Environmental influences on the spatial ecology and spawning behaviour of an estuarine-resident fish, Macquaria colonorum

Estuarine, Coastal and Shelf Science 118 (2013) 60e71

Contents lists available at SciVerse ScienceDirect

Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss

Environmental influences on the spatial ecology and spawning behaviour of an estuarine-resident fish, Macquaria colonorum C.T. Walsh a, b, c, *, I.V. Reinfelds d, M.C. Ives b, C.A. Gray b, R.J. West c, D.E. van der Meulen b a

NSW Department of Primary Industries, Batemans Bay Fisheries Centre, PO Box 17, Batemans Bay, NSW 2536, Australia NSW Department of Primary Industries, Cronulla Fisheries Research Centre of Excellence, PO Box 21, Cronulla, NSW 2230, Australia c University of Wollongong, School of Biological Sciences, NSW 2522, Australia d NSW Office of Water, PO Box 53, Wollongong, NSW 2500, Australia b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 March 2012 Accepted 28 December 2012 Available online 8 January 2013

Estuarine-resident fishes are highly susceptible to the effects of environmental and anthropogenic impacts on their assemblages and habitats. We investigated the distribution, movement and spawning behaviour of estuary perch, Macquaria colonorum, in response to selected environmental variables using an acoustic telemetry array in a large tidal river in south-eastern (SE) Australia. Adult M. colonorum were monitored for up to two years, covering two consecutive spawning periods between September 2007 and 2009. Salinity, water temperature and river flows all had a significant relationship with their estuarine distribution. In particular, large-scale movements were influenced by large freshwater inflow events and the resultant reduction in salinity levels, together with the seasonal cooling and warming trends in water temperatures associated with spawning behaviour. During the winter months, male and female M. colonorum migrated from their upper estuarine home ranges to the lower estuarine spawning grounds in synchrony, with numbers of individual visits by both sexes consistently higher in the ‘wetter’ winter/spring period of 2008. Location, arrival, departure and occupation time within the spawning grounds were similar between sexes and years. Both resident and migrating M. colonorum exhibited strong diel, and to a lesser extent, tidal behavioural patterns, with fish more likely to be detected at night and during the ebb tides. It is postulated that the effect of environmental fluctuations on the distribution and movement of M. colonorum is influenced by behavioural mechanisms in response to osmoregulatory stress, predatoreprey interactions and reproductive activity. The results also demonstrate the importance of accounting for autocorrelation inherent in telemetry data, and for developing management strategies that are more robust to the effect of future climate trends on estuarine fish populations. Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.

Keywords: acoustic array biological rhythms environmental factors estuarine-resident Macquaria colonorum spawning migrations

1. Introduction Estuarine environments are characterised by marked temporal and spatial fluctuations in abiotic conditions (e.g., salinity, water temperature and turbidity) due to competing influences of freshwater inflows and tidal flow (Bennett and Branch, 1990; Kimmerer, 2002; Heupel and Simpendorfer, 2008). Strong relationships exist between the distributions of estuarine fish and such environmental factors (Blaber and Blaber, 1980; Whitfield, 1994; Marshall and Elliott, 1998). Salinity has long been recognised as a primary factor influencing the use, movement and community composition of fishes in estuaries (Harrison and Whitfield, 2006; Selleslagh and

* Corresponding author. NSW Department of Primary Industries, Batemans Bay Fisheries Centre, PO Box 17, Batemans Bay, NSW 2536, Australia. E-mail address: [email protected] (C.T. Walsh).

Amara, 2008), while water temperature has been found to control key physiological, biochemical and life-history processes (Beitinger and Fitzpatrick, 1979). Fish in estuaries have evolved strategies to deal with fluctuating environmental conditions. For example, sedentary estuarine fishes tend to osmoregulate over a broad range of salinities, whereas more mobile fishes may move to areas that are more suitable as environmental conditions change (Whitfield, 1994; Heupel and Simpendorfer, 2008). Biotic factors, including the distributions and behaviour of predator and prey species also influence the movement of fish in estuaries (Marshall and Elliott, 1998; Elliott et al., 2007). This is not unusual as the behavioural rhythms of fish and invertebrates have been found to be similarly linked to cyclical events such as diel, tidal and moon phase (Morgan, 2001). Many estuarine-resident, including estuarine-dependent fishes have well defined home ranges for most of their routine activities before migrating to distant discrete spawning areas (Pittman and McAlpine, 2001; Crook et al., 2010; Walsh et al., 2012b). The

0272-7714/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ecss.2012.12.009

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

factors determining spawning can be complex and usually relate to one or more cues over a range of temporal scales, including time of day, state of the tide, lunar period and season (Pittman and McAlpine, 2001). Although the spawning migrations of many freshwater and estuarine fishes have been linked to environmental cues (McDowall, 1988; Pusey et al., 2004), the precise ‘triggers’ for these migrations are still not fully understood. Increasing and decreasing photoperiod has been found to be a major predictive, proximate factor indicating seasonal spawning migrations in fishes (Smith, 1984). However, the annual variation of timing of migration is more likely to be caused by differences in environmental conditions such as river discharge and water temperature (Jonsson, 1991; Svendsen et al., 2004). This has implications for regulated river systems as ‘unseasonal’ periodic flow releases or periods of flow suppression can desynchronise temperature and flow conditions, potentially resulting in a loss of cues for spawning and dispersal, as well as restricting fish passage (Whitfield, 1994; Gehrke et al., 1999; Reinfelds et al., 2010). The estuary perch, Macquaria colonorum is an estuarineresident fish found predominantly within the tidal reaches of rivers, lakes and coastal lagoons of south-eastern (SE) Australia (Harris and Rowland, 1996). Although on occasions they have been captured in coastal waters, previous studies suggest they complete their life cycle within a specific estuary (Williams, 1970; McCarraher and McKenzie, 1986; Walsh et al., 2012b). Specifically, Walsh et al. (2011) reported increasing gonad development of both sexes within the lower estuaries of rivers during the austral winter months, with ripe females carrying multiple size-class ‘batches’ of eggs. Moreover, salinity levels considered for optimal M. colonorum spawning and larval development are consistent with those found within these marine dominated lower estuarine areas (Beckman, 1999). Similar to other estuarine-resident fishes (e.g., mulloway, Argyrosomus japonicus, Ferguson et al., 2008; black bream,

61

Acanthopagrus butcheri, Jenkins et al., 2010) the high inter-annual variability in recruitment success of M. colonorum associated with occurrences of ‘dry and ‘wet’ spawning years suggests that freshwater inflows may play a role in influencing spawning behaviour and success (Walsh et al., 2010). Although the spatial and temporal distribution patterns and reproductive ecology of M. colonorum have been investigated (Walsh et al., 2011, 2012b), the influence of environmental factors on their residence, movement and spawning behaviour has not been examined in detail. Recent developments in acoustic tagging technologies have increased our capacity to directly link the distribution and movement of fish with key abiotic and biotic factors (Hindell et al., 2008; Childs et al., 2008a; Sakabe and Lyle, 2010). Such work has enhanced our understanding of the ecological relationships of, and potential effects of anthropogenic impacts, on estuarine fish species. The present study investigates the distribution, movement and spawning behaviour of Macquaria colonorum in response to selected environmental variables using an acoustic telemetry array in a regulated coastal river. Specifically, our hypotheses were: (1) salinity, water temperature and river flow influenced M. colonorum location, movement and spawning behaviour; (2) there was no effect of diel, tidal state and moon phase on M. colonorum detectability, and (3) there were no sex-related differences in the timing, frequency and duration of M. colonorum spawning events. 2. Material and methods 2.1. Study area and environmental characterisation The Shoalhaven River (
34 530 S, 150 450 E) is ca.327 km long and has a total catchment area of 7300 km2 (Roy et al., 2001) (Fig. 1). The Tallowa Dam impounds the Shoalhaven River 75 km upstream of the Pacific Ocean with mean and median annual

Fig. 1. Study site location and detailed map showing the position of acoustic listening stations in the Shoalhaven River. Filled circles represent receivers used in this study, while open circles represent additional receivers deployed in the freshwater reaches. Lower estuary (LE), lower middle estuary (LME), upper middle estuary (UME) and upper estuary (UE) zonation based on salinity and geomorphological characteristics (Roy et al. 2001; Walsh et al. 2012b).

62

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

inflows to the dam (calculated for 1909e2004, Reinfelds et al., 2010) of 1043 GL and 631 GL, respectively. Due to the small storage capacity relative to annual inflows, the dam frequently overtops during wet years creating unmanaged spills of water flowing to the river downstream thereby regulated environmental flow releases (Reinfelds et al., 2012). The estuarine section of the river (48 km in length upstream of mouth) has a tidal range of approximately 1.2 m and temperatures typically range between 8  C and 25  C annually. Average, minimum and maximum salinities recorded for the lower estuary (LE), lower middle estuary (LME), upper middle estuary (UME) and upper estuary (UE) (Fig. 1) were 32.7 (8.7e34.5), 21.6 (0e29.7), 13.27 (0e24.6) and 3.9 (0e11.4), respectively. Hence, the LE zone is typically marine (euhaline), while the middle two zones represent the estuarine or mixing zone between river and sea (mesohaline and polyhaline) and the UE zone is freshwater dominated (oligohaline). Throughout the study freshwater inflows (river flow) were continuously recorded at Grassy Gully (gauge: 215,216) upstream of the tidal limits, while salinity and water temperature were recorded via Odyssey data loggers deployed at seven locations within the estuary (Fig. 1). A calibrated estuarine hydrodynamic salinity model (Miller et al., 2006) was extended to cover the study period, thus enabling salinity measurements and tidal state predictions to be estimated for each receiver location at hourly increments.

2.3. Research procedure

2.2. Fish tagging and acoustic telemetry Nineteen mature adult Macquaria colonorum were implanted with Vemco V13 transmitters and released between September and December 2007 (see Table 1). To avoid the possibility that movement patterns may be influenced by the location of tag and release sites, fish were captured at several sites among representative reaches. Fish were angled using hook and line and operated on in situ. Fish were placed in an anaesthetic bath containing 50 mg L
1 of Aqui-S solution until their operculum rate became slow and irregular (2e3 min). Fish were then weighed (1 g), measured (fork length, 1 mm) and placed ventral-side up in a V-shaped surgical cradle. During surgery, 25 mg L
1 of Aqui-S solution was irrigated over the gills of the fish using a battery operated pump. Vemco V131L/13 TP (13 mm in diameter) transmitters, with a battery life of two years and weighing <2% of the body mass of the fish, was

Fish ID

Sex

Fork length (mm)

Tagging date

Last detection

No. of days monitored

Analysed

1035952 1035953 1035954 1035955 1035956 1035957 1035958 1035960 1036284 1036285 1036286 1036287 1036288 1036289 1036290 1039619 1039758 1039759 1039761

F F M M F * F * F M F F F F M M M M M

360 373 358 366 376 327 403 335 300 321 315 322 312 339 331 312 321 314 314

24/09/2007 23/10/2007 21/09/2007 21/09/2007 21/09/2007 14/12/2007 14/12/2007 14/12/2007 21/09/2007 24/09/2007 24/09/2007 24/09/2007 24/09/2007 23/10/2007 24/09/2007 24/09/2007 14/12/2007 14/12/2007 14/12/2007

02/12/2009 24/01/2008 12/09/2009 16/11/2009 06/10/2009 03/02/2010 25/12/2007 11/02/2010 28/09/2008 22/12/2009 21/12/2009 29/08/2009 30/11/2009 02/02/2009 09/12/2009 21/12/2009 23/09/2009 14/02/2010 20/05/2009

801 94 723 788 747 783 12 791 374 821 820 706 799 469 808 820 650 794 524

Y N Y Y Y Y N Y Y Y Y Y Y Y Y Y Y Y Y

Sex could not be determined.

Data from receivers were downloaded bi-monthly and stored in a Microsoft SQL Server database. Prior to analyses, ‘false’ detections and all single hit data found on receivers were discarded (see Clements et al., 2005). Once a tagged fish was detected within the array, it was allocated to a receiver for every second of the day until its last detection. This was determined by halving the time it took the fish to move between receivers and allocating an equal portion of the time to the receivers that the fish departed from and arrived at, respectively (Cowley et al., 2008; Walsh et al., 2012b). The mean daily location of individuals (i.e., distance from sea) was calculated based on the proportion of time spent within the vicinity of a receiver (a proxy for known river distance, Walsh et al., 2012b). Secondly, all fish and receiver location data were binned into hourly increments, including whether a fish was detected (presence/ absence) during the particular time period. 2.4. Analysis of distribution and environmental parameters

Table 1 Summary of M. colonorum tagged and released in the Shoalhaven River.

*

inserted into the peritoneal cavity of each fish via a 20 mm incision on the ventral mid-line between the pelvic fins and the anus. The incision was closed with two synthetic absorbable sutures (Ethicon Vicryl 3e0) and tied with a double surgeons knot. Post surgery, fish were injected with an antibiotic (oxytetracycline) at a dose rate of 75 mg kg
1 fish weight. All fish were further identified with an external T-bar anchor tag (Hallprint) inserted below the dorsal fin. Fish were allowed to recover in holding pens until they were able to swim normally, after which they were released at their point-ofcapture within 24 h. A linear array of 28 acoustic receivers (Vemco VR2W) was deployed in the estuarine section of the Shoalhaven River (Fig. 1). Methods for deploying the receivers are described in Walsh et al. (2012a). All receivers were deployed by September 2007 and continuously monitored until February 2010. Receivers recorded the time, date, identity and, when available, the water temperature and depth of the tagged fish that swam within range of a unit. Receivers were single frequency (69 kHz) omnidirectional units and had mean detection ranges of 350 m (range, 280e420 m). The location of receivers within the array ensured that there was minimal chance of tagged Macquaria colonorum escaping detection (0.4%) when swimming past a receiver, resulting in the continuous extensive monitoring of fish present within the study area (Walsh et al., 2012a).

Regression analyses and generalised linear models (GLM) were used to investigate whether daily mean salinity, water temperature and river flow influenced the location of Macquaria colonorum. The Gypsy Point logger station was taken as the reference station for all regression and GLM analyses, and daily mean salinity and water temperature values calculated using data measured at the location (Fig. 1). This approach was taken as the aim was to examine how the fish reacted to environmental variation, not how well they managed to minimise exposure to this variation through movement. The inherent autocorrelation (serial correlation) found in movement data has been well recognised in the telemetry literature (De Solla et al., 1999; Childs et al., 2008a). This is a problem as the location of an animal is not independent of where they were in the previous time step (Aarts et al., 2008). As the data used in this study were averaged over each day, thus reducing the chance of dependence between successive data points (Swihart and Slade, 1985), we fitted the data using a non-autoregressive model then tested for autocorrelation using both the Durbin-Watson and BreuscheGodfrey tests. Once the autocorrelation in the dependent variable was confirmed a number of autoregressive (AR) models

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

3. Results Tagged Macquaria colonorum were monitored within the Shoalhaven estuary for a period of 12e821 days (mean ¼ 649 days) (Table 1). Two of the 19 tagged individuals were not detected after 100 days and were excluded from all analyses. Of the remaining 17 fish, all were present after one year, with 14 fish detected almost two years after initial capture. Overall, these individuals were present within the estuary across all months, and were exposed to a wide range of environmental conditions. Tagged M. colonorum were detected in salinities ranging between 0 and 35 (median ¼ 15.4) and in water temperatures ranging between 10 and 27  C (median ¼ 20.2). River flow was relatively constant throughout the study period with four minor floods recorded, including a 1 in 1.5 year flood event (>250 m3 s
1 mean daily flow rate) in February 2008 (see Figs. 2c and 3).

a)

20

15

15

10

10

5

5

30

b)

30

25

25

20

20

15

15

10

10

5

5

30

c)

250

25

200

20

150

15

30

10

20

5

10

Oct

Temperature (C)

25

20

Salinity (ppt.)

30

25

Feb

Jun

Oct

Feb

-1

30

3

Temporal periodicity in Macquaria colonorum detections (i.e., whether fish visited locations randomly or at recurring time intervals), were determined using Fast Fourier Transform (FFT) analysis. This method decomposes time period data into a finite sum of sine and cosine waves of different frequencies, allowing biological cycles in fish movement to be identified (Hartill et al., 2003; Reyier et al., 2011). Each fish detected by the array was recorded as “present”, while those undetected recorded as “absent”. FFT can only be performed on data series lengths that are truncated to a power of two (i.e., 2, 4, 8, 16. 4096, 8192 etc.). As a result, 8192 h of data were available for 14 fish; and 4096 h of data for three fish. Daily variations in detection rates were displayed graphically by determining the mean proportion of total detections of all fish into hourly bins. As day length varies throughout the year, detections were further classified as ‘dawn’ (sunrise  1 h), ‘day’, ‘dusk’ (sunset  1 h) and ‘night’. A chi-squared test was used to determine whether fish were detected more or less often than expected by chance during these time periods. All FFT calculations and chi-squared tests were performed using MS Excel, and analysed using the software package Statistica 6 (Statsoft Inc., 2010). The influence of diel, tidal state and moon phase on Macquaria colonorum spawning migrations was examined using circular statistics (Batschelet, 1981; Childs et al., 2008b). The mean time of day (hour), tidal state (hours after high tide) and moon phase (day of the lunar month) for each fish was calculated as theta (q), the mean direction of the resultant vector (measured in radians). In order to not contravene the assumption of independence (Grafen and Hails, 2002), q for each fish was then used to calculate mean q or time of day, tidal state and moon phase that each tagged M. colonorum ‘arrived’ at, or ‘departed’ from, the identified ‘spawning grounds’. Spawning grounds were defined as the area monitored by the first five upstream receivers within the LE (Fig. 1). This was based on spatial and temporal studies that examined seasonal trends in population abundance and reproductive status of M. colonorum, as well as from recent egg surveys (Williams, 1970; Walsh et al., 2011; D. van der Meulen unpl. data). The Rayleigh test of randomness (Batschelet, 1981) was used to test whether the timing of these migrations were random or whether they exhibited ‘directedness/ non-randomness’ towards a specific time of day, tidal state, or moon phase.

The spawning migration behaviour of male and female Macquaria colonorum were characterised by the date of arrival, date of departure and residency time within the spawning grounds. The dates of arrival and departure were defined as the first and last day a tagged M. colonorum spent a full 24 h within the designated area. A non-parametric KruskaleWallis ANOVA was used to test the null hypothesis that there was no significant difference in median number of days spent within the spawning grounds between male and female M. colonorum for 2008 and 2009. A Kolmogorove Smirnov (KS) maximum difference statistic test (Sokahl and Rohlf, 1981) was used to test for any significant difference in the timing and utilisation (i.e., cumulative distribution) of M. colonorum within the spawning grounds, within and between years, by sex.

Flow (m .s )

2.5. Analysis of temporal periodicity and patterns of detections

2.6. Analysis of spawning migration and behaviour

Mean distance from sea (Km)

were tested against the data, with each model fitted using generalised least squares fit by maximum likelihood (‘gls’ function in nlme library of the R software, Ihaka and Gentleman, 1996). The autoregressive model with factors that produced the lowest Akaike Information Criterion (AIC) was considered the best fitting and most parsimonious model (Aarts et al., 2008). We tested models up to the twelfth order where the AIC values began to rise again. The dependent variable used was the daily mean distance of Macquaria colonorum individuals from the sea (Distance from Sea). Explanatory variables that where tested in the model included daily water temperature (Temp), salinity (Salinity) and log transformed river flow (LogRiverFlow) as well as gender (Sex) and two Boolean variables representing high river discharge (HRD) events (>100 m3 s
1) and low water temperature (LowTemp) days (13  C). The latter was incorporated into the analysis as a substitute for seasonal biological induced movement associated with the timing of annual M. colonorum spawning behaviour (Walsh et al., 2011). Although correlated with river flow, the HRD variable was included in an attempt to test the hypothesis that high river discharge events are the only changes in flow that are relevant to the movement of M. colonorum (Loneragan and Bunn, 1999; Robins et al., 2005).

63

Jun

Month (2007-9) Fig. 2. Mean distance from sea (black lines) of tagged M. colonorum and mean daily a) salinity, b) water temperature, and c) river flow (all grey lines). Daily salinity and water temperatures measured at Gypsy point. River flow recorded at Grassy Gully.

27

10

26

150

8 6 4

100

2 0

200

3

-1

Flow (m .s )

250

12

25 24 23 22

o

a) 300

Temperature ( C)

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

Salinity (ppt.)

64

21

50

20

Proproportion of duration

b) 1.0

X D ata

UE

0.8

UME

0.6

LME

0.4

LE

0.2

F eb

M ar

A pr

Table 2 Comparison of models constructed for estimating the influence of environmental factors on the daily mean distance of M. colonorum from the sea. Models are grouped by a) non autoregressive b) first order autoregressive and c) eleventh order autoregressive. Models with the lowest AIC value indicate the best fitting and most parsimonious models (in bold). Explanatory variables include daily water temperature (Temp), salinity (Salinity), log transformed river flow (LogRiverFlow), gender (Sex), high river discharge (HRD) (>100 m3 s
1) and low water temperature (LowTemp) days (13  C). a) Non AR Models Salinity þ Temp þ LowTemp þ HRD þ Sex þ LogRiverFlow Salinity þ Temp þ LowTemp þ HRD þ Sex

AIC 78610 78631

b) AR1 Models Salinity þ Temp þ LowTemp þ HRD þ Sex þ LogRiverFlow Salinity þ Temp þ LowTemp þ HRD þ LogRiverFlow Salinity þ Temp þ LowTemp þ HRD Salinity þ LowTemp þ HRD Salinity þ LowTemp Salinity

AIC 48827 48823 48822 48821 48821 48822

c) AR11 Models Salinity þ Temp þ LowTemp þ HRD þ Sex þ LogRiverFlow Salinity þ Temp þ LowTemp þ HRD þ LogRiverFlow Salinity þ Temp þ LowTemp þ HRD Salinity þ LowTemp þ HRD Salinity þ HRD

46741 46737 46736 46735 46741

MONTH (2008) Fig. 3. a) Mean daily river flow (grey bars), salinity (dashed line) and water temperature (grey line); and b) proportion of daily time spent by tagged M. colonorum in each river zone (LE, lower estuary; LME, lower middle estuary; UME, upper middle estuary; UE, upper estuary) during January to April 2008.

3.1. Distribution and environmental parameters Regression analysis demonstrated there was a positive correlation between mean location of Macquaria colonorum individuals within the estuary and salinity (slope ¼ 0.78, r2 ¼ 0.42, P < 0.0001). This was particularly evident during the warmer months (NovembereMarch) when individuals moved further upriver with increasing salinity, and during high river discharge events (e.g., November 2007 and February 2008) when fish migrated downstream in response to the reduction in salinity (Figs. 2a and 3). Mean fish location was positively correlated with water temperature (slope ¼ 0.49, r2 ¼ 0.65, P < 0.0001), particularly during spawning periods (JuneeAugust) when fish spent considerable time in the LE (Fig. 2b). Although statistically significant, the negative relationship between M. colonorum location and river flow, was weak (slope ¼
0.02, r2 ¼ 0.09, P < 0.01). Individuals were observed to only migrate downstream towards the river mouth in response to river flow events with mean daily flow rates >100 m3 s
1, with no other real patterns evident for the remainder of the study (Figs. 2c and 3). The non-autoregressive model with the best AIC score included all of the environmental variables tested (Table 2a). However, as expected a high level of autocorrelation was found in the model (DurbineWatson d ¼ 0.0923, P < 0.001, BreuscheGodfrey P < 0.001). This was confirmed in the best first order autoregressive model showing a 91% correlation between successive each fish-location time steps (Table 2b, 41 ¼ 0.906). The eleventh order autoregression model Distance from Sea w Salinity þ LowTemp þ HRD was found to be the most parsimonious of all the models tested (Table 2c, AIC ¼ 46735). Despite this high level of autocorrelation the model indicated that salinity, low water temperatures and high river discharge events all had a significant influence on fish location (Table 3c, P < 0.005). Accounting for the autocorrelation removed a number of variables present in the best fitting non-autoregressive model (Table 3a) including Temperature, Sex and LogRiverFlow. There was no significant correlation found between Salinity, LowTemp and HRD (all r < 0.05) which suggests that these

environmental conditions have separate influences on the distributions of Macquaria colonorum. 3.2. Temporal periodicity and patterns of detections Spectral analyses were performed on the 17 fish detected by the array for more than 365 days. Based on presence/absence of detection data, the periodograms of tagged female and male Macquaria colonorum were similar and indicated two main biological or behavioural rhythms - diel (females: median 23.9 h; males: 24.4 h) and tidal (12.4 h for both sexes) (Fig. 4). For all fish, there was an overriding diel behavioural rhythm, with a tidal rhythm of lesser

Table 3 Summary statistics for the most parsimonious models fit to the daily mean distance from sea that tagged M. colonorum spent using a) no autoregression (Non AR) b) first order autoregression (AR1) and c) eleventh order autoregression (AR11). AR 4’s are the autoregression parameters. Coefficients

Estimate

s.e.

t-value

P-value

a) Best Non AR Model: Distance From Sea w Salinity þ LowTemp þ HRD þ Temp þ Sex þ LogRiverFlow (Intercept) 18.647 0.454 41.080 <0.001 Salinity 0.160 0.013 12.207 <0.001 LowTemp
5.507 0.325 16.890 <0.001 HRD
3.427 0.721
4.753 <0.001 Temp 0.250 0.017 14.467 <0.001 Sex 0.207 0.139 1.493 0.136 LogRiverFlow 0.100 0.087 1.148 0.251 Durbin-Watson BreuscheGodfrey d ¼ 0.0923, P < 0.001 P < 0.001 b) Best AR1 Model: Distance From Sea w Salinity þ LowTemp (Intercept) 22.606 0.635 35.584 Salinity 0.149 0.028 5.280 LowTemp 0.222 0.140 1.578 0.966 AR 41 c) Best AR11 Model: Distance From Sea w Salinity (Intercept) 22.526 Salinity 0.177 LowTemp
0.310 HRD 0.792 AR 4’s 1.346,
0.465, 0.059,
0.033, 0.010, 0.010, 0.017,
0.019, 0.020,
0.025, 0.049

<0.001 <0.001 0.115

þ LowTemp þ HRD 0.645 34.884 <0.001 0.028 6.184 <0.001 0.113
2.748 <0.005 0.202 3.923 <0.001

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

65

a) 6 0.05 7

0.04 0.03

Depth (m)

Proportion of detections

0.06

8

0.02 0.01

0

2

4

6

8

10

12 14

16

18

20

22

Hour

b) Proportion of detections

0.6

flood ebb

0.5 0.4 0.3 0.2 0.1

LE

Fig. 4. Fourier periodicity analysis of presence/absence per hour (8192 continuous hours of detection) for tagged M. colonorum a) males and b) females.

strength being detected in 12 of 17 individuals. There was no evidence of any lunar behavioural (detection) cycles in any of the individuals examined. M. colonorum were detected predominantly during the night, followed by the morning and evening crepuscular periods (i.e., dawn and dusk), and less during the day (chi-squared: c2 ¼ 4305.1, d.f. ¼ 3, P < 0.0001). In particular, 65% of fish were ‘present’ between 2000 and 0500 h, with the least detections recorded between 0700 and 1700 h (26%) (Fig. 5a). Reflecting these diel patterns in detectability were the mean depths recorded by tagged M. colonorum. The shallowest swimming depths were between 0700 and 1700 h, before a crepuscular increase to maximum swimming depths between 1900 and 0400 h (Fig. 5a). In addition, M. colonorum were detected most often on the outgoing tide, with probability of detection consistently higher (>52%) for this tidal phase across all four estuarine zones (Fig. 5b). There were significant trends in relation to the arrival and departure of Macquaria colonorum migrating to and from the spawning grounds. The mean time of day that fish arrived in the defined spawning area was 01:54  03:01 (i.e., 1:54 am) (Rayleigh test: r ¼ 0.68, n ¼ 17, P < 0.001), with 71% of all mean arrival times between the hours of 21:00 and 05:00. The mean time that fish departed the spawning grounds was also during the early evening (21:37  03:36, Rayleigh test: r ¼ 0.88, n ¼ 15, P < 0.001), with the timing of 73% of the return trips between 19:00 and 01:00. The majority of tagged M. colonorum utilised the outgoing tide (69%) during their downstream spawning migration, arriving at the spawning grounds 5.0 h  1.48 after high tide (Rayleigh test: r ¼ 0.46, n ¼ 17, P ¼ < 0.05). In contrast, M. colonorum departed the spawning grounds 8.43  1.48 h after high tide (Rayleigh test:

LME UME River zone

UE

Fig. 5. a) Mean proportion of detections (grey bars) and mean depth (C) recorded by M. colonorum for each hour of the day, irrespective of river zone. b) Mean frequency of M. colonorum detections recorded during the incoming ‘flood’ and outgoing ‘ebb’ tidal phases for all river zones (LE, lower estuary; LME, lower middle estuary; UME, upper middle estuary; UE, upper estuary).

r ¼ 0.48, n ¼ 17, P < 0.05), with 65% of fish utilising the incoming tide. There were no significant trends detected between moon phase (day of the lunar month) and the timing of fish arrival to, and departure from, the spawning grounds (arrival: 16.83  5.9 days, P > 0.1; departure: 18.93 d  5.9 days, P > 0.1). 3.3. Spawning migration and behaviour All tagged Macquaria colonorum that were still being detected at the end of each spawning period migrated to the lower estuary on at least one occasion in 2008 and 2009 (Fig. 6). In general, all migrations and their subsequent return journeys back to their original resident sites (up to 30 km in distance) were rapid and completed within 12e24 h. Migration times for each individual included periods of ‘rest’, i.e., fish were often detected for several hours within the vicinity of ‘in between’ receivers (i.e., between resident sites and the spawning grounds). The frequency of these annual migrations ranged between 1 and 5 visits, with population medians identical between sexes, but not years (Table 4). For example, 80% of females and 74% of males visited the spawning grounds on at least one more occasion in 2008 than in 2009. Overall, the majority of both sexes spent less than 5 days within the spawning grounds during each visit. In contrast, two males and one female each made only one extended visit (between 88 and 112 days) to the spawning grounds in both years (Fig. 6b,d,l). The median number of days spent by males within the lower estuary during the spawning periods was significantly higher than females in 2008 (ANOVA: H1, 15 ¼ 4.34, P ¼ 0.03), but not in 2009

66

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

40

a)

b)

c)

30 20 10 1035952(F)

40

d)

1035954(M)

1039761(M)

e)

f)

30 20 10 1035955(M)

40

g)

1035957(U)

1035956(F)

h)

i)

Distance upstream (km)

30 20 10 1035960(U)

j)

40

1036285(M)

1036284(F)

k)

l)

30 20 10 1036286(F)

40

m)

1036288(F)

1036287(F)

n)

o)

30 20 10 1036289(F)

40

p)

1036290(M)

1039619(M)

q)

O D F A J AO D FA J A

30 20 10 1039758(M)

O D FA J AO D FA J A

1039759(M)

O D FA J AO D FA J A

Month (2007-9) Fig. 6. (aeq) Mean daily river location of tagged M. colonorum individuals at liberty between September 2007 and October 2009 (Transmitter number and sex noted in each plot).

(P > 0.1). There was no significant difference in mean number of days spent for each sex between years (females, P > 0.05; males, P > 0.1). Cumulative distributions of residency, based on time spent within the spawning grounds were also similar between sexes and among years (KeS test: P > 0.05, Table 4; Fig. 7). 4. Discussion 4.1. Effects of salinity, temperature and river flow All tagged Macquaria colonorum remained in the estuary throughout the study, however, there was substantial spatial and

temporal variation in utilisation of hydrographic habitats. Salinity, high river discharge events and water temperatures all had a significant effect on the distribution of M. colonorum. Similar responsive behaviour has been reported for other estuarineresident and dependent fishes in the southern hemisphere (e.g., spotted grunter, Pomadasys commersonnii, Childs et al., 2008a; tupong, Pseudaphritis urvillii, Crook et al., 2010; black bream, Acanthopagrus butcheri, Sakabe and Lyle, 2010) and northern hemisphere (e.g., brown trout, Salmo trutta, Svendsen et al., 2004; cownose ray, Rhinoptera bonasus, Collins et al., 2008; Carcharhinus leucas, Heupel and Simpendorfer, 2008). High individual site fidelity (Walsh et al., 2012b), combined with the relatively low

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

67

Table 4 Frequency and duration of visits by female and male M. colonorum, as well as probabilities of there being no differences in their cumulative residency time on the spawning grounds in 2008 and 2009. No. of visits (d) Median Females 2008 3 2009 2 Males 2008 3 2009 2 KS test (females vs males)

Visit duration (d)

Cumm. distribution

KS test

Range

Median

Range

Median

5the95th%

(2008 vs 2009)

1e5 1e3

4 5

2e82 2e89

15Jul 14Jul

04June17Aug 04June16Aug

D ¼ 0.014, P > 0.05

1e5 1e3

8 6

1e119 1e95 2008 2009

19Jul 20Jul D ¼ 0.12 D ¼ 0.11

31Maye26Aug 12June28Aug P > 0.1 P > 0.1

D ¼ 0.012 P > 0.1

spatial resolution of passive acoustic telemetry arrays (Heupel et al., 2006), meant that fish distribution changes in relation to subtle changes in environment were difficult to detect. Nevertheless, we were successful in modelling the relationship of environmental fluctuations as predictor variables for the location of tagged fish present within the estuary. Telemetry researchers should take note that other biotic and abiotic factors found to be significant predictors when using non-autoregressive models, were removed once autoregressive models were employed. Such results demonstrate the importance of accounting for autocorrelation in fitting models involving telemetry-based location data when examining spatio-environmental relationships in fish behaviour. The significant relationship found between salinity, high river discharge and fish location was evident during the February 2008 event (Fig. 3) resulting in the large-scale downstream movement of the of Macquaria colonorum to the lower estuary. This indicates, similar to that found in other estuarine-dependent fishes (e.g., Macquaria novemaculeata, Reinfelds et al., 2012) the significance of high episodic river discharge events, as opposed to smaller or gradual increases in river flows in triggering large-scale migrations. Individuals remained in the lower estuary for over a month until salinity levels increased back to more mesohaline conditions

(salinity range between 5 and 18), upon which fish returned to their original resident home ranges. Potentially, such movements may be an evolutionary adaptation aimed at reducing energy costs associated with osmoregulation. Osmoregulation has been found to require up to 10% of the total energy budget in some estuarine fishes (e.g., Killifish, Fundulus heteroclitus, Kidder et al., 2006), especially when moving from freshwater to sea water. Therefore M. colonorum migrating down river during a high river discharge event would allow individuals to remain in more mixed saline (brackish) water, thus alleviating the osmo- and thermo- regulatory stress associated with the more ‘fresh’ and cooler flood waters (Ortega et al., 2009). A similar downstream migratory response to increased freshwater flows has also been observed in a major prey species (i.e., school prawn, Metapenaeus macleayi). Indeed, 47% of M. colonorum dietary intake in the Shoalhaven River has been found to constitute M. macleayi (Williams, 1970). Known for their intolerance to low levels of salinity, large volumes of these ‘migrating’ penaeids are regularly caught by commercial fishers in the lower estuary of the Shoalhaven River (W. Ganderton pers. comm.) and other large tidal rivers in SE Australia after high rainfall events (Ruello, 1973; Glaister, 1978). Therefore, we postulate that freshwater inflows influenced the movement of M. colonorum through: (1) the direct effect of

a)

b)

c)

d)

0.6

Proportion of population

0.4

0.2

0.6 0.4

0.2

Jun

Jul

Aug

Month - 2008

Sep

Jun

Jul

Aug

Sep

Month - 2009

Fig. 7. Daily proportion of tagged female (grey bars e a, b) and male (black bars e c, d) M. colonorum present on the spawning grounds (within the first five upstream receiver stations) during the winter periods of 2008 (a, c) and 2009 (b, d).

68

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

salinity on the individual, thereby causing their movement to seek refuge to areas of more suitable environmental conditions; (2) and/ or the effect of salinity on the distribution of prey species that may be intolerant to environmental change. There was a significant positive relationship between the distribution of Macquaria colonorum and low water temperature throughout the study. In particular, large-scale downstream movements were detected in response to decreasing water temperatures. Fish also migrated downstream when water temperatures declined in association with periods of high freshwater inflows. Distinct seasonal water temperature patterns were observed throughout the study, with lower temperatures during the winter months (Junee August) and higher temperatures during the austral spring to autumn period. Mirroring these trends were the downstream spawning and upstream post-spawning migrations made by M. colonorum, respectively. This timing coincides precisely with the peaks in gonadosomatic indices of captured mature (>250 mm FL) male and female M. colonorum (Walsh et al., 2011). Moreover, decreasing water temperatures (to <15  C) have recently been associated with the initiation of downstream spawning migrations in the closely related Macquaria novemaculeata in the Shoalhaven River (Reinfelds et al., 2012). Harris (1986) attributes this behaviour as being the main environmental cue driving their spermatogenesis and ovary conditioning. Certainly it does appear that the spawning and post-spawning migrations made by tagged M. colonorum were in response to overall decreasing and increasing water temperatures in both years, with the emphasis being on the cooling and warming trend rather than any absolute temperature. 4.2. Diel and tidal activity Detection of tagged Macquaria colonorum followed rhythmic behavioural rhythmic patterns. In particular, fish detections were predominantly nocturnal for all individuals, across all locations, with the lowest detection rates recorded during the middle of the day. Diel behaviour patterns have been observed in many estuarinedependent fishes, with some fish more active at night (e.g., Acanthopagrus butcheri, Sakabe and Lyle, 2010; Argyrosomus japonicus, Taylor et al., 2006) while others exhibit increased movement during the day (e.g., tautog, Tautoga onitis, Arendt et al., 2001). Possible explanations for reduced detection rates in fish during the day time include: (1) tagged fish moved away from receiver detection range; or (2) detectability is reduced when fish retreat to areas of high structural heterogeneity, hence masking tag transmissions (Arendt et al., 2001; Hartill et al., 2003). Previous studies (Williams, 1970; McCarraher and McKenzie, 1986), indicate that M. colonorum, similar to other percichthyids (e.g., Macquaria novemaculeata, Harris, 1983; trout cod, Maccullochella macquariensis, Thiem et al., 2008) utilise structural components of habitats (e.g., submerged trees, boulders, undercut banks etc.) during the day. Moreover, tagged fish were detected at shallower depths during the middle of the day, suggesting that diel behaviour in depth selection may be based on the use of habitat closer to the shoreline, rather than directed movement up and down in the water column. While further fine scale continuous tracking is needed to test this hypothesis, we suggest that M. colonorum likely occupy the deeper sections of main channel at night, with fish retreating to the adjacent shallower, structurally-complex habitats during the day. There was strong evidence of Macquaria colonorum exploiting particular tides. Tidal flows in the Shoalhaven River reach velocities up to 1 m s
1 (Miller et al., 2006), and therefore it is not surprising that tidal biorhythms were observed in 71% of all fish assessed. Many estuarine-dependent fishes (e.g., Acanthopagrus butcheri, Sakabe and Lyle, 2010; bonefish, Albula vulpes, Humston et al., 2005; white perch, Morone americana, McGrath and Austin, 2009) venture

into shallow habitats, such as tidal flats, areas of submerged macrophytes, and complex structure during inundation on high tides, while retreating into deep channels during low tides. A similar cyclical behaviour to utilise particular habitats may also be occurring for M. colonorum in response to the dynamics of tidal flow, possibly for foraging or predator avoidance (Almeida, 1996; Humston et al., 2005). M. colonorum exhibits high site fidelity, a behavioural trait that promotes the increased knowledge of locations with high prey densities and shelter areas (McGrath and Austin, 2009). Although, there have been no diel or tidal based dietary studies done on M. colonorum, previous work has shown that penaeids and other decapods comprise a large proportion of their diet (Williams, 1970; Howell et al., 2004). Diel cycles of emergence and movement are common in burrowing decapods, with most species burrowing during the day and emerging at night (Vance, 1992; Honculada Primavera and Lebata, 1995). Moreover, penaeids have been shown to have strong tidal-based patterns of behaviour (Dall et al., 1990). Therefore, it is possible that biorhythmic patterns displayed by highly site-attached species such as M. colonorum are ‘mimicking’ the cyclical behaviour patterns exhibited by their prey. 4.3. Spawning behaviour Spawning migrations by Macquaria colonorum individuals (up to 30 km in distance) were completed within 24 h, with most fish stopping for extended periods during their downstream journey. Considerable osmoregulatory effort is required to accommodate abrupt increases/decreases in salinity levels (Langdon, 1987; Kidder et al., 2006). Therefore, it is likely that ‘resting’ fish may be acclimatising to the higher salinity regimes (salinities > 30) found in the lower estuary. All but three fish made multiple visits (up to 5) to one area within the lower estuary during the austral winter periods of both 2008 and 2009. The frequency of these visits was similar between sexes and years. However, the visitation rate between years for both sexes was higher in 2008 than in 2009 and this may be related to river flow. Previous studies have shown that Macquaria novemaculeata require increased river flows to stimulate and facilitate their spawning migrations (Harris, 1983; Reinfelds et al., 2012). Moreover, Gillson et al. (2011) found the highest correlations between fisheries catch per unit effort (CPUE) and freshwater flow coincided with the spawning periods of several Australian estuarine-dependent fishes. The mean monthly total river discharge in the Shoalhaven River during the winter period of 2008 was more than twice the amount for the same period in 2009, with high freshwater inflow events (>100 m3 s
1) occurring during August and early September 2008. While all M. colonorum were capable of carrying out spawning migrations in each of the two years, ‘freshes’ in the river at critical spawning periods, especially during times of low flows, may trigger increased visitation frequency. Male and female Macquaria colonorum migrated to the spawning grounds of the lower estuary in synchrony. Males remained on the spawning grounds for as long as females in both years of the study, with significant longer residency recorded in 2008. Similar observations have been made for the American percichthyid striped bass Morone saxatilus (Carmichael et al., 1998; Douglas et al., 2009). Carmichael et al. (1998) hypothesised that females and males migrated according to their maturity, while Douglas et al. (2009) postulated that the exhaustion of reproductive materials may be a cue for M. saxatilus to leave the spawning grounds of the Miramichi River, resulting in the immediate departure of postspawning females, but a prolonged stay of males. Spent female fish and males continuing to produce large quantities of milt, have regularly been observed in both M. colonorum and the Macquaria novemaculeata populations in the Shoalhaven River during postspawning periods (pers. obs.). From an evolutionary perspective,

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

69

Fig. 8. Conceptual model describing the environmental influences on the spatial ecology and spawning behaviour of M. colonorum.

this ensures there are always sufficient ripe males on the spawning grounds to breed with any ‘late’ arriving females. Similar to their resident biorhythmic behaviour, there were significant diel and tidal patterns detected in tagged Macquaria colonorum during annual spawning migrations. M. colonorum predominantly travelled at night, with fish migrating downstream on the ebb tide and arriving at the spawning grounds just prior to the low tide. On return migrations, the majority of fish departed the spawning grounds at night immediately post low tide, travelling back to their resident summer locations on the incoming flood tides. Selective use of tides is a migration strategy typically used by estuarine-dependent fishes (e.g., Morone saxatilus, Windgate and Secor, 2007; Pomadasys commersonnii, Childs et al., 2008b; Pseudaphritis urvilli, Crook et al., 2010). This has been hypothesised to conserve energy by allowing individuals to move via tidal transport rather than expend energy swimming, as well as a means of prey utilisation (Almeida, 1996; Smith and Smith, 1997; Heupel and Simpendorfer, 2008). In the Shoalhaven River, the largest of the annual spring ebb tides (peak water velocities) occur during the night in the austral winter (Miller et al., 2006). This period coincided with peak migration activity of tagged M. colonorum and with peak abundances of eggs sampled from ichtyoplankton surveys done within the spawning grounds in 2009 (D. van der Meulen unpl. data). We suggest that M. colonorum has adapted this biorhythmic behaviour, not only for residency and estuarine movements, but also as part of its reproductive strategy.

Secondly, combined with an overall trend of increasing atmospheric and water temperatures, climate change is likely to have ramifications for estuarine-resident species with specific spawning cues, together with critical temperature requirements required for optimal egg and larval survival (Beckman, 1999). The higher frequency of spawning visits made by Macquaria colonorum individuals during ’wetter’ winter/spring years, together with evidence of episodic recruitment found amongst populations (Walsh et al., 2010), indicates similar to that found in Macquaria novemaculeata (Harris, 1983; Reinfelds et al., 2012), the importance of increases in river flow as a stimulatory cue for their spawning migrations and any subsequent recruitment success. This has implications, particularly in regulated rivers such as the Shoalhaven and elsewhere across SE Australia. Reinfelds et al. (2010) found that during periods of water supply augmentation in times of severe drought, floods equivalent to the 1.1-year annual maximum event (similar to that registered in winter/spring 2008) can be fully captured by Tallowa Dam. Any large changes in the magnitude and frequency of drought climatic cycles affecting freshwater inflows could suppress or remove migratory stimulatory cues and exacerbate the already highly variable inter-annual recruitment patterns found in M. colonorum and other estuarine fish and invertebrate species (Robins et al., 2005; Ferguson et al., 2008; Walsh et al., 2010). Appropriate water management strategies must therefore be considered when maintaining the reproductive potential of estuarine fishes in regulated river systems.

4.4. Potential impacts of climate change

5. Conclusions

Macquaria colonorum has high site fidelity with specific habitat and spawning requirements. The knowledge of, and the ability to model with confidence, the effects of abiotic and biotic variables on their distribution, movement and spawning behaviour is critical to the management of the species. This is particularly relevant considering future changes in environmental conditions, combined with increasing anthropogenic stressors on estuaries (Gillanders et al., 2011). We propose two likely scenarios to occur with a changing climate, for M. colonorum populations in SE Australia. Firstly, a predicted increase in the timing and magnitude of freshwater inflows (Nicholls, 2004; Cai and Cowan, 2008) will likely increase variability in estuary salinity regimes, altering preferred habitats for predator and/or their prey species that prefer brackish waters (Gillson, 2011).

Based on the use of acoustic tags and receiver arrays the current understanding of the linkages between resident Macquaria colonorum distributions, movements, spawning and environmental processes is summarised in the conceptual model in Fig. 8. Overall, salinity, water temperature and river flow all had a significant influence on the estuarine distribution of M. colonorum, with the likely drivers of large scale non-spawning and spawning movements being flood events and seasonal cooling and warming trends in water temperatures, respectively. Location, visitation rates and occupation of the spawning grounds were similar among sexes. Behavioural rhythms were consistent over broad temporal and spatial scales, with movement between depth-related habitats that were significantly influenced by diel and tidal state. We postulate

70

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71

that the distribution and movement of M. colonorum is not only an innate response to environmental fluctuations, but is also suggestive of complex underlying behavioural mechanisms, such as reproductive behaviour and predatoreprey interactions. A continuous tracking component within this acoustic array, that accurately pinpoints the position of M. colonorum over a full range of spatial habitats, will provide further insight into the finer scale (high resolution) movements, and the environmental influences that act upon them. Acknowledgements Southern Bass and Basin Lure and Fly fishing clubs, as well as Steve Starling, are gratefully acknowledged for assisting with the capture of fish. We thank Jim Craig for providing valuable assistance with data collation and Amber Childs for statistical advice and helpful discussions. Gavin Butler, Matt Taylor and anonymous referees greatly enhanced the final manuscript through their constructive criticism. This research was funded by NSW Department of Primary Industries, NSW Office of Water Flow Response Monitoring Program, NSW Saltwater Recreational Fishing Trust and University of Wollongong and was done under a Section 37 permit (Fisheries Management Act 1994) issued by NSW Department of Primary Industries. This study complied within the requirements of the Animal Research Act 1985 (ACEC # 06/03). References Aarts, G., MacKenzie, M., McConnell, B., Fedak, M., Matthiopoulos, J., 2008. Estimating space-use and habitat preference from wildlife telemetry data. Ecography 31, 140e160. Almeida, P.R., 1996. Estuarine movement patterns of adult thin-lipped grey mullet, Liza ramada (Risso) (Pisces, Mugilidae), observed by ultrasonic tracking. Journal of Experimental Marine Biology and Ecology 202, 137e150. Arendt, M.D., Lucy, J.A., Munro, T.A., 2001. Seasonal occurrence and site-utilisation patterns of adult tautog, Tautoga onitis (Labridae), at manmade and natural structures in lower Chesapeake Bay. Fisheries Bulletin 99, 519e527. Batschelet, E., 1981. Circular Statistics in Biology. Academic Press, London, 371 pp. Beckman, D.C., 1999. Effect of salinity on hatchability and early development of estuary perch (Macquaria colonorum). BSc. Honours thesis. Deakin University, Warrnambool, Victoria. Beitinger, T.L., Fitzpatrick, L.C., 1979. Physiological and ecological correlates of preferred temperature in fish. Integrative and Comparative Biology 19, 319e329. Bennett, B.A., Branch, G.M., 1990. Relationships between production and consumption of prey species by resident fish in the Bot, a cool temperate South African estuary. Estuarine, Coastal and Shelf Science 39, 139e155. Blaber, S.J.M., Blaber, T.G., 1980. Factors affecting the distribution of juvenile estuarine and inshore fish. Journal of Fish Biology 17, 143e162. Cai, W., Cowan, T., 2008. Evidence of impacts from rising temperatures on inflows to the MurrayeDarling Basin. Geophysical Research Letters 35, L07701. http:// dx.doi.org/10.1029/2008 GL033390. Carmichael, J.T., Haeseker, S.L., Hightower, J.E., 1998. Spawning migration of telemetered striped bass in the Roanoke River, North Carolina. Transactions of the American Fisheries Society 97, 320e343. Childs, A.R., Cowley, P.D., Naesje, T.F., Booth, A.J., Potts, W.M., Thorstad, E.B., Økland, F., 2008a. Do environmental factors influence the movement of estuarine fish? A case study using acoustic telemetry. Estuarine, Coastal and Shelf Science 78, 227e236. Childs, A.R., Cowley, P.D., Naesje, T.F., Booth, A.J., Potts, W.M., Thorstad, E.B., Økland, F., 2008b. Estuarine use by spotted grunter Pomadasys commersonnii in a South African estuary, as determined by acoustic telemetry. African Journal of Marine Science 30, 123e132. Clements, S., Jepsen, D., Karnowski, M., 2005. Optimization of an acoustic telemetry array for detecting transmitter-implanted fish. North American Journal of Fisheries Management 25, 429e436. Collins, A.B., Heupel, M.R., Simpfendorfer, C.A., 2008. Spatial distribution and longterm movement patterns of cownose rays Rhinoptera bonasus within an estuarine river. Estuaries and Coasts 31, 1174e1178. Cowley, P.D., Kerwath, S.E., Childs, A.R., Thorstad, E.B., Økland, F., Naesje, T.F., 2008. Estuarine habitat use by juvenile dusky kob Argyrosomus japonicus (Sciaenidae), with implications for management. African Journal of Marine Science 30, 247e253. Crook, D.A., Koster, W.M., Macdonald, J.I., Nicol, S.J., Belcher, C.A., Dawson, D.R., O’Mahony, D.J., Lovett, D., Walker, A., Bannam, L., 2010. Catadromous migrations by female tupong (Pseudaphritis urvillii) in coastal streams in Victoria, Australia. Marine and Freshwater Research 61, 474e483.

Dall, W., Hill, B.J., Rothlisburg, P.C., Sharples, D.J., 1990. The biology of the Penaeidae. Advances in Marine Biology 27, 1e489. De Solla, S.R., Bonduriansky, R., Brooks, R.J., 1999. Eliminating autocorrelation reduces biological relevance of home range estimates. Journal of Animal Ecology 68, 221e234. Douglas, S.G., Chaput, G., Hayward, J., Sheasgreen, J., 2009. Prespawning, spawning, and postspawning behaviour of striped bass in the Miramichi River. Transactions of the American Fisheries Society 138, 121e134. Elliott, M., Whitfield, A.K., Potter, I.C., Blaber, S.J.M., Cyrus, D.P., Nordlie, F.G., Harrison, T.D., 2007. The guild approach to categorizing estuarine fish assemblages: a global review. Fish and Fisheries 8, 241e268. Ferguson, G.J., Ward, T.M., Geddes, M.C., 2008. Do recent age structures and historical catches of mulloway, Argyrosomus japonicus (Sciaenidae), reflect freshwater inflows in the remnant estuary of the Murray River, South Australia? Aquatic Living Resources 21, 145e152. Gehrke, P.C., Astles, K.L., Harris, J.H., 1999. Within catchment effects of flow alteration on fish assemblages of the HawkesburyeNepean River system, Australia. Regulated Rivers: Research and Management 15, 181e198. Gillanders, B.M., Elsdon, T.S., Halliday, I.A., Jenkins, G.P., Robins, J.B., Valesini, F.J., 2011. Potential effects of climate change on Australian estuaries and fish utilising estuaries: a review. Marine and Freshwater Research 62, 1115e1131. Gillson, J., 2011. Freshwater flow and fisheries production in estuarine and coastal systems: where a drop of rain is not lost. Reviews in Fisheries Science 19, 168e186. Glaister, J.P., 1978. Impact of river discharge on distribution and production of the school prawn Metapenaeus macleayi (Haswell) (Crustacea: Penaeidae) in the Clarence River region, northern New South Wales. Australian Journal of Marine and Freshwater Research 29, 311e323. Grafen, A., Hails, R., 2002. Modern Statistics for the Life Sciences. Oxford University Press, New York, 351 pp. Harris, J.H., 1983. The Australian bass, Macquaria novemaculeata. PhD. thesis. University of New South Wales, NSW, Australia. Harris, J.H., 1986. Reproduction of the Australian bass, Macquaria novemaculeata (Perciformes: Percichthyidae) in the Sydney Basin. Australian Journal of Marine and Freshwater Research 37, 209e235. Harris, J.H., Rowland, S.J., 1996. Australian cods and basses. In: McDowall, R. (Ed.), Freshwater Fishes of South-eastern Australia. Reed Books, Sydney, pp. 150e163. Harrison, T.D., Whitfield, A.K., 2006. Temperature and salinity as primary determinants influencing the biogeography of fishes in South African estuaries. Estuarine, Coastal and Shelf Science 66, 335e345. Hartill, B.W., Morrison, M.A., Smith, M.D., Boubee, J., Parsons, D.M., 2003. Diurnal and tidal movements of snapper (Pagrus auratus, Sparidae) in an estuarine environment. Marine and Freshwater Research 54, 931e940. Heupel, M.R., Simpendorfer, C.A., 2008. Movement and distribution of young bull sharks Carcharhinus leucas in a variable estuarine environment. Aquatic Biology 1, 277e289. Heupel, M.R., Semmens, J.M., Hobday, A.J., 2006. Automated acoustic tracking of aquatic animals: scales, design and deployment of listening station arrays. Marine and Freshwater Research 57, 1e13. Hindell, J.S., Jenkins, G.P., Womersley, B., 2008. Habitat utilization and movement of black bream Acanthopagrus butcheri (Sparidae) in an Australian estuary. Marine Ecology Progress Series 366, 219e229. Honculada Primavera, J., Lebata, J., 1995. Diel activity in Metapenaeus and Penaeus juveniles. Hydrobiologia 295, 295e302. Howell, T., Laurenson, L.J., Myers, J.H., Jones, P.L., 2004. Spatial, temporal and sizeclass variation in the diet of estuary perch (Macquaria colonorum) in the Hopkins River, Victoria, Australia. Hydrobiologia 515, 29e37. Humston, R., Ault, J.S., Larkin, M.F., Luo, J., 2005. Movements and site fidelity of the bonefish Albula vulpes in northern Florida keys determined by acoustic telemetry. Marine Ecology Progress Series 291, 237e248. Ihaka, R., Gentleman, R., 1996. R: a language for data analysis and graphics. Journal of Computational and Graphical Statistics 5, 299e314. Jenkins, G.P., Conron, S.D., Morison, A.K., 2010. Highly variable recruitment in an estuarine fish is determined by salinity stratification and freshwater flow: implications of a changing climate. Marine Ecology Progress Series 417, 249e261. Jonsson, N., 1991. Influence of water flow, water temperature and light on fish migration in rivers. Nordic Journal of Freshwater Research 66, 20e35. Kidder, G.W., Petersen, C.W., Preston, R.L., 2006. Energetics of osmoregulation: II. Water flux and osmoregulatory work in the euryhaline fish, Fundulus heteroclitus. Journal of Experimental Zoology 305, 318e327. Kimmerer, W.J., 2002. Effects of freshwater flow on abundance of estuarine organisms: physical effects or trophic linkages? Marine Ecology Progress Series 243, 39e55. Langdon, J.S., 1987. Active osmoregulation in the Australian bass, Macquaria novemaculeata (Steindachner), and the golden perch, Macquaria ambigua (Richardson) (Percichthyidae). Australian Journal of Marine and Freshwater Research 38, 771e776. Loneragan, N.R., Bunn, S.E., 1999. River flows and estuarine ecosystems: implications for coastal fisheries from a review and a case study of the Logan River, southeast Queensland. Austral Ecology 24, 431e440. Marshall, S., Elliott, M., 1998. Environmental influences on the fish assemblage of the Humber Estuary, UK. Estuarine, Coastal and Shelf Science 46, 175e184. McCarraher, D.B., McKenzie, J.A., 1986. Observations on the Distribution, Growth and Diet of Estuary Perch (Macquaria colonorum) in Victorian Waters. Arthur Rylah Institute for Environmental Research, Technical Report Series No. 42, 21 pp.

C.T. Walsh et al. / Estuarine, Coastal and Shelf Science 118 (2013) 60e71 McDowall, R.M., 1988. Diadromy in Fishes: Migrations between Freshwater and Marine Environments. Croom Helm, London, 308 pp. McGrath, P., Austin, H.A., 2009. Site fidelity, home range, and tidal movements of white perch during the summer in two small tributaries of the York River, Virginia. Transactions of the American Fisheries Society 138, 966e974. Miller, B.M., Hawker, K.M., Badenhop, A.M., 2006. Environmental Flow Modelling of Salinity Structure in the Shoalhaven Estuary. The University of New South Wales School of Civil and Environmental Engineering, Water Research Laborotory, WRL Technical Report 2006/23, 93 pp. Morgan, E., 2001. The moon and life on earth. Earth Moon Planets 85-86, 279e290. Nicholls, N., 2004. The changing nature of Australian droughts. Climatic Change 63, 323e336. Ortega, L.A., Heupel, M.R., Beynen, P.V., Motta, P.J., 2009. Movement patterns and water quality preferences of juvenile bull sharks (Carcharhinus leucas) in a Florida estuary. Environmental Biology of Fishes 84, 361e373. Pittman, S.J., McAlpine, C.A., 2001. Movements of marine fish and decapod crustaceans: process, theory and application. Advances in Marine Biology 44, 205e294. Pusey, B.J., Kennard, M.J., Arthington, A.H., 2004. Freshwater Fishes of North-eastern Australia. CSIRO Publishing, Collingwood, Victoria, 400 pp. Reinfelds, I.V., Lincoln-Smith, M., Haeusler, T., Ryan, D., Growns, I.O., 2010. Hydraulic assessment of environmental flow regimes to facilitate fish passage through natural riffles: Shoalhaven River below Tallowa Dam, New South Wales, Australia. River Research and Applications 26, 589e604. Reinfelds, I.V., Walsh, C.T., van der Meulen, D.E., Growns, I.O., Gray, C.A., 2012. Magnitude, frequency and duration of instream flows to stimulate and facilitate catadromous fish migrations: Australian bass (Macquaria novemaculeata Perciformes, Percichthyidae). Published online in Wiley Online Library. River Research and Applications. http://dx.doi.org/10.1002/rra.1611. wileyonline library.com. Reyier, E.A., Lowers, R.H., Scheidt, D.M., Adams, D.H., 2011. Movement patterns of adult red drum, Sciaenops ocellatus, in shallow Florida lagoons as inferred through autonomous acoustic telemetry. Environmental Biology of Fishes 90, 343e360. Robins, J.B., Halliday, I.A., Staunton-Smith, J., Mayer, D.G., Sellin, M.J., 2005. Freshwater-flow requirements of estuarine fisheries in tropical Australia: a review of the state of knowledge and application of a suggested approach. Marine and Freshwater Research 56, 343e360. Roy, P.S., Williams, R.J., Jones, A.R., Yassini, R., Gibbs, P.J., Coates, B., West, R.J., Scanes, P.R., Hudson, J.P., Nichol, S., 2001. Structure and function of south-east Australian estuaries. Estuarine, Coastal and Shelf Science 53, 351e384. Ruello, N.V., 1973. Influence of rainfall on the distribution and abundance of the school prawn Metapenaeus macleayi (Haswell) in the Hunter River region, Australia. Marine Biology 23, 221e228. Sakabe, R., Lyle, J.M., 2010. The influence of tidal cycles and freshwater inflow on the distribution and movement of an estuarine resident fish Acanthopagrus butcheri. Journal of Fish Biology 77, 643e660. Selleslagh, J., Amara, R., 2008. Environmental factors structuring fish composition and assemblages in a small macrotidal estuary (eastern English Channel). Estuarine, Coastal and Shelf Science 79, 507e517.

71

Smith, R.J.F., 1984. The Control of Fish Migration. Springer-Verlag, Berlin, 240 pp. Smith, I.P., Smith, G.W., 1997. Tidal and diel timing of river entry by adult Atlantic salmon returning to the Aberdeenshire Dee, Scotland. Journal of Fish Biology 50, 463e474. Sokahl, R.R., Rohlf, F.J., 1981. Biometry: The Principles and Practice of Statistics in Biological Research, second ed. Freeman, San Francisco, 776 pp. StatSoft, Inc, 2010. Electronic Statistics Textbook. StatSoft, Tulsa, OK. http://www. statsoft.com/textbook/. Svendsen, J.C., Koed, A., Aarestrup, K., 2004. Factors influencing the spawning migration of female anadromous brown trout. Journal of Fish Biology 64, 528e540. Swihart, R.K., Slade, N.A., 1985. Testing for independence of observations in animal movements. Ecology 4, 1176e1184. Taylor, M.W., Laffan, S.D., Fielder, S.D., Suthers, I.M., 2006. Key habitat and home range of mulloway Argyrosomous japonicus in a south-east Australian estuary: finding the estuarine niche to optimise stocking. Marine Ecology Progress Series 328, 237e247. Thiem, J.D., Ebner, B.C., Broadhurst, B.T., 2008. Diel activity of the endangered tout cod (Maccullochella macquariensis) in the Murrumbidgee River. Proceedings of the Linnean Society of New South Wales 129, 167e173. Vance, D.J., 1992. Activity patterns of juvenile penaeid prawns in response to artificial tidal and day-night cycles: a comparison of three species. Marine Ecology Progress Series 87, 201e213. Walsh, C.T., Gray, C.A., West, R.J., van der Meulen, D.E., Williams, L.F.G., 2010. Growth, episodic recruitment and age truncation in populations of a catadromous percichthyid, Macquaria colonorum. Marine and Freshwater Research 61, 397e407. Walsh, C.T., Gray, C.A., West, R.J., Williams, L.F.G., 2011. Reproductive biology and spawning strategy of a catadromous percichthyid, Macquaria colonorum (Günther, 1863). Environmental Biology of Fishes 91, 471e486. Walsh, C.T., Reinfelds, I.V., West, R.J., Gray, C.A., van der Meulen, D.E., 2012a. Distribution and movement of catadromous fish: design and implementation of a freshwater-estuarine acoustic telemetry array. In: Mckenzie, J., Phelps, Q., Kopf, R., Mesa, M., Parsons, B., Seitz, A. (Eds.), Advances in Fish Tagging and Marking Technology. American Fisheries Society Symposium 76, Bethesda, Maryland, pp. 251e264. Walsh, C.T., Reinfelds, I.V., Gray, C.A., West, R.J., van der Meulen, D.E., Craig, J.R., 2012b. Seasonal residency and movement patterns of two co-occurring catadromous percichthyids within a south-eastern Australian river. Ecology of Freshwater Fish 21, 145e159. Whitfield, A.K., 1994. Abundance of 0þ juvenile marine fishes in the lower reaches of three South African estuaries with differing freshwater inputs. Marine Ecology Progress Series 105, 257e267. Williams, N.J., 1970. A comparison of the two species of the genus Percalates Ramsey and Ogilby (Percomorphi: Macquariidae) and their taxonomy. NSW State Fisheries Research Bulletin 11, 60. Windgate, R.L., Secor, D.H., 2007. Intercept telemetry of the Hudson River striped bass resident contingent: migration and homing patterns. Transactions of the American Fisheries Society 136, 95e104.