from target crayfish that occupied deeper habitat in the river and when the target was located beneath a substantial amount of boulder and cobble material. When the latter condition occurred it was often very difficult to detect a signal from the target using the directional antenna, and this impacted significantly on the ability to get a reasonable bearing. When the target was situated in shallow water and beneath smaller sized material, the signal was often strongest and it was consequently easiest to obtain a definite bearing. As a result, these locations tended to be more accurate.
The results of the assessment clearly show that crayfish location data that is obtained using the directional antenna has a significant error when used alone for the purpose of identifying crayfish habitat use at the meso-scale within a river system. In this case, data accuracy is substantially improved by verifying location using a low-gain, stripped coaxial cable antenna. Where the aim is to understand medium to large-scale movement, the accuracy of the directional antenna approach is considered less of an issue. It should therefore be kept in mind that when viewing data on crayfish movement that has been obtained using directional antennae, smaller scale movements (<2 m) may be due to error in locating animals rather than small-scale movements by crayfish.
31 Table B1: Summary statistics of difference between crayfish locations recorded using the ‘directional 12 antenna’ and ‘stripped coaxial cable’ approaches. 10 8 Summary statistic Mean 1.395 6
Median 1.2 Frequency 4 Mode 1 2 Standard Error 0.105 Confidence Level (95%) 0.213 0 0 - 0.5 0.5 - 1.0 1.0 - 1.5 1.5 - 2.0 2.0 - 2.5 >2.5 Standard Deviation 0.612 Location difference (m) Sample Variance 0.374 Kurtosis 1.017 Skewness 0.904 Figure B1. Frequency histogram of difference Minimum 0.5 between crayfish locations recorded using the Maximum 3.2 ‘directional antenna’ and ‘stripped coaxial Count 34 cable’ approaches.
32 APPENDIX C: Habitat use data for juvenile A. gouldi
25 18 16 20 14 12 15 10 8 10 6
5 4 Frequency of observations of Frequency Frequency of observations of Frequency 2
0 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Depth (m) Velocity (m/s)
16 18
14 16 14 12 12 10 10 8 8 6 6 4 4 Frequency of observations of Frequency Frequency of observations 2 2 0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 %Boulder %Cobble
50
45 40
35 Control Reach
30 Histograms showing the habitat utilisation 25 of crayfish radio-tracked at the ‘control’
20 reach of the Emu River during 10-29 March, 15 2000. Size Categories: 10 Frequency of observations Boulder = >256 mm 5 Cobble = 64-256 mm 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Pebble = 8-64 mm %Pebble Gravel = 2-8 mm
33 18 14
16 12 14 10 12
10 8
8 6 6 4 4 Frequency Frequency of observations Frequency ofFrequency observations 2 2
0 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Depth (m) Velocity (m/s)
16 25
14 20 12
10 15
8
6 10
4 5 Frequency Frequency of observations Frequencyofobservations 2
0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 %Cobble %Pebble
45 40 Test Reach 35 Histograms showing the habitat utilisation 30 of crayfish radio-tracked at the ‘test’
25 reach of the Emu River during 10-29 March, 2000. Size Categories: 20 Boulder = >256 mm 15 Cobble = 64-256 mm 10 Pebble = 8-64 mm Frequency Frequency of observations 5 Gravel = 2-8 mm 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
%Gravel
34