Download Abiotic and biotic factors influencing the assemblage of tadpoles

Abiotic and biotic factors influencing the assemblage of
tadpoles and adult anurans in coastal wallum habitats
of eastern Australia
Clay Alan Simpkins, BSc.
Submitted in fulfilment of the requirements for the degree of
Master of Philosophy
Environmental Futures Centre, Griffith School of Environment, Griffith University,
Gold Coast, Queensland, Australia
Submitted November 2012
Statement of originality
This work has not previously been submitted for a degree or diploma at any university.
To the best of my knowledge, the thesis contains no material previously published or
written by another person except where due reference and acknowledgement is made in
the thesis itself.
Clay Alan Simpkins
ii
16th November 2012
Abstract
The emergence of the global amphibian crisis has seen the extinction of 122 species
worldwide, with 18.8% of Australia’s 213 amphibian species being threatened. Despite
these declines, little is known about the biology and ecology of certain Australian
threatened species. Hence, successful conservation and management of threatened
amphibian species cannot be fully realised.
Several environmental variables may influence amphibian adult or tadpole assemblages.
These variables include, but are not limited to, water chemistry factors (i.e. pH, salinity,
turbidity), predation, competition, hydroperiod and water flow. These variables will
influence individual species differently, with each species displaying differences in
tolerance to these specific variables.
The coastal wallum vegetation along the eastern coast of Australia is the primary habitat
for four specialist frog species (Litoria olongburensis, Litoria freycineti, Litoria
cooloolensis and Crinia tinnula) that are listed as Vulnerable under the IUCN Red List.
All species are referred to as ‘acid’ frogs due to their association with low pH waters.
‘Acid’ frog populations within protected areas are believed to be stable. However,
populations of ‘acid’ frogs occurring outside of protected areas are at risk from ongoing
habitat loss and fragmentation. It is therefore vital that conservation managers know
which environmental factors influence ‘acid’ frogs to ensure these environmental
variables remain constant and populations remain stable. Furthermore, it is imperative
to determine if these environmental variables are the same within anthropogenic
waterbodies and if ‘acid’ frogs utilise anthropogenic waterbodies. This knowledge
would assist in the future prioritisation of waterbodies for conservation. However, the
factors influencing ‘acid’ frog species tadpole and adult relative abundance and
occupancy within protected and non-protected wallum heathland waterbodies have not
been reported.
Therefore, this thesis aims to determine what environmental variables influence these
assemblages of ‘acid’ and ‘non-acid’ frog species within and around wallum vegetation
of eastern Australia, in both natural and anthropogenic waterbodies.
iii
Overall, five tadpole and 14 adult amphibian species were found within surveyed
wallum heathland. Several environmental variables influenced the relative abundance
and occupancy of L. olongburensis tadpoles and adults. For tadpoles, these variables
included pH, water depth and turbidity while variables for adults included pH, water
depth, salinity and sedge cover. Environmental variables influencing C. tinnula tadpole
occupancy included predatory fish, water depth and turbidity. Several environmental
variables influenced adults of competitive species such as L. fallax, indicating that this
species is a generalist within the surveyed environment.
Water chemistry variables and the adult amphibian assemblage differed between natural
and anthropogenic/compensatory waterbodies. The specialist ‘acid’ frog species had
higher relative abundance and reproduced predominantly within natural waterbodies.
These patterns are explained by the ideal environmental variables for these species in
these natural habitats. The lower relative abundance of generalist ‘non-acid’ frog
species in natural waterbodies could be explained by their intolerance to environmental
variables, such as low pH. It was therefore possible to differentiate between ‘acid’ frog
and ‘non-acid’ frog assemblages in waterbodies using multivariate analyses.
The presence of predatory fish did not influence the relative abundance of L.
olongburensis tadpoles or adults. However, the relative abundance of predatory fish was
either low or absent in waterbodies where L. olongburensis occurred. Additionally,
exotic fish have been proposed as influencing the amphibian assemblage more than
other native predatory species. However, predation experiments completed in this study
showed that native predators had higher or equal predation rates for tadpoles of L.
olongburensis, Limnodynastes peronii and Litoria fallax.
This thesis demonstrates that several environmental variables need to be considered
when conservation of ‘acid’ frog species (primarily C. tinnula and L. olongburensis) is
undertaken. However, if conservation of all amphibian assemblages within and around
wallum heathland areas is the objective, then both anthropogenic and natural
waterbodies should be conserved.
iv
Table of Contents
Abstract ……………………....................……………………………………….…...
Table of Contents …………....................…………………………………………….
List of Figures …………....................……………………………………………......
List of Tables …………....................…………………………………………….......
Acknowledgements …………....................………………………………..…………
Chapter 1 - Introduction …………………...………………………………………...
1.1. The Importance of Amphibians .…………………..………………..……….....
1.2. Assemblages and Communities ………………………………………………..
1.3. Factors Influencing Tadpoles and Adult Amphibian Assemblages…………….
1.3.1. The Adult Assemblage ……………………………………………………...
1.3.2. Water Quantity and Chemistry ……………………………………………..
1.3.2.1. Water Quantity / Hydroperiod ………….......…………………………...
1.3.2.2. Water Chemistry ………………………………………………………...
1.3.2.2.1. Water pH ……………………………………………………………...
1.3.2.2.2. Natural Organic Acids (NOA) ………………………………………..
1.3.2.2.3. Salinity ………………………………………………………………..
1.3.2.2.4. Turbidity and Eutrophication …………………………………………
1.3.2.2.5. Dissolved Oxygen …………………………………………………….
1.3.2.2.6. Water Temperature ……………………………………………………
1.3.3. Competition …………………………………………………………………
1.3.4. Predation ……………………………………………………………………
1.4. Study Area and Study Species …………………………………………………
1.4.1. Study Area ………………………………………………………………..
1.4.2. Study Fauna ………………………………………………………………
1.4.2.1. Litoria olongburensis ……………………………………………………
1.4.2.2. Crinia tinnula ……………………………………………………………
1.5. Study Aims …………………………………………………………………..
1.6. References ……………………………………………………………………...
Chapter 2 - Environmental variables associated with the distribution and occupancy
of tadpoles in naturally acidic, oligotrophic waterbodies………………………..
2.1. Abstract ………………………………………………………………………...
2.2. Introduction …………………………………………………………………….
2.3. Methods ………………………………………………………………………...
2.4. Results ………………………………………………………………………….
2.5. Discussion ……………………………………………………………………...
2.6. References …………………………………………………………………..….
Chapter 3 - Battling habitat loss: Suitability of anthropogenic waterbodies for
amphibians associated with naturally acidic, oligotrophic environments……….
3.1. Abstract ………………………………………………………………………...
3.2. Introduction …………………………………………………………………….
3.3. Methods ………………………………………………………………………...
3.4. Results ………………………………………………………………………….
v
iii
v
vii
x
xii
1
1
2
3
3
5
5
6
7
8
8
9
10
10
10
12
15
15
16
19
21
24
25
39
39
40
42
49
57
63
69
69
70
72
77
3.5. Discussion ……………………………………………………………………...
3.6. References …………………………………………………………………..….
Chapter 4 - Compensatory ponds provide poor habitats for the conservation of frogs
associated with naturally oligotrophic, acidic environments……………………
4.1. Abstract …..……………………………………………………..……………...
4.2. Introduction …………………………………………………………………….
4.3. Methods ………………………………………………………………………...
4.4. Results ………………………………………………………………………….
4.5. Discussion ……………………………………………………………………...
4.6. References …………………………………………………………………..….
Chapter 5 - Comparison of predation rates between the introduced mosquito fish
(Gambusia holbrooki) and native aquatic predators on L. olongburensis,
L. fallax and Limnodynastes peronii tadpoles……………………………………
5.1. Abstract ………………………………………………………………………...
5.2. Introduction …………………………………………………………………….
5.3. Methods ………………………………………………………………………...
5.4. Results ………………………………………………………………………….
5.5. Discussion ……………………………………………………………………...
5.6. References …………………………………………………………………..….
Chapter 6 - General Conclusions ................................................................................
6.1 Chapter Overviews ...............................................................................................
6.1.1 Chapter 2 – Variable influencing wallum heathland tadpole assemblages......
6.1.2 Chapter 3 – Usage of anthropogenic waterbodies and variables influencing
adult and amphibian assemblages. ..................................................................
6.1.3 Chapter 4 - Compensatory pond usage by wallum heathland amphibians
and variables influencing adult amphibian assemblages .......................................
6.1.4 Chapter 5 – Predation experiments with G. holbrooki and natural
predators ..........................................................................................................
6.2. Management Outcomes ......................................................................................
6.3. Future Priorities for Research .............................................................................
6.4. References .......................................................................
Chapter 7 - Appendices: Publications on ‘acid’ frogs published during candidature...
Appendix 1: Long-range movement in the rare Cooloola sedgefrog Litoria
cooloolensis………………………………………………………………...………..
vi
85
91
96
96
97
98
102
108
116
120
120
121
122
126
128
135
140
140
140
141
142
143
143
145
146
144
147
List of Figures
Figure 1.1: The distribution of the ‘acid’ or ‘wallum’ frog species as indicated by
red circles. Regional boundaries are indicated by grey lines. State and territory
boundaries indicated by solid black circles. Records sourced from the Australian
Museum, Queensland Museum, South Australian Museum, Environmental
Protection Agency/Queensland Parks and Wildlife Service WildNet database,
New South Wales Dept of Environment and Conservation Wildlife
Atlas database, and various biologists. Figure obtained from Meyer et al. 2006........
17
Figure 1.2: The four Australian ‘acid’ frog species – 1. Litoria
olongburensis; 2. Crinia tinnula; 3. Litoria freycineti; 4. Litoria
cooloolensis……..........................................................................................................
20
Figure 1.3: Distribution of L. olongburensis as indicated by red and blue circles.
Red circles indicate records obtained between 1995-2004.Blue circles indicate
records obtained before 1995. Regional boundaries are indicated by grey lines.
Solid line represents the Queensland / New South Wales state boundary. Records
sourced from EPA/QPWS, NSWDEC, the Australian Museum, Queensland
Museum, South Australian Museum, and various biologists. Figure obtained from
Meyer et al. 2006..........................................................................................................
22
Figure 1.4: Distribution of C. tinnula as indicated by red and blue circles. Red
circles indicate records obtained between 1995-2004.Blue circles indicate records
obtained before 1995. Regional boundaries are indicated by grey lines. Solid line
represents the Queensland / New South Wales state boundary. Records sourced
from EPA/QPWS, NSWDEC, the Australian Museum, Queensland Museum, South
Australian Museum, and various biologists. Figure obtained from Meyer et al.
2006............................................................................................................................... 23
Figure 2.1: Localities of survey sites, with numbers representing the following
localities: 1 – Cooloola Section of the Great Sandy National Park; 2- Noosa
National Park; 3 – Mooloolah National Park; 4 – Beerwah Scientific Reserve; 5 –
Tyagarah Nature Reserve; 6 – Lennox Heads; 7 – Bunjalung National Park; 8 –
Yuragir National Park (North); 9 – Yuragir National Park (South). Black dots
represent Litoria olongburensis record localities from EPA/QPWS, NSWDEC, the
Australian Museum, Queensland Museum, South Australian Museum, and various
biologists (Meyer et al. 2006). Solid lines represent Australian coastline and the
Queensland / New South Wales state border. Map of Australia shows enlarged area
within the rectangle, with solid lines representing the Australian coastline and the
Australia’s state and territory borders........................................................................... 43
Figure 2.2: ‘Jitter’ plots for quantile regressions of the 0.85(a) and 0.65(b) quantiles
(solid line) and the 95% confidence intervals (dotted lines) for mean waterbody pH
vii
and relative abundance of Litoria olongburensis tadpoles. Circles represent
waterbody transects.......................................................................................................
54
Figure 2.3: ‘Jitter’ plots for quantile regressions of the 0.85(a) and 0.65(b) quantiles
(solid line) and the 95% confidence intervals (dotted lines) for mean waterbody
depth and relative abundance of Litoria olongburensis tadpoles. Circles represent
waterbody transects. .....................................................................................................
55
Figure 3.1: Amphibian species richness and species presence within each
waterbody type. Colours/patterns indicate individual species......................................
79
Figure 3.2: Proportion of natural and anthropogenic waterbodies occupied for each
recorded anuran species. Records are combined for both visual and acoustic
records...........................................................................................................................
80
Figure 3.3: nMDS ordination of waterbodies for anuran species where a relative
abundance measurement was calculated. Stress associated with 4 dimensions used
in MDS ordination was 0.0268. Species ordinations are overlaid. Environmental
variables significantly influencing the community structure are displayed. Circles
represent waterbodies. ..................................................................................................
81
Figure 3.4: ‘Jitter’ plot for relative abundance counts of (a) L. olongburensis and
(b) L. fallax in natural and anthropogenic waterbodies. Abbreviations on the x-axis
represent the first surveys at natural (NW1), artificial lakes (AL1), road side ditches
(RD1) and golf course waterbodies (GCW1) and the second surveys at natural
(NW2), artificial lakes (AL2), road side ditches (RD2) and golf course waterbodies
(GCW2).........................................................................................................................
83
Figure 4.1: nMDS ordination of amphibian species composition using Axis 1 and 2
from the MDS amphibian species abundance matrix. Black dots represent
compensatory ponds while white dots represent established ponds. Species
positions within the matrix are displayed ...................................................................
105
Figure 4.2: Gradient analysis using average pH as a gradient with abudance of each
species recorded across the survey period. N represents a natural pond while C
represents a compensatory pond...................................................................................
106
Figure 5.1: Percentage of predators that consumed (black bars) or attacked (white
bar) Litoria olongburensis tadpoles for experiments where one individual L.
olongburensis was used in each experiment. Number of replicates/experiments is
presented above each predatory species.......................................................................
124
viii
Figure 5.2: Number of tadpoles consumed for each predatory species. Symbolys
represent the number of tadpoles consumed for an individual experiment. ‘o’
represents Limnodynastes peronii, ‘Δ’ represents small Litoria fallax, ‘x’ represents
large L. fallax and ‘+’ represent L. olongburensis........................................................ 127
ix
List of Tables
Table 1.1: ‘Acid’ frog conservation status from Queensland, New South Wales and
Australian legislation and the IUCN Red List. V = Vulnerable; NT = Near
Threatened; - = no status; N/A = not applicable (species not occurring within the
state of the Act). (Adapted from Meyer et al. 2006). ..................................................
18
Table 2.1: Comparison of habitat characteristics for surveyed waterbodies in
wallum habitats of eastern Australia. Spearman correlation coefficients (SCC) were
compared for 37 waterbody transects. None of the variables were considered highly
correlated (SCC ≥ 0.7) .................................................................................................
48
Table 2.2: Comparison of waterbody characteristics associated with the relative
abundance and occupancy of L. olongburensis or C. tinnula tadpoles in eastern
Australia. Aikiki models with Δi values less than 4 are presented. + indicates a
positive relationship while – indicates a negative relationship to L. olongburensis or
C. tinnula tadpole relative abundance or occupancy. Variables with a 2 indicate a
unimodal distribution with L. olongburensis or C. tinnula tadpole relative
abundance or occupancy...............................................................................................
51
Table 2.3: Relative importance of waterbody characteristics associated with the
relative abundance and occupancy of L. olongburensis or C. tinnula tadpoles in
eastern Australia. Model averaged coefficients and relative importance of each
environmental predictor for models where Δi < 4 for L. olongburensis relative
abundance and occupancy and C. tinnula occupancy are displayed............................
52
Table 2.4: Coefficients of the 0.85 and 0.65 regression quantiles where the
independent factors were mean pH and mean water depth. Litoria olongburensis
tadpoles were the dependant factor within the regression quantile models.................. 56
Table 3.1: Measured variable averages and ranges between the four waterbody
types surveyed and for waterbodies with L. olongburensis and L. fallax..................... 78
Table 3.2: Correlations (R2 values) between nMDS axis 1 and 2 and environmental
variables influencing assemblage structure, with significant correlations (Pr (> r))
highlighted in bold........................................................................................................
82
Table 3.3: Models with a Δi value < 4 for L. olongburensis and L. fallax adult
relative abundance per metre for 2011 surveys. (+) indicates a positive relationship
while (-) indicates a negative relationship between relative abundance and the
model variable............................................................................................................... 85
x
Table 3.4: Estimates for model averaged coefficients, standard error (SE),
confidence interval (CI) and relative variable importance (RI) for each parameter in
models where Δi < 4 for L. olongburensis and L. fallax tadpole relative abundance.
(+) indicates a positive relationship while (-) indicates a negative relationship
between relative abundance and the model variable..................................................... 86
Table 4.1: Total number of individuals per species detected over the survey period
for compensatory and established waterbodies. * indicate threatened species and ^
indicate introduced species listed under the Australian EPBC Act 1999...................... 104
Table 4.2: Correlations to the MDS Axis 1-4 with variables playing a significant
influence on assemblage structure highlighted in bold. A significant influence was
considered a variable that had a p value less than 0.05. A* indicates significant
variables while a # indicates a variable nearing significance (p = 0.052).....................
107
Table 4.3: Models with a Δi value < 4 for L. olongburensis and C. tinnula calling
activity and relative abundance. + indicates a positive relationship while – indicates
a negative relationship to L. olongburensis or C. tinnula calling activity for the
variable within the model.............................................................................................. 109
Table 4.4: Model averaged coefficients for models where Δi < 4 for L.
olongburensis and C. tinnula calling activity and relative abundance. Relative
importance of each environmental predictor variable is displayed..............................
110
Table 5.1: Number of experiments conducted for each tadpole predator species for
multiple prey experiments............................................................................................. 125
Table 5.2: Average number of tadpoles consumed for each predator species during
multiple prey experiments............................................................................................. 128
xi
Acknowledgements
Numerous people need to be acknowledged for assistance throughout the duration of
this thesis. It is difficult to put in words my appreciation for those following people who
contributed towards my study. However, I shall attempt to express my appreciation.
I thank my primary supervisor, Associate Professor Jean-Marc Hero, for his advice and
sharing his knowledge throughout this entire process. I would especially like to thank
my associate supervisor, Dr Guy Castley, for providing valuable guidance and support
throughout my entire candidature and for providing invaluable feedback on the final
draft of this thesis.
Special thanks go out to my family. To my aunty, Elaine Emery, who provided cheap
accommodation in her townhouse and to Mum and Dad for providing a roof over my
head when I was without a scholarship. I would also like to give a special thanks to my
partner’s parents who provided weekly Sunday dinners and for putting up with my
‘froggish’ antics. Additionally, I thank my partner, Amanda Winzar, who helped me
through this process when my morale was low and gave me the incentive to ‘slug it out’
with special ‘slug it out’ brownies, cookies and general ‘bad for you but it tastes so
good’ food.
I would like to thank my field assistants – Jodie-Lee Hills, Chays Ogston, Chris Dahl,
Diana Virkki, James Bone, Chris Tuohy, Tempe Parnell, Billy Ross, Matt Davies,
Donna Treby, Katrin Lowe, Alan Kerr, Gregory Lollback, Nick Clarke and Amanda
Winzar. Special thanks go to Jon Shuker for assistance in the field. Without Jons ‘bush
bashing’ abilities in the wallum heathland I am unsure if this study would have been
possible. Together, Jon and I surveyed the entire distributional range of an amphibian
species and, without each other, probably would have succumbed to insanity. I also
wish to thank Alan Kerr from the Bribie Island Environmental Protection Society for
providing accommodation during fieldtrips.
Michael Arthur, Jon Shuker, Clare Morrison, Gregory Lollback, Donna Treby, Katrin
Lowe, Diana Virkki, Ryan Hughes, James Bone, and Sonya Clegg all provided valuable
xii
advice on either earlier drafts of this thesis, or on statistical issues (primarily on how to
use the ‘R’ statistical program (lovingly known as the ‘Pirate Stats Program’)).
I would like to thank Margie Carsburg and Belinda Hachem for their assistance with
administrative matters that arose throughout the duration of this thesis. I would also like
to thank John Robertson for general guidance and always asking how I was doing. A
big thank-you to Jutta Masterton who helped with obtaining field survey equipment –
even when the equipment was meant to be forever ‘dead’. I also thank numerous School
of Environment staff, including, but not limited to, Tony Carroll, James Furse,
Catherine Pickering, Clare Morrison, Sonya Clegg and Hamish McCallum, who were
there to listen to my problems, concerns and dilemmas.
I wish to thank the funding bodies that made this thesis possible. Firstly, FKP Pty. Ltd.,
who provided funding for data collection that contributed towards Chapter 2. Secondly,
the Griffith School of Environment, who provided funding for data collection for the
remaining thesis chapters. For half of my candidature, I also received a living allowance
through the Australian Postgraduate Award scheme.
xiii
1.0 Introduction
The emergence of the global amphibian crisis has seen the disappearance of 122 species of
amphibians (Stuart et al., 2004), with 18.8% of Australia’s 213 species being threatened (Hero and
Morrison, 2004). Despite these declines, little is known about the population dynamics, biology
and ecology of certain Australian threatened species (Hines et al., 1999; Hero et al. 2006).
Understanding what environmental variables influence amphibians within the landscape is
essential if conservation management is to be conducted successfully. Since most amphibians
occur in different ecological niches during different stages of their lifecycle (Wells, 2007) it is
imperative to determine what environmental factors influence amphibian distributions during all
lifecycle stages.
1.1 The Importance of Amphibians
Amphibian reproductive modes are numerous, with amphibian larvae developing in both the
aquatic and terrestrial environment (Haddad and Prado, 2005; Wells, 2007). The non-reproductive
larval stages (tadpoles) occur in different ecological niches compared with the adult stages (Wells,
2007; Halliday, 2008; McDiarmid and Altig, 2010). The tadpole stage is pivotal within the
amphibian lifecycle and has been described as having ‘the potential to have the greatest impact on
the continuing persistence of the (amphibian) population’ (Lane and Mahony, 2002).
Tadpole composition and abundance heavily influence the structure of many aquatic communities.
Sediment dynamics (Flecker et al., 1999; Ranvestal et al., 2004) and the assemblage and
abundance of algae (Morin, 1995; Ranvestal et al., 2004) and zooplankton (Mokany, 2007) are
influenced by the assemblage and abundance of tadpoles. Therefore, freshwater aquatic
communities rely heavily upon amphibian tadpoles in maintaining ecosystem equilibrium.
Adult anurans also play a key role in ecosystem function as they are food sources for numerous
predators and prey on numerous fauna (Duellmann and Trueb, 1986; Wells, 2007; Crump, 2010).
The adult assemblage is also important in determining the tadpole assemblage as tadpoles cannot
occur in areas where adults fail to deposit eggs. Additionally, human society has benefited from
1
amphibians with the discovery and isolation of chemical compounds from adult anurans (Crump,
2010). Therefore, the importance of amphibians to human society and environmental processes
cannot be underestimated.
Amphibian adult density has been linked to larval survivorship in certain species. For example,
densities of adult Rana sylvatica have been found to be dependent on the survival of R. sylvatica
larvae (Berven, 1990). This is also evident in Bufo calamita where the density of adults was
positively correlated with B. calamita metamorph density (Beebee et al., 1996). Therefore, factors
that influence amphibian larval assemblages will also influence the adult amphibian assemblages.
1.2 Assemblages and Communities
For the purpose of this thesis, an assemblage can be described as a group of species that occur
within a particular environment where interactions amongst individuals do not have to occur
(Retallick, 2000). A community is defined as individuals of different species occurring within a
particular environment that interact with each other (Hickman Jr. et al., 1998).
Interactions occurring between individuals within communities may be positive, negative or
neutral and can potentially influence a species’ distribution and population structure. For example,
fish predation on tadpoles has a negative interaction on the tadpole but a positive interaction for
the fish by providing nutrition. These interactions may exclude or reduce specific amphibian
species from waterbodies (Kats et al., 1988; Hecnar and M'Closkey, 1997; Hero et al., 1998; Kats
and Ferrer, 2003; Vonesh et al., 2009) and thus structure the overall amphibian tadpole assemblage
occurring within a waterbody.
Interspecific competition and predation are two important biotic interactions that can structure an
assemblage or community (Schoener, 1983), but these are also influenced by other environmental
factors. The influence of environmental factors on individual species will differ as species have
variable responses to these factors (Cushman, 2006). The environmental effects may also differ
between populations of the same species at different spatial scales (Pierce, 1985; Grand and
Cushman, 2003). Furthermore, low levels of disturbance are believed to aid in maintaining high
species diversity within an assemblage (Death and Winterborn, 1995).
2
The importance of environmental factors in influencing amphibian assemblage and community
structure has been noted within the literature (see 1.3 of this thesis). Arguments that favour
multiple factors structuring communities and assemblages are likely correct as assemblages and
communities contain multiple species that will co-exist with different predators and competitors
and be tolerant to differing ranges of environmental variables.
1.3 Factors influencing Tadpole and Adult Amphibian Assemblages
There are numerous environmental factors that have the potential to influence amphibian adult and
tadpole assemblages. These factors include, but are not limited to, predation (Kats et al., 1988;
Hecnar and M'Closkey, 1997; Hero et al., 1998; Gillespie and Hero, 1999; Vonesh et al., 2009),
competition (Wiltshire and Bull, 1977; Hickman Jr. et al., 1998; Twomey et al., 2008), water
chemistry (Gosner and Black, 1957; Pierce, 1985; Freda, 1986; Freda and Taylor, 1992; Smith et
al., 2007; Sparling, 2010), water quantity (i.e. hydroperiod) (Wilbur, 1987; Snodgress et al., 2000;
Baber et al., 2004; Moreira et al., 2010) and water flow (Richards, 2002). Additionally, factors that
influence the adult assemblage may influence the tadpole assemblage and vice versa. Amphibian
adult density has been linked to larval survivorship in certain species. For example, densities of
adult Rana sylvatica have been found to be dependent on the survival of R. sylvatica larvae
(Berven, 1990). This is also evident in Bufo calamita where the density of adults was positively
correlated with B. calamita metamorph density (Beebee et al., 1996). These factors may exclude
certain species from particular waterbodies or be tolerated at different levels, with the tolerance
level depending on the individual species (Gosner and Black, 1957; Pierce, 1985; Freda, 1986;
Freda and Taylor, 1992; Meyer, 2004; Smith et al., 2007; Sparling, 2010).
1.3.1 The Adult Assemblage
Vegetation cover (Gibbs, 1998; Girish and Krishna-Murthy, 2009) and habitat size (Kolozsvary
and Swihart, 1999) influence the distribution of adult amphibians. Factors influencing adult
distributions may also be correlated with other variables ultimately influencing adult distributions.
For example, Girish and Krishna-Murthy (2009) found increased light intensity as a result of
decreased forest cover affected air and water temperatures. Furthermore, the abundance or
3
emergence of particular vegetation species (Lemckert et al., 2006; Shuker, 2012), or the proportion
of the water margin with emergent vegetation (Hazell et al., 2004) Lemckert et al., 2006, may also
influence amphibian usage or species abundances within waterbodies. Pond isolation may also
negatively influence adult amphibian species richness (Smallbone et al., 2011), individual species
usage (reviewed in Marsh and Trenham 2001) or breeding success (Marsh et al., 1999; reviewed in
Marsh and Trenham 2001).
Temperature and rainfall will affect the adult contribution towards the tadpole assemblage as
calling and breeding may not occur when temperatures and water levels are inadequate
(Duellmann and Trueb, 1986; Oseen and Wassersug, 2002; Wells 2007). For example, at high
altitudes or in temperate environments, adult amphibians will often hibernate when temperatures
fall below ideal conditions (Pearson and Bradford 1976; Carey 1978; Pinder et al., 1992; Wells
2007), while calling and the selection of spawning sites will peak when optimal water
temperatures and depth are reached (Oseen and Wassersug, 2002; Goldberg et al., 2006).
Adult females have the ability to influence the tadpole assemblage by choosing oviposition sites
and the number of eggs that are deposited (Resetarits Jr and Wilbur, 1989). Furthermore, by
avoiding waterbodies that contain predatory cues when depositing eggs (Binckley and Resetarits
Jr., 2003; Orizaola and BraÑa, 2003), females may influence the tadpole assemblage. Site
selection can also be influenced by the female’s ability to detect competitors (Resetarits Jr and
Wilbur, 1989) and intolerant water chemistry levels (Haramura, 2008).
It is imperative to note that, despite the importance of the adult assemblage, the presence of adults
does not indicate a site of reproduction (Mazerolle, 2005) and thus tadpoles of the adult species
recorded at a waterbody may be absent. Additionally, Girish and Krishna-Murthy (2009) found
factors influencing the occurrence of tadpoles also influenced adult occurrence. Therefore,
scenarios and environmental factors that affect the tadpole assemblage may also influence the
adult assemblage.
4
1.3.2 Water Quantity and Chemistry
Amphibians are largely dependent on water throughout all stages of their lifecycle (Halliday,
2008; Sparling, 2010), with the egg and larval stages of numerous amphibian species spent within
the aquatic environment (Duellmann and Trueb, 1986; McDiarmid and Altig, 2010). Adult stages
are not restricted to waterbodies and have the freedom to move within the terrestrial landscape
(Johnson et al., 2007; Simpkins et al., 2011). Nevertheless, water is an essential commodity in
both quantity (hydroperiod) and quality (chemistry).
1.3.2.1 Water Quantity / Hydroperiod
The time water is present in a waterbody (hydroperiod) influences the structure of anuran tadpole
assemblages either directly (Wilbur, 1987; Snodgress et al., 2000; Baber et al., 2004; Moreira et
al., 2010), or indirectly (Herrmann et al., 2005). Hydroperiod may indirectly influence amphibian
larval assemblages by influencing predator and competitor assemblages and abundance
(Woodward, 1983; Richter-Boix et al., 2007), or water chemistry variables (Herrmann et al.,
2005).
Hydroperiod influences assemblage and abundance via natural selection for species that can
successfully reproduce in temporary, permanent or both types of waterbodies. For example,
ephemeral amphibian breeders are often excluded from permanent waterbodies due to the presence
of different predator assemblages (Welborn et al., 1996; Hero et al., 1998; Richter-Boix et al.,
2007) and, potentially, predator size, which is often larger in permanent waterbodies (reviewed in
Welborn et al., 1996; Richter-Boix et al., 2007). Ephemeral breeders, therefore, need to
metamorphose quickly before aquatic predators can become established. Overall success is
therefore higher in species that arrive in ephemeral waterbodies early (Wilbur, 1997). The opposite
may occur for permanent breeders when they are excluded from ephemeral waterbodies due to
shorter hydroperiod lengths and inadequate metamorphosis time (Skelly, 1995). The effects of
aquatic predators on structuring tadpole assemblages are outlined in the ‘Predation’ section of this
chapter.
5
In addition to hydroperiod, a decrease in water level can result in a decrease in food availability
(Loman, 1999). This may be problematic in ephemeral waterbodies where limited food resources
could restrict development (Newman 1994), as tadpoles need to metamorphose before waterbody
desiccation occurs. Tadpoles occurring in ephemeral waterbodies have evolved to metamorphose
early in response to decreasing water levels (Lane and Mahony, 2002; Márquez-García et al.,
2010), despite a reduction in resources (Loman, 1999). Additionally, temperature influences the
developmental rate of tadpoles (Duellmann and Trueb, 1986; Orizaola and Laurila, 2009), with
temperature potentially being elevated in ephemeral waterbodies (Noland and Ultsch, 1981). Early
metamorphosis ensures the short-term survival of the individual but often results in decreased
juvenile body size, potentially affecting the individual’s long-term survival (Smith, 1987; Lane
and Mahony, 2002; Altwegg and Reyer, 2003). To avoid early metamorphosis and waterbody
desiccation anurans that breed within ephemeral waterbodies often breed after extensive rainfall
when water levels are sufficient to ensure optimal chances of metamorphosis.
Shorter hydroperiods and decreasing water levels also increase the amount of UV-B radiation.
Exposure to UV-B radiation increases fungal infection in eggs where amphibians were forced to
deposit eggs in waterbodies with lower than average water levels (Kiesecker et al., 2001). Finally,
water chemistry also changes in response to varying hydroperiod lengths as reported by Herrmann
et al. (2005), where conductivity was significantly lower in ephemeral waterbodies when
compared with permanent waterbodies. This is just one example of the effects of water chemistry
that are outlined in further detail below.
1.3.2.2 Water Chemistry
The permeability of anuran skin facilitates the uptake of water through osmosis (Shoemaker and
Nagy, 1977; Bentley and Yorio, 1979), which in turn is influenced by differing levels of particular
water chemistry variables (Wells, 2007). Water chemistry is therefore important for amphibian
larval and embryo survival.
Water chemistry factors can be tolerated at different levels (tolerance range) depending on the
anuran species (Gosner and Black, 1957; Pierce, 1985; Freda and Taylor, 1992; Chinathamby et
al., 2006; Persson et al., 2007; Smith et al., 2007; Rios-López, 2008; Barth and Wilson, 2010;
6
Sparling, 2010) and the lifestage of the individual (Strahan, 1957; Freda, 1986). The tolerance
range can be categorised into sub-lethal, lethal or non-lethal/optimal with the effects varying
between and within species. Water chemistry can also influence anuran tadpole assemblages by
influencing parasite, predator and competitor assemblages (Sparling, 2010).
Water chemistry factors that may include conductivity (caused by the quantity of anions and
cations) (Smith et al., 2007; Sparling, 2010), pH (caused by hydrogen ion concentration)
(Sparling, 2010), salinity (concentration of chloride salts) (Sparling, 2010), turbidity, (influenced
by particle suspension of inorganic and organic matter) (Sparling, 2010), dissolved oxygen
(Sparling, 2010) and natural organic acids (Steinberg et al., 2006). Dissolved pollutants have also
been shown to affect anuran tadpole assemblages (Sparling, 2010). These factors may influence
amphibian assemblages and are further discussed in the sections below.
1.3.2.2.1 Water pH
Amphibians that breed successfully in acidic aquatic environments can tolerate lower pH levels
than those breeding in non-acidic environments (Gosner and Black, 1957; Freda, 1986; Meyer,
2004). Despite this, amphibians have been known to breed in water bodies that were either outside
or near their pH tolerance level where rainfall temporarily increases pH levels (Sadinski and
Dunson, 1992). Additionally, intraspecific tolerance may differ amongst populations (Pierce,
1985; Glos et al., 2003; Persson et al., 2007). For example, populations of Rana sylvatica, an
anuran tolerant of low pH levels, differ in their pH tolerance range depending on the level of
acidity exposure, which varies across geographical locations (Pierce, 1985).
Sub-lethal pH levels can produce tadpoles hatching with abnormalities (Gosner and Black, 1957;
Andrén et al., 1988), increase time to metamorphosis, reduction in body size (Cummings, 1986)
and, indirectly, a reduction in clutch and egg size (Räsänen et al., 2008). Lethal effects of pH can
result in tadpoles failing to hatch from eggs (Gosner and Black, 1957; Sadinski and Dunson, 1992;
Meyer, 2004), or inhibit fertilisation due to reduced movement or death of sperm (Schlichter,
1981). However, jelly membranes on eggs may also act as a buffer to acidic waters (Picker et al.,
1993). To combat these effects some amphibian tadpoles have adapted mechanisms to detect pH
levels and can actively avoid exposure to unsuitable pH levels (Freda and Taylor, 1992). Certain
7
low pH waterbodies can be heterogeneous, in relation to pH (Freda and Taylor, 1992), and thus
detection of unsuitable pH levels may influence the distribution of tadpole species within
waterbodies. Additionally, tolerance to varying pH levels increases with increased development
stages (Pierce, 1985; Freda, 1986).
1.3.2.2.2 Natural Organic Acids (NOA)
Naturally occurring organic acids (NOA) are derived from decomposing organic matter and are
often referred to as dissolved humic substances. The dark brown coloration that occurs in
waterbodies in wallum heathland of Eastern Australia and in the ‘blackwaters’ of Rio Negro in the
Amazon are attributed to the chemical properties of the waters; that is low or absent in magnesium
and/or calcium (soft-water), low buffering capacity and high organic acids (Barth and Wilson,
2010). In some isolated waterbodies, natural organic acids can decrease pH levels (Sparling,
2010).
It is also believed that NOA can influence the faunal community independent of pH (Steinberg et
al., 2006), however, current knowledge on how humic/organic acids influence tadpole
communities is lacking (Barth and Wilson, 2010). While hatching success can be reduced in low
pH waters with high levels of NOA, this may be dependent on individual species tolerance levels
(Picker et al., 1993). Within soft-water, humic substances can either protect (Wood et al., 2003;
Steinberg et al., 2006), or expose (Steinberg et al., 2006), other non-amphibian aquatic fauna (i.e.
fish) to ion-loss. Under these conditions humic substances can impose negative stresses on aquatic
invertebrates (Timofeyev et al., 2006). Therefore, NOA have the potential to structure tadpole
assemblages directly, by selecting for tadpole species that can tolerate high levels of NOA within
the waterbody, or indirectly, by influencing the structure of the aquatic predator assemblage.
1.3.2.2.3 Salinity
Amphibians are rarely detected in waters with high salt concentrations due to their inability to
efficiently osmoregulate under these conditions (Gomez-Mestre et al., 2004). Despite the majority
of amphibians being intolerant to saline waters some can live in waterbodies with salinity levels
close to seawater (e.g. Fejervarya cancrivora (crab-eating frog)). Adult F. cancrivora regulate the
8
osmotic process in response to salinity by increasing the amount of urea, sodium and chloride
present within their bodies (Wells, 2007). Tadpoles of this species displayed signs of increased
sodium and chloride but regulated the osmotic process by excreting salts via the gill membranes
(Wells, 2007). A similar increase in sodium, chloride and calcium present within tadpoles in saline
waters was observed in Bufo calamita, although there was no mention of salt being excreted across
the gill membranes (Gomez-Mestre et al., 2004).
Salinity tolerance levels will differ among species (Smith et al., 2007). Sub-lethal levels of water
salinity can cause an increased time to metamorphosis (Christy and Dickman, 2002; GomezMestre et al., 2004; Chinathamby et al., 2006), a reduction in body weight (Christy and Dickman,
2002; Gomez-Mestre et al., 2004) and retardation of external features (Rios-López, 2008). Lethal
effects of salinity can cause death to individual tadpoles and failure to metamorphose (Christy and
Dickman, 2002; Chinathamby et al., 2006). The lethal effects of salinity can decrease with older
developmental stages in some tadpoles (Strahan, 1957). To overcome the effects of sub-lethal and
lethal levels of salinity coastal frog species (e.g. Buergeria japonica) have evolved the ability to
detect water salinity levels and will actively choose their oviposition sites in non-saline water
(Haramura, 2008).
1.3.2.2.4 Turbidity and Eutrophication
The effects of turbidity on tadpoles are not well known (Schmutzer et al., 2008), but may influence
tadpole assemblages by decreasing predation levels in turbid water. For example, tadpoles of
Phrynomantis microps increased their schooling density and size when waters were less turbid and
this was attributed to an increased risk of predation in clearer water (Spieler, 2003).
Other studies have shown that nitrate levels can influence the survival of amphibian larvae by
inhibiting growth and development (Mann and Bidwell, 1999). Therefore, Meyer et al. (2006)
proposed that increasing nitrate levels in waterbodies along Australia’s eastern seaboard (i.e.
nutrient poor wallum heathland waterbodies) could alter the viability of that habitat for ‘native’
species. Additionally, the increase in nitrates could alter vegetation community structure towards a
vegetation community that would be unsuitable for native species (Meyer et al., 2006).
9
1.3.2.2.5 Dissolved Oxygen
Dissolved oxygen (D.O.) levels differ depending on the waterbody in question. Well mixed lotic
waterbodies are usually higher in D.O. compared to lentic waterbodies where D.O. usually
decreases within increasing depth (Sparling, 2010). Therefore, tadpoles living in low oxygenated
(hypoxic) waterbodies rely on oxygen being taken up primarily through their lungs by surfacing to
the top of the water (Wassersug and Seibert, 1975; Sparling, 2010). Surfacing can come at a cost
as it increases predation risk from visually orientated predators (Feder, 1983).
The proportion of dissolved oxygen needed for tadpole survival differs among species (Sparling,
2010). Furthermore, while some anuran eggs hatch earlier when exposed to low oxygenated
waters, low oxygen can also be lethal and result in death of eggs (Seymour et al., 2000).
1.3.2.2.6 Water Temperature
Amphibians lack the ability to produce their own body heat. Therefore, environmental temperature
is extremely important for behaviour, metabolic rates and other physiological processes (Sparling,
2010). As mentioned previously, water temperature can influence the developmental process and
growth rates of tadpoles (Duellmann and Trueb, 1986; Orizaola and Laurila, 2009). Low
temperatures will often result in slow development (Duellman and Trueb, 1986) and therefore be a
key factor for tadpole survival in ephemeral waterbodies where hydroperiod can be short.
Additionally, water temperature may influence oviposition timing and location (Goldberg et al.,
2006). Water temperature may also influence levels of D.O. with concentrations decreasing as
water temperatures increase (Sparling, 2010).
1.3.3 Competition
Darwin considered competition an important factor in ecological communities as it reduces fitness
of the weaker competitor (Hickman Jr. et al., 1998). Competition can occur within (intraspecific)
or amongst (interspecific) species when a resource being shared is scarce and will result in
10
competitive exclusion or competitive coexistence (Hickman Jr. et al., 1998; Twomey et al., 2008),
thereby structuring communities (Wiltshire and Bull, 1977).
Interspecific competition among amphibian species has the potential to alter rates of tadpole
growth and development (Morin, 1986; Wilbur, 1987; Relyea, 2004; Twomey et al., 2008), as well
as aquatic, non-amphibian species (Mokany and Shine, 2003). Slower growth rates can be lethal in
ephemeral waterbodies if metamorphosis does not occur before waterbody desiccation.
Furtermore, delaying metamorphosis may also prolong exposure to aquatic predators and increase
the predator-prey interaction time.
High competition can result in smaller body size at metamorphosis (Semlitsch and Ryer, 1992;
Rudolf and Rödel, 2007) and can lead to higher mortality as a metamorph, lower reproductive
success as an adult (Lane and Mahony, 2002; Altwegg and Reyer, 2003) or increased time to
sexual maturity (Smith, 1987; Scott, 1994). The reduction in growth and development under
competition can potentially be related to food availability which would be lower under increased
competition. Decreased food availability has also been shown to negatively influence the growth
of tadpoles (Griffiths et al., 1993; Mokany and Shine, 2003) and the size at metamorphosis
(Newman, 1994).
Interspecific competition can be minimised both spatially and temporally within the tadpole
assemblage. Temporally, the phenology of anurans contributes towards reducing interspecific
competition when eggs are deposited at different time intervals (Heyer, 1973; Toft, 1985; Wells,
2007; Crossland et al., 2009). Larger tadpoles have been noted to outcompete and kill smaller
tadpoles (Toft, 1985) and earlier hatching can provide older tadpoles with first choice of resources
and the potential to outgrow their competitors. Additionally, hatching at different times will
minimise interaction with other species and reduce resource competition.
The spatial structuring of the tadpole assemblage in different columns of a waterbody can remove
or minimise interspecific competition (Heyer, 1973). Predation can additionally reduce inter- and
intraspecific competition by minimising the number of individuals present (Wilbur, 1997) and
removal of species that are susceptible to predation (Kats et al., 1988). Female choice of
11
oviposition sites will further reduce the risk of competition with some species actively avoiding
sites where more competitive species are present (Resetarits Jr and Wilbur, 1989).
Morphology of tadpoles may change under interspecific competition. To maximise success of
obtaining food, tadpoles of Rana sylvatica and Rana pipiens, have increased their mouth width by
up to 10% and 5%, respectively, when competing with each other (Relyea, 2000). Alternatively,
tadpoles may shift their dietary preference to reduce competition of food resources through
phenotypic plasticity (Pfennig and Murphy, 2002). Therefore, tadpoles which display lower levels
of phenotypic plasticity are less likely to adapt to competitive pressures, having an increased
chance of exclusion.
Ultimately, tadpoles should ideally occur in waterbodies that maximise their ability to grow,
develop and avoid any negative interactions that results from competition (Retallick, 2000).
1.3.5 Predation
Predation is often an underlying factor determining tadpole assemblages within waterbodies (Kats
et al., 1988; Hecnar and M'Closkey, 1997; Hero et al., 1998; Vonesh et al., 2009) and is arguably
the most important biotic factor influencing tadpole assemblages both spatially and temporally
(Heyer et al., 1975). Predators of tadpoles include fish (Hero et al., 1998; Baber and Babbitt, 2003;
Vonesh et al., 2009), aquatic invertebrates (Heyer et al., 1975; Fox, 1978; Stoneham, 2001; Jara,
2008; Álvarez and Nicieza, 2009), salamanders (Morin, 1995) and other tadpoles (Heyer et al.,
1975; Álvarez and Nicieza, 2009).
Amphibian tadpoles have evolved anti-predator strategies to coexist with their natural predators.
These include unpalatability or chemical defences (Kats et al., 1988; Hero et al., 2001; Gunzburger
and Travis, 2005), behavioural avoidance (Skelly, 1994; Relyea, 2003; Gregoire and Gunzburger,
2008; Saidapur et al., 2009; Smith and Awan, 2009), morphological adaptations (Hecnar and
M'Closkey, 1997; McCollum and Leimberger, 1997; Touchon and Warkentin, 2008) and
grouping/schooling (Watt et al., 1997). Tadpoles can also detect predators through chemical cues.
In a classical co-evolutionary arms race, predators respond to tadpoles’ evolutionary tactics by
12
evolving abilities of their own to overcome the tadpoles’ anti-predator strategies (Hickman Jr. et
al., 1998; Brodie III and Brodie Jr., 1999).
Predatory fish can exclude some species of tadpoles that lack adequate anti-predator strategies
from waterbodies (Kats et al., 1988). The level of predation on a tadpole will be determined by a
number of factors, including the tadpole’s vulnerability to the predator assemblage and predator
abundance (Hossie and Murray, 2010). Predator assemblage and abundance will often vary
depending on biotic and abiotic factors (Woodward, 1983; Babbitt et al., 2003). For example,
permanent waterbodies will contain predators that require permanent water to survive. Conversely,
in ephemeral waterbodies predators may be absent in the ‘starting’ period of the waterbody, or be
of a smaller size when compared with predators in permanent waterbodies (Richter-Boix et al.,
2007). Thus, it may be beneficial for tadpoles in ephemeral waterbodies to develop quickly due to
the risk of predation increasing with time (Duellmann and Trueb, 1986; Relyea, 2007).
If rapid development is not possible tadpoles can co-exist with their predators by using refuges
(Babbitt and Tanner, 1998; Kopp et al., 2006; Saidapur et al., 2009). An increase or decrease in
use of refuges is often dependant on the type of predator being avoided (Morin, 1986; Smith et al.,
2008; Smith and Awan, 2009). For example, the effectiveness of habitat refugia will often depend
on the size of the predator, with larger predators being avoided more successfully than smaller
predators (Babbitt and Tanner, 1998). Furthermore, predators, like Dytiscus sp (diving water
beetles), may adopt different hunting strategies under different levels of habitat complexity and
thus use of refuges may not decrease the overall risk of predation (Michel and Adams, 2009).
Additionally, refugia are not always beneficial and complex environments have been found to
increase predation in fast swimming tadpoles by hindering the tadpoles swimming ability
(Saidapur et al., 2009).
Tadpoles may reduce time spent foraging and/or moving to reduce the risk of predation (RichterBoix et al. 2007, Relyea 2007, Saidapur et al. 2009, Smith and Awen 2009), with altered
behaviour often influenced by the threat level that is associated with a predator. For example,
experiments performed on Rana sylvatica showed a decrease in movement when exposed to
‘fresh’ predatory cues, but increased when predation threat levels were lower (Ferrari and Chivers
2008). Furthermore, some tadpoles have the ability to learn, which may influence movement
13
behaviour based on past predator experiences (Shah et al. 2010). A decrease in
movement/foraging time can potentially come at a cost of metamorphic weight, size and growth
rate which can have detrimental effects during later lifecycle stages (Skelly 1995).
Tadpoles may form groups or schools in an attempt to reduce predation (Rödel and Linsenmair,
1997; Watt et al., 1997; Spieler, 2003; Stav et al., 2007). An increase in schooling size will reduce
the probability of predation on individuals, despite an increase in the number of overall attacks by
predators as schooling size increases (Watt et al., 1997).
The risk of predation is lower for larger tadpoles than small tadpoles (Brodie Jr. and Formanowicz
Jr., 1987; Jara, 2008; Arendt, 2009), but is also affected by predator size. Some species also show
adaptive plasticity allowing them to grow larger in deeper waterbodies compared to shallow
waterbodies (Loman and Claesson, 2003), potentially reducing predation in deeper, more
permanent, waterbodies where larger predators occur (Richter-Boix et al., 2007). In the absence of
a large body size some tadpoles have evolved the ability to ‘sprint’ (i.e. swim fast), to actively
evade predators (Arendt, 2009; Saidapur et al., 2009).
Colouration can also be used as a camouflage defence against predators that see primarily in
particular light spectrums (Touchon and Warkentin, 2008), or for camouflage within the natural
environment (McCollum and Leimberger, 1997). It has been noted that some tadpole species have
the ability to change their colouration when exposed to predators (Touchon and Warkentin, 2008)
and can centre colouration to particular areas in an attempt to focus attacks away from important
vulnerable areas (Van Buskirk et al., 2004).
Particular tadpole species have been noted to be unpalatable (Heyer et al., 1975; Henrikson, 1990;
Lawler and Hero, 1997; Jara and Perotti, 2008) due to secretions that are toxic to particular
predators. The level of palatability may also change as tadpoles develop (Lawler and Hero, 1997),
however, unpalatable tadpoles may still be attacked and suffer injuries from predators (Jara and
Perotti, 2008), that may result in cannibalism from other tadpoles (Álvarez and Nicieza, 2009).
Additionally, in an attempt to reduce tadpole movement, tadpole tail-nipping by predatory fish has
been recorded (Baber and Babbitt, 2003) in an attempt to allow for easier consumption regardless
of tadpole size.
14
Predators may use visual (Rödel and Linsenmair, 1997; Saidapur et al., 2009), or chemical
(Saidapur et al., 2009), cues to hunt prey and thus the effectiveness of one anti-predator defence
may not work on all predators within a system (Hero et al., 2001; Saidapur et al., 2009). Decreased
palatability, for example, has been known to work against Pyrrhulina sp. for tadpoles of Hyla
boans but is ineffective against the odonate naiad Gynacantha membranalis (Hero et al., 2001). To
overcome these scenarios a combination of anti-predator defences may be employed by tadpoles
(Kats et al., 1988). Different strategies may not occur together but could shift with ontogeny (e.g. a
reduction in activity as unpalatbility defence decreases with increase development) (Jara and
Perotti, 2008).
Introduction of exotic predators into an aquatic ecosystem can have significant impacts upon the
naturally occurring tadpole assemblage (Gillespie and Hero, 1999; Kats and Ferrer, 2003), as
tadpoles may not have evolved any effective anti-predator strategies against introduced predators
(Kats and Ferrer, 2003). Alternatively, native tadpole species may fail to detect predatory cues
from non-native predatory species (Polo-Cavia et al., 2010) and may therefore be heavily predated
upon.
Predators may therefore influence the amphibian community by removing amphibian species that
lack adequate anti-predator defences and selecting for amphibian species that can co-exist with
their predators. Species that lack these defences will be presented with higher exclusion pressures
within an ecosystem. Alternatively, temporal separation will select for tadpoles that avoid
predators completely.
1.4 Study Area and Study Species
1.4.1 Study Area
The study area encompassed waterbodies within wallum vegetation along the eastern coast of
Australia, between Rainbow Beach, QLD, and Wooli, N.S.W. (Figure 1.1). Wallum vegetation
occurs along the coastal lowlands of eastern Australia between Newcastle, N.S.W., and
Rockhampton, QLD (Griffith et al., 2008). For the purposes of this study ‘wallum’ is described as
15
the vegetation communities that include Banksia woodland, sedgeland, heathland and Melaleuca
swamps (Hines et al., 1999; Griffith et al., 2003) occurring within the coastal lowlands of South
East Queensland (SEQLD) and north east New South Wales (Griffith et al., 2003). Within the
study area wallum vegetation communities occur along ‘dunefields, beach ridge plains and sandy
barrier flats’ with soils that are often acidic (pH 3.4-5.4) (Griffith et al., 2008), sandy (Hines et al.,
1999) and low in nitrogen and phosphorus (Groves, 1981).
Waterbodies associated with wallum communities are also low in nutrients and contain acidic
water (pH <5.5) (Meyer et al., 2006), as a result of decaying detritus/organic matter (Barth and
Wilson, 2010). The dark brown colouration that occurs in waterbodies of wallum heathland are
attributed to the chemical properties of the waters; low or absent in magnesium and/or calcium
(soft-water), low buffering capacity and high organic acids (Barth and Wilson, 2010).
Wallum vegetation communities are highly flammable and fire is believed to play an integral part
in certain wallum ecosystems (Specht, 1981). Drought and flooding are also common within
wallum communities as seasonal rainfall can fluctuate considerably
(Griffith et. al., 2004). Within SE QLD rainfall can be largely dependent on cyclones and
thunderstorms (Coaldrake, 1961).
1.4.2 Study Fauna
Occurring within coastal wallum vegetation along the eastern coast of Australia (Figure 1.1) are
four species of anurans that have been described as ‘acid’ or ‘wallum’ frogs (Ingram and Corben,
1975; Hines et al., 1999; Meyer et al., 2006). The terming ‘acid’ frog is based around these species
ability to occur in waters with low pH (Ingram and Corben, 1975) and include Litoria
cooloolensis, Litoria freycineti and, the primary study species, Litoria olongburensis and Crinia
tinnula (Meyer et al., 2006) (Figure 1.2).
The main threat facing ‘acid’ frogs is habitat loss for agricultural, residential or infrastructure
development (Hines et al., 1999; Meyer et al., 2006). Threat severity is increased as the majority
of the ‘acid’ frog species distributions overlap with areas where human population growth rates are
highest (Hines et al., 1999). Other threatening processes to ‘acid’ frog populations include habitat
16
Figure 1.1: The distribution of the ‘acid’ or ‘wallum’ frog species as indicated by red circles.
Regional boundaries are indicated by grey lines. State and Territory boundaries indicated by solid
black circles. Records sourced from the Australian Museum, Queensland Museum, South
Australian Museum, Environmental Protection Agency/Queensland Parks and Wildlife Service
WildNet database, New South Wales Dept of Environment and Conservation Wildlife Atlas
database, and various biologists. Figure obtained from Meyer et al., 2006.
17
degradation, invasive flora species, disease (i.e. Chytrid fungus), inappropriate fire management,
hydrological alteration, and introduction or translocation of fish species (Gillespie and Hero, 1999;
Hines et al., 1999; Meyer et al., 2006). Chemicals used to control mosquitoes and weeds may also
pose a risk (Meyer et al., 2006). Despite these threats, it is believed that populations of ‘acid’ frogs
occurring within protected areas (i.e. National Parks) are stable (Hines et al., 1999; Lewis and
Goldingay, 2005).
All ‘acid’ frog species are listed as ‘Vulnerable’ or ‘Near Threatened’ under the Queensland
Nature Conservation Act 1992 (NCA 1992) and the New South Wales Threatened Species
Conservation Act 1995 (TSC Act 1995). Litoria olongburensis is the only species currently listed
under the Commonwealth Environment Protection and Biodiversity Conservation Act 1999
(EPBC Act 1999). Internationally, the World Conservation Union lists the ‘acid’ frog species as
either ‘Endangered’ or ‘Vulnerable’ (Table 1.1).
Table 1.1: ‘Acid’ frog conservation status from Queensland, New South Wales and Australian
legislation and the IUCN Red List of Threatened Species. V = Vulnerable; NT = Near Threatened;
- = no status; N/A = not applicable (species not occurring within the state of the Act). (Adapted
from Meyer et al., 2006).
Species
NCA1 1992
TSC2 Act
EPBC3 Act
1995
1999
IUCN4
Litoria olongburensis
V
V
V
V
Litoria coolooensis
NT
N/A
-
E
Litoria freycineti
V
-
-
V
Crinia tinnula
V
V
-
V
18
1.4.2.1 Litoria olongburensis
One target study species, L. olongburensis, belongs to the Hylidae family and is often referred to
as the Wallum Sedge Frog, Olongburra Frog or the Sharp-Snouted Reed Frog (Tyler, 1997; Meyer
et al., 2006). The species can range between 25-31mm SVL and varies in coloration, with the
dorsum of different individuals being recorded as grey-brown, beige or bright green (Meyer et al.,
2006; Lowe and Hero, 2012). Ventral coloration is usually white or cream with the thighs being
blue or purple blue with red or orange (Figure 1.2) (Meyer et al., 2006).
On the mainland, distribution of L. olongburensis occurs between Rainbow Beach, QLD, and
Woolgoolga, N.S.W. (Figure 1.3) in well vegetated, low nutrient, acidic, tannin stained, ephemeral
swamps that consist of reeds, sedges and emergent ferns (Tyler, 1997; Robinson 2002, Meyer et
al., 2006) Fragmentation of habitat occurs throughout this species range (Tyler, 1997, Meyer et al.
2006).
Breeding occurs during spring, summer and autumn periods when males call and eggs are
deposited at the base of sedges or reed stems (Anstis, 2002; Meyer et al., 2006). Tadpoles can
grow to 37mm in tail length and 13mm in body length at Gosner stage 37 (Anstis, 2002). Tadpoles
are located in the mid or surface water and are well camouflaged against the tannin stained waters
(Anstis, 2002), or can be found foraging or resting on matted sedges (Meyer et al., 2006).
Metamorphosis is likely to occur during the summer or autumn period (Anstis, 2002).
Litoria olongburensis is the most threatened ‘acid’ frog species, and is listed in QLD, N.S.W. and
Australian legislation as Vulnerable (Table 1.1). In addition to those threatening processes outlined
previously Meyer et al. (2006) indicates that the mosquito fish (Gambusia holbrooki) may threaten
populations and that competition from L. fallax may occur in disturbed areas. Despite these
threats, the amount of peer-reviewed literature on the biology and ecology of L. olongburensis is
limited, particularly in relation to the non-breeding habitat requirements or factors that influence
this species’ distribution (Hines et al., 1999; Meyer et al., 2006). Only a single publication
provides data on habitat preference of L. olongburensis in north-eastern N.S.W. (Lewis and
Goldingay, 2005).
19
Figure 1.2: The four Australian ‘acid’ frog species – 1. Litoria olongburensis; 2. Crinia tinnula; 3. Litoria freycineti; 4. Litoria
cooloolensis
20
1.4.2.2 Crinia tinnula
The second study species, C. tinnula, belongs to the Myobatrachidae family and is often
referred to as the Wallum Froglet or the Tinkling Frog (Meyer et al., 2006). The species can
range between 16-22mm SVL and vary in coloration, with the dorsum of different
individuals being recorded as beige, red or dark-brown with irregular markings or stripes
(Figure 1.2) (Meyer et al., 2006). Ventral coloration is also variable and may be white with
dark grey flecking, peppered grey with black and white or white flecking on dark grey
(Meyer et al., 2006). There will usually be a pale stripe that runs from the throat through the
middle of the belly (Meyer et al., 2006) and occasionally another stripe running from
armpit to armpit (Robinson 2002).
The mainland distribution of C. tinnula occurs between Littabella National Park, QLD, to
South Kurnell, Sydney, N.S.W. (Figure 1.4) in acidic Melaleuca (paperbark) and sedge
swamps (Meyer et al., 2006). Records have also been located in disturbed sites (i.e. pine
plantations, burnt heath) with sister species co-occurrence (i.e. Crinia signifera) (Simpkins,
per. obs.)
Breeding occurs in spring, late summer, autumn and winter when males call (Anstis, 2002;
Meyer et al., 2006). Tadpoles can grow to 36.7mm in tail length and 11.4mm in body
length at Gosner stage 35 (Anstis, 2002). Tadpoles are usually found at the bottom of
waterbodies (Anstis, 2002) in shallow waters (< 1m depth) (Meyer et al., 2006).
Metamorphosis is likely to occur after about six months, however, this was observed in
captivity over winter (Anstis, 2002).
Crinia tinnula is the second most threatened ‘acid’ frog species, being listed in both QLD
and N.S.W. as vulnerable (Table 1.1). There is currently no peer-reviewed literature on
non-breeding habitat requirements or factors that influence this species’ distribution (Hines
et al., 1999).
21
Figure 1.3: Distribution of L. olongburensis as indicated by red and blue circles. Red
circles indicate records obtained between 1995-2004.Blue circles indicate records obtained
before 1995. Regional boundaries are indicated by grey lines. Dotted line represents the
Queensland / New South Wales state boundary. Records sourced from EPA/QPWS,
NSWDEC, the Australian Museum, Queensland Museum, South Australian Museum, and
various biologists. Figure obtained from Meyer et al., 2006.
22
Figure 1.4: Distribution of C. tinnula as indicated by red and blue circles. Red circles
indicate records obtained between 1995-2004.Blue circles indicate records obtained before
1995. Regional boundaries are indicated by grey lines. Dotted line represents the
Queensland / New South Wales state boundary. Records sourced from EPA/QPWS,
NSWDEC, the Australian Museum, Queensland Museum, South Australian Museum, and
various biologists. Figure obtained from Meyer et al., 2006.
23
1.5 Study Aims
This study aimed to identify those environmental factors influencing the tadpole and adult
amphibian assemblages (with primary focus on L. olongburensis) in the wallum heathland
of eastern Australia. Hypotheses were derived from current information suggesting that
fish (i.e. introduced species), competition and water chemistry (i.e. pH) influence the
distribution and assemblage of these tadpoles and adult amphibians.
The first three chapters of this thesis focus on determining which environmental variables
are important in structuring the adult and tadpole amphibian assemblage within and around
wallum heathland areas. Furthermore, the first of these three chapters will focus on
determining how biotic and abiotic variables are influencing the two most threatened ‘acid’
frog species, L. olongburensis and C. tinnula. The forth chapter aimed at determining
predation rates of L. olongburensis and two other amphibian tadpole species to both native
and exotic aquatic predators through an experimental manipulation.
The results from this study are urgently needed for conservation/management purposes as
there is little peer-reviewed information on any wallum associated amphibian species. This
knowledge will allow managers and scientists to evaluate suitable habitat for threatened
species associated with wallum heathland, and prioritise areas for conservation. This
research is also essential for environmental consultants performing Environmental Impact
Assessments (EIA) to identify important breeding habitat for L. olongburensis and C.
tinnula. Futhermore, the study will determine if anthropogenic habitats can be constructed
in an attempt to mitigate against habitat loss when develeopment occurs in areas where L.
olongburensis are present.
Each chapter has been written in draft format for publication purposes. Consequently,
individual chapters have been formatted to meet the style requirements of target peerreviewed journals.
24
1.6 References
Altwegg, R., Reyer, H.-U., 2003. Patterns of Natural Selection on Size at Metamorphosis in
Water Frogs. Evolution 57 (4), 872-882.
Álvarez, D., Nicieza, A. G., 2009. Differential success of prey escaping predators: tadpole
vulnerability or predator selection? Copeia 2009, 453-457.
Andrén, C., Henrikson, L., Olsson, M., Nilson, G., 1988. Effects of pH and aluminium on
embryonic and early larval stages of Swedish brown frogs Rana arvalis, R.
temporaria and R. dalmatina. . Ecography 11 (2), 127-135.
Anstis, M. 2002. Tadpoles of South-eastern Australia: a guide with keys, New Holland
Publishers, Sydney.
Arendt, J. D., 2009. Infuence of sprint speed and body size on predator avo in New
Mexican spadefoot toads (Spea multiplicata). Oecologia 159, 455-461.
Babbitt, K. J., Baber, M. J., Tarr, T. L., 2003. Patterns of larval amphibian distribution
along a wetland hydroperiod gradient. Canadian Journal of Zoology 1552, 15391552.
Babbitt, K. J., Tanner, G. W., 1998. Effects of cover and predator size on survival and
development of Rana utricularia tadpoles. Oecologia 114, 258-262.
Baber, M. J., Babbitt, K. J., 2003. The relative impacts of native and introduced predatory
fish on a temporary wetland tadpole assemblage. Oecologia 136, 289-295.
Baber, M. J., Fleishman, E., Babbitt, K. J., Tarr, T. L., 2004. The relationship between
wetland hydroperiod and nestedness patterns in assemblages of larval amphibians
and predatory macroinvertebrates. Oikos 107, 16-27.
Barth, B. J., Wilson, R. S., 2010. Life in acid: interactive effects of pH and natural organic
acids on growth, development and locomotor performance of larval striped marsh
frogs (Limnodynastes peronii). The Journal of Experimental Biology 213, 12931300.
Beebee, T. J. C., Denton, J. S., Buckley, J., 1996. Factors affecting population densities of
adult Natterjack Toads Bufo calamita in Britain. Journal of Applied Ecology 33,
263-268.
Bentley, P.J., Yorio, T., 1979. Do frogs drink? Journal of Experimental Biology 79, 41-46.
25
Berven, K. A., 1990. Factors affecting population fluctuations in larval and Adult Stages of
the Wood Frog (Rana sylvatica). Ecology 71, 1599-1608.
Binckley, C. A., Resetarits Jr., W. J., 2003. Functional equivalence of non-lethal effects:
generalized fish avoidance determines distribution of gray treefrog, Hyla
chrysoscelis, larvae. Oikos 102, 623-629.
Brodie III, E. D., Brodie Jr., E. D., 1999. Predator-Prey Arms Races. Bioscience 49, 557568.
Brodie Jr., E. D., Formanowicz Jr., D. R., 1987. Antipredator mechanisms of larval
anurans: protection of palatable individuals. Herpetologica 43, 369-373.
Carey, C. (1978). Factors affecting body temperatures of toads. Oecologia 35, 197-219.
Chinathamby, K., Reina, R. D., Bailey, P. C. E., Lees, B. K., 2006. Effects of salinity on
the survival, growth and development of tadpoles of the brown tree frog, Litoria
ewingii. Australian Journal of Zoology 54, 97-105.
Christy, M. T., Dickman, C. R., 2002. Effects of salinity on tadpoles of the green and
golden bell frog (Litoria aurea). Amphibia-Reptilia 23, 1-11.
Coaldrake, J.E., 1961. The eco-system of the coastal lowlands (Wallum) of southern
Queensland. C.S.I.R.O. Bulletin. No. 283.
Crossland, M. R., Alfrod, R. A., Shine, R., 2009. Impact of the invasive cane toad (Bufo
marinus) on an Australian frog (Opisthodon ornatus) depends on minor variation in
reproductive timing. Oecologia 158, 625-632.
Crump, M. L. (2010) Amphibian diversity and life history. In: Dodd Jr. CK (ed) Amphibian
Ecology and Conservation: A Handbook of Techniques. Oxford University Press,
New York.
Cummings, C. P., 1986. Effects of aluminum and low pH on growth and development in
Rana temporaria tadpoles. . Oecologia 69 (2), 248-252.
Cushman, S., 2006. Effects of habitat loss and fragmentation on amphibians: A review and
prospectus. Biological Conservation 128, 231-240.
Death , R. G., Winterborn, M. J., 1995. Diversity patterns in stream benthic invertebrate
communities: The influence of habitat stability. Ecology 76, 1446-1460.
Duellmann, W. E., Trueb, L. 1986. Biology of amphibians, John Hopkins University Press,
Baltimore.
26
Feder, M. E., 1983. The relation of air breathing and locomotion to predation on tadpoles,
Rana berlandieri, by turtles. Physiological Zoology 56 (4), 522-531.
Flecker, A. S., Feifarek, B. P., Taylor, B. D., 1999. Ecosystem engineering by a tropical
tadpoles: density-dependant effects on habitat structure and larval growth rates.
Copeia 1999 (2), 495-500.
Fox, P. J., 1978. Caddis larvae (Trichoptera) as predators of fish eggs. Freshwater Biology
8, 343-345.
Ferrari, M. C. O., Chivers, D.P., 2008. Temporal variability, threat sensitivity and
conflicting information about the nature of risk: understanding the dynamics of
tadpole antipredator behaviour. Animal Behaviour 78, 11-16.
Freda, J., 1986. The influence of acidic pond water on amphibians: A review. Water, Air
and Soil Pollution 30, 439-450.
Freda, J., Taylor, D. H., 1992. Behavioral response of amphibian larvae to acidic water.
Journal of Herpetology 26, 429-433.
Gibbs, J. P., 1998. Distribution of woodland amphibians along a forest fragmentation
gradient. Landscape Ecology, 263-268.
Gillespie, G., Hero, J.-M. (1999) Potential impacts of introduced fish and fish
translocations on Australian amphibians. In: Campbell A (ed) Declines and
Disappearances of Australian Frogs. Environment Australia, Canberra, pp 131-144.
Girish, K. G., Krishna-Murthy, S. V. B., 2009. Distribution of tadpoles of large wrinkled
frog Nyctibatrachus major in central Western Ghats: influence of habitat variables.
Acta Herpetologica 4, 153-160.
Glos, J., Grafe, T.U., Rödel, M.-O., Linsenmair, K.W., 2003. Geographic variation in pH
tolerance of two populations of the European common frog, Rana temporaria.
Copeia 2003, 650-656.
Goldberg, F. J., Quinzio, S., Vaira, M., 2006. Oviposition-site selection by the toad
Melanophryniscus rubriventris in an unpredictable environment in Argentina.
Canadian Journal of Zoology 84, 699-705.
Gomez-Mestre, I., Tejedo, M., Ramayo, E., Estepa, J., 2004. Developmental Alterations
and osmoregulatory physiology of a larval anuran under osmotic stress.
Physiological and Biochemical Zoology 77, 267-274.
27
Gosner, K. L., Black, I. H., 1957. The effects of acidity on the development and hatching of
New Jersey Frogs. Ecology 38, 256-262.
Grand, J., Cushman, S. A., 2003. A multiple-scale analysis of species-habitat relationships:
breeding birds in a pitch pine-scrub oak (Pinus rigida-Quercus ilicifolia)
community. Biological Conservation 112 (3), 307-317.
Gregoire, D. R., Gunzburger, M. S., 2008. Effects of predatory fish on survival and
behavior of Larval Gopher Frogs (Rana capito) and Southern Leopard Frogs (Rana
sphenocephala). Journal of Herpetology 42, 97-103.
Griffith, S. J., Bale, C., Adam, P., 2008. Environmental correlates of coastal heathland and
allied vegetation. Australian Journal of Botany 56, 512-526.
Griffith, S. J., Bale, C., Adam, P., Wilson, R., 2003. Wallum and related vegetation on the
NSW North Coast: description and phytosociological analysis. Cunninghamia 8 (2),
202-252.
Griffith, S.J., Bale, C., Adam, P., 2004. The influence of fire and rainfall on seedling
recruitment in sand mass (wallum) heathland of north-eastern New South Wales.
Australian Journal of Botany 52, 93-118.
Griffiths, R. A., Denton, J., Wong, A. L.-C., 1993. The effect of food level on competition
in tadpoles: interference mediated by Protothecan algae? Journal of Animal Ecology
62, 274-279.
Groves, R. H., 1981. Heathland soils and their fertility status. In Specht, R.L. Ecosystems
of the world 9B: heathlands and related shrublands; analytical studies. Elservier
Scientific Publishing Company, Amsterdam. Pp143-149.
Gunzburger, M. S., Travis, J., 2005. Critical literature review of the evidence of
unpalatability of amphibian eggs and larvae. Journal of Herpetology 39, 547-571.
Haddad, C. F. B., Prado, C. P. A., 2005. Reproductive modes in frogs and their unexpected
diversity in the Atlantic forest in Brazil. Bioscience Biotechnology and
Biochemistry 55 (3),
Halliday, T. R., 2008. Why amphibians are important. International Zoo Yearbook 42, 714.
Haramura, T., 2008. Experimental test of spawning site selection by Buergeria japonica
(Anura: Rhacophoridae) in response to salinity level. Copeia 2008, 64-67.
28
Hecnar, S. J., M'Closkey, R. T., 1997. The effects of predatory fish on amphibian species
richness and distribution. Biological Conservation 79, 123-131.
Henrikson, B. I., 1990. Predation on amphibian eggs and tadpoles by common predators in
acidified lakes. Holartic Ecology 13, 201-206.
Hero, J.-M., Gascon, C., Magnusson, W. E., 1998. Direct and indirect effects of predation
on tadpole community structure in the Amazon rainforest. Australian Journal of
Ecology 23, 474-482.
Hero, J.-M., Magnusson, W. E., Rocha, C. F. D., Catterall, C. P., 2001. Antipredator
defenses influence the distribution of amphibian prey species in the central Amazon
rain forest. Biotropica 33, 131-141.
Hero, J.-M., Morrison, C., 2004. Frog declines in Australia: Global implications.
Herpetological Journal 14 (4), 175-186.
Hero, J.-M., Morrison, C., Gillespie, G., Roberts, J.D., Newell, D., Meyer, E., McDonald,
K., Lemckert, F., Mahony, M., Osborne, W., Hines, H., Richards, S., Hoskin, C.,
Clarke, J., Doak, N. and Shoo, L. (2006). Overview of the conservation status of
Australian Frogs. Pacific Conservation Biology 12, 313-320.
Herrmann, H. L., Babbitt, K. J., Baber, M. J., Congalton, R. G., 2005. Effects of landscape
characteristics on amphibian distribution in a forest-dominated landscape.
Biological Conservation 123, 139-149.
Heyer, W. R., 1973. Ecological Interactions of Frog Larvae at a Seasonal Tropical Location
in Thailand. Journal of Herpetology 7, 337-361.
Heyer, W. R., McDiarmid, R. W., Weigmann, D. L., 1975. Tadpoles, predation and pond
habitats in the tropics. Biotropica 7, 100-111.
Hickman Jr., C. P., Roberts, L. S., Larson, A. 1998. Biology of amphibians, McGraw-Hill
Companies, United States of America.
Hines, H., Mahony, M., McDonald, K. (1999) An assessment of frog declines in wet
subtropical Australia. In: Campbell A (ed) Declines and Disappearances of
Australian Frogs. Environment Australia, Canberra, pp 44-63.
Hossie, T. J., Murray, D. L., 2010. You can't run but you can hide: refuge use in frog
tadpoles elicits density-dependent predation by dragonfly larvae. Oecologia 163,
395-404.
29
Ingram, G. J., Corben, C. J., 1975. The frog fauna of North Stradbroke Island, with
comments on the 'acid' frogs of the wallum. The Proceedings of the Royal Society
of Queensland 86, 49-54.
Jara, F. G., 2008. Tadpole–odonate larvae interactions: influence of body size and diel
rhythm. Aquatic Ecology 42, 503-509.
Jara, F. G., Perotti, M. G., 2008. Toad tadpole responses to predator risk: ontogenetic
change between constitutive and inducible defenses. Journal of Herpetology 43 (1),
82-88.
Johnson, J. R., Knouft, J. H., Semlitsch, R. D., 2007. Sex and seasonal differences in the
spatial terrestrial distribution of gray treefrog (Hyla versicolor) populations.
Biological Conservation 140, 250-258.
Kats, L. B., Ferrer, R. P., 2003. Alien predators and amphibian declines: review of two
decades of science and the transition to conservation. Diversity and Distributions
2003, 99-110.
Kats, L. B., Petranka, J. W., Sih, A., 1988. Antipredator defenses and the persistence of
amphibian larvae with fishes. Ecology 69, 1865-1870.
Kiesecker, J. M., Blausien, A. R., Swihart, R. K., 2001. Complex causes of amphibian
population declines. Nature 410, 681-684.
Kolozsvary, M. B., Swihart, R. K., 1999. Habitat fragmentation and the distribution of
amphibians: patch and landscape correlates in farmland. Canadian Journal of
Zoology 77,
Kopp, K., Wachlevski, M., Eterovick, P. C., 2006. Environmental complexity reduces
tadpole predation by water bugs. Canadian Journal of Zoology 84, 136-140.
Lane, M. J., Mahony, S. J., 2002. Larval anurans with synchronous and asynchronous
development periods: contrasting responses to water reduction and predator
presence. Journal of Animal Ecology 71, 780-792.
Lawler, K. L., Hero, J.-M., 1997. Palatability of Bufo marinus tadpoles to a predatory fish
decreases with development. Wildlife Research 24, 327-334.
Lemckert, L., Haywood, A., Brassil, T., Mahony, M. (2006). Correlations between frogs
and pond attributes in central New South Wales, Australia: What makes a good
pond? Applied Herpetology 3, 67-81.
30
Lewis, B. D., Goldingay, R. L., 2005. Population monitoring of the vulnerable wallum
sedge frog (Litoria olongburensis) in north-eastern New South Wales. Australian
Journal of Zoology 53, 185-194.
Loman, J., 1999. Early metamorphosis in common frog Rana temporaria tadpoles at risk of
drying: an experimental demonstration. Amphibia-Reptilia 20, 421-430.
Loman, J., Claesson, D., 2003. Plastic response to pond drying in tadpoles Rana
temporaria: test of cost models. Evolutionary Ecology and Research 5, 179-194.
Lowe, K., Hero, J.-M., 2012. Sexual dimorphism and color polymorphism in the Wallum
Sedge Frog (Litoria olongburensis). Herpetological Review 43, 236-240.
Mann, R., Bidwell, J. (1999) Toxicological issues for amphibians in Australia. In:
Campbell A (ed) Declines and Disappearances of Australian Frogs. Environment
Australia, Canberra, pp 185-201
Márquez-García, M., Correa-Solís, M., Méndez, M. A., 2010. Life-history trait variation in
tadpoles of the warty toad in response to pond drying. Journal of Zoology 281, 105111.
Marsh, D.M., Fegraus, E.H. and Harrison, S. (1999). Effects of breeding pond isoloation on
spatial and temporal dynamics of pond use by the tungara frog, Physalaemus
pustulosus. Journal of Animal Ecology 68, 804-814.
Marsh, D.M. and Trenham, P.C. (2001). Metapopulation dynamics and amphibian
conservation. Conservation Biology 15, 40-49.
Mazerolle, M. J., 2005. Peatlands and green frogs: A relationship regulated by acidity?
Ecoscience 12, 60-67.
McCollum, S. A., Leimberger, J. D., 1997. Predator-induced morphological changes in an
amphibian: predation by dragonflies affects tadpole shape and color. Oecologia 109,
615-621.
McDiarmid, R. W., Altig, R. (2010) Morphology of amphibian larvae. In: Dodd Jr. CK (ed)
Amphibian Ecology and Conservation: A Handbook of Techniques. Oxford, New
York, pp 39-53
Meyer, E. 2004. Acid adaptation and mechanisms for softwater acid tolerance in larvae of
anuran species native to the 'Wallum' of east Australia. PhD Thesis. University of
Queensland.
31
Meyer, E., Hero, J.-M., Shoo, L., Lewis, B. (2006) National recovery plan for the wallum
sedgefrog and other wallum-dependent frog species. Report to the Department of
the Environment and Water Resources, Canberra. Queensland Parks and Wildlife
Service, Brisbane,
Michel, M. J., Adams, M. M., 2009. Differential effects of structural complexity on
predator foraging behavior. Behavioral Ecology 20, 313-317.
Mokany, A., 2007. Impacts of tadpoles and mosquito larvae on ephemeral pond structure
and processes. Marine and Freshwater Research 20 (2), 313-317.
Mokany, A., Shine, R., 2003. Competition between tadpoles and mosquito larvae.
Oecologia 135, 615-620.
Moreira, L. F. B., Machado, I. F., Garcia, T. V., Maltchik, L., 2010. Factors influencing
anuran distribution in coastal dune wetlands in southern Brazil. Journal of Natural
History 44, 1493-1507.
Morin, P. J., 1986. Interactions between intraspecific competition and predation in an
amphibian predator-prey system. Ecology 67 (3), 713-720.
Morin, P. J., 1995. Functional redundancy, non-additive interactions, and supply-side
dynamics in experimental pond communities. Ecology 76, 133-149.
Newman, R. A., 1994. Effects of Changing Density and Food Level on Metamorphosis of a
Desert Amphibian, Scaphiopus couchii. . Ecology 75 (4), 1085-1096.
Noland, R., Ultsch, G. R., 1981. The roles of temperature and dissolved oxygen in
micorhabitat selection by the tadpoles of a frog (Rana pipiens) and a toad (Bufo
terrestris). Copiea 1981 (3), 645-652.
Orizaola, G., BraÑa, F., 2003. Do predator chemical cues affect oviposition site selection in
newts? Herpetological Journal 13, 189-193.
Orizaola, G., Laurila, A., 2009. Intraspecific variation of temperature-induced effects on
metamorphosis in the pool frog (Rana lessonae). Canadian Journal of Zoology 87,
581-588.
Oseen, K. L., Wassersug, R. J., 2002. Environmental factors influencing calling in
sympatric anurans. Oecologia 133, 616-625.
Pearson, O.P. and Bradford, D.F. (1976). Thermoregulation of lizards and toads at high
altitudes in southern Peru. Copeia 176, 155-170.
32
Persson, M., Räsänen, K., Laurila, A., Merilä, J., 2007. Maternally determined adaptation
to acidity in Rana arvalis: Are laboratory and field estimates of embryonic stress
tolerance congruent? Canadian Journal of Zoology 85, 832-838.
Pfennig, D. W., Murphy, P. J., 2002. How fluctuating competition and phenotypic plasticity
mediate species divergence. Evolution 56, 1217-1228.
Picker, M. D., McKenzie, C. J., Fielding, P., 1993. Embryonic Tolerance of Xenopus
(Anura) to Acidic Blackwater. Copeia 4, 1072-1081.
Pierce, B. A., 1985. Acid tolerance in amphibians. Bioscience 35, 239-243.
Pinder, A. W., Storey, K. B., Ultsch, G. R. (1992) Estivation and hibernation. In: Feder ME
and Burggren, W.W (ed) Environmental physiology of the amphibia. Chicago
Press, Chicago, pp 250-276
Polo-Cavia, N., Gonzalo, A., López, P., Martín, J., 2010. Predator recognition of native but
not invasive turtle predators by naïve anuran tadpoles. Animal Behaviour 80, 461466.
Ranvestal, A. W., Lips, K. R., Pringle, C. M., Whiles, M. R., Bixby, R. J., 2004.
Neotropical tadpoles influence stream benthos: evidence for the ecological
consequences of decline in amphibian populations. Freshwater Biology 49, 274285.
Räsänen, K., Söderman, F., Laurila, A., Merilä, J., 2008. Geographic variation in maternal
investment: acidity affects egg size and fecundity in Rana arvalis. Ecology 89,
2553-2562.
Relyea, R. a., 2003. Predators Come and Predators Go: the Reversibility of PredatorInduced Traits. Ecology 84, 1840-1848.
Relyea, R. A., 2007. Getting out alive: how predators affect the decision to metamorphose.
Oecologia 152, 389-400.
Resetarits Jr, W. J., Wilbur, H. M., 1989. Choice of oviposition site by Hyla chrysoscelis:
role of predators and competitors. Ecology 70, 220-228.
Retallick, R. W. R. 2000. Determinants of the assemblage structure of tadpoles in the
streams of Eungella National Park. James Cook University
Richards, S. J., 2002. Influence of flow regime on habitat selection by tadpoles in an
Australian rainforest stream. Journal of Zoology 257, 273-279.
33
Richter-Boix, A., Llorente, G. A., Montori, A., 2007. A comparative study of predatorinduced phenotype in tadpoles across a pond permanency gradient. Hydrobiologia
583, 43-56.
Rios-López, N., 2008. Effects of increased salinity on tadpoles of two anurans from a
Caribbean coastal wetland in relation to their natural abundance. Amphibia-Reptilia
29, 7-18.
Robinson, M., 2002. A field guide to frogs of Australia: from Port Augusta to Fraser Island
including Tasmania. Reed New Holland, Australia.
Rödel, M.-O., Linsenmair, K. E., 1997. Predator-induced swarms in the tadpoles of an
African savanna frog, Phyzomuntis mimps. Ethology 103, 902-914.
Rudolf, V. H. W., Rödel, M.-O., 2007. Phenotypic plasticity and optimal timing of
metamorphosis under uncertain time constraints. Evolutionary Ecology 21, 121142.
Sadinski, W. A., Dunson, W. J., 1992. A multilevel study of effects of low pH on
amphibians of temporary ponds. Journal of Herpetology 26, 413-422.
Saidapur, S. K., Veeranagoudar, D. K., Hiragond, N. C., Shanbhag, B. A., 2009.
Mechanism of predator–prey detection and behavioral responses in some anuran
tadpoles. Chemoecology 19, 21-28.
Schlichter, L., 1981. Low pH affects the fertilization and development of Rana pipiens
eggs. Canadian Journal of Zoology 59, 1693-1699.
Schmutzer, A. C., Gray, M. J., Burton, E. C., Miller, D. L., 2008. Impacts of cattle on
amphibian larvae and the aquatic environment. Freshwater Biology 53, 2613-2625.
Schoener, T. W., 1983. Field experiments on interspecific competition. The American
Naturalist 122, 240-285.
Scott, D. E., 1994. The effect of larval density on adult demographic traits in Ambystoma
opacum. Ecology 75, 1383-1396.
Semlitsch, R. D., Ryer, H. U., 1992. Performance of tadpoles from the hybridogenetic Rana
esculenta complex: interactions with pond drying and interspecific competition.
Evolution 46 (3), 665-676.
Seymour, R. S., Roberts, J. D., Mitchell, N. J., Blaylock, A. J., 2000. Influence of
environmental oxygen on development and hatching of aquatic eggs of the
34
Australian frog, Crinia georgiana. Physiological and Biochemical Zoology 73 (4),
501-507.
Shah, A.A., Ryan, M.J., Bevilacqua, E., Schlaepfer, M.A., 2010. Prior experience
alters the behavioral response of prey to a nonnative predator. Journal of
Herpetology 44, 185-192.
Shoemaker, V.H., Nagy, K.A., 1977. Osmoregulation in amphibians and reptiles. Annual
Review of Physiology 1977, 449-471
Simpkins, C. A., Meyer, E., Hero, J.-M., 2011. Long-range movement in the rare Cooloola
sedgefrog Litoria cooloolensis. Australian Zoologist 35, 977-978.
Skelly, D. K., 1994. Activity level and susceptibility of anuran larvae to predation. Animal
Behaviour 47, 465-468.
Skelly, D. K., 1995. A Behavioral Trade-Off and Its Consequences for the Distribution of
Pseudacris Treefrog Larvae. Ecology 76, 150-164.
Smallbone, LT., Luck, G.W. and Wassens, S. (2011). Anuran species in urban landscapes:
relationships with biophysical, built environment ans socio-economic factors.
Landscape and Urban Palnning 101, 43-51.
Smith, A. R., Awan, G. R., 2009. The roles of predator identity and group size in the
antipredator responses of American toad (Bufo americanus) and bullfrog (Rana
catesbeiana) tadpoles. Behaviour 146, 225-243.
Smith, D. C., 1987. Adult recruitment in chorus frogs: effects of size and date at
metemorphosis. Ecology 68, 344-350.
Smith, G. R., Burgett, A. A., Temple, K. G., Sparks, K. A., Winter, K. E., 2008. The ability
of three species of tadpoles to differentiate among potential fish predators. Ethology
114, 701-710.
Smith, M. J., Schreiber, E. S. G., Scroggie, M. P., Kohout, M., Ough, K., Potts, J., Lennie,
R., Turnbill, D., Jin, C., Clancy, T., 2007. Associations between anuran tadpoles
and salinity in a landscape mosaic of wetlands impacted by secondary salinisation.
Freshwater Biology 52, 75-84.
Snodgress, J. W., Komoroski, M. J., Bryan Jr., A. L., Burger, J., 2000. Relationships among
isolated wetland size, hydroperiod, and amphibian species richness: implications for
wetland regulations. Conservation Biology 14, 414-419.
35
Sparling, D. W. (2010) Water-quality criteria for amphibians. In: Dodd Jr. CK (ed)
Amphibian Ecology and Conservation: A Handbook of Techniques. Oxford
University Press, Oxford, pp 105-120.
Spieler, M., 2003. Risk of predation affects aggregation size: a study with tadpoles of
Phrynomantis microps (Anura: Microhylidae). Animal Behaviour 65, 179-184.
Specht, R.L., 1981. Conservation: Australian Heathlands. In Specht, R.L. Ecosystems of
the World 9B: Heathlands and Related Shrublands; Analytical Studies. Elsevier
Scientific Publishing Company, Amsterdam, pg 235-240.
Stav, G., Kotler, B. P., Blaustein, L., 2007. Direct and indirect effects of dragonfly (Anax
imperator) nymphs on green toad (Bufo viridis) tadpoles. Hydrobiologia 579, 85-93.
Steinberg, C. E. W., Kamara, S., Prokhotskaya, V. Y., Manusadzianas, L., Karasyova, T.
A., Timofeyev, M. A., Jie, Z., Paul, A., Meinelt, T., Farjalla, V. F., Matsuo, A. Y.
O., Burnison, B. K., Menzel, R., 2006. Dissolved humic substances – ecological
driving forces from the individual to the ecosystem level? Freshwater Biology 51,
1189-1210.
Stoneham, M. 2001. The influence of stream-dwelling predators on the distribution and
density of Mixophyes tadpoles in Southeast Queensland. Unpublished Honours
Thesis, Griffith University.
Strahan, R., 1957. The Effect of Salinity on the Survival of Larvae of Bufo melanostictus
Schneider. Copeia 1957, 146-147.
Stuart, S. N., Chanson, J. S., Cox, N. A., Young, B. E., Rodrigues, A. S. L., Fischman, D.
L., Waller, R. W., 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 306, 1783-1786.
Tarr, T. L., Babbitt, K. J., 2002. Effects of habitat complexity and predator identity on
predation of Rana clamitans larvae. Amphibia-Reptilia 23, 13-20.
Timofeyev, M. A., Shatilina, Z. M., Bedulina, D. S., Menzel, R., Steinberg, C. E. W., 2006.
Natural organic matter (NOM) has the potential to modify and multixenobiotic
resistance (MXR) activity in freshwater amphipods Eulimnogammarus cyaneus and
E. verrucosus. Comparative Biochemistry and Physiology, Part B 146, 496-503.
Toft, C. A., 1985. Resource Partitioning in Amphibians and Reptiles. Copeia 1985 (1), 121.
36
Touchon, J. C., Warkentin, K. M., 2008. Fish and dragonfly nymph predators induce
opposite shifts in color and morphology of tadpoles. Oikos 117, 634-640.
Twomey, E., Morales, V., Summers, K., 2008. Evaluating condition-specific and
asymmetric competition in a species-distribution context. Oikos 117, 1175-1184.
Tyler, M. J., 1997. The Action Plan for Australian Frogs. Widlife Australia Canberra, ACT.
Van Buskirk, J., Aschwanden, J., Buckelmüller, I., Reolon, S., Rüttiman, S., 2004. Bold tail
coloration protects tadpole from dragonfly strikes. 2004, 599-602.
Vonesh, J. R., Kraus, J. M., Rosenberg, J. S., Chase, J. M., 2009. Predator effects on
aquatic community assembly: disentangling the roles of habitat selection and postcolonization processes. Oikos 118, 1219-1229.
Wassersug, R. J., Seibert, E. A., 1975. Behavioral responses of amphibian larvae to
variation in dissolved oxygen. Copeia 1975 (1), 86-103.
Watt, P. J., Nottingham, S. F., Young, S., 1997. Toad tadpole aggregation behaviour:
evidence for a predator avoidance function. Animal Behaviour 54, 865-872.
Welborn, G. A., Skelly, D. K., Werner, E. E., 1996. Mechanisms creating community
structure across a freshwater habitat gradient. Annual Review of Ecology and
Systematics 27, 337-363.
Wells, K. D. 2007. The Ecology and Behavior of Amphibians, The University of Chicago
Press, USA
Wilbur, H. M., 1987. Regulation of structure in complex systems: experimental temporary
pond communities. Ecology 68, 1437-1452.
Wilbur, H. M., 1997. Experimental ecology of food webs: compleex systems in temporary
ponds. Ecology 78, 2279-2302.
Wiltshire, D. J., Bull, C. M., 1977. Potential competitive interactions between larvae of
Pseudophryne bibroni and P. semimarmorata (Anura: Leptodactylidae). Australian
Journal of Zoology 25, 449-454.
Wood, C. M., Matsuo, A. Y. O., Wilson, R. W., Gonzalez, R. J., Patrick, M. L., Playle, R.
C., Val, A. L., 2003. Protection by natural blackwater against disturbances in ion
fluxes caused by low pH exposure in freshwater stingrays endemic to the Rio
Negro. Physiological and Biochemical Zoology 76, 12-27.
37
Woodward, B. D., 1983. Predator-prey interactions and breeding-pond use of temporarypond species in a desert anuran community. Ecology 64, 1549-1555.
38
Chapter 2 - Environmental variables associated with the
distribution and occupancy of tadpoles in naturally acidic,
oligotrophic waterbodies
2.1 Abstract
Environmental factors play an integral role, either directly or indirectly, in structuring
faunal assemblages. Water chemistry, predation, hydroperiod and competition influence
tadpole assemblages within waterbodies. I surveyed aquatic predators, habitat refugia,
water height and water chemistry variables (pH, salinity and turbidity) at 37 waterbodies
over an intensive 22 day field survey to determine which environmental factors influence
the relative abundance and occupancy of anuran tadpole species in naturally acidic,
oligotrophic waterbodies within eastern Australian wallum communities. The majority of
tadpoles found were of Litoria olongburensis (Wallum Sedge Frog) and Crinia tinnula
(Wallum Froglet) species, both habitat specialists that are associated with wallum
waterbodies and listed as Vulnerable under the IUCN Red List. Tadpoles of two other
species (Litoria fallax (Eastern Sedge Frog), and Litoria cooloolensis (Cooloola Sedge
Frog)) were recorded from two waterbodies. Tadpoles of Litoria gracilenta (Graceful
Treefrog) were recorded from one waterbody. Relative abundance and occupancy of L.
olongburensis tadpoles were associated with pH and water depth. Additionally, L.
olongburensis tadpole relative abundance was negatively associated with turbidity.
Waterbody occupancy by C. tinnula tadpoles was negatively associated with predatory fish
and water depth and positively associated with turbidity. Variables associated with relative
abundance of C. tinnula tadpoles were inconclusive and further survey work is required to
identify these environmental factors. These results show that the ecology of tadpole species
associated with wallum waterbodies is complex and species specific. Therefore, several
environmental factors require consideration for successful management of waterbodies
where the conservation of threatened wallum amphibian communities is a priority. These
results may also be relevant in assisting scientists and managers in determining how
39
environmental variables are influencing tadpole distributions for species that are associated
with naturally acidic, oligotrophic waters around the globe.
2.2 Introduction
Environmental factors play an integral role, either directly or indirectly, in structuring
faunal assemblages (Krebs 2009). Different species will have varying tolerances to
environmental factors (e.g. reviewed in Pierce 1985; Berkelmans & Willis 1999; Schofield
& Nico 2009), with organisms occurring outside of their tolerance range often resulting in
death or deformities (Gosner & Black 1957; Sadinski & Dunson 1992; Schofield & Nico
2009). Therefore, organisms only occur in areas where environmental variables are within
their tolerance limits.
The anuran tadpole is the non-reproductive larval stage that occurs in different ecological
niches to the adult stage (Wells 2007) and is critical in the amphibian lifecycle with ‘the
potential to have the greatest impact on the continuing persistence of a population’ (Lane &
Mahony 2002). It is imperative for amphibian conservation to determine which
environmental factors influence individual abundance and occupancy of species within
amphibian larval assemblages. Several factors may structure tadpole assemblages,
including: intolerances to water chemistry (Smith et al. 2007; Wells 2007), hydroperiod
(Snodgrass et al. 2000; Baber et al. 2004; Moreira et al. 2010), predation (Hero et al. 1998;
Hero et al. 2001; Vonesh et al. 2009) and competition (Wiltshire & Bull 1977; Twomey et
al. 2008).
Water chemistry and hydroperiod can influence amphibian assemblages, with numerous
studies focusing on either water chemistry tolerance limits under laboratory conditions or
water chemistry variables influencing amphibian assemblages. These variables include
pH/acidity (reviewed in Freda 1986; Freda & Taylor 1992; Persson et al. 2007; Barth &
Wilson 2010) salinity (Christy & Dickman 2002; Chinathamby et al. 2006; Rios-López
2008) and turbidity, as represented by suspended sediment (Knutson et al. 2004).
Hydroperiod can structure the tadpole assemblage by excluding amphibian species with
40
long-lived tadpole stages, as these species can fail to metamorphose before desiccation
occurs from temporary waterbodies (Welborn et al. 1996)..
Aquatic predators have also been suggested as a primary factor structuring tadpole
assemblages (Heyer et al. 1975; Hero et al. 2001; Vonesh et al. 2009). The risk of
predation can be determined by a number of factors, including the tadpole’s vulnerability to
the predator assemblage or ability to avoid predators (i.e. via refugia usage) (Kopp et al.
2006; Saidapur et al. 2009). Competition may also structure tadpole communities by
altering rates of tadpole growth and development (Wilbur 1987; Mokany & Shine 2002;
Twomey et al. 2008), potentially leading to prolonged exposure to aquatic predators.
The coastal wallum vegetation along the eastern coast of Australia is the primary habitat
for two frog species (Litoria olongburensis and Crinia tinnula) that are listed as
Vulnerable under the IUCN Red List (IUCN 2011). Both species are referred to as ‘acid’
frog due to their association with low pH waters (Ingram and Corben 1975). For the
purposes of this study ‘wallum’ will be referred to as the vegetation communities within
the study area including Banksia woodland, sedgeland, heathland and Melaleuca
swamps (Hines et al 1999, Griffith et al. 2003) that contain soils that are often acidic
(pH 3.4-5.4) and low in nitrogen (Griffith et al. 2008) and phosphorus (Groves 1981).
‘Acid’ frog populations within protected areas are believed to be stable (Hines et al.
1999; Lewis and Goldingay 2005). However, populations of ‘acid’ frogs occurring
outside of protected areas are at risk from habitat loss (Hines et al. 1999), with more than
30% of original heathland and Melaleuca cover in south-east Queensland cleared
between 1974-1989 (Catterall and Kingston 1993, cited in Hines et al. 1999). It is
therefore vital that mangers know which environmental factors influence ‘acid’ frogs
within protected areas to ensure these environmental variables remain constant and
populations remain stable. This knowledge would also aid managers, scientists and
environmental consultants when evaluating suitable waterbodies for successful breeding
of L. olongburensis and C.tinnula and, thus, be able to prioritize areas for conservation
with greater accuracy. However, surveys to determine factors influencing L.
olongburensis and C. tinnula tadpole relative abundance and occupancy within protected
41
wallum heathland waterbodies have never been published in peer-reviewed journals.
Despite this, it has been suggested that pH (Ingram & Corben 1975) and introduced
predatory fish (i.e. Gambusia holbrooki) (Meyer et al. 2006) are two important factors
that may influence tadpole assemblages within these waterbodies.
I therefore investigated the influence of water chemistry factors (pH, salinity, turbidity),
water depth, aquatic predators (predatory fish and aquatic invertebrates), competitors
(tadpoles of other species) and tadpole refuge availability (represented by habitat
complexity), on the occupancy and relative abundance of L. olongburensis and C.
tinnula tadpoles across available habitats within protected areas of eastern Australia. I
also examine the optimal ranges for factors that were observed to influence the relative
abundance of L. olongburensis tadpoles.
2.3 Methods
Study Site Selection and Sampling Design
Nine national parks and reserves where L. olongburensis were known to occur were
selected between the Great Sandy National Park, Queensland (QLD) (26.014° S, 153.024°
E), and Wooli, New South Wales (NSW) (29.853° S, 153.263° E), in coastal wallum
vegetation communities on mainland Australia. Eleven ‘survey transects’ were established
across the national parks and reserves (one at each national park or reserve for all areas
except Bundjalung National Park and Yuraygir National Park which contained two survey
transect lines) to select waterbodies (Figure 2.1). With the exception of areas where no
national parks or protected areas occur (i.e. major cities), the survey transects were
established evenly across the known distributional range of L. olongburensis (Meyer et al.
2006) (Figure 2.1). Wallum vegetation communities often consist of several vegetation
habitats (Ingram & Corben 1975, Hines et al. 1999). Therefore, survey transects were
placed using QLD regional ecosystem and NSW vegetation G.I.S. layers to ensure all
vegetation communities were intersected at least once from one of the survey transects.
Additionally, historical locality records of L. olongburensis were used to ensure each
42
Figure 2.1: Localities of survey sites, with numbers representing the following localities: 1
– Cooloola Section of the Great Sandy National Park; 2- Noosa National Park; 3 –
Mooloolah National Park; 4 – Beerwah Scientific Reserve; 5 – Tyagarah Nature Reserve; 6
– Lennox Heads; 7 – Bunjalung National Park; 8 – Yuragir National Park (North); 9 –
Yuragir National Park (South). Black dots represent Litoria olongburensis record localities
from EPA/QPWS, NSWDEC, the Australian Museum, Queensland Museum, South
Australian Museum, and various biologists (Meyer et al. 2006). Solid lines represent
Australian coastline and the Queensland / New South Wales state border. Map of Australia
shows enlarged area within the rectangle, with solid lines representing the Australian
coastline and the Australia’s state and territory borders.
43
survey transect intersected a minimum of one waterbody where L. olongburensis had been
recorded. Survey transects spanned from one perimeter of the protected area to the opposite
perimeter and, therefore, varied in length. Anthropogenic disturbances (i.e. urban
development) outside of national parks may influence environmental factors like water
depth/hydrology, increase in nutrients or change in water chemistry (Meyer et al. 2006).
Therefore, national parks and reserves were selected for surveying in an attempt to
minimize the influence of anthropogenic disturbances, which are apparent throughout the
range of L. olongburensis (Meyer et al. 2006).
Survey transects were then uploaded onto a Trimble® Juno™ SC unit for ground-based
surveys targeting waterbodies. A maximum of two days was allowed for surveying of each
survey transect.
Waterbodies were sampled if the survey transect intersected the waterbody or if the
waterbody could be seen from the survey transect. Each waterbody had a ‘waterbody
transect’ that was perpendicular to the edge of the waterbody, and went from one edge of
the waterbody to the opposite edge. Where practical, the waterbody transect was placed
through the deepest section of the waterbody. Waterbody transects varied in length
depending on the waterbody size, with the shortest and longest waterbody transect being 48
meters and 200 meters, respectively.
Sites with large waterbodies containing heterogeneous vegetation had multiple waterbody
transects to capture this variation. In these sites, primarily Bundjalung National Park and
Yuraygir National Park, separate transects were placed a minimum of 10m apart in each
distinct waterbody vegetation type to reduce edge effects and maintain independence.
All waterbody transects were surveyed over a 22 day period, in March 2010, to minimize
the influence of temporal variation. This time of year was chosen as peak activity calling
for L. olongburensis was predicted to occur in February and March (Hopkins unpublished
data) and it was assumed that tadpoles could be present from eggs deposited in February.
44
This time also coincided with another survey that aimed at determining what environmental
factors influenced the relative abundance of L. olongburensis adults (Shuker 2012).
It must be acknowledged that, due to the timing of the survey, winter breeding species may
have been underepresented or missed during this study.
Dipnetting of aquatic fauna and measurements of water and vegetation characteristics were
undertaken at five evenly distributed sampling points along the waterbody transects.
Sampling point placement began 5 meters from the waterbody transect edge. Diurnal
dipnetting was conducted using a circular net, with an aperture of approximately 30
centimetres (cm) in diameter and with mesh size less than 0.5 millimetres to capture
tadpoles and aquatic predators. In an attempt to sample the entire water column three water
column levels (bottom, middle and top) were dipnetted, with five sweeps at each level. To
ensure no recaptures occurred animals were not released until dip-netting had been
completed at each sampling point. It was assumed sampling points were sufficiently
spaced (minimum distance between sampling points was 9.5 m) to prevent recaptures
between successive sampling points.
Waterbody pH, salinity, turbidity, depth (as a representation of hydroperiod) and percent
vegetation cover were also measured. A handheld TPS Aqua-CPA Conductivity-TDS-pHTemperature Meter (version 1.2) was used to measure pH and salinity. Calibration occurred
between survey transects to ensure accurate measurements. A 2 point calibration curve was
used at 4.00 and 6.88 for pH calibration, while 2.00 parts per thousand (ppt) and 0.00 ppt
were used for salinity calibration. Turbidity was measured by placing a black and white
marker into the bottom of a transparent turbidity tube. Water was then added to the
turbidity tube until the black and white marker could not be seen when looking directly
down into the tube. The height of the water within the turbidity tube was then correlated to
the turbidity value for the waters height. Water for turbidity measurements was sampled
approximately 10 cm below the water’s surface. Water was sampled as close to the waters
surface as possible when water depth fell below 10cm. Water depth was measured to the
nearest centimetre in the centre of each sampling point. I compare waterbody size using the
45
waterbody transect lengths due to inadequate detail in available GIS layers, the size of the
waterbodies and dense vegetation making it impractical to measure waterbody area during
the surveys.
Aquatic sedge and herb cover was visually estimated using an estimated 5m x 5m quadrate
centered on each sampling point. Percent cover was combined for each waterbody transect
and divided by the number of sampling points (5) to give representation of percent cover
for each waterbody. Aquatic sedge and herb cover was defined as the percentage of the
quadrate occupied by the vertical projection of foliage and was used as an estimate of
habitat complexity.
Rainfall data between December 2009 – March 2010 was obtained from the Australian
Bureau of Meteorology Weather Stations located at Double Island Point Lighthouse
(Weather Station # 040068), Beerburrum Forest Reserve (Weather Station # 040284),
Byron Bay Cap Byron Lighthouse (Weather Station # 058009) and Wooli Beach (Weather
Station # 058080). The rainfall averages for each month were obtained for all years that the
weather stations had been in service.
Statistical Analysis
Water depth, water chemistry factors and all species of aquatic predators obtained from
dipnetting were combined for each waterbody transect. One waterbody located next to an
estuarine creek system contained an extreme mean salinity level of 420 parts per million
and had recently been disturbed by fire, and was removed from the analysis. Water depth
and water chemistry factors were divided by the number of sampling points (5) to give a
mean per waterbody transect for each water chemistry variable. Tadpole species were
considered present from a waterbody transect if a species was recorded from one of the five
sampling points. Tadpole numbers were added together for each waterbody transect for
each species for relative abundance analysis. Tadpoles used for analysis were of Gosner
Stage 25 or later due to difficulties identifying tadpoles in earlier Gosner Stages. Fish from
the species Gambusia holbrooki (Mosquito Fish), Rhadinocentrus ornatus (Ornate
46
Rainbow Fish), Hypseleotris galii (Firetail Gudgeon) and Nannoperca oxleyana (Oxleyan
Pygmy Perch) were grouped into the ‘predatory fish’ category to estimate relative fish
abundance for each waterbody transect. Predatory aquatic invertebrates (Belostomatidae
(Giant Water Bugs) and Aeshnidae (Dragonfly families) were excluded from analysis as
they were only detected from two transects.
To determine which variables were correlated a Spearman Rank Correlation Test was
performed in IBM SPSS Statistics Version 19 (SPSS, Inc., 2009, Chicago). Variables that
had a correlation coefficient value greater than or equal to 0.7 (sensu Babbitt et al. 2003;
Garden et al. 2007) were identified as being highly correlated. No variables were highly
correlated, and thus all variables were used in the analyses (Table 2.1).
Models focusing on the influence of environmental variables on abundance and occupancy
of L. olongburensis and C. tinnula were constructed a priori. Generalized linear models
were used to assess the importance of environmental variables on relative abundance
models (using a Poisson link function) and occupancy models (using a binomial link
function). I used a generalized ‘rule of thumb’ of n/3 (where n = number of waterbodies
sampled) to obtain the maximum number of predictor variables to use in each model
(Crawley 2007). Predictor variables included in L. olongburensis and C. tinnula abundance
and occupancy models included vegetation cover, predatory fish abundance, pH, salinity,
turbidity and depth.
Predation levels can be influenced by the availability of aquatic refuge (i.e. vegetation
cover) (Babbitt & Tanner 1998; Kopp et al. 2006).Additionally, it has been proposed that
water clarity may impact on predation success (Spieler 2003). Therefore, interaction
models between these three variables were included in analysis. Tadpoles of other species
were excluded from the models due to the low number of waterbodies in which they were
detected. Factors that relate to physiological tolerances often produce a unimodal
distribution (e.g. Austin 1999). To detect any unimodal responses, some models included
quadratic terms for pH and depth.
47
Table 2.1: Comparison of habitat characteristics for surveyed waterbodies in wallum
habitats of eastern Australia. Spearman correlation coefficients (SCC) were compared for
37 waterbody transects. None of the variables were considered highly correlated (SCC ≥
0.7).
pH
Salinity
Depth
Turbidity
Vegetation
Cover
Predatory
Fish
0.290
-0.331
0.576
-0.168
0.025
pH
Salinity
Depth
Turbidity
-0.281
0.178
-0.357
-0.108
0.016
0.426
-0.110
-0.042
0.206
-0.086
Generalised linear models using Akaike’s Information Criterion (AICc to adjust for small
sample size (n =37)) were performed in the freeware statistical package R (R Core
Development Team , 2011) to determine model ranking and selection. The ‘best’ model
was the model with the lowest AICc value (Burnham & Anderson 2002). To determine the
ranking of the models, Δi values were calculated, where higher Δi values indicated less
accurate models for the given data (Burnham & Anderson 2002; Johnson & Omland 2004).
If a model had a Δi ≤ 2, then there was considerable evidence that the model could be the
“best” model, given the data (Johnson & Omland 2004). If a model had a Δi 2-4 then there
was considered to be moderate evidence that the model could be the “best” model, given
the data. Akaike Weights (wi) enables greater interpretation of the relative likelihood of a
model given the data (Burnham & Anderson 2002; Johnson & Omland 2004). Therefore,
each model was assigned a wi, which was used to determine the “probability that model i is
the best model for the observed data, given the candidate set of models” (Johnson &
Omland 2004). The closer the wi was to 1, the closer the model for the given data (Burnham
& Anderson 2002). To determine the relative importance of variables within models where
Δi < 4 the wi values were summed from all models where the variable of interest occurred
(Grueber et al. 2011). The closer the variable of interest was to 1 the higher the importance
of the variable. Twenty models were chosen to model L. olongburensis and C. tinnula
tadpole relative abundance and occupancy.
48
Quantile regressions (Cade & Noon 2003; Lancaster & Belyea 2006) were performed in the
statistical program R (R Core Development Team, 2011) using the quantreg package
(version 4.62) in an attempt to determine the optimal range for factors having a unimodal
distribution that were related to L. olongburensis tadpole relative abundance. The number
of waterbodies sampled limited the upper and lower quantiles that could be fitted to the
data. Therefore, the 0.85 quantile was used to predict the upper limits for relative
abundance with 95% confidence intervals. The 0.65 quantile was also plotted to determine
shape consistency with the 0.85 quantile. A Gaussian bell-shaped response was assumed for
each model.
2.4 Results
Waterbody Characteristics and Rainfall Conditions
A total of 37 waterbody transects were surveyed. Five tadpole species were encountered
throughout the survey: L. olongburensis, C. tinnula, Litoria gracilenta, Litoria fallax and
Litoria cooloolensis. Waterbody transects surveyed had mean pH ranging between 3.00 –
5.04, salinity ranging between 3.32-108.3 parts per million (ppm), turbidity ranging
between 0 – 40 nephelometric turbidity units (NTU), mean water depth ranging between
6.2 – 36.2 cm and relative fish abundance between 0 – 67. Tadpoles of L. olongburensis
were recorded from 11 waterbodies where the maximum number of individual L.
olongburensis tadpoles caught for an individual waterbody transect was nine. Waterbodies
containing L. olongburensis tadpoles had a mean pH ranging from 3.40 – 4.34, salinity
between 35.82 - 93.72 ppm, turbidity between 0 – 20.4 NTU, mean water depth ranging
between 10.2 – 30 cm and relative fish abundance between 0 – 4 individuals. Tadpoles of
C. tinnula were recorded from 14 waterbody transects where the maximum number of
individual C. tinnula tadpoles caught for an individual waterbody transect was 44.
Waterbodies with C. tinnula tadpoles had mean pH ranging between 3.35 – 4.84, salinity
between 35.82 – 99.10 ppm, turbidity between 0 - 26.2 NTU, mean water depth ranging
between 7.2 – 27.4 cm and relative fish abundance between 0 – 4 individuals. Gambusia
holbrooki occurred at four waterbody transects. Tadpoles of L. olongburensis and C.
tinnula were not recorded in waterbodies with G. holbrooki. Both L. fallax and L.
49
cooloolensis were recorded from two separate waterbodies. Litoria gracilenta occurred at
one waterbody. Tadpoles of L. fallax were not recorded with L. cooloolensis, L.
olongburensis or C. tinnula tadpoles. Tadpoles of L. cooloolensis and C. tinnula were
found co-occurring in 3 waterbodies.
The average rainfall for the months between December 2009 – March 2010 differed
between weather stations, with above average rainfall being recorded for the Beerburrum
Forest Reserve (+219mm) and Byron Bay Cap Byron Lighthouse (+202.2mm) weather
stations and below average rainfall for the Double Island Point Lighthouse (-210.9mm) and
Wooli Beach (-87.9mm) weather stations.
Abundance
Three models for L. olongburensis relative abundance had a Δi < 2. The weighting of the
best models for L. olongburensis tadpole relative abundance was 99.8%, suggesting that the
other models compared poorly (Table 2.2). The best model contained pH, depth and
turbidity, with all variables having strong relative importance (Table 2.3). Turbidity was the
only factor within the best model that was negatively associated with the relative
abundance of L. olongburensis tadpoles (Table 2.2). The other two models contained all
variables measured. Other variables within these models had lower relative variable
importance when compared with pH, depth and turbidity (Table 2.3).
Unlike L. olongburensis, C. tinnula tadpole relative abundance was best explained
(weighting of 100%), by the model with all predictor factors (Table 2.2).
Occupancy
Two models for L. olongburensis tadpole occupancy had a Δi < 2 while only one model had
a Δi between 2-4. The combined weighting of the models with a Δi < 2 for L. olongburensis
occupancy was 70%, suggesting that the other models compared poorly given the data.
50
Table 2.2: Comparison of waterbody characteristics associated with the relative abundance
and occupancy of L. olongburensis or C. tinnula tadpoles in eastern Australia. Aikiki
models with Δi values less than 4 are presented. + indicates a positive relationship while –
indicates a negative relationship to L. olongburensis or C. tinnula tadpole relative
abundance or occupancy. Variables with a 2 indicate a unimodal distribution with L.
olongburensis or C. tinnula tadpole relative abundance or occupancy.
Model
Litoria olongburensis tadpole relative
abundance
pH – pH2 + Depth – Depth2 – Turbidity
pH - pH2 + Salinity + Depth - Depth2 Turbidity - % Cover
pH - pH2 + Salinity + Depth - Depth2 Turbidity - Predatory Fish - % Cover
Litoria olongburensis tadpole
occupancy
pH – pH2
pH – pH2 + Depth – Depth2
Depth – Depth2
Crinia tinnula tadpole relative
abundance
pH - pH2 - Salinity + Depth - Depth2 +
Turbidity - Predatory Fish - % Cover
Crinia tinnula tadpole occupancy
(-) Predatory Fish
(-) Depth
(-) Predatory Fish + Turbidity
Depth – Depth2
Turbidity
Predatory Fish * Depth
Salinity
Predatory Fish * Turbidity
(-) % Cover
(-) % Cover – Predatory Fish + Turbidity
51
AICc
Δi
wi
126.2
126.9
0
0.64
0.444
0.322
127.5
1.30
0.232
43.2
44.5
46.9
0
1.34
3.69
0.463
0.237
0.073
472.4
0
1
48.1
48.8
49.9
50.2
51.7
51.8
51.9
51.9
52
52.1
0
0.67
1.78
2.10
3.52
3.66
3.77
3.8
3.84
3.99
0.257
0.184
0.106
0.09
0.044
0.041
0.039
0.038
0.038
0.035
Table 2.3: Relative importance of waterbody characteristics associated with the relative
abundance and occupancy of L. olongburensis or C. tinnula tadpoles in eastern Australia.
Model averaged coefficients and relative importance of each environmental predictor for
models where Δi < 4 for L. olongburensis relative abundance and occupancy and C. tinnula
occupancy are displayed.
Variable
Litoria olongburensis
abundance
pH
pH2
Depth
Depth2
Turbidity
% Cover
Salinity
Predatory Fish
Litoria olongburensis
occupancy
pH
pH2
Depth
Depth2
Crinia tinnula occupancy
Predatory Fish
Depth
Depth2
Turbidity
Salinity
% Cover
Predatory Fish * Turbidity
Predatory Fish * Depth
Estimate
S.E. †
Confidence
Interval
Rel. var.
imp.‡
59.5
-7.47
0.4
-0.01
-0.026
-0.017
0.0243
-0.137
16.5
2.08
0.127
-0.003
-0.03
0.009
0.012
-0.096
26.02, 92.994
-11.7, -3.247
-0.143, 0.663
-0.018, -0.004
-0.087, 0.034
-0.036, 0.002
-0.0002, 0.049
-0.334, 0.06
1
1
1
1
1
0.55
0.55
0.23
54.689
-7.33
0.625
-0.018
34.155
4.158
0.393
0.009
-14.24, 123.62
-15.76, 1.1
-0.167, 1.418
-0.037, -0.004
0.76
0.72
0.36
0.33
-0.395
-0.013
-0.006
0.031
0.01
-0.008
-0.027
0.022
0.525
0.166
0.007
0.038
0.016
0.015
0.054
0.041
-1.459, 0.67
-0.346, 0.319
-0.02, 0.007
-0.046, 0.108
-0.023, 0.043
-0.038, 0.022
-0.137, 0.082
-0.061, 0.104
0.48
0.38
0.11
0.28
0.07
0.07
0.04
0.04
† Standard Error ‡ Relative variable importance
52
Models with a Δi < 2 contained pH and depth, with pH having the highest relative
variable importance (Table 2.3). Both variables were unimodel in their influence on L.
olongburensis occupancy (Table 2.2).
Three models for C. tinnula tadpole occupancy had a Δi < 2 while seven models had a
Δi between 2-4. The combined weighting for models with a Δi < 2 for C. tinnula
tadpole occupancy was 54.7%. These best models contained predatory fish, turbidity
and depth. Predatory fish and depth had the highest relative variable importance when
compared with other variables (Table 2.3). Predatory fish and depth were negatively
associated with C. tinnula tadpole occupancy, while turbidity was positively associated
with C. tinnula tadpole occupancy (Table 2.2).
Optimal Range
Determining an optimal range for factors influencing L. olongburensis tadpole
abundance is difficult due to large 95% confidence intervals in the maximum (0.85
quantile) abundance models. However, quantile regressions showed a unimodal
response between L. olongburensis tadpole relative abundance and water depth and pH
(Figure 2.2 and 2.3). Confidence intervals for pH are large for the 0.85 quantile and are
lowest towards the lower and upper pH ranges. Additionally, confidence intervals are
smallest towards the deeper water depths for the 0.85 quantile and largest towards
shallower water depths.
The Gaussian distribution fits well for the 0.85 and 0.65 pH quantiles, with p < 0.001
for all co-efficient values. The intercept of the 0.85 and the 0.65 water depth quantiles
have p > 0.05, indicating that intercepts for these quantiles do not fit a Gaussian
distribution. However the Depth and Depth2 for the 0.85 quantiles fit a Gaussian
distribution with co-efficient values having p < 0.05. The 0.65 water depth quantile
lacks a Gaussian distribution fit, with p > 0.05 for all co-efficient values (Table 2.4).
53
Figure 2.2: ‘Jitter’ plots for quantile regressions of the 0.85(a) and 0.65(b) quantiles
(solid line) and the 95% confidence intervals (dotted lines) for mean waterbody pH and
relative abundance of Litoria olongburensis tadpoles. Circles represent waterbody
transects.
54
Figure 2.3: ‘Jitter’ plots for quantile regressions of the 0.85(a) and 0.65(b) quantiles
(solid line) and the 95% confidence intervals (dotted lines) for mean waterbody depth
and relative abundance of Litoria olongburensis tadpoles. Circles represent waterbody
transects.
55
Table 2.4: Coefficients of the 0.85 and 0.65 regression quantiles where the independent
factors were mean pH and mean water depth. Litoria olongburensis tadpoles were the
dependant factor within the regression quantile models.
Quantile values for Litoria olongburensis and pH
S.E.†
Parameter
Estimate
t
0.85 quantile
Intercept
-29.74462
7.03260
-4.22953
pH
15.81659
3.40438
4.64595
pH2
-1.96724
0.40236
-4.88921
0.65 quantile
Intercept
-29.28921
12.66667
-2.31231
pH
15.09734
6.11008
2.47089
2
pH
-1.84790
0.71455
-2.58611
Quantile values for Litoria olongburensis and water depth
Parameter
0.85 quantile
Intercept
Depth
Depth2
0.65 quantile
Intercept
Depth
Depth2
† Standard Error
56
p-value
0.00017
0.00005
0.00002
0.02695
0.01865
0.01416
Estimate
SE†
t
p-value
-1.66692
0.35474
-0.00853
1.05192
0.08616
0.00167
-1.58465
4.11721
-5.11025
0.12230
0.00023
0.00001
-0.79584
0.15177
-0.00378
0.50358
0.08050
0.00186
-1.58036
1.88529
-2.02956
0.12328
0.06796
0.05029
2.5 Discussion
I present the first quantitative assessment of environmental variables associated with the
relative abundance and occupancy of tadpoles within waterbodies of wallum heathland
of eastern Australia. As expected, waterbodies in wallum vegetation were dominated by
tadpoles of the “acid frog” (L. olongburensis and C. tinnula), with the abundance and
occupancy of these species associated with either water chemistry variables or predatory
fish. Additionally, tadpoles of L. fallax and L. cooloolensis were also recorded from two
independent waterbodies. The low occupancy recorded for L. cooloolensis is likely due
to surveys only being conducted within a proportion of L. cooloolensis distributional
range. However, the low occupancy of L. fallax tadpoles is likely explained by this
species inability to successfully metamorphose in acidic waters (pH = 3.5) (Meyer
2004).
Variables influencing abundance and occupancy
Waterbody pH and water depth were associated with the relative abundance and
occupancy of L. olongburensis tadpoles. The range of pH values of waterbodies where
L. olongburensis tadpoles were found falls within the pH ranges of other studies that
found L. olongburensis adults between pH 3.8 – 4.6 (Hopkins unpublished data), 3.5 –
5.2 (Hero unpublished data) and 3.11 – 5.02 (Shuker unpublished data). While L.
olongburensis tadpole pH tolerance has never been tested, two other ‘acid’ frogs (C.
tinnula and L. cooloolensis) have successfully metamorphosed when exposed to pH
waters of 3.5, 4.5 and 6.5 (Meyer 2004). It is expected that L. olongburensis tadpoles
would also be able to metamorphose when exposed to these pH levels as tadpoles of L.
olongburensis co-occurred with C. tinnula and L. cooloolensis tadpoles within this
study.
Large confidence intervals and the quantile ranges extending over a majority of the
response variables tested (pH and depth) give a guide to the optimal range for L.
olongburensis tadpole abundance. Despite large confidence intervals, the 0.85 and 0.65
quantile for pH showed a unimodal response, suggesting that pH is influencing L.
olongburensis as a quadratic, nonlinear response. The unimodal relationship observed
57
suggests that competition, predation or a physiological intolerance is limiting this
species at the upper or lower bounds of the unimodal relationship. Acidity intolerance
has been recorded for numerous amphibian species, where death occurs to tadpoles
outside of their pH tolerance range (reviews in Pierce 1985 & Freda 1986; Meyer 2004).
Therefore, physiological intolerance of L. olongburensis tadpoles to pH levels occurring
at the lower limits of the pH range within this study is probable. However, experimental
tests on L. olongburensis tadpole tolerance using pH levels lower than those used by
Meyer (2004) are required to confirm this conclusion. Towards the upper limits of the
pH quantile regressions it is possible that competition from ‘non-acid’ frog species (i.e.
L. fallax) may influence the abundance of L. olongburensis tadpoles. This is difficult to
confirm in this study as only two waterbodies recorded a potential competitor (L. fallax)
and pH within waterbodies surveyed did not exceed 5.21. It is unlikely that ‘non-acid’
frog species would successfully compete and exclude or reduce L. olongburensis
tadpoles in the lower pH ranges as ‘non-acid’ frog tadpoles (i.e. L. fallax and Crinia
parinsignifera) have 100% hatchling mortality or 0% hatching success when exposed to
pH levels of 3.5 (Meyer 2004).
The 0.85 quantile for water depth showed a unimodal response, suggesting that water
depth is influencing L. olongburensis in a quadratic and not a linear response. Quantile
regression models for L. olongburensis tadpoles and water depth showed that tadpole
relative abundance will increase with water depth until water depth reaches
approximately 22cm, after which tadpole relative abundance will decrease. Water depth
was used as an indicator of hydroperiod, with deeper waterbodies representing
waterbodies with longer hydroperiods. The reduced number of L. olongburensis
tadpoles in deeper waterbodies is likely associated with an increase predatory fish,
which need a more permanent hydroperiod for population persistence, or a change in the
predator assemblage. Predatory fish were not in the best models explaining L.
olongburensis tadpole abundance or occupancy. However, predatory fish had the
highest correlation coefficient with depth when compared with other measured
variables. Additionally, predatory fish only co-occurred with L. olongburensis tadpoles
in two waterbodies, suggesting either coexistence between predatory fish and L.
olongburensis tadpoles is low or abundance of L. olongburensis tadpoles is lower when
coexisting with predatory fish. Aquatic fauna assemblages (Welborn et al. 1996;
58
Babbitt et al. 2003; Jocqué et al. 2007; Richter-Boix et al. 2007; Fernandes et al. 2010)
and an increase in predator size (Welborn et al. 1996; Richter-Boix et al. 2007) are
associated with increased hydroperiod or water depth. Therefore, an increase in
hydroperiod/depth may result in the number of L. olongburensis tadpoles being reduced
due to inadequate anti-predator strategies to larger predators (i.e. tadpole size may be an
inadequate anti-predator strategy to predators with large mouth-gapes) occurring in
deeper sections of a waterbody. Alternatively, deeper waterbodies may have reduced the
probability of detecting tadpoles due to an increase in the volume of water. Decreasing
hydroperiod has been shown to induce early metamorphosis (Loman 1999; Lane &
Mahony 2002; Loman 2002; Márquez-García et al. 2010) and tadpoles of L.
olongburensis may have metamorphosed in shallower waterbodies in an attempt to
metamorphose before waterbody desiccation. Alternatively, shallower waterbodies may
never have contained tadpoles of L. olongburensis. Water height has been found to
influence oviposition-site selection (Goldberg et al. 2006) and adult L. olongburensis
may have chosen deeper waterbodies with a lower desiccation risk to lay their eggs.
Turbidity was positively associated with C. tinnula tadpole occupancy and negatively
associated with L. olongburensis tadpole abundance. Numerous factors may have
caused turbidity, including natural organic acids (NOA), which produce dark, tanninstained waters (Barth & Wilson 2010), and/or suspended sediment (Walling 1977).
NOA has been attributed to an increase in growth and locomotor performance in
Limnodynastes peronii tadpoles (Barth & Wilson 2010) and other protective
mechanisms in various aquatic species (Gonzalez et al. 2002; Wood et al. 2003) when
exposed to acidic conditions. Similar, positive, processes could be occurring with C.
tinnula tadpoles. Alternatively, reduced tadpole numbers have been observed or
proposed with an increased turbidity in previous surveys (Gillespie 2002; Knutson et al.
2004; Schmutzer et al. 2008), with several hypothesis suggested as to the cause of the
decline, including suffocation of tadpole eggs (hypothesised from suffocation of fish
eggs in waters with high turbidity) or reduced ability to forage (Schmutzer et al. 2008).
Similar, negative, processes could be occurring with L. olongburensis tadpoles.
There were several waterbody transects that failed to detect L. olongburensis tadpoles
despite being within the optimal pH and water depth ranges. The adult surveys that
59
coincided with these surveys found a majority of waterbodies to have L. olongburensis
adults present (Shuker unpublished data). The absence of tadpoles from these
waterbodies suggests that eggs may have been deposited but did not survive long
enough, or were in such low densities that they were not detected. As mentioned
previously, females also have the ability to choose their oviposition site (Goldberg et al.
2006) and females may have chosen alternative spawning sites due to an unsuitable
environmental factor. Alternatively, tadpoles occurring in larger, deeper waterbodies
may have been missed due to an increase in the sampling area in larger waterbodies and
tadpoles potentially being dispersed over a larger area.
Predatory fish were low in occupancy and abundance throughout the survey. Hence, the
following conclusions must be interpreted with some caution. Having said this, the
given data showed that predatory fish were negatively associated with C. tinnula
tadpole occupancy. Predatory fish have the ability to exclude tadpole species that lack
adequate anti-predator defences from waterbodies (Kats et al. 1988; Hero et al. 1998,
2001; Gillespie 2001). Species that lack efficient anti-predatory fish defences (e.g.
unpalatability (Kats et al. 1988; Hero et al. 2001) or reduced mobility (Woodward
1983) may therefore be confined to waterbodies with ephemeral hydroperiods where
fish are often excluded due to waterbody desiccation (Heyer et al. 1975). Therefore, the
negative association between predatory fish and C. tinnula tadpole occupancy is likely
explained by lack of efficient anti-predatory fish strategies.
The model that best explained C. tinnula relative abundance was the model with all
variables, suggesting there are complex interactions occurring between all variables and
C. tinnula tadpole relative abundance. Further survey work and experimentation are
required to determine the key factors influencing C. tinnula tadpole relative abundance.
Implications for Conservation
The lack of co-occurrence of C. tinnula tadpoles and low co-occurrence of L.
olongburensis with fish suggests that introduction of fish (both native and exotic) into
wallum waterbodies would be detrimental to populations of these two threatened ‘acid’
frog species. Therefore, it is imperative that introduction of fish into wallum
60
waterbodies is avoided. For example, establishment of permanent waterbodies, where
fish would likely have the highest chance of population persistence, may result in fish
populations spreading to fish naive waterbodies during flooding events. This scenario
has to be taken into consideration if construction of ‘permanent’ waterbodies are
undertaken to reduce the effects of climate change (Shoo et al. 2011) on ‘acid’ frog
populations. Additionally, establishment of permanent waterbodies is of particular
concern if waterbodies contain G. holbrooki, a known introduced, predatory fish within
Australia (Pyke & White 2000; Reynolds 2009; Hunter et al. 2011), which is capable of
spreading between waterbodies in surface water runoff (Baber et al. 2002). The impact
of G. holbrooki on L. olongburensis and C. tinnula abundance and presence is difficult
to assess from this study, as only four waterbodies contained G. holbrooki. However, G.
holbrooki did not occur in the same waterbodies with L. olongburensis and C. tinnula
tadpoles, suggesting that G. holbrooki could negatively influence these amphibian
populations.
The low occupancy of L. fallax tadpoles across waterbodies indicates L. fallax do not
breed successfully in most waterbodies within wallum heathland. This is of particular
importance for conservation of waterbodies within wallum heathland, as changed
waterbody chemistry could result in conditions that are favourable for the generalist L.
fallax, which has been suggested as a competitive species to the threatened specialist L.
olongburensis (Meyer et al. 2006). Stormwater and runoff from urban areas and golf
courses have been proposed as potentially altering wallum waterbody chemistry (Meyer
et al. 2006) and, therefore, these anthropogenic disturbances need to be monitored and
diverted away from wallum waterbodies to maintain natural waterbody chemistry.
Waterbody transect length gave an indication of waterbody size, with the smallest
waterbody surveyed being no less than 48 meters from one waterbody edge to the other.
Waterbodies may fluctuate in size during the year and results from the study can only be
attributed to scenarios that were occurring within the survey period. However, smaller
waterbodies then those detected in this study outside of protected areas or within natural
habitats at different times of year may also be suitable for L. olongburensis and C.
tinnula. Furthermore, additional surveys are required to detect winter breeding species,
and determine how abundances and occupancy of the species outside of protected areas.
61
Despite these limitations, these results demonstrate a strong association with low pH
and the threatened amphibian species recorded. Additional environmental variables
associated with the distribution and abundance of amphibian tadpole assemblages
within oligotrophic, acidic waterbodies are complex and varies between species.
Therefore, conservation requirements of wallum heathland amphibian species need to
be assessed separately for each species, and encompass a range of waterbodies where
environmental variables are within the species optimal tolerance range. Furthermore, pH
and water depth influencing relative abundance and occupancy of L. olongburensis
tadpoles indicate that either measurement can be used when assessing waterbody
suitability for tadpoles of the specialist species, L. olongburensis.
Acknowledgements
I would like to thank the Griffith School of Environment and FKP Ltd for providing
project funding for this chapter. I also thank Sonya Clegg, Katrin Lowe, Donna Treby,
two anonymous reviewers and the associate editor of Austral Ecology for feedback on
this chapter (sections of which have been accepted for publication in Austral Ecology*).
The submitted manuscript of this chapter was co-authored by Jon Shuker, Gregory
Lollback, Guy Castley and Jean-Marc Hero. Griffith Animal Ethics number ENV/17/08,
New South Wales Department of Environment and Climate Change Scientific License
number S21717 and Queensland Department of Environment and Resource
Management permit number WITK05620308 were obtained for this work conducted in
this chapter.
*Simpkins, C.A., Shuker, J.D., Lollback, G.W., Castley, J.G. and Hero, J.-M. (2013).
Environmental variables associated with the distribution and occupancy of habitat
speclialist tadpoles in naturally acidic, oligotrophic waterbodies. Austral Ecology, In
press.
62
2.6 References
Austin M.P. (1999) The potential contribution of vegetation ecology to biodiversity
research. Ecography 22, 465-484.
Babbitt K.J. & Tanner G.W. (1998) Effects of cover and predator size on survival and
development of Rana utricularia tadpoles. Oecologia, 114.
Babbitt K.J., Baber M.J. & Tarr T.L. (2003) Patterns of larval amphibian distribution
along a wetland hydroperiod gradient. Can. J. Zool. 81, 1539-1552.
Baber M.J., Childers D.L., Babbitt K.J. & Anderson D.H. (2002) Controls on fish
distribution and abundance in temporary wetlands. Can. J. Fish. Aquat. Sci. 59,
1441-1450.
Baber M.J., Fleishman E., Babbitt K.J. & Tarr T.L. (2004) The relationship between
wetland hydroperiod and nestedness patterns in assemblages of larval amphibians
and predatory macroinvertebrates. Oikos 107, 16-27.
Barth B.J. & Wilson R.S. (2010) Life in acid: interactive effects of pH and natural
organic acids on growth, development and locomotor performance of larval striped
marsh frogs (Limnodynastes peronii). J. Exp. Biol. 213, 1293-1300.
Berkelmans R. & Willis B.L. (1999) Seasonal and local spatial patterns in the upper
thermal limits of corals on the inshore Central Great Barrier Reef. Coral Reefs 18,
219-228.
Burnham K.P. & Anderson D.R. (2002) Model Selection and Multimodel Inference: A
Practical Information-Theoretic Approach. Springer, New York.
Cade B.S. & Noon B.R. (2003) A gentle introduction to quantile regression for
ecologists. Front. Ecol. Environ. 1, 412-420.
Chinathamby K., Reina R.D., Bailey P.C.E. & Lees B.K. (2006) Effects of salinity on
the survival, growth and development of tadpoles of the brown tree frog, Litoria
ewingii. Aust. J. Zool. 54, 97-105.
Christy M.T. & Dickman C.R. (2002) Effects of salinity on tadpoles of the green and
golden bell frog (Litoria aurea). Amphibia-Reptilia 23, 1-11.
Crawley M.J. (2007) The R Book. Wiley, England.
Fernandes I.M., Machado F.A. & Penha J. (2010) Spatial pattern of a fish assemblage in
a seasonal tropical wetland: effects of habitat, herbaceous plant biomass, water
depth, and distance from species sources. Neotrop. Ichthyol. 8, 289-298.
Freda J. (1986) The influence of acidic pond water on amphibians: A review . Water Air
Soil Poll. 30, 439-450.
63
Freda J. & Taylor D.H. (1992) Behavioral response of amphibian larvae to acidic water.
J. Herpetol. 26, 429-433.
Garden J.G., McAlpine C.A., Possingham H.P. & Jones D.N. (2007) Habitat structure is
more important than vegetation composition for local-level management of native
terrestrial reptile and small mammal species living in urban remnants: A case study
from Brisbane, Australia. Austral Ecol. 32, 669-685.
Gillespie G.R. (2001) The role of introduced trout in the decline of the spotted tree frog
(Litoria spenceri) in south-eastern Australia. Biol. Conserv. 100, 187-198.
Gillespie, G.R. (2002). Impacts of sediment loads, tadpole density, and food type on the
growth and development of tadpoles of the spotted tree frog Litoria spenceri: an
in-stream experiment. Biol. Conser. 106, 141-150.
Goldberg F.J., Quinzio S. & Vaira M. (2006) Oviposition-site selection by the toad
Melanophryniscus rubriventris in an unpredictable environment in Argentina. Can.
J. Zool. 84, 699-705.
Gonzalez R.J., Wilson R.W., Wood C.M., Patrick M.L. & Val A.L. (2002) Diverse
strategies for ion regulation in fish collected from the ion-poor, acidic Rio Negro.
Physiol. Biochem. Zool. 75, 37-47.
Gosner K.L. & Black I.H. (1957) The effects of acidity on the development and
hatching of New Jersey Frogs. Ecology 38, 256-262.
Grueber C.E., Nakagawa S., Laws R.J. & Jamieson I.G. (2011) Multimodel inference in
ecology and evolution: challenges and solutions. J. Evolution. Biol. 24, 699-711.
Hecnar S.J. & M’Closkey R.T. (1997) The effects of predatory fish on amphibian
species richness and distribution. Biol. Conserv. 79, 123-131.
Hero J.-M., Gascon C. & Magnusson W.E. (1998) Direct and indirect effects of
predation on tadpole community structure in the Amazon rainforest. Aust. J. Ecol.
23, 474-482.
Hero J.-M., Magnusson W.E., Rocha C.F.D. & Catterall C.P. (2001) Antipredator
defenses influence the distribution of amphibian prey species in the Central
Amazon rain forest. Biotropica 33, 131-141.
Heyer W.R. (1973) Ecological interactions of frog larvae at a seasonal tropical location
in Thailand. J. Herpetol. 7, 337-361.
Heyer W.R., McDiarmid R.W. & Weigmann D.L. (1975) Tadpoles, predation and pond
habitats in the tropics. Biotropica 7, 100-111.
Hossie T.J. & Murray D.L. (2010) You can’t run but you can hide: refuge use in frog
tadpoles elicits density-dependent predation by dragonfly larvae. Oecologia 163,
395-404.
64
Hunter D.A., Smith M.J., Scroggie M.P. & Gilligan D. (2011) Experimental
examination of the potential for three introduced fish species to prey on tadpoles of
the endangered Booroolong Frog, Litoria booroolongensis. J. Herpetol. 45, 181185.
Ingram G.J. & Corben C.J. (1975) The frog fauna of North Stradbroke Island, with
comments on the “acid” frogs of the wallum. Proc. R. Soc. Queensl. 86, 49-54.
IUCN 2011. (2011) IUCN Red List of threatened species. Available from
http://www.iucnredlist.org (accessed July 2011).
Jocqué M., Graham T. & Brendonck L. (2007) Local structuring factors of invertebrate
communities in ephemeral freshwater rock pools and the influence of more
permanent water bodies in the region. Hydrobiologia 592, 271-280.
Johnson J.B. & Omland K.S. (2004) Model selection in ecology and evolution. Trends
Ecol. Evol. 19, 101-8.
Kats L.B., Petranka J.W. & Sih A. (1988) Antipredator defenses and the persistence of
amphibian larvae with fishes. Ecology 69, 1865-1870.
Knutson M.G., Richardson W.B., Reineke D.M., Gray B.R., Parmelee J.R. & Weick
S.E. (2004) Agricultural ponds support amphibian populations. Ecol. Appl. 14,
669-684.
Kopp K., Wachlevski M. & Eterovick P.C. (2006) Environmental complexity reduces
tadpole predation by water bugs. Can. J. Zool. 84, 136-140.
Krebs C.J. (2009) Ecology: Sixth Edition. Benjamin Cummings, United States of
America.
Lancaster J. & Belyea L.R. (2006) Defining the limits to local density: alternative views
of abundance-environment relationships. Freshwater Biol. 51, 783-796.
Lane M.J. & Mahony S.J. (2002) Larval anurans with synchronous and asynchronous
development periods: contrasting responses to water reduction and predator
presence. J. Anim. Ecol. 71, 780-792.
Lewis B.D. & Goldingay R.L. (2005) Population monitoring of the vulnerable wallum
sedge frog (Litoria olongburensis) in north-eastern New South Wales. Aust. J.
Zool. 53, 185-194.
Loman J. (1999) Early metamorphosis in common frog Rana temporaria tadpoles at
risk of drying: an experimental demonstration. Amphibia-Reptilia 20, 421-430.
Loman J. (2002) Temperature, genetic and hydroperiod effects on metamorphosis of
brown frogs Rana arvalis and R. temporaria in the field. J. Zoo. 258, 115-129.
65
Meyer E. (2004) Acid adaptation and mechanisms for softwater acid tolerance in larvae
of anuran species native to the “Wallum” of east Australia. PhD Thesis,
University of Queensland.
Meyer E., Hero J-M., Shoo L. & Lewis B. (2006) National recovery plan for the
wallum sedgefrog and other wallum-dependent frog species. Report to Department
of the Environment and Water Resources, Canberra. Queensland Parks and
Wildlife Service, Brisbane.
Mokany A. & Shine R. (2002) Competition between tadpoles and mosquito larvae.
Oecologia 135, 615-620.
Moreira L.F.B., Machado I.F., Garcia T.V. & Maltchik L. (2010) Factors influencing
anuran distribution in coastal dune wetlands in southern Brazil. J. Nat. Hist. 44,
1493-1507.
Morin P.J. (1986) Interactions between intraspecific competition and predation in an
amphibian predator-prey system. Ecology 67, 713-720.
Márquez-García M., Correa-Solís M. & Méndez M.A. (2010) Life-history trait variation
in tadpoles of the warty toad in response to pond drying. J. Zoo. 281, 105-111.
Persson M., Räsänen K., Laurila A. & Merilä J. (2007) Maternally determined
adaptation to acidity in Rana arvalis: Are laboratory and field estimates of
embryonic stress tolerance congruent? Can. J. Zool. 85, 832-838.
Pierce B.A. (1985) Acid tolerance in amphibians. BioScience 35, 239-243.
Pyke A.W. & White G.H. (2000) Factors influencing predation on eggs and tadpoles of
the endangered Green and Golden Bell Frog Litoria aurea by the introduced
Plague Minnow Gambusia holbrooki. Aust. Zoo. 31, 496-505.
Reynolds S.J. (2009) Impact of the introduced Poeciliid Gambusia holbrooki on
amphibians in southwestern Australia. Copeia 2009, 296-302.
Richter-Boix A., Llorente G.A. & Montori A. (2007) A comparative study of predatorinduced phenotype in tadpoles across a pond permanency gradient. Hydrobiologia
583, 43-56.
Rios-López N. (2008) Effects of increased salinity on tadpoles of two anurans from a
Caribbean coastal wetland in relation to their natural abundance. Amphibia-Reptilia
29, 7-18.
Sadinski W.A. & Dunson W.J. (1992) A multilevel study of effects of low pH on
amphibians of temporary ponds. J. Herpetol. 26, 413-422.
Saidapur S.K., Veeranagoudar D.K., Hiragond N.C. & Shanbhag B.A. (2009)
Mechanism of predator–prey detection and behavioral responses in some anuran
tadpoles. Chemoecology 19.
66
Schmutzer A.C., Gray M.J., Burton E.C. & Miller D.L. (2008) Impacts of cattle on
amphibian larvae and the aquatic environment. Freshwater Biol. 53, 2613-2625.
Schofield P.J. & Nico L.G. (2009) Salinity tolerance of non-native Asian swamp eels
(Teleostei: Synbranchidae) in Florida, USA: comparison of three populations and
implications for dispersal. Environ. Biol. Fish. 85, 51-59.
Shuker, J.D., Hero, J.-M., 2012. Perch substrate use by the threatened wallum sedge
frog (Litoria olongburensis) in wetland habitats of mainland eastern Australia.
Aust. J. Zoo. http://dx.doi.org/10.1071/ZO12030.
Shoo L.P., Olson D.H., McMenamin S.K., Murray K.A., Van Sluys M., Donnelly M.A.,
et al. (2011) Engineering a future for amphibians under climate change. J. Appl.
Ecol. 48, 487-492.
Smith M.J., Schreiber E.S.G., Scroggie M.P., Kohout M., Ough K., Potts J., et al.
(2007) Associations between anuran tadpoles and salinity in a landscape mosaic of
wetlands impacted by secondary salinisation. Freshwater Biol. 52, 75-84.
Snodgrass J.W., Bryan Jr. A.L. & Burger J. (2000) Development of expectations of
larval amphibian assemblage structure in southeastern depression wetlands. Ecol.
Appl. 10, 1219-1229.
Spieler, M. (2003). Risk of predation affects aggregation size: a study with tadpoles of
Phrynomantis microps (Anura: Microhylidae). Anim. Behav.65, 179-184.
Twomey E., Morales V. & Summers K. (2008) Evaluating condition-specific and
asymmetric competition in a species-distribution context. Oikos 117, 1175-1184.
Vonesh J.R., Kraus J.M., Rosenberg J.S. & Chase J.M. (2009) Predator effects on
aquatic community assembly: disentangling the roles of habitat selection and postcolonization processes. Oikos 118, 1219-1229.
Walling D.E. (1977) Assessing the accuracy of suspended sediment rating curves for a
small basin. Water Resour. Res. 13, 531-538.
Welborn G.A., Skelly D.K. & Werner E.E. (1996) Mechanisms creating community
structure across a freshwater habitat gradient. Annu. Rev. Ecol. Syst. 27, 337-363.
Wells K.D. (2007) The Ecology and Behavior of Amphibians. The University of
Chicago Press, United States of America.
Wilbur H.M. (1987) Regulation of structure in complex systems: experimental
temporary pond communities. Ecology 68, 1437-1452.
Wiltshire D.J. & Bull C.M. (1977) Potential competitive interactions between larvae of
Pseudophryne bibroni and P. Semimarmorata (Anura: Leptodactylidae). Aust. J.l
of Zool. 25, 449-454.
67
Wood C.M., Matsuo A.Y.O., Wilson R.W., Gonzalez R.J., Patrick M.L., Playle R.C., et
al. (2003) Protection by natural blackwater against disturbances in ion fluxes
caused by low pH exposure in freshwater stingrays endemic to the Rio Negro.
Physiol. Biochem. Zool.76, 12-27.
Woodward B.D. (1983) Predator-prey interactions and breeding-pond use of temporarypond species in a desert anuran community. Ecology 64, 1549-1555.
68
Chapter 3 - Suitability of anthropogenic waterbodies for
amphibians associated with naturally acidic, oligotrophic
environments and environmental variables influencing their
distribution
3.1 Abstract
Habitat destruction is a key threatening process to amphibians outside of protected
areas. Anthropogenically modified or waterbodies can be used to compensate for habitat
loss. Several amphibian species utilise artificial waterbodies as compensatory habitats,
or opportunistically. Use of anthropogenic/modified waterbodies by adult and tadpoles
of amphibian species that are both associated and not associated with naturally acidic,
oligotrophic waterbodies was measured. Additionally, the environmental factors
influencing the anuran assemblage, with a focus on Litoria olongburensis (a threatened
species associated with naturally acidic, oligotrophic waterbodies) and Litoria fallax (a
common, potentially competitive species of L. olongburensis) was examined. Nine road
trenches/ditches, eight artificial ‘lakes’, six abandoned golf course waterbodies and
thirteen natural waterbodies were surveyed for anuran adults and tadpoles during the
summer/spring period of 2011/2012. Water chemistry, aquatic predators and vegetation
types were also measured. Water chemistry and anuran species richness differed among
waterbody types, with two of three threatened anuran species present in both natural and
anthropogenic/modified waterbodies. The anuran assemblage was influenced by several
environmental variables, including pH, turbidity, salinity and % cover of certain
vegetation types. Relative abundance of adults and tadpole occupancy of L.
olongburensis was highest in natural waterbodies, while adult relative abundance of L.
fallax was highest within artificial lakes. Additionally, L. olongburensis relative
abundance was positively associated with increased sedge density and negatively
associated with increased water pH. This study demonstrates that differing levels of
water chemistry factors and vegetation density influence the amphibian assemblage
within these environments. Effective conservation of all anuran species would be
enhanced by conserving a variety of waterbody types, however, natural waterbodies
provide the best conservation protection for threatened species including L.
69
olongburensis and C. tinnula. Environmental managers creating habitat offsets should
therefore critically assess the quality of constructed habitat for specialist anuran species.
3.2 Introduction
Amphibian species are declining globally (Stuart et al., 2004), with habitat loss and
modification being one of the main threatening factors (Collins and Storfer, 2003;
Beebee and Griffiths, 2005). Despite this, modified or anthropogenic waterbodies can
be used to compensate for habitat loss (e.g. Mazerolle et al., 2006; Ruhí et al., 2012),
with several amphibian species within Australia (reviewed in Hazell, 2003; Hazell et al.,
2004; Lemckert et al., 2006), North America (Monello and Wright, 1999; Brand and
Snodgrass, 2010; Brown et al., 2012) and Europe (Rannap et al., 2009; Brown et al.,
2012; Ruhí et al., 2012) utilising these habitats. However, species richness and
assemblages can differ between these and natural waterbodies (Hazell et al., 2004),
potentially due to differences in environmental factors that influence amphibian species
distributions within waterbodies.
Several environmental factors can influence the distribution of amphibians within
waterbodies, including aquatic predators (Kats et al., 1988; Hecnar and M’Closkey,
1997; Hero et al., 2001; Vonesh et al., 2009) or competitors (Morin, 1986; Wilbur,
1987; Mokany and Shine, 2002; Twomey et al., 2008). Tadpole intolerance to water
chemistry variables, including pH (reviewed in Pierce, 1985; Sparling, 2010) and
salinity (Strahan, 1957; Christy and Dickman, 2002; Chinathamby et al., 2006; RiosLópez, 2008) may also exclude amphibian species from a waterbody due to water
chemistry intolerances. Additionally, turbidity has been found to influence amphibian
species richness (Hecnar and M’Closkey, 1996) and may influence the structure of other
amphibian assemblages. Furthermore, the abundance or emergence of particular
vegetation species (Lemckert et al., 2006; Shuker, 2012), or the proportion of the water
margin with emergent vegetation (Hazell et al., 2004) Lemckert et al., 2006, may also
influence amphibian usage or species abundances within waterbodies. Pond isolation
may also influence amphibian species richness within Australian ponds (Smallbone et
al., 2011).
70
Reductions in the availability of coastal waterbodies have occurred globally (e.g.
Turner, 1990; Levin et al., 2009), due to human population growth and urban
expansion. Within Australia, the majority of freshwater, coastal waterbodies situated
along the eastern seaboard between Fraser Island, QLD, and Jervis Bay, NSW, are
unique as they are both naturally oligotrophic (Hines et al., 1999) and acidic (Coaldrake,
1961; Griffith et al., 2008, Hines and Meyer, 2011). The ‘wallum’ waterbodies of these
coastal areas, characterised by Banksia woodlands, sedgeland, heathland and Melaleuca
swamps (Hines et al., 1999; Griffith et al., 2003), also support stable populations of
threatened ‘acid’ frog species (Ingram and Corben, 1975; Hines et al., 1999; Lewis and
Goldingay, 2005). However, populations occurring outside of protected areas are at risk
from habitat loss (Hines et al., 1999), with more than 30% of original heathland and
Melaleuca cover in south-east QLD cleared between 1974-1989 (Catterall and
Kingston, 1993, cited in Hines et al., 1999). Furthermore, areas within this range have
one of the highest human growth rates within Australia (Hines et al., 1999; Garden et
al., 2010). Wallum habitats also contain modified and anthropogenic waterbodies
(Simpkins pers. obs.) which may be utilised by ‘acid’ frog species and aid in reducing
the impacts of habitat loss. Despite the numerous threats to these species and the
presence of anthropogenic or modified waterbodies in their habitat, no peer-reviewed
studies have been undertaken to determine if anurans normally associated with naturally
acidic, oligotrophic waterbodies use and successfully reproduce within anthropogenic or
modified waterbodies.
Consequently, this chapter aims to compare use of natural and anthropogenic
waterbodies by adult and tadpole anuran species around and within coastal wallum of
eastern Australia. Furthermore, the study examines which environmental variables
(aquatic predators, water chemistry variables, vegetation type) influence the anuran
assemblage and the relative abundance of the threatened wallum sedge frog, Litoria
olongbureneis, and a potential competitor, Litoria fallax (Meyer et al., 2006), across the
natural-modified landscape. Information obtained from this study will aid
environmental managers creating habitat offsets to critically assess the quality of
constructed habitat for specialist anuran species that are associated with naturally acidic,
oligotrophic waterbodies. This chapter also provides information that will inform
71
stakeholders if human intervention through anthropogenic waterbody construction can
compensate for habitat loss of specialist wallum frogs.
3.3 Methods
Study Site Selection and Sampling Design
Waterbodies were located within and around Tyagarah Nature Reserve (NR)
(28.6067°S, 153.5693°E) and the southern section of Bribie Island National Park (NP)
(27.0732°S, 153.1774°E). Waterbodies were selected using Google Earth satellite
imagery to cover the range of waterbodies available at each site. All waterbodies were
surveyed twice, firstly between 18 October 2011 and 5 December 2011, and again
between 7-21 February 2012. Thirteen natural (n = 5 Tyagarah, n = 8 Bribie Island) and
twenty-five anthropogenic waterbodies (total n = 38) were surveyed for adult anurans,
aquatic predators and tadpoles. Anthropogenic waterbodies consisted of roadside
trenches/ditches (n = 4 Tyagarah; n = 5 Bribie Island), artificial ‘lakes’ (n = 1
Tyagarah; n = 8 Bribie Island) and old golf course waterbodies (n = 6 Tyagarah). With
the exception of golf course waterbodies, all waterbodies contained natural vegetation
around the vast majority (> 80%) of the waterbody perimeter.
Roadside trenches were waterbodies that had been constructed next to roads or
firebreaks where earth material had been removed to aid construction of the road or
firebreak. Artificial lakes were constructed waterbodies with the majority of the
waterbody (> 80%) as open water with vegetation fringing the perimeter. Golf course
waterbodies were constructed waterbodies within the golf course boundaries.
Waterbodies within wallum heathland can be composed of heterogeneous vegetation
types (refer to Chapter 2). Therefore, one large natural waterbody had multiple transects
established a minimum of 300 m apart to ensure transect independence. Survey
transects were established 5 m from the waterbody perimeter and ran parallel to the
waterbody perimeter. Transect length varied between 50 – 100 m, depending on
waterbody size.
72
Rainfall data was collected from the Australian Bureau of Meteorology Weather Station
#058216 for Tyagarah NR and #040998 for Bribie Island NP. Sampling was conducted
in 2011, with 246.9 millimetres (mm) recorded at Bribie Island and 373.4 mm recorded
for Tyagarah NR during the three months prior to the date of the first survey. The 2012
surveys recorded 936.7 mm at Bribie Island and 405.2 mm for Tyagarah NR during the
three months prior to the first survey date of the second survey. All waterbodies
surveyed had water present during both the 2011 and 2012 surveys.
Diurnal aquatic predator traps (38cm length x 25cm width x 25cm height, with two
square shaped entrances both 5cm x 5cm) were placed at the beginning, middle and end
of each transect. Traps were baited with Orca Floating Fish Food Pellets and were left
for approximately 30 minutes before being collected. Diurnal dip-netting for predatory
fauna and tadpoles was conducted using a circular net, with an aperture of
approximately 30 cm in diameter and mesh size less than 0.5 mm. Five ‘sweeps’ of the
dip-net and measurements of pH and salinity were taken at 10 m intervals along each
transect. Each ‘sweep’ included three water column levels (bottom, middle and top) to
capture any variation in tadpole species richness or abundance that may occur along the
water depth gradient (Heyer, 1973).
Water pH and salinity were measured using a TPS Aqua-CPA Conductivity-TDS-pHTemperature Meter (version 1.2) approximately 15 cm below the surface of the water at
every 10 m sampling point along the transect. Two 25 mL surface water samples were
collected at the beginning and the end of each transect. Turbidity was measured using a
HACH DREL 2000 Direct Reading Spectrophotometer within 24 hours of sample
collection. Water chemistry measurements and samples were taken before dip-netting
and trapping were conducted. Waterbody area was calculated using Google Earth
satellite imagery.
Vegetation was categorized into five vegetation types; sedge, grass, lilly, Gahnia spp.
and Melaleuca spp. Percent cover of each vegetation type was visually estimated using
a 1 m x 1 m quadrat placed at each 10 m sampling interval.
73
Nocturnal anuran surveys were conducted by walking and visually searching the
transect using a Princeton Apec headtorch. The species and the relative number of
anurans seen for each species were recorded for each transect. For C. tinnula, numbers
of individuals were determined using acoustic calls due to visual observations being
absent or low at the survey transects. The vegetation type an individual anuran was first
encountered on was also recorded. Substrate records were only conducted on the first
survey.
Data Analysis
Relative abundances of L. olongburensis and L. fallax adults were higher during the
2011 survey. The cumulative number of L. olongburensis visually recorded during the
first survey for each transect were divided by the transect length to obtain a relative
abundance of L. olongburensis per meter. This was repeated for L. fallax, L. tyleri,
Rhinella marina, and Limnodynastes peronii.
Incidental visual observations and frogs heard calling less than one meter outside the
waterbody perimeter or calling within the waterbody during the first or second survey
were recorded as occupying the waterbody. Tadpole species were considered present
within a waterbody if they were recorded from any of the sampling points along
transects.
Predatory fish caught in aquatic traps were used in data analysis due to low counts of
predatory fish caught during dip-netting. The total number of predatory fish in each
waterbody included individuals from the species Gambusia holbrooki (Eastern
mosquito fish), Rhadinocentrus ornatus (Ornate rainbow fish), Hypseleotris galii
(Firetail gudgeon), Hypseleotris compressa (Empire gudgeon) and Hypseleotris sp.
(Midgleys carp gudgeon). The number of fish recorded for a given transect was divided
by the number of aquatic traps used within the waterbody to obtain a relative fish
abundance for each transect.
Salinity, turbidity and pH were averaged for each transect. Mean cover of each
vegetation type for each transect was calculated by dividing the total percentage cover
74
of each vegetation type from each sampling point by the number of sampling points
along the transect.
A Spearman Rank Correlation Test was performed in IBM SPSS Statistics Version 19
(SPSS, Inc., 2009, Chicago) to determine correlated variables. Highly correlated
variables were considered to be variables that had a correlation coefficient value greater
than or equal to 0.7 (sensu Babbitt et al., 2003; Garden et al., 2007). No variables were
highly correlated and all variables were used in analyses.
A One-Way ANOVA with Tukey’s post-hoc analysis was used to determine significant
differences in environmental variables between waterbody types. One way ANOVAs
were also used to compare the relative abundance of L. olongburensis and L. fallax in
the four different waterbody types.
The anuran community assemblage was analysed (using visual counts for L.
olongburensis, L. fallax, L. peronii, R. marina and L. tyleri, and acoustic counts for C.
tinnula) using non-metric Multidimensional Scaling (nMDS). These species were
chosen as they were encountered in at least two waterbodies and were highest in relative
abundance when compared with other species. Only waterbodies which contained one
of these six species were used in nMDS analysis (n = 27). Bray-Curtis distance
measures were used to determine waterbody similarities for the relative abundance of
individual species. Four dimensions were used to minimise stress based on 999 random
permutations to determine which environmental variables were significantly influencing
the amphibian assemblage (Oksanen, 2011). Analyses were performed using the
statistical program R (R-core Development Team 2011) using the vegan (version 2.04)
package (Oksanen et al., 2012)
Twenty-eight models focusing on the influence of variables on the relative abundance of
L. olongburensis and L. fallax were constructed a priori. To obtain the maximum
number of predictor variables to use in each model, a generalised ‘rule of thumb’ of n/3
(where n = number of waterbodies sampled) (Crawley, 2007) was applied. Due to this
rule failing to be adhered to by some models small sample Akaike’s Information
Criterion (AICc) was used to account for the number of model failing to meet this ‘rule’
75
(Burnham and Anderson, 2002). Predictor variables in L. olongburensis models were
mean pH, average salinity, average turbidity, waterbody size, fish relative abundance, L.
fallax relative abundance, % sedge cover, % Gahnia cover, % grass cover, % lily cover
and % Melaleuca cover. Predictor variables in L. fallax models were the same as the L.
olongburensis models, except L. fallax relative abundance was replaced with L.
olongburensis relative abundance. Some variables (i.e. pH) were non-normally
distributed. Therefore, Generalised Additive Mixed Effects models, using a QuasiPoisson link function to account for overdispersion and location as a covariate, were
used to determine the importance of the predictor variables on L. olongburensis and L.
fallax relative abundance
Small sample Akaike’s Information Criterion was used for model selection, with the
‘best’ model having the lowest AICc value (Burnham and Anderson, 2002). To
determine the ranking of the models, ΔAICci values were calculated, where higher
ΔAICi values indicated less accurate models for the given data (Burnham and Anderson,
2002; Johnson and Omland, 2004). If a model had a Δi ≤ 2, then there was considerable
evidence that the model could be the “best” model, given the data (Johnson and
Omland, 2004). Models with a Δi ≤ 2-4 were considered to have moderate support for
being the “best” model, given the data. Each model was assigned a model weight (wi)
which was used to determine the “probability that model I is the best model for the
observed data, given the candidate set of models” (Johnson and Omland, 2004). The
closer the wi was to 1, the closer the model for the given data (Burnham and Anderson,
2002). To determine the relative importance of variables within models where Δi < 4 the
wi values were summed from all models where the variable of interest occurred
(Grueber et al. 2011). The closer the variable of interest was to 1 the higher the
importance of the variable. All models were run in the freeware statistical package R (R
Core Development Team, 2011) using the MuMIn (version 1.7.2) (Barton, 2012) and
vegan (version 2.04) (Oksanen et al., 2012) packages.
76
3.4 Results
Waterbody Characteristics
Few waterbody characteristics varied among waterbody types (Table 3.1). Variables
that were significantly different between waterbody types were pH (df = 36, F2/34= 5.36,
p = 0.004), turbidity (df = 36, F2/34 = 3.925, p = 0.017) and waterbody size (df = 36,
F2/34 = 6.001, p = 0.002). Tukeys post hoc analysis revealed pH was significantly higher
in artificial lakes than in natural waterbodies (p = 0.005) and roadside ditches (p =
0.011). Tukeys post hoc analysis also revealed that turbidity was significantly higher in
natural waterbodies than in golf course waterbodies (p = 0.011) and natural waterbodies
were significantly larger than road side ditches (p = 0.003) and golf course waterbodies
(p = 0.024).
Anuran Assemblage / Occupancy
A total of 10 species were encountered; six recorded in natural waterbodies, six in
artificial lakes, five in roadside ditches and six in golf course waterbodies (Figure 3.1).
Litoria olongburensis and C. tinnula occupancy was highest within natural waterbodies
while L. fallax, Litoria tyleri and R. marina occupancy was highest within artificial
lakes (Figure 3.2). Litoria freycineti was only recorded from roadside ditches while L.
gracilenta was only recorded from natural waterbodies. Additionally, one species of
Uperolia sp. was recorded from one golf course waterbody. Litoria fallax was the only
species to occupy all four waterbody types (Figures 3.1 and 3.2).
Water pH, salinity, % lily cover and % sedge cover significantly influenced the
amphibian assemblage (Table 3.2). All species were separated on the nMDS plot, with
L. olongburensis, C. tinnula and L. tyleri falling within close proximity of each other
(Figure 3.3).
77
Table 3.1: Measured variable averages and ranges between the four waterbody types surveyed and for waterbodies with L. olongburensis and L. fallax.
Variable
Natural
Roadside ditches
waterbodies
pH (average)
Golf course
Artificial lakes
waterbodies
L. olongburensis
L. fallax
waterbodies
waterbodies
3.96
3.96
4.73
5.41
4.07
4.81
3.7 – 4.85
3.64 – 4.57
4.32 – 5.91
3.79 – 7.97
3.43 – 5.83
3.79 – 6.84
Salinity (ppm)
119.75
100.57
4684
153.89
90.8
109
Salinity (ppm)
50.22 – 108.67
73.56 – 206.67
40.6 – 27700
96.1 – 317.5
47.78 – 209.7
64.64 – 198.17
Turbidity (FTU)
395
229
54
261.1
323
230
Turbidity (FTU)
150.5 – 889.33
21 – 217.33
30 – 100.33
16.7 – 724
21 – 889.33
16.6 – 724
22968
1196
337
11056
15574
8185
1726 – 51787
66 – 2666
750 – 5564
415 – 43351
376 – 51787
791 – 41129
55
19
5
22
52
28
% Sedge (range)
0 – 100
0 – 61
0 – 16
0 – 88
2 – 100
0 – 88
Fish abundance
2.13
2.09
3.06
2.9
3.07
3.42
0 – 13
0 – 8.3
0.33 – 9
0 – 9.33
0 – 13
0 – 9.33
pH (range)
(range)
(range)
Area (m2)
Area (m2) (range)
% Sedge
Fish (range)
78
Figure 3.1: Amphibian species richness and species presence within each waterbody type. Colours/patterns indicate individual species.
79
Figure 3.2: Proportion of natural and anthropogenic waterbodies occupied for each recorded anuran species. Records are combined for both visual and
acoustic records.
80
Figure 3.3: nMDS ordination of waterbodies for anuran species where a relative
abundance measurement was calculated. Stress associated with 4 dimensions used in
MDS ordination was 0.0268. Species ordinations are overlaid. Environmental variables
significantly influencing the community structure are displayed. Circles represent
waterbodies.
81
Table 3.2: Correlations (R2 values) between nMDS axis 1 and 2 and environmental
variables influencing assemblage structure, with significant correlations (Pr (> r))
highlighted in bold.
NMDS1
NMDS2
R2
Pr (> r)
pH
0.724
0.69
0.293
0.018
Salinity
0.046
0.999
0.258
0.033
Turbidity
-0.988
-0.152
0.151
0.135
Area
-0.998
0.057
0.091
0.316
Fish
0.706
0.708
0.002
0.975
-0.8
-0.599
0.251
0.039
Melaleuca
-0.925
0.38
0.079
0.369
Fern
-0.05
-0.999
0.105
0.241
Lily
0.941
0.339
0.311
0.016
Gahnia
-0.022
-0.999
0.211
0.061
Grass
-0.126
0.992
0.077
0.414
Variable
Sedge
Litoria olongburensis and L. fallax relative abundance
Water chemistry variables differed among waterbodies containing L. olongburensis and
L. fallax, with pH being 0.74 higher and sedge cover being 24% lower in waterbodies
with L. fallax (Table 3.1). There was a significant difference between waterbody type
for L. olongburnensis relative abundance (df = 36, F1/34 = 3.558, p = 0.025) but not for
L. fallax relative abundance (df = 36, F1/34 = 1.682, p = 0.19). However, L. fallax was
only recorded from two natural waterbodies while the highest abundance of L.
olongburensis was recorded from natural waterbodies (Figure 3.4). Additionally, no L.
olongburensis were recorded from golf courses while the highest relative abundance of
L. fallax was recorded from artificial lakes (Figure 3.4).
82
Figure 3.4: ‘Jitter’ plot for relative abundance counts of (a) L. olongburensis and (b) L.
fallax in natural and anthropogenic waterbodies. Abbreviations on the x-axis represent
the first surveys at natural (NW1), artificial lakes (AL1), road side ditches (RD1) and
golf course waterbodies (GCW1) and the second surveys at natural (NW2), artificial
lakes (AL2), road side ditches (RD2) and golf course waterbodies (GCW2).
83
Litoria olongburensis were recorded perched on sedge (n = 153), Gahnia sp. (n = 14),
Melaleuca sp. (n = 3), fern (n = 4) and grass (n = 2). Litoria fallax were recorded
perched on sedges (n = 28), Gahnia sp. (n = 15), lilies (n = 14), Melaleuca sp. (n = 2)
and ferns (n = 1).
Tadpoles of L. olongburensis were recorded in 11 natural waterbodies and two roadside
ditches, which were adjacent to natural waterbodies. Tadpoles of L. fallax were only
recorded from one natural waterbody adjacent to a roadside ditch. Tadpoles were only
recorded during the summer survey in 2012 despite tadpole surveys being conducted
during the 2011 and 2012 surveys.
One model for L. olongburensis relative abundance had a Δi ≤ 2 while no models had a
Δi ≤ 2-4. The weighting of the best model was 78.7%, indicating that the other models
compared poorly (Table 3.3). This model contained % sedge cover and pH as key
parameters. Litoria olongburensis abundance had a positive relationship with increasing
% sedge cover and a negative relationship with increasing pH. Both % sedge cover and
pH had high relative variable importance. However, pH had a confidence interval that
intersected zero, diminishing the strength of the influence that this variable had on L.
olongburensis relative abundance (Table 3.4).
Five models for L. fallax relative abundance had a Δi < 2 and five models had a Δi ≤ 24. The combined weighting for models with Δi ≤ 2 was 61.8%. The combined weighting
for models with Δi ≤ 2-4 was 28.6%. This gave a total weighting of 90.4% for all
models with Δi ≤ 4, indicating that the other models compared poorly (Table 3.3).
Models with a Δi < 2 contained % sedge cover, L. olongburensis relative abundance,
waterbody size, % fern density and % grass cover. All of these variables, with the
exception of % grass cover, were negatively associated with L. fallax relative
abundance. Models with a Δi ≤ 2-4 contained % Melaleuca cover, % Gahnia cover, %
sedge cover, turbidity and predatory fish. Predatory fish and % Gahnia cover were the
only variables within models where Δi ≤ 2-4 that was positively associated with L.
fallax relative abundance. Sedge cover had the highest relative variable importance
compared with the other variables within models where Δi ≤ 4. However, all variables
84
Table 3.3: Models with a Δi value < 4 for L. olongburensis and L. fallax adult relative
abundance per metre for 2011 surveys. (+) indicates a positive relationship while (-)
indicates a negative relationship between relative abundance and the model variable.
AICC
Δi
wi
150.8
0.00
0.787
(-) % Sedge
159.5
0.00
0.219
(-) L. olongburensis abundance
160.6
1.08
0.127
(-) Waterbody Size
160.9
1.46
0.105
(-) % Fern
161.4
1.88
0.085
% Grass
161.4
1.96
0.082
(-) % Melaleuca
161.8
2.32
0.069
(-) Turbidity
161.8
2.34
0.068
% Garnia
162.1
2.64
0.058
(-) % Sedge + % Garnia
162.3
2.85
0.053
Predatory Fish abundance
163.0
3.52
0.038
Model
Litoria olongburensis
% Sedge – pH
Litoria fallax
had a confidence interval that included zero, indicating reduced evidence that these
variables did not have a strong influence on L. fallax relative abundance (Table 3.4).
3.5 Discussion
Anuran Assemblages
There was considerable overlap in the anuran species recorded from natural and
anthropogenic waterbodies in wallum habitats of eastern coastal Australia. These results
are similar to the majority of surveys conducted within the northern hemisphere
(predominantly North America and Europe) that found anuran species richness to be
higher or equal in artificial, restored waterbodies (reviewed in Brown et al., 2012). This
is in contrast to a previous Australian study conducted in New South Wales comparing
farm dams and natural waterbodies outside of wallum heathland areas, where natural
85
Table 3.4: Estimates for model averaged coefficients, standard error (SE), confidence
interval (CI) and relative variable importance (RI) for each parameter in models where
Δi < 4 for L. olongburensis and L. fallax tadpole relative abundance. (+) indicates a
positive relationship while (-) indicates a negative relationship between relative
abundance and the model variable.
Parameter
Estimate
SE
CI
RI
% Sedge
3.66
0.615
2.44, 4.86
0.94
pH
-0.6
1.16
-2.88, 1.67
0.85
(-) %Sedge
-0.81
1
-2.78, 1.16
0.33
(-) L. olongburensis abundance
-3.95
4.38
-12.53, 4.62
0.13
-0.00004
-0.00003
-0.0001, 0.00002
0.11
(-) % Fern
-7.01
7.76
-22.22, 8.19
0.09
% Grass
0.39
3.22
-5.92, 6.7
0.11
(-) % Melaleuca
-0.84
4.61
-9.87, 8.19
0.07
-0.0014
-0.0017
-0.004, 0.002
0.08
1.64
2.65
-3.55, 6.83
0.14
-0.023
-0.09
-0.15, 0.2
0.05
Litoria olongburensis
Litoria fallax
(-) Waterbody Size
(-) Turbidity
% Garnia
Predatory Fish abundance
waterbodies had higher anuran species richness when compared with farm dams (Hazell
et al., 2004). However, presence of an amphibian species at a waterbody is
predominately associated with environmental variables, rather than the ‘status’ (i.e.
natural / anthropogenic) of the waterbody (Mazerolle et al., 2005; Hagman and Shine
2006; Rannap, 2009).
The differences and similarities of these findings to previous studies are likely
explained by the ability of individual species to respond to differences in the
environmental variables that were found to be associated with the anuran assemblage
(i.e. pH, % sedge cover, % lily cover). Vegetation variables significantly associated
with the anuran assemblage are likely a function of particular species favouring certain
varieties of vegetation. This was observed for L. olongburensis, which was found
86
predominantly on sedge species and supports past studies that found L. olongburensis
prefers perching on certain sedge species (Shuker 2012). Intolerance to varying water
chemistry levels by tadpoles of different anuran species explains the influence of water
chemistry variables on the anuran assemblage, as adults have been shown to avoid
depositing eggs in waters where water chemistry variables would likely be unfavourable
for successful reproduction (Takahashi, 2007; Hamamura, 2008). For example, post
Gosner stage 25 tadpoles of L. fallax failed to metamorphose when exposed to acidic
waters (i.e. pH 3.5.), while acid-water adapted tadpoles species (i.e. C. tinnula)
successfully metamorphosed in pH waters of 3.5, 4.5 and 5.0 (Meyer, 2004). This was
further seen in Chapter two, where pH was associated with the relative abundance and
occupancy of L. olongburensis tadpoles.
Intolerance to water chemistry variables only explains abundance and occupancy
patterns for some species. Conversely, populations of species that are able to tolerate a
wide range of water chemistry variable levels may be excluded from, or depressed
within, waterbodies due to competition or predation. It has been hypothesised that
competition may explain the parapatric distributions of L. fallax and L. olongburensis
(Meyer et al., 2006), as shown by previous studies revealing competitive interactions
between other amphibian species (Wiltshire and Bull, 1977; Twomey et al., 2008).
Surprisingly, the acid frog species L. freycineti was only recorded from roadside
ditches. This species is infrequently encountered in natural waterbodies within coastal
wallum systems; however, it is occasionally recorded from disturbed sites (i.e. drainage
lines (Meyer et al., 2006), on roads away from wetlands (Hero pers. obs.) and fire trails
near water (Simpkins pers. obs.)). These disturbed areas often have lower vegetation
cover, which would likely increase detectability of this species when compared with
natural waterbodies, where vegetation cover is often dense. Therefore, the absence of
this species from natural waterbodies is potentially erroneous, complicating and the
interpretation of natural habitat usage for this species. However, results presented in this
chapter indicate that use of artificial lakes and golf course waterbodies by this species is
low, possibly due to higher pH levels and competition from the non-acid frog species L.
nasuta, which was recorded at these sites. Litoria nasuta has previously been proposed
as a potential competitor of L. freycineti (Meyer et al., 2006).
87
Variables influencing L. olongburensis and L. fallax relative abundance
The variables most strongly associated with both L. olongburensis relative abundance
was a high percent sedge cover and low pH. Sedge cover positively influenced L.
olongburensis abundance with individuals predominantly found perching on sedge
species. Amphibians may have coloration that assists with camouflage in their natural
environment (Norris and Lowe, 1964; Toledo and Haddad, 2009). This may also occur
with L. olongburensis, with sedge and dorsal colouration or patterning possibly aiding
with camouflage (as suggested by Lowe and Hero, 2012). These results are consistent
with a past survey that showed L. olongburensis to be habitat specialists using
ecological niches within acidic (pH < 5) coastal wallum waterbodies containing sedges
(Lewis and Goldingay, 2005; Shuker, unpublished data).
There were several variables that were associated with L. fallax relative abundance but
it is difficult to identify discriminatory variables since those were within the top models
had low relative variable importance and had confidence intervals that included zero.
This suggests that a variable that is strongly influencing L. fallax relative abundance
was not measured or that L. fallax is a generalist species, with the variables measured
having the same relative influence on L. fallax relative abundance across the natural and
modified landscape. Despite this, L. fallax numbers and occupancy within natural
waterbodies (where pH was low and sedge cover high) were low and were highest
within artificial lakes (where pH was high and sedge cover was intermediate). As
mentioned previously, this is likely a result of L. fallax tadpole intolerance to low pH
waters (Meyer, 2004). Additionally, these results indicate that L. fallax do not require
waterbodies where sedge cover is high.
Implications for conservation
A range of waterbody types, both natural and anthropogenic/modified, are required to
effectively conserve representative anuran assemblages within and around waterbodies
that are naturally oligotrophic and acidic. However, conservation efforts should be
focused primarily on natural waterbodies, where occupancy and reproduction of the
88
threatened, specialist species is highest. Conserving natural waterbodies would therefore
increase the persistence of threatened species populations.
The presence of L. olongburensis in artificial lakes and roadside ditches and the
presence of C. tinnula in artificial lakes and golf course waterbodies indicate that these
waterbodies can provide habitat for adults of wallum associated threatened anuran
species. These results corroborate recommendations that the establishment of artificial
waterbodies with longer hydroperiods can be used to combat the effects of climate
change in amphibian populations (Shoo et al., 2011). However, the low number of
anthropogenic waterbodies with L. olongburensis tadpoles suggests that the majority of
anthropogenic waterbodies may be unfavourable for breeding by this species.
Subsequently, while providing potential refuge habitat, these waterbodies may be
ecological traps that would not be able to sustain continuing L. olongburensis
populations. Furthermore, the road ditches with L. olongburensis tadpoles were adjacent
to natural waterbodies. Under these circumstances, road side ditches may be able to
facilitate recruitment as well as connectivity between waterbodies.
With the exception of golf course waterbodies, the vast majority of waterbodies
surveyed were surrounded by wallum heathland habitat. Therefore, results from this
study may only be applicable to waterbodies where wallum heathland habitat is present.
One ‘acid’ frog species (Litoria cooloolensis) has been located up to 1.3 kilometres
away from natural waterbodies within undisturbed environments (Simpkins et al., 2011
- Appendix 1). This suggests dispersal of Litoria ‘acid’ frog species (i.e. L.
olongburensis) could occur over equal distances and, provided that natural habitat is
still intact, modified/disturbed waterbodies that are established away from natural
waterbodies may still receive recruitment from natural waterbodies. Further study,
where wallum heathland is not present around modified/disturbed waterbodies, is
required to determine if ‘acid’ frogs can utilise waterbodies across disturbed landscapes.
Natural waterbodies where pH and sedge cover are within the ideal ranges for L.
olongburensis should be given top priority for conservation. If construction or
conservation of artificial waterbodies is undertaken for habitat loss compensation or in
89
an attempt to reduce the risks of climate change, then the variables that highly influence
L. olongburensis (high sedge cover, low pH) must be attained.
It must be noted that habitat fragmentation and pond isolation may influence the
assemblage of species within a waterbody. Pond isolation could not be measured for
this study due to ponds being hidden from satellite images due to thick vegetation cover.
Fragmentation was also difficult to measure because of this reason. Therefore, further
study measuring pond isolation and pond fragmentation are needed to determine if these
variables are influencing the amphibian assemblages within these ponds.
Acknowledgments
I would like to thank Clare Morrison, Evan Pickett, Jean-Marc Hero, Guy Castley,
Katrin Lowe and Dianna Virki, who provided valuable comments on drafts of this
chapter. We would also like to thank Edward Meyer for advice on pH data analysis and
sampling techniques. We would like to thank Alan Kerr from the Bribie Island
Environmental Protection Association who provided accommodation during fieldwork
and Bayshore Resort in Byron Bay, New South Wales, Australia for giving us access to
their land. We also thank Chris Touey, Tempe Parnell, Billy Ross, Matt Davies and
Nick Clarke for assisting in data collection. Work conducted for this chapter was
authorized by Griffith Animal Ethics number ENV/28/11/AEC, Queensland Department
of Environment and Resource Management ECOACCESS permit number
WITK10631812, and New South Wales Department of Environment Scientific License
number S21717.
90
3.6 References
Babbitt, K.J., Baber, M.J., Tarr, T. L., 2003. Patterns of larval amphibian distribution
along a wetland hydroperiod gradient. Canadian Journal of Zoology 81: 1539–
1552.
Barton, K., 2012. MuMIn: Multi-model inference. R package, version 1.7.2. Page
http://CRAN.R-project.org/package=MuMIn
Beebee, T.J.C., Griffiths, R.A., 2005. The amphibian decline crisis: A watershed for
conservation biology? Biological Conservation 125: 271–285.
Brand, A. B., Snodgrass, J.W., 2010. Value of artificial habitats for amphibian
reproduction in altered landscapes. Conservation Biology 24: 295–301.
Brown, D. J., Street, G.M., Nairn, R.W., Forstner, M.R.J., 2012. A place to call home:
amphibian use of created and restored wetlands. Journal of Ecology 2012,
doi:10.1155/2012/989872.
Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference: A
Practical Information-Theoretic Approach. Springer, New York.
Catterall, C.P., Kingston, M., 1993. Remnant Bushland of South East Queensland in the
1990’s: Its Distribution, Loss, Ecological Consequences, and Future Prospects.
Institute of Applied Environmental Research, Griffith University and Brisbane
City Council, Brisbane, Australia.
Chinathamby, K., Reina, R.D., Bailey, P.C.E., Lees, B.K., 2006. Effects of salinity on
the survival, growth and development of tadpoles of the brown tree frog, Litoria
ewingii. Australian Journal of Zoology 54: 97–105.
Christy, M.T., Dickman, C.R., 2002. Effects of salinity on tadpoles of the green and
golden bell frog (Litoria aurea). Amphibia-Reptilia 23: 1–11.
Coaldrake, J.E., 1961. The eco-system of the coastal lowlands (Wallum) of southern
Queensland. C.S.I.R.O Bulletin. No. 283.
Collins, J.P., Storfer, A., 2003. Global amphibian declines: sorting the hypotheses.
Diversity and Distributions 9: 89–98.
Crawley, M.J., 2007. The R Book. Wiley, England.
Freda, J., Taylor, D.H., 1992. Behavioral response of amphibian larvae to acidic water.
Journal of Herpetology 26: 429–433.
Garden, J.G., McAlpine, C.A., Possingham, H.P., 2010. Multi-scaled habitat
considerations for conserving urban biodiversity: native reptiles and small
mammals in Brisbane, Australia. Landscape Ecology 25: 1013–1028.
91
Garden, J.G., McAlpine, C.A., Possingham, H.P., Jones, D.N., 2007. Using multiple
survey methods to detect terrestrial reptiles and mammals: what are the most
successful and cost-efficient combinations? Wildlife Research 34: 218–227.
Griffith, S.J., Bale, C., Adam, P., 2008. Environmental correlates of coastal heathland
and allied vegetation. Australian Journal of Botany 56: 512–526.
Griffith, S.J., Bale, C., Adam, P., Wilson, R., 2003. Wallum and related vegetation on
the NSW north coast: description and phytosociological analysis. Cunninghamia 8,
202–252.
Grueber, C.E., Nakagawa, S., Laws, R.J., Jamieson, G., 2011. Multimodel inference in
ecology and evolution: challenges and solutions. Journal of Evolutionary Biology
24: 699-711.
Haramura, T., 2008. Experimental test of spawning site selection by Buergeria japonica
(Anura: Rhacophoridae) in response to salinity levels. Copeia 2008: 64-67.
Hagman, M., Shine, R. 2006. Spawning site selection by feral cane toads (Bufo
marinus) at an invasion front in tropical Australia. Austral Ecology 31: 551-558.
Hazell, D. (2003). Frog ecology in modified Australian landscapes: a review. Wildlife
Research 30: 193-205.
Hazell, D., Hero, J.-M., Lindenmayer, D., Cunningham, R., 2004. A comparison of
constructed and natural habitat for frog conservation in an Australian agricultural
landscape. Biological Conservation 119: 61–71.
Hecnar, S.J., M’Closkey, R.T., 1996. Amphibian species richness and distribution in
relation to pond water chemistry in south-western Ontario, Canada. Freshwater
Biology 36: 7-15.
Hecnar, S.J. , M’Closkey, R.T., 1997. The effects of predatory fish on amphibian
species richness and distribution. Biological Conservation 79: 123–131.
Hero, J.-M., Magnusson, W.E., Rocha, C.F.D., Catterall, C.P., 2001. Antipredator
defenses influence the distribution of amphibian prey species in the central
Amazon rain forest. Biotropica 33: 131–141.
Heyer W.R., 1973 Ecological interactions of frog larvae at a seasonal tropical location
in Thailand. Journal of Herpetology 7: 337-361.
Hines, H., Mahony, M. McDonald, K., 1999. An assessment of frog declines in wet
subtropical Australia. Pp. 44- 63 In Campbell, A. (ed.) Declines and
Disappearances of Australian Frogs. Environment Australia, Canberra.
Hines, H.B., Meyer, E.A., 2011. The frog fauna of Bribie Island: an annotated list and
comparison with other Queensland dune islands. Proceedings of the Royal Society
of Queensland 117: 261-274.
92
Ingram, G.J., Corben, C.J., 1975. The frog fauna of North Stradbroke Island, with
comments on the “acid” frogs of the wallum. The Proceedings of the Royal Society
of Queensland 86: 49–54.
Johnson, J. B., Omland, K.S., 2004. Model selection in ecology and evolution. Trends
in Ecology & Evolution 19: 101–108.
Kats, L.B., Petranka, J.W., Sih, A., 1988. Antipredator defenses and the persistence of
amphibian larvae with fishes. Ecology 69: 1865–1870.
Lemckert, L., Haywood, A., Brassil, T., Mahony, M. (2006). Correlations between frogs
and pond attributes in central New South Wales, Australia: What makes a good
pond? Applied Herpetology 3: 67-81.
Levin, N., Elron, E., Gasith, A., 2009. Decline of wetland ecosystems in the coastal
plain of Israel during the 20th century: Implications for wetland conservation and
management. Landscape and Urban Planning 92: 220–232.
Lewis, B. D., Goldingay, R.L., 2005. Population monitoring of the vulnerable wallum
sedge frog (Litoria olongburensis) in north-eastern New South Wales. Australian
Journal of Zoology 53: 185–194.
Lowe, K., Hero, J.-M., 2012. Sexual dimorphism and color polymorphism in the
wallum sedge frog (Litoria olongburensis). Herpetological Review 43: 236–240.
Mazerolle, M.J., Desrochers, A., Rochefort, L. 2005. Landscape characteristics
influence pond occupancy by frogs after accounting for detectability. Ecological
Applications 15: 825-834.
Mazerolle, M. J., Poulin, M., Lavoie, C., Rochefort, L., Desrochers, A., Drolet, B.,
2006. Animal and vegetation patterns in natural and man-made bog pools:
implications for restoration. Freshwater Biology 51: 333–350.
Meyer, E., Hero, J-M., Shoo, L. and Lewis, B., 2006 National recovery plan for the
wallum sedgefrog and other wallum-dependent frog species. Report to Department
of the Environment and Water Resources, Canberra. Queensland Parks and
Wildlife Service, Brisbane.
Mokany, A., Shine, R., 2002. Competition between tadpoles and mosquito larvae.
Oecologia 135: 615–620.
Monello, R.J., Wright, R.G., 1999. Amphibian habitat preferences among artificial
ponds in the Palouse region of northern Idaho. Journal of Herpetology 33: 298–
300.
Morin, P.J., 1986. Interactions between intraspecific competition and predation in an
amphibian predator-prey system. Ecology 67: 713–720.
93
Oksanen, J., 2011. Multivariate analysis of ecological communities in R: vegan tutorial.
http://ccoulufi/~jarioksa/opetus/metodi/vegantutorpdf, 20 September 2012.
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B.,
Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2012. Vegan:
Community Ecology Package, version 2.0-4. Page http://vegan.r-forge.rproject.org/, 20 September 2012.
Orizaola, G., BraÑa, F., 2003. Do predator chemical cues affect oviposition site
selection in newts? Herpetological Journal 13: 189-193
Pierce, B.A., 1985. Acid tolerance in amphibians. BioScience 35: 239–243.
Rannap, R., Lõhmus, A., Briggs, L., 2009. Restoring ponds for amphibians: a success
story. Hydrobiologia 634: 87–95.
Rios-López, N., 2008. Effects of increased salinity on tadpoles of two anurans from a
Caribbean coastal wetland in relation to their natural abundance. AmphibiaReptilia 29: 7–18.
Ruhí, A., Sebastian, O.S., Feo, C., Franch, M., Gascón, S., Richter-Boix, A., Boix, D.,
Llorente, G., 2012. Man-made Mediterranean temporary ponds as a tool for
amphibian conservation. International Journal of Limnology 48: 81–93.
Shuker, J.D., Hero, J.-M., 2012. Perch substrate use by the threatened wallum sedge
frog (Litoria olongburensis) in wetland habitats of mainland eastern Australia.
Australian Journal of Zoology. doi.org/10.1071/ZO12030.
Shoo, L. P. et al., 2011. Engineering a future for amphibians under climate change.
Journal of Applied Ecology 48: 487–492.
Simpkins, C. A., Meyer, E., Hero, J.-M., 2011. Long-range movement in the rare
Cooloola sedgefrog Litoria cooloolensis. Australian Zoologist 35: 977–978.
Smallbone, LT., Luck, G.W. and Wassens, S. (2011). Anuran species in urban
and scapes: relationships with biophysical, built environment ans socio-economic
factors. Landscape and Urban Planning 101: 43-51.
Smith, M.J., Schreiber, E.S.G., Scroggie, M.P., Kohout, M., Ough, K., Potts, J., Lennie,
R., Turnbill, D., Jin, C., Clancy, T., 2007. Associations between anuran tadpoles
and salinity in a landscape mosaic of wetlands impacted by secondary salinisation.
Freshwater Biology 52:75–84.
Sparling, D.W., 2010. Water-quality criteria for amphibians. Pp 105-120. in Dodd, Jr.,
C.K. (ed.) Amphibian Ecology and Conservation: A Handbook of Techniques.
Oxford University Press, New York.
94
Strahan, R., 1957. The effect of salinity on the survival of larvae of Bufo melanostictus
Schneider. Copeia 1957: 146–147.
Stuart, S.N., Chanson, J.S., Cox, N.A., Young, B.E., Rodrigues, A.S.L., Fischman,
D.L., Waller, R.W., 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 306: 1783–1786.
Takahashi, M., 2007. Oviposition site selection: pesticide avoidance by gray treefrogs.
Environmental Toxicology and Chemistry 26: 1476-1480.
Toledo, L.F., Haddad, C.F.B., 2009. Colors and some morphological traits as defensive
mechanisms in anurans. International Journal of Zoology 2009: 1–12.
Turner, R.E., 1990. Landscape development and coastal wetland losses in the northern
Gulf of Mexico. American Zoologist 30: 89–105.
Twomey, E., Morales, V., Summers, K., 2008. Evaluating condition-specific and
asymmetric competition in a species-distribution context. Oikos 117: 1175–1184.
Vonesh, J.R., Kraus, J.M., Rosenberg, J.S., Chase, J.M., 2009. Predator effects on
aquatic community assembly: disentangling the roles of habitat selection and postcolonization processes. Oikos 118: 1219–1229.
Wilbur, H.M., 1987. Regulation of structure in complex systems: experimental
temporary pond communities. Ecology 68: 1437–1452.
Wiltshire, D.J., Bull, C.M., 1977. Potential competitive interactions between larvae of
Pseudophryne bibroni and P. semimarmorata (Anura: Leptodactylidae).
Australian Journal of Zoology 25: 449–454
95
Chapter 4 - Compensatory ponds provide poor habitats for
the conservation of frogs associated with naturally
oligotrophic, acidic environments
4.1 Abstract
Habitat loss is a key threatening process for many amphibians. Developers attempt to
compensate for habitat loss by either restoring degraded habitat or artificially creating
new habitats. Waterbodies constructed to aid in conservation of amphibians are quickly
colonised by common species, however their utility for the conservation of threatened
species has rarely been evaluated. Four compensatory and four established ponds in
coastal wallum heathland of eastern Australia were surveyed over 20 months to
determine if compensatory ponds were colonised by species from amphibian
assemblages in adjacent established ponds. Additionally, the environmental variables
associated with calling activity and relative abundance of two threatened species
(Litoria olongburensis and Crinia tinnula) within the study area were determined.
Established ponds contained different adult amphibian assemblages compared to
compensatory ponds, with compensatory ponds being less suitable for the threatened
amphibian species L. olongburensis and C. tinnula. Water pH significantly influenced
amphibian assemblages within the landscape. Air temperature was the strongest variable
positively influencing L. olongburensis calling activity, while daily and monthly rainfall
were the strongest variables positively influencing calling activity by C. tinnula.
Furthermore, pH and salinity were positively associated with the relative abundance of
L. olongburensis while salinity, pH, water depth and minimum hydroperiod length were
positively associated with the relative abundance of C. tinnula. Information on the
optimal conditions for detecting L. olongburensis and C. tinnula and their preferred
habitat characteristics need to be considered for restoring and constructing
compensatory habitat for threatened amphibian species associated with naturally
oligotrophic, acidic environments.
96
4.2 Introduction
Habitat loss is a key threatening process of global significance for the majority of
vertebrate taxa (see Wilcove et al. 1998; Brooks et al. 2002; Cushman 2006).
Compensatory, or artificial, habitats are often advocated and implemented when
attempting to mitigate habitat loss or when restoring degraded habitat (Morris et al.
2006; Seaman 2007; reviewed in Brown et al. 2012). Additionally, compensatory
habitats can be implemented as attempts to aid populations during extreme
environmental events (i.e. extended droughts through climate change) (reviewed in
Shoo et al. 2011). Compensatory habitats can be placed into two categories; ‘habitat
creation’, where the type of habitat being created has not previously occurred within the
construction area, or ‘habitat re-creation’, where the type of habitat being created is to
mimic the former habitat (Morris et al. 2006). However, to be effective, compensatory
habitats should mimic the original environment to replace the loss of the original
environment. Unfortunately, there is little evidence to show that all environmental
factors and abundance and richness of species can be accurately/successfully recreated
(Morris et al. 2006; reviewed in Moreno-Mateos et al. 2012). Importantly, the time
taken for compensatory habitats to reach a state that resembles the original environment
can differ among habitat types (Morris et al. 2006; reviewed in Moreno-Mateos et al.
2012). This is of particular concern if compensatory habitats are to protect species that
are already declining in an area.
Despite habitat re-creation being difficult, anthropogenic waterbodies designed to aid in
conservation of amphibians, as well as those that are not, are still utilised by numerous
amphibian species across different environments (Babbitt & Tanner 2000; Hazell et al.
2004; Barry et al. 2008; Rannap et al. 2009; Brand & Snodgrass 2010; reviewed in
Brown et al. 2012). The majority of these studies focus on the northern hemisphere,
with comparatively little research being conducted on southern hemisphere amphibians
(reviewed in Brown et al. 2012), particularly within Australia. Therefore, the value of
compensatory habitats for Australian amphibians is poorly understood and could impact
on their successful design and implementation.
97
Numerous environmental variables may influence successful colonisation of
anthropogenic waterbodies by amphibian species, including waterbody size (Babbitt &
Tanner 2000), waterbody isolation, or habitat suitability (Lehtinen & Galatowitsch
2001). Additionally, environmental variables that influence amphibian assemblages
within natural waterbodies (e.g. predators (Azevedo-Ramos et al. 1999; Hero et al.
2001) or hydroperiod (Babbitt et al. 2003; Babbitt 2005; Moreira et al. 2010)) may also
influence amphibian usage of anthropogenic waterbodies. Environmental variables can
be tolerated at different levels depending on the amphibian species (Gosner & Black
1957; reviewed in Pierce 1985; Christy & Dickman 2002). Therefore, the influence of
environmental variables structuring amphibian assemblages within anthropogenic
waterbodies will be species specific (Brown et al. 2012). Compensatory habitats may
not provide suitable habitat for all species, particularly species that have a narrow
tolerance to environmental factors.
With amphibians experiencing worldwide declines (Stuart et al. 2004) it is imperative
that compensatory waterbodies are monitored to determine their efficiency in providing
suitable habitats for amphibians. This chapter aimed to determine if newly constructed
compensatory ponds contained similar frog assemblages when compared with
established ponds over a two year study period within naturally oligotrophic, acidic
environments. Compensatory habitats were hypothesised to be unsuitable for two
threatened frog species (Litoria olongburensis and Crinia tinnula) that are associated
with low pH waters (hereafter referred to as ‘acid’ frogs) and that habitat suitability
would be a function of water chemistry. Furthermore, the study aimed to determine
which environmental variables influenced the calling activity and relative abundance of
L. olongburensis and C. tinnula to develop an optimised detection method for these two
threatened species to assist in assessing habitat suitability of compensatory ponds.
4.3 Methods
Pond Description
Four compensatory ponds were constructed during the development of a four lane
highway within south-eastern Queensland and north-eastern New South Wales,
98
Australia. Pond construction was aimed to compensate for removal of coastal wallum
heathland that provided habitat for two threatened ‘acid’ frog species, Litoria
olongburensis and Crinia tinnula. Four compensatory ponds (Ponds 1-4) were created
within 30 meters of the roadside construction perimeter. Three established ponds (Ponds
6-8) had minimal anthropogenic disturbance occurring around their perimeter at the
beginning of the survey period while one established pond (Pond 5) was isolated from
native vegetation.
Survey Technique
Monthly surveying at the established and compensatory ponds commenced after
completion of the compensatory ponds on the 20th September 2006 and finished on 10th
May 2008. Surveys in the months of September 2007 and November 2007 were not
conducted due to accessibility issues. Consequently, two additional surveys were
conducted at the beginning and end of October 2007 to compensate for lost surveys. All
ponds had a total of 20 surveys conducted over the study period, with the exception of
compensatory pond 4 that had only 17 completed.
Acoustic and visual surveys were conducted for 10 minutes around the perimeter of
each pond between 16hrs00min and 21hrs15min to maximise the probability of L.
olongburensis being acoustically recorded (Hopkins unpublished data). During each
survey, three to four acoustic playbacks of L. olongburensis and C. tinnula calls were
conducted at randomly positioned points around each pond, with the random playback
points alternating between surveys. Surveys were conducted, where possible, 24 hours
(Hopkins unpublished data) to one week (Lewis & Goldingay 2005) after heavy rainfall
to increase the probability of visually detecting L. olongburensis adults.
Abiotic Pond Characteristics
Abiotic pond characteristics (water pH, salinity and maximum depth) were measured for
each pond before surveying commenced. Water chemistry parameters and water depth
were measured from the deepest section of the ponds on the first survey. This point was
used to measure water depth and water chemistry for all subsequent surveys. Water
99
chemistry was measured in the field until July 2007, after which water was obtained in
the field and analysed back in the laboratory no more than 24 hours after the sample had
been collected. Pond area estimates were calculated using Google Earth satellite
photography by estimating the maximum length multiplied by maximum width of each
waterbody. Distance to nearest vegetation patch was also calculated using Google Earth
satellite photography. Air temperature was measured at each pond either at the
beginning or during the survey period. Monthly, weekly and daily rainfall were
obtained from the Australian Bureau of Meteorology Coolangatta Airport Rainfall
Monitoring Station (Station number 040717), which is located adjacent to the study site.
Statistical Analysis
Water chemistry variables (pH and salinity) were combined separately for each pond
and divided by the number of times water chemistry variables were surveyed for each
individual pond. Minimum and maximum hydroperiod were determined by the lowest
and highest number of consecutive surveys water was present within a pond. Air
temperature was averaged from each pond for each survey period. The total number of
adult frogs detected acoustically over the entire study period was combined for each
species for each pond for analysis of variables influencing relative abundance of C.
tinnula and L. olongburensis. For each survey period the number of calling C. tinnula
and L. olongburensis encountered for each pond was used to determine the influence of
environmental variables on calling activity of C. tinnula and L. olongburensis.
A one-way Analysis of Variance (ANOVA) was used to determine if there were
significant differences in number of C. tinnula individuals between compensatory and
established ponds. One-way ANOVAs were also used to determine if there was a
significant difference in pond area, pH and salinity levels between compensatory and
established ponds.
To determine which environmental variables were correlated, a Spearman Rank
Correlation Test was performed in IBM SPSS Statistics Version 19 (SPSS, Inc., 2009,
Chicago). Variables that had a correlation coefficient value greater than or equal to 0.7
(sensu Babbitt et al. 2003; Garden et al. 2007) were identified as being highly
100
correlated. No variables were highly correlated and thus all variables were used for data
analysis.
Amphibian assemblages for each pond were represented using non-metric
Multidimensional Scaling (NMDS). The Bray-Curtis distance measure was used to
determine pond similarities for individual species relative abundance. Four dimensions
were used to determine stress minimisation. Linear regression was used to determine
significant environmental variables influencing the four MDS axes. Analysis was
performed in the statistical program R (R-core Development Team 2011) using the
vegan (version 2.03) package (Oksanen 2011).
Models focusing on the influence of environmental variables on calling activity and
models focusing on the influence of environmental variables on relative abundance of L.
olongburensis and C. tinnula were constructed a priori. Twenty-three models were
constructed to model L. olongburensis and C. tinnula calling activity and relative
abundance. Predictor variables included in L. olongburensis and C. tinnula calling
activity models included water depth, air temperature and daily, weekly and monthly
rainfall. Predictor variables included in L. olongburensis and C. tinnula relative
abundance models included water depth, pH, salinity, pond size, distance to nearest
vegetated habitat and maximum and minimum hydroperiod encountered during
surveying. Generalised Additive Mixed-effects Models (GAMMs) were used to assess
the importance of environmental variables on calling activity of L. olongburensis and C.
tinnula (using a Poisson link function) with pond as a random effect and survey date as
an auto-correlated structure. Generalised Linear Models (GLM) were used to assess the
importance of environmental variables on relative abundance of L. olongburensis and C.
tinnula (using a Poisson link function).
Akaike’s Information Criterion (AICc) was used to determine model ranking and
selection. The best model was the model with the lowest AICc value (Burnham &
Anderson 2002). To determine the ranking of the models, Δi values were calculated,
where higher Δi values indicated less accurate models for the given data (Burnham &
Anderson 2002; Johnson & Omland 2004). If a model had a Δi ≤ 2, then there was
considerable evidence that the model could be the “best” model, given the data (Johnson
101
& Omland 2004). If a model had a Δi 2-4 then there was considered to be moderate
evidence that the model could be the “best” model, given the data. Akaike Weights (wi)
were assigned to each model to enable greater interpretation of the relative likelihood of
a model (Burnham & Anderson 2002; Johnson & Omland 2004). The closer the wi was
to 1, the stronger the model for the given data (Burnham & Anderson 2002). To
determine the relative importance of variables within models where Δi < 4 the wi values
were summed from all models where the variable of interest occurred (Grueber et al.
2011). The closer the variable of interest was to 1 the higher the importance of the
variable. Models were run in R version 2.14.0 (R Core Development Team, 2011) using
the ‘MuMIn’ (version 1.7.2) package.
4.4 Results
Waterbody Characteristics
Water pH was significantly higher (F1/95 = 39.513, p < 0.001) between compensatory
ponds (4.00 – 7.54; average = 5.94 +/- 0.80) and established ponds (3.04 – 7.31; average
= 4.68 +/- 1.09). There was no significant difference in salinity levels (F1/77 = 1.283, p
= 0.261) between compensatory ponds (0.02ppm – 0.62ppm, average = 0.16ppm +/0.17ppm) and established ponds (0.01ppm – 0.71ppm, average = 0.15ppm +/0.17ppm). Similarly, pond size did not differ significantly between compensatory and
established ponds (F1/6 = 6.093, p = 0.195) with the largest and smallest compensatory
pond being 206.09m2 and 82.51m2, respectively, while the largest and smallest
established ponds were 1408.1m2 and 208.8m2, respectively.
Water was artificially added to the compensatory ponds in an attempt to ensure water
was present within compensatory ponds, altering the ‘natural’ hydroperiods of the
ponds. Artificial watering is believed to have resulted in water being present over a
majority of the surveys for compensatory ponds one and three (17/20 surveys and 20/20
surveys). However, compensatory ponds two and four failed to maintain water for
similar extended periods (11/20 surveys and 3/17 surveys). Established ponds did not
receive artificial watering. Water length varied between established ponds, with water
present during 17/20, 17/20, 11/20 and 12/20 surveys.
102
Minimum average air temperature recorded for the surveys was 14°C while maximum
average temperature was 28°C. Minimum monthly rainfall was 9mm while maximum
monthly rainfall was 279.2mm. Minimum weekly rainfall was 0.2mm while maximum
weekly rainfall was 141mm. Minimum daily rainfall was 0mm while maximum daily
rainfall was 24mm.
Compensatory and Established Pond Comparisons
Ten amphibian species (Crinia signifera, Crinia tinnula, Litoria fallax, Litoria nasuta,
Litoria olongburensis, Litoria tyleri, Litoria peroni, Limnodynastes peronii, Opisthodon
ornatus and Rhinella marina) were recorded from both established and compensatory
ponds. An additional two species (Litoria rubella and Limnodynastes terraereginae)
were only recorded at established ponds (Table 4.1). Individual species abundance
varied between established and compensatory ponds, with threatened specialist species
(L. olongburensis and C. tinnula) abundance being higher in established ponds (Table
4.1). Crinia tinnula abundance was significantly lower in compensatory waterbodies
when compared with established waterbodies (F1/150 = 0.410, p = 0.003).
Non-metric multidimensional scaling using a Bray-Curtis distance measure was used to
determine pond similarities for species assemblages between ponds (Figure 4.1). pH
(49.7%) and depth (40.8%) explained the largest portion of MDS axis 1 (Table 4.2).
However, only pH was significant in explaining the variation in species assemblages
between ponds while depth was nearing significance (Table 4.2). The stress value
associated with the NMDS plot was 0.001 (Figure 4.1).The influence of pH on the
amphibian assemblage structure is clearly seen when plotted against the raw amphibian
abundance data for each species (Figure 4.2).
103
Table 4.1: Total number of individuals per species detected over the survey period for compensatory and established waterbodies. *
indicate threatened species and ^ indicate introduced species listed under the Australian EPBC Act 1999.
Compensatory Ponds
Species
Established Ponds
Pond 1
Pond 2
Pond 3
Pond 4
Total
Pond 5
Pond 6
Pond 7
Pond 8
Total
Crinia signifera
2
0
1
0
3
0
2
1
2
5
Crinia tinnula*
18
5
16
5
44
32
51
64
28
175
Litoria fallax
26
5
8
1
40
2
6
0
3
11
Litoria nasuta
0
0
0
1
1
0
0
4
5
9
Litoria olongburensis*
1
0
0
0
1
9
7
15
21
52
Litoria tyleri
0
0
1
0
1
11
25
3
0
39
Litoria rubella
0
0
0
0
0
0
1
0
0
1
Limnodynastes peronii
10
12
0
0
22
0
0
6
9
15
Limnodynastes
0
0
0
0
0
4
3
1
1
9
Opisthodon ornatus
1
0
0
0
1
0
1
0
0
1
Rhinella marina^
3
12
8
1
23
0
6
1
1
8
terraereginae
104
Figure 4.1: nMDS ordination of amphibian species composition using Axis 1 and 2
from the MDS amphibian species abundance matrix. Black dots represent compensatory
ponds while white dots represent established ponds. Species positions within the matrix
are displayed.
105
Figure 4.2: Gradient analysis using average pH as a gradient with abudance of each
species recorded across the survey period. N represents a natural pond while C
represents a compensatory pond.
106
Table 4.2: Correlations to the MDS Axis 1-4 with variables playing a significant
influence on assemblage structure highlighted in bold. A significant influence was
considered a variable that had a p value less than 0.05. A* indicates significant variables
while a # indicates a variable nearing significance (p = 0.052).
Variable
MDS Axis 1
MDS Axis 2
MDS Axis 3
MDS Axis 4
0.4973*
-0.0819
-0.1421
-0.1038
-0.1
-0.1365
-0.1353
0.0928
0.265
-0.0626
-0.1204
-0.02292
-0.1574
-0.0566
0.1815
-0.0229
Depth
0.4087#
0.1643
-0.0761
-0.0898
Distance to
0.1387
0.1991
-0.1658
-0.0652
-0.0641
-0.0237
0.1247
-0.03957
pH
Salinity
Pond Size
Hydroperiod
(min)
nearest habitat
Hydroperiod
(max)
Factors influencing relative abundance of L. olongburensis and C. tinnula
One model for L. olongburensis abundance had a Δi < 2. The weighting of this best
model was 78.1%, suggesting that all other models compared poorly (Table 4.3). The
best model contained pH and salinity (Table 4.3), with both these variables having
strong relative importance (Table 4.4). One model also had a Δi 2-4 and a weighting of
16.4%. This model contained pH, salinity and depth. Combined, models with a Δi < 4
had a weighting of 94.5% (Table 4.3). However, relative variable importance
confidence intervals for depth included zero, reducing evidence that this variable had a
strong association with L. olongburensis abundance (Table 4.4).
Two models for C. tinnula abundance had a Δi < 2. The weighting of these best models
was 97.6%, suggesting that the other models compared poorly (Table 4.3). The best
models contained pH, salinity, depth and minimum hydroperiod length (Table 4.3).
With the exception of pH, all of these variables had strong relative variable importance
(Table 4.4).
107
L. olongburensis and C. tinnula calling activity
One model for L. olongburensis calling activity had a Δi < 2. The weighting of this best
model was 86.1%, suggesting that the other models compared poorly (Table 4.3). The
best model contained air temperature and monthly survey, which were both positively
associated with calling activity of L. olongburensis (Table 4.3). Both of these variables
had strong relative importance (Table 4.4), however, relative variable importance
confidence intervals for these variables included zero, reducing the evidence that these
variables were strongly associated with L. olongburensis calling. All other models had a
Δi > 4.
One model for C. tinnula calling activity had a Δi < 2. The weighting of this best model
was 43.5% where monthly rainfall, daily rainfall and monthly survey were all positively
associated with C. tinnula calling activity. Four models for C. tinnula calling activity
had a Δi 2-4 with a combined weighting of 39.8%. The total weighting of all models
with Δi < 4 was 83.3%, suggesting other models compared poorly. Variables within
models with a Δi 2-4 were daily rainfall, weekly rainfall, monthly rainfall, water depth
and survey date, with all variables except weekly rainfall being positively related to C.
tinnula calling activity (Table 4.3). The relative importance of variables that were in
models with a Δi 2-4 and not the model with a Δi < 2 were low when compared with
variables in the model with a Δi < 2 (Table 4.4). Additionally, with the exception of
monthly rainfall, all variables had relative variable importance confidence intervals that
included zero, indicating reduced evidence that these variables have a strong association
on C. tinnula calling activity (Table 4.4).
4.5 Discussion
Influence of environmental variables on the amphibian assemblage
Established and compensatory ponds supported different amphibian assemblages, with
pH significantly associated with amphibian assemblages across the landscape. Frog
assemblages within compensatory ponds were primarily influenced by high pH.
Compensatory pond species exclusion or reduction in abundance from established
108
Table 4.3: Models with a Δi value < 4 for L. olongburensis and C. tinnula calling
activity and relative abundance. + indicates a positive relationship while – indicates a
negative relationship to L. olongburensis or C. tinnula calling activity for the variable
within the model.
AICC
ΔAICC
w
534.7
0.00
0.861
pH + salinity
38.4
0.00
0.781
pH + salinity + depth
41.5
3.13
0.164
363.1
0.00
0.435
365.1
2.10
0.152
366.1
2.99
0.097
Water Depth + Monthly Survey
366.3
3.25
0.086
Daily Rainfall + Water Depth + Monthly
366.9
3.85
0.063
Depth + Salinity + Minimum Hydroperiod
67.6
0.00
0.698
Depth + Salinity + pH
69.5
1.85
0.278
Model
Litoria olongburensis calling activity
Air Temperature + Monthly Survey
Litoria olongburensis abundance
Crinia tinnula calling activity
Monthly Rainfall + Daily Rainfall + Monthly
Survey
Monthly Rainfall + Daily Rainfall + Water
Depth + Monthly Survey
Monthly Rainfall – Weekly Rainfall + Daily
Rainfall + Monthly Survey
Survey
Crinia tinnula abundance
109
Table 4.4: Model averaged coefficients for models where Δi < 4 for L. olongburensis
and C. tinnula calling activity and relative abundance. Relative importance of each
environmental predictor variable is displayed.
Variable
Estimate
S.E. †
Confidence
Rel. var.
Interval
imp.‡
Litoria olongburensis
calling activity
Monthly Survey
0.102
0.055
-0.005, 0.209
1
Air Temperature
0.077
0.068
-0.056, 0.21
0.87
-14.204
0.436
-26.551, -1.859
1.00
pH
-2.03
5.063
-3.155, -0.905
0.99
Depth
0.192
0.128
-0.163, 0.546
0.17
Monthly Survey
0.033
0.03
-0.027, 0.092
1
Daily Rainfall
0.015
0.009
-0.004, 0.034
0.78
Monthly Rainfall
0.004
0.001
0.001, 0.007
0.76
Water Depth
0.012
0.007
-0.002, 0.026
0.43
Weekly Rainfall
-0.001
0.003
-0.006, 0.005
0.16
Salinity
-4.584
0.8449
-6.899, -2.269
1.00
Water Depth
0.085
0.0266
-1.112, -0.203
0.98
Minimum Hydroperiod
-0.2172
0.0551
-0.37, -0.064
0.70
pH
-0.6578
0.1755
-1.112, -0.203
0.30
Litoria olongburensis
abundance
Salinity
Crinia tinnula calling
activity
Crinia tinnula abundance
† Standard Error ‡ Relative variable importance
110
ponds is likely due to species tolerances to differing levels of pH. For example, postGosner Stage 25 tadpoles of C. tinnula successfully metamorphose when exposed,
under laboratory conditions, to pH levels of 3.5 (Meyer 2004). Additionally, adults of
species comprising the majority of established pond assemblages (i.e. C. tinnula and L.
olongburensis) successfully breed, in previous surveys, in waterbodies where minimum
pH was between 3.2-3.4 (Meyer 2004). The majority of species occurring within
compensatory ponds (i.e. L. fallax and L. peronii) were recorded from reproductive
waterbodies where minimum pH was 5.0 (Meyer 2004). Therefore, the failure of
compensatory pond species occurring in higher or equal numbers in the established
ponds is likely due to intolerance to low pH levels. However, the low abundance of
species comprising the established pond assemblage when compared with compensatory
ponds is unlikely to be attributed to an intolerance of high pH waters as some species
(C. tinnula and L. terraereginae) can successfully metamorphose when exposed to pH
waters of 6.5 (Meyer 2004). The low abundance or exclusion of established pond
species (i.e. L. olongburensis) from compensatory ponds could be contributed to
competition from compensatory pond species. While no studies have been conducted to
support this theory, it has been hypothesised that competition occurs between L. fallax
and L. olongburensis (Meyer et al. 2006). Furthermore, studies on other amphibian
species have revealed competitive interactions (Wiltshire & Bull 1977; Twomey et al.
2008), demonstrating competition can occur between amphibian species. Additionally,
data from Chapter 2 indicates that competition may be restricting L. olongburensis at
the upper limits of the ‘ideal’ pH range.
Variables influencing relative abundance of L. olongburensis and C. tinnula
Salinity and pH were both negatively associated with L. olongburensis and C. tinnula
abundance while depth was positively associated with L. olongburensis and C. tinnula
abundance. High salinity levels have been found to negatively influence survival of
other amphibian tadpoles (Christy & Dickman 2002; Chinathamby et al. 2006; RiosLópez 2008). The negative relationship with salinity and C. tinnula and L.
olongburensis suggests that these species may be avoiding higher salinity levels
encountered during this survey due to tadpole intolerance to high salinity levels.
111
However, salinity tolerance experiments on these species are required to fully discern
this.
As discussed previously, the negative association between pH and L. olongburensis and
C. tinnula abundance can be attributed to the these species having a broader tolerance to
low pH conditions as opposed to sister species, which are less tolerate to these
conditions. Furthermore, in high pH waters, competition from non-acid frog species is
likely excluding these two acid-frog species from occurring in ponds with high pH
while non-acid frog species tadpole intolerance is excluding amphibians within low pH
ponds. This would allow acid tolerant species to occur in low pH ponds without the
negative effect of competition.
Minimum hydroperiod was negatively associated with C. tinnula abundance. Crinia
tinnula are often associated with ephemeral waterbodies (Meyer et al. 2006). Shorter
hydroperiods would result in different predator assemblages when compared with
waterbodies with longer hydroperiods (Welborn et al. 1996; Hero et al. 1998; Babbitt et
al. 2003; Jocqué et al. 2007; Richter-Boix et al. 2007; Fernandes et al. 2010).
Therefore, the negative association between C. tinnula and hydroperiod is likely due to
the absence of more permanent aquatic tadpole predators, where C. tinnula would be
unlikely to co-exist with. This can be seen in Chapter 2 of this thesis, that would inhibit
coexistence of C. tinnula (see Chapter 2).
The positive association between water depth and L. olongburensis and C. tinnula is
likely due to the species reproductive life history. Both species have tadpoles that occur
in the water (Anstis 2007) and would need to deposit their eggs when risk of pond
desiccation was low. For example, amphibian species may select their oviposition sites
(Resetarits Jr & Wilbur 1989; Resetarits Jr 1996), with spawning occurring in ponds
when water levels are at their peak (Goldberg et al. 2006), or in waterbodies that have
the longest hydroperiod (Spieler & Linsenmair 1997).
The positive association between water depth and then negative association of minimum
hydroperiod and C. tinnula could be perceived as ‘cancelling’ each other out.
Dependant on the scenario, C. tinnula may be to seek a trade-off between pond
112
reliability and high risk of predation (increased water depth) and pond unreliability and
safety (ephemeral/minimum hydroperiod).
Calling activity of L. olongburensis and C. tinnula
Air temperature was positively associated with L. olongburensis calling activity while
previous daily and previous monthly rainfall were the variables with the highest relative
importance for C. tinnula calling activity. Previous studies have shown that calling
would not occur if air temperature fell below a certain threshold (Howard 1980; Saenz
et al. 2006; Van Sluys et al. 2012), or where calling would be associated with, amongst
other environmental variables, time of year and rainfall (Lemckert 2001; Saenz et al.
2006). The association between increased calling activity in L. olongburensis and air
temperature can be related to amphibian biology. Amphibians are ectotherms and are
likely to increase activity levels with increased temperature (Wells 2007). Daily and
monthly rainfall increasing calling activity for C. tinnula is likely a reproductive
response as the species attempts to breed when waters are peaking and risk of pond
desiccation would be at its lowest.
Suitability of compensatory ponds for threatened species
The results presented here demonstrate these compensatory ponds are unable to mirror
amphibian species assemblages of established ponds. Furthermore, these compensatory
ponds were less effective in maintaining populations of the threatened, specialist
species, C. tinnula and L. olongburensis, when compared with established ponds.
Compensatory pond pH ranges were within the pH tolerance range (3.5 - 6.5 (Meyer
2004)) for C. tinnula tadpoles, suggesting that successful breeding could occur.
However, the lack of any C. tinnula tadpole records within compensatory ponds (C.
Simpkins, unpublished data) reduces the chances that successful breeding of C. tinnula
occurred within compensatory ponds. Previous reports have noted C. tinnula association
with disturbed waterbodies in wallum heathland (Meyer et al. 2006). However, this
report did not mention if these populations successfully reproduced or if population
numbers were the same in non-disturbed waterbodies (Meyer et al. 2006).
113
In this study, compensatory ponds were ineffective at maintaining populations of L.
olongburensis and contained higher numbers of L. fallax when compared with
established ponds. Sympatric co-existence between L. fallax and L. olongburensis rarely
occurs, with L. fallax being proposed as a competitor of L. olongburensis (Meyer et al.
2006). Previous studies have found L. olongburensis adults occurring in ponds with pH
between 3.8 – 4.6 (Hopkins unpublished data), 3.5 – 5.2 (Hero unpublished data) and
3.11 – 5.02 (Shuker unpublished data). Furthermore, the minimum pH of waterbodies
within wallum where L. fallax have been found to occur is 5.0, with L. fallax eggs or
tadpole death occurring when experimentally exposed to water of pH 3.5 (Meyer 2004).
The mean pH of the compensatory waterbodies fell outside of the ranges where L.
olongburensis have been detected while established pond pH averages fell within these
ranges. Therefore, reduced numbers of L. olongburensis across compensatory ponds
may be from L. fallax competition while the low abundance of L. fallax within
established ponds is likely due to L. fallax larval intolerance to low pH waters.
Conclusions and implications for future establishment of compensatory habitats
The compensatory ponds surveyed were inadequate in their provision of suitable
replacement habitats for wallum based amphibian species. The pH levels of the
compensatory ponds are unlikely to exclude competitive species and/or their successful
reproduction. Any attempts for future construction of compensatory habitats for
threatened amphibian species need to take this into consideration and construct ponds
where pH is kept within the optimal range of the species concerned. Additionally, pond
depth needs to be within the deeper sections of the range detected for this survey. If
construction of compensatory habitats for ‘acid’ frog species in eastern Australia is
undertaken, post construction monitoring is recommended to determine if compensatory
habitats are being utilised by C. tinnula and L. olongburensis. To maximise acoustic
detection of C. tinnula and L. olongburensis within both compensatory and established
ponds surveys should be conducted towards the upper limits of daily and/or monthly
rainfall and the upper ranges of temperatures encountered during surveys. These results
may only apply to L. olongburensis and C. tinnula populations within the study area and
additional surveys of L. olongburensis and C. tinnula are required across the entire
114
distributional range of these species to determine if the variables within this study
influence calling activity within other populations.
Acknowledgments
I would like to thank Jean-Marc Hero for providing me with the opportunity to work on
the project that produced this chapter. I would also like to thank Stephen Lamb and Jon
Shuker for helping with data collection. I would like to thank Queensland Main Roads
for providing funding for data collection. I would like to thank Katrin Lowe, Jean-Marc
Hero and Guy Castley for feedback on drafts of this chapter.
115
4.6 References
Anstis, M. (2002). Tadpoles of south-eastern Australia: a guide with keys. New Holland
Publishers, Sydney.
Azevedo-Ramos, C., W. E. Magnusson, and P. Bayliss. 1999. Predation as the key
factor structuring tadpole assemblages in a savanna area in Central Amazonia.
Copeia 1999:22-33.
Babbitt, K. J. 2005. The relative importance of wetland size and hydroperiod for
amphibians in southern New Hampshire, USA. Wetlands Ecology and
Management 13:269-279.
Babbitt, K. J., M. J. Baber, and T. L. Tarr. 2003. Patterns of larval amphibian
distribution along a wetland hydroperiod gradient. Canadian Journal of Zoology
81:1539-1552.
Babbitt, K. J., and G. W. Tanner. 2000. Use of temporary wetlands by anurans in a
hydrologically modified landscape. Wetlands 20:313-322.
Barry, D. S., T. K. Pauley, and J. C. Maerz. 2008. Amphibian use of man-made pools
on clear-cuts in the Allegheny Mountains of West Virginia, USA. Applied
Herpetology 5:121-128.
Brand, A. B., and J. W. Snodgrass. 2010. Value of artificial habitats for amphibian
reproduction in altered landscapes. Conservation Biology 24:295-301.
Brooks, T. M. et al. 2002. Habitat loss and extinction in the hotspots of biodiversity.
Conservation Biology 16:909-923.
Brown, D. J., G. M. Street, R. W. Nairn, and M. R. J. Forstner. 2012. A place to call
home: amphibian use of created and restored wetlands. Journal of Ecology 2012:
doi:10.1155/2012/989872.
Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference:
a practical information-theoretic approach. Springer, New York.
Chinathamby, K., R. D. Reina, P. C. E. Bailey, and B. K. Lees. 2006. Effects of salinity
on the survival, growth and development of tadpoles of the brown tree frog, Litoria
ewingii. Australian Journal of Zoology 54:97-105.
Christy, M. T., and C. R. Dickman. 2002. Effects of salinity on tadpoles of the green
and golden bell frog (Litoria aurea). Amphibia-Reptilia 23:1-11.
Cushman, S. 2006. Effects of habitat loss and fragmentation on amphibians: a review
and prospectus. Biological Conservation 128:231-240.
Fernandes, I. M., F. A. Machado, and J. Penha. 2010. Spatial pattern of a fish
assemblage in a seasonal tropical wetland: effects of habitat, herbaceous plant
116
biomass, water depth, and distance from species sources. Neotropical Ichthyology
8:289-298.
Garden, J. G., C. A. McAlpine, H. P. Possingham, and D. N. Jones. 2007. Using
multiple survey methods to detect terrestrial reptiles and mammals: what are the
most successful and cost-efficient combinations? Wildlife Research 34:218-227.
Goldberg, F. J., S. Quinzio, and M. Vaira. 2006. Oviposition-site selection by the toad
Melanophryniscus rubriventris in an unpredictable environment in Argentina.
Canadian Journal of Zoology 84:699-705.
Gosner, K. L., and I. H. Black. 1957. The effects of acidity on the development and
hatching of New Jersey Frogs. Ecology 38:256-262.
Hazell, D., J.-M. Hero, D. Lindenmayer, and R. Cunningham. 2004. A comparison of
constructed and natural habitat for frog conservation in an Australian agricultural
landscape. Biological Conservation 119:61-71.
Hero, J.-M., C. Gascon, and W. E. Magnusson. 1998. Direct and indirect effects of
predation on tadpole community structure in the Amazon rainforest. Australian
Journal of Ecology 23:474-482.
Hero, J.-M., W. E. Magnusson, C. F. D. Rocha, and C. P. Catterall. 2001. Antipredator
defenses influence the distribution of amphibian prey species in the Central
Amazon Rain Forest. Biotropica 33:131-141.
Howard, R. D. 1980. Mating behaviour and mating success in woodfrogs, Rana
sylvatica. Animal Behaviour 28:705-716.
Jocqué, M., T. Graham, and L. Brendonck. 2007. Local structuring factors of
invertebrate communities in ephemeral freshwater rock pools and the influence of
more permanent water bodies in the region. Hydrobiologia 592:271-280.
Johnson, J. B., and K. S. Omland. 2004. Model selection in ecology and evolution.
Trends in Ecology & Evolution 19:101-108.
Lehtinen, R. M., and S. M. Galatowitsch. 2001. Colonization of restored wetlands by
amphibians in Minnesota. The American Midland Naturalist 145:388-145.
Lemckert, F. 2001. The influence of micrometeorological factors on the calling activity
of the frog Crinia signifera (Anura: Myobatrachidae). Australian Zoologist 31:625
- 631.
Lewis, B. D., and R. L. Goldingay. 2005. Population monitoring of the vulnerable
wallum sedge frog (Litoria olongburensis) in north-eastern New South Wales.
Australian Journal of Zoology 53:185-194.
117
Meyer, E. 2004. Acid adaptation and mechanisms for softwater acid tolerance in larvae
of anuran species native to the “Wallum” of east Australia. PhD Thesis, University
of Queensland.
Moreira, L. F. B., I. F. Machado, T. V. Garcia, and L. Maltchik. 2010. Factors
influencing anuran distribution in coastal dune wetlands in southern Brazil. Journal
of Natural History 44:1493-1507.
Moreno-Mateos, D., M. E. Power, F. A. Comı, and R. Yockteng. 2012. Structural and
functional loss in restored wetland ecosystems. PLoS Biology 10:
doi:10.1371/journal.pbio.1001247..
Morris, R. K. a., I. Alonso, R. G. Jefferson, and K. J. Kirby. 2006. The creation of
compensatory habitat—can it secure sustainable development? Journal for Nature
Conservation 14:106-116.
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B.,
Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2012. Vegan:
Community Ecology Package, version 2.0-4. Page http://vegan.r-forge.rproject.org/
Pierce, B. A. 1985. Acid tolerance in amphibians. BioScience 35:239-243.
Rannap, R., A. Lõhmus, and L. Briggs. 2009. Restoring ponds for amphibians: a
success story. Hydrobiologia 634:87-95.
Resetarits Jr, W. J. 1996. Oviposition site choice and life history evolution. American
Zoologist 36:205-215.
Resetarits Jr, W. J., and H. M. Wilbur. 1989. Choice of oviposition site by Hyla
chrysoscelis: role of predators and competitors. Ecology 70:220-228.
Richter-Boix, A., G. A. Llorente, and A. Montori. 2007. A comparative study of
predator-induced phenotype in tadpoles across a pond permanency gradient.
Hydrobiologia 583:43-56.
Rios-López, N. 2008. Effects of increased salinity on tadpoles of two anurans from a
Caribbean coastal wetland in relation to their natural abundance. AmphibiaReptilia 29:7-18.
Saenz, D., L. A. Fitzgerald, K. A. Baum, and R. N. Conner. 2006. Abiotic correlates of
anuran calling phenology: the importance of rain, temperature, and season.
Herpetological Monographs 20:64-82.
Seaman, W. 2007. Artificial habitats and the restoration of degraded marine ecosystems
and fisheries. Hydrobiologia 580:143-155.
Shoo, L. P. et al. 2011. Engineering a future for amphibians under climate change.
Journal of Applied Ecology 48:487-492.
118
Van Sluys, M., R. V. Marra, L. Boquimpani-Freitas, and C. F. D. Rocha. 2012.
Environmental factors affecting calling behavior of sympatric frog species at an
Atlantic Rain Forest Area, Southeastern Brazil. Journal of Herpetology 46:41-46.
Spieler, M., and K. E. Linsenmair. 1997. Choice of optimal oviposition sites by
Hoplobatrachus occipitalis (Anura: Ranidae) in an unpredictable and patchy
environment. Oecologia 109:184-199.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodrigues, D. L.
Fischman, and R. W. Waller. 2004. Status and trends of amphibian declines and
extinctions worldwide. Science 306:1783-1786.
Twomey, E., V. Morales, and K. Summers. 2008. Evaluating condition-specific and
asymmetric competition in a species-distribution context. Oikos 117:1175-1184.
Welborn, G. A., D. K. Skelly, and E. E. Werner. 1996. Mechanisms creating
community structure across a freshwater habitat gradient. Annual Review of
Ecology and Systematics 27:337-363.
Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying
threats to imperiled species in the United States. BioScience 48:607-615.
Wiltshire, D. J., and C. M. Bull. 1977. Potential competitive interactions between larvae
of Pseudophryne bibroni and P. semimarmorata (Anura: Leptodactylidae).
Australian Journal of Zoology 25:449-454.
119
Chapter 5 - Comparison of predation rates between the
introduced mosquito fish (Gambusia holbrooki) and native
aquatic predators on L. olongburensis, L. fallax and
Limnodynastes peronii tadpoles
5.1 Abstract
Predation by aquatic predators is often an underlying factor in structuring tadpole
assemblages. Predation effects may increase when tadpole communities are exposed to
‘exotic’ predators, as evolved anti-predator strategies may be ineffective. Few studies have
compared predation rates of introduced and native tadpole predators on tadpole species.
This chapter tested and compared predator effectiveness of an introduced predator
(Gambusia holbrooki) with one native fish species (Hypseleotris galii) and four native
aquatic invertebrates (Belostomatidae (water scorpion), Ashnidae (dragonfly larvae),
Zygoptera (mayfly larvae) and Cherax sp. (crayfish)) on ‘small’ and ‘large’ tadpoles of
Limnodynastes peronii (Striped Marsh Frog), Litoria fallax (Eastern Sedge Frog) and
Litoria olongburensis (Wallum Sedge Frog). Predator effectiveness was dependent on the
tadpole and predator species; Belostomatidae were the most effective predators within
single prey experiments while Cherax sp. were the most effective predators within multiple
prey experiments. Surprisingly, in the single tadpole experiments, the native fish H. galii
did not attack any tadpoles. However, the introduced predator G. holbrooki attacked 3 of 12
tadpoles. Within the multiple prey experiments, H. galii consumed an average of 4.87
tadpoles while G. holbrooki consumed an average of 2.23 tadpoles. However, the predator
species that consumed the highest average number of tadpoles differed between tadpole
species. These findings indicate that while G. holbrooki were either less effective or equal
predators of tadpole. However, these results indicate that native predatory species may have
greater potential to influence the tadpole assemblage than G. holbrooki.
120
5.2 Introduction
Predation by aquatic predators is often an underlying factor structuring tadpole assemblages
(Hero et al., 1998; Hero et al., 2001; Vonesh et al., 2009), with aquatic tadpole predators
including, but not limited to, predatory fish (Hero et al., 2001; Baber and Babbitt, 2003;
Gregoire and Gunzburger, 2008; Nelson et al., 2011) crayfish (Axelsson et al., 1997;
Stoneham, 2001), Odonate larvae (Jara, 2008; Álvarez and Nicieza, 2009) and predacious
tadpoles (Heyer et al., 1975; Wells, 2007).
Tadpole predation risk is determined by a number of factors, but primarily predator
assemblages and predator abundance, which varies among waterbodies (Woodward, 1983;
Hero et al., 1998). Furthermore, a predators ability to exclude or reduce tadpole populations
will also be influenced by the tadpoles anti-predator strategies, which can include, but are
not limited to, unpalatability/chemical defences (Kats et al., 1988; Crossland, 2001; Hero et
al., 2001; Gunzburger and Travis, 2005), behavioural avoidance (Skelly, 1994; Saidapur et
al., 2009; Smith and Awan, 2009) or morphological adaptations (Hecnar and M'Closkey,
1997; McCollum and Leimberger, 1997; Touchon and Warkentin, 2008).
Anti-predator strategies may only be effective against specific predators. Unpalatability, for
example, may work at deterring predatory fish species but be ineffective against some
aquatic invertebrate predators (Azevedo-Ramos et al., 1992; Hero et al., 2001).
Additionally, certain strategies will be more effective at particular stages of development,
with palatability found to decrease with development in Rhinella marina (previously Bufo
marinus) tadpoles (Lawler and Hero, 1997). Increased tadpole body size can decrease the
risk of attacks from predatory species (Jara, 2008), thereby influencing tadpole predation
rates.
The introduction of non-native predators into waterbodies has been hypothesised as
contributing towards amphibian declines (Morgan and Buttemer, 1996; Gillespie and Hero,
1999; Kats and Ferrer, 2003; Pyke, 2008), due to tadpole species not evolving adequate
anti-predator defences to prevent predation by the introduced predator (Kats and Ferrer,
121
2003). Hence, determining the effectiveness of an introduced predator on prey populations
cannot be fully discerned without comparing the influence of native predators on the prey
population. To my knowledge, only one study has conducted preliminary experiments to
determine if native aquatic predators of tadpoles have the same predation levels as G.
holbrooki (Pyke and White, 2000), an introduced fish species in several countries. This
study found that predation levels on Litoria aurea tadpoles were lower for native predators
when compared with G. holbrooki (Pyke and White, 2000).
To expand on this previous work this chapter examines the predation rates for native
aquatic predators (Cherax sp., Hypseleotris galii, Belostomatidae, Aeshnidae and
Zygoptera) and one introduced aquatic predator (G. holbrooki) on larvae of a threatened
‘acid’ tadpole species (Litoria olongburensis) and a ‘non-acid’ species (Litoria fallax),
which has been identified as a potential competitor of L. olongburensis (Meyer et al.,
2006). We also tested predator effectiveness on tadpoles of Limnodynastes peronii, a
widespread species known to occur within tadpole assemblages containing both L. fallax
and L. olongburensis. Results of this study can aid in determining the influence of G.
holbrooki on amphibian communities in waterbodies where G. holbrooki have been
introduced. It should be noted that results obtained for this study are from an artificial
environment and interpretation of these results into the ‘real world’ environment needs to
be extended with future studies.
5.3 Methods
Collection of egg and larvae
Fertilised eggs from L. fallax were obtained by placing one adult male and one adult female
frog into a plastic zip-lock bag (22 cm x 25 cm) that contained approximately 500mL of
water. Water was obtained from the waterbody where the adults were caught. Sedges or
other forms of aquatic vegetation were also placed in the bag. The pair of frogs were then
left overnight to facilitate spawning. This was repeated several times using different
individual frogs until a sufficient number of eggs had been obtained to conduct predation
122
experiments. Several clutches of eggs from L. fallax were obtained from an ephemeral
wetland (-27.965712°, 153.379338°) on the 23rd September 2011 and the 18th January 2012.
Two additional L. fallax egg clutches were also obtained from a permanent wetland at
Currumbin Wildlife Sanctuary, Currumbin, Queensland, Australia (-28.139091°,
153.484030°) on the 15th November 2011. Eggs from L. peronii were obtained from three
foam egg masses found in a permanent waterbody at Musgrave Park, Southport,
Queensland, Australia (-27.957785°, 153.394208°) on the 25th August 2011. Eggs were
placed into an eight litre plastic bucket with water from the waterbody where their parents
were captured or from where eggs were collected. Eggs hatched in the bucket and were fed
frozen lettuce until tadpoles reached lengths between 8-13mm (hereafter referred to as
‘small’) and 16-27mm (hereafter referred to as ‘large’).
Several attempts using the described methods for obtaining eggs from L. fallax failed to
produce any L. olongburensis adult pairs spawning eggs. Therefore, tadpoles were dipnetted from a semi-permanent waterbody at Bribie Island National Park (-27.074697°,
153.178290°). Tadpoles from L. olongburensis were between 21-34 mm for experiments
containing multiple tadpoles and 8-38 mm for experiments containing one tadpole.
Predation Experiments
White plastic containers (43 cm length x 34 cm width x 17 cm height) holding
approximately eight litres of pond water were used for predation experiments. A secure
fibreglass mesh lid (mesh size 0.1cm x 0.1cm) was placed over each container to ensure
no interference from ‘outside’ fauna occurred and to ensure that ‘climbing’ predators
(i.e. crayfish) could not escape. Water was collected from the surface of the waterbody to
reduce the amount of detritus that was collected. Predator-prey experiments were
conducted next to waterbodies where the predators were collected to minimise the risks
of transporting predators and to expose experiments to natural climatic conditions.
Potential predators were captured using aquatic traps (38cm length x 25cm width x 25cm
height, with two square shaped entrances both 5cm x 5cm) and were baited with
123
approximately 20 grams of floating fish food pellets. Aquatic invertebrates were caught
using dip-netting techniques. Individual predators and tadpoles were measured to the
nearest millimetre using a 30 centimetre ruler before placement into experimental
containers. Fish were measured from the mouth to the tail tip, crayfish from the rostrum to
the telson, and tadpoles from the mouth to the tip of the flagellum.
Individual L. olongburensis tadpole predation experiments
Due to the difficulty in finding tadpoles of L. olongburensis, only one L. olongburensis
tadpole and one predator were placed into a plastic container at 1200hrs and signs of
predation were checked at 1800hrs the following day. This allowed for 30 hours of
predator-prey interaction time. These individual interactions were replicated for each
predator, with replication being dependant on the number of predators captured (Figure
1). A maximum of 20 experiments were conducted at a time. Variation in time between
predator capture and placement within the containers varied between one to two hours.
No tadpoles or predators were re-used in any experiments and were released at the site of
capture at the end of each experiment. Tadpoles were categorised as attacked if visual
signs of predatory damage were visible on the flagellum and categorised as
consumed/eaten if no tadpole was present at the end of the experiment.
The percentage of consumed tadpoles was used to determine predator effectiveness in
single tadpole experiments. Percentages for each predatory species was obtained by adding
the number of experiments where the tadpole was consumed and dividing by the total
number of experiments used for the predatory species in question.
Multiple tadpole experiments
Aquatic predators were placed into white plastic containers and starved for twelve hours
prior to the addition of 10 tadpoles from a single anuran species into each container. The
number of tadpoles surviving were checked and recorded every 12 hours over a 48 hour
period. Tadpoles were considered to be consumed if a tadpole was absent. Consumed
124
tadpoles were replaced with new tadpoles every 12 hours. The total number of tadpoles
consumed within one container over the 48 hour period was considered as one replicate
when undertaking data analysis. This was replicated for each tadpole species and
predator species, with the number of experiments dependent on the availability of
tadpoles and predators (Table 5.1). No tadpoles or predators were re-used in experiments
and all surviving tadpoles were released at the site of capture at the end of each
experiment.
Predator size can influence predation rates on tadpoles (Webb and Joss, 1997). Therefore,
for multiple-prey experiments, a One-Way ANOVA was used to determine if there were
significant differences in size of predator species between tadpole species. Tukey’s honest
significant difference (HSD) post hoc t-tests were then used to determine which tadpole
species were influencing the significant differences.
An ANCOVA was used to examine the relationship between the total number of tadpoles
consumed and the independent variables; tadpole species, predator species and predator
size as a covariate. ANCOVAs were also used to determine if the number of tadpoles that
were predated upon were significantly different among predator species.
Table 5.1: Number of experiments conducted for each tadpole predator species for multiple
prey experiments.
Tadpole Type
Cherax sp.
H. galii
G. holbrooki
Total
Litoria olongburensis
4
8
6
18
Litoria fallax small
6
10
8
24
Litoria fallax large
4
3
5
12
Limnodynastes peronii
10
10
11
31
Total
24
31
30
85
125
5.4 Results
Individual Tadpole Experiments
The efficiency of predators varied amongst taxa (Figure 5.1). The predacious beetles, in
the family Belostomatidae, were the most effective predators, consuming 90% (9/10
tadpoles) of L. olongburensis tadpoles offered to them. The next most effective predators
were members of the Aeshnidae family and members within the Cherax genus, which
consumed 27% (3/11 tadpoles) and 10% (1/10 tadpoles) of L. olongburensis tadpoles
offered to them, respectively. Individuals of the species G. holbrooki attacked, but did
not consume, 25% (3/12 tadpoles) of L. olongburensis tadpoles offered to them. All
tadpoles attacked by G. holbrooki were alive at the end of the study. Individuals of the
native fish species H. galii and larvae of the sub-order Zygoperta neither consumed nor
attacked any L. olongburensis tadpoles (Figure 5.1).
Multiple Prey Experiments
There was a significant difference in size of Cherax sp. (df = 23, F3/20=18.025, p = 0.000)
and H. galii (df = 30, F3/27 =3.381, p = 0.033) used among tadpole species. Tukeys post-hoc
t-tests revealed that the significant differences for Cherax sp. size were between L.
olongburensis treatments and all other tadpole species (L. fallax large, (M = 34, 95% CI
[20.21, 47.79], p =0.000); L. fallax small, (M = 28.25, 95% CI [13.14, 43.36], p =0.000);
Lim. peronii, (M = 28.5, 95% CI [15.86, 41.14], p =0.000). Tukeys post-hoc t-tests for H.
galii size revealed that the size difference between L. olongburensis and Lim. peronii (M =
8.8, 95% CI [-18.14, 0.54], p =0.7) and between L. olongburensis and ‘small’ L. fallax (M
= 9.2, 95% CI [-18.54, 0.14], p =0.55).
There was a significant interaction between tadpole species and predator species (df = 6, F
= 2.8, p = 0.017) for the total number of consumed tadpoles. Overall, Cherax sp. consumed
the largest number of tadpoles while G. holbrooki consumed the lowest number of tadpoles
(Table 5.2).
126
Figure 5.1: Percentage of predators that consumed (black bars) or attacked (white bar)
Litoria olongburensis tadpoles for experiments where one individual L. olongburensis was
used in each experiment. Number of replicates/experiments is presented above each
predatory species.
There was a significant difference in the number of tadpoles consumed between tadpole
species for Cherax sp. (df = 3, F3/20 = 4.232, p =0.019) and H. galii (df = 3, F3/27 = 7.38, p =
0.001), with Cherax sp. and H. galii consuming the highest number of Lim. peronii
tadpoles and the lowest number of L. olongburensis tadpoles (Figure 5.2 and Table 5.2).
Despite G. holbrooki consuming the highest number of tadpoles in large L. fallax and Lim.
peronii experiments,
there were no significant differences in the number of tadpoles
consumed between tadpole species for G. holbrooki (df = 3, F3/26 = 1.36, p = 0.279). With
the exception of small L. fallax tadpoles, G. holbrooki consumed, on average, fewer
tadpoles when compared with the other two native predatory species.
127
Table 5.2: Average number of tadpoles consumed for each predator species during multiple
prey experiments.
Tadpole Type
Cherax sp.
H. galii
G. holbrooki
Litoria olongburensis
4.5(+/- 5.8)
0.6 (+/- 0.89)
0.55 (+/- 1.66)
Litoria fallax small
8 (+/- 2.28)
3.2 (+/- 2.4)
4.13 (+/- 1.73)
Litoria fallax large
7.75 (+/- 4.65)
2.67 (+/- 1.53)
0.8 (+/- 0.45)
14 (+/- 4.32)
10.8 (+/- 8.68)
2.27 (+/- 5.29)
9.88 (+/- 5.46)
4.87 (+/- 6.57)
2.23 (+/ - 3.59)
Limnodynastes peronii
Total Consumed
There was a significant difference in the number of Lim. peronii tadpoles consumed (df = 2,
F2/28 = 4.37, p = 0.023) and the number of small L. fallax consumed (df = 2, F2/21 = 11.74, p
= 0.000) among predator species. Cherax sp. consumed the highest number of tadpoles in
both Lim. peronii and small L. fallax experiments (Figure 5.2 and Table 5.2). There was no
significant difference in the number of large L. fallax tadpoles consumed (df = 2, F2/15=
3.552, p = 0.079) or the number of L. olongburensis (df = 2, F2/15=0.746, p = 0.492)
tadpoles consumed among predator species.
5.5 Discussion
This study is the first to quantify differences in Australian tadpole predation levels
between G. holbrooki and native aquatic predators using an experimental approach. As
expected, predation rates varied between tadpole species and predatory species, with G.
holbrooki predation levels being approximately equal to or less then predation levels of
native predatory species.
128
Figure 5.2: Number of tadpoles consumed for each predatory species. Symbolys
represent the number of tadpoles consumed for an individual experiment. ‘o’ represents
Limnodynastes peronii, ‘Δ’ represents small Litoria fallax, ‘x’ represents large L. fallax
and ‘+’ represent L. olongburensis
129
Gambusia holbrooki predation rates
These results are the first, to my knowledge, to show that G. holbrooki are either less or
equally effective as tadpole predators when compared with native predatory species. Only
one other study has provided preliminary comparative data for G. holbrooki predation
levels with four Australian native predatory fish species (Pyke and White, 2000). Pyke and
White (2000) showed that native fish predators had lower tadpole predation rates when
compared with G. holbrooki. Results from the current study contrast with these as native
fish within the multiple prey experiments consumed a lower number of tadpoles when
compared with G. holbrooki. Furthermore, crayfish and some aquatic invertebrate predators
had higher consumption rates when compared with G. holbrooki in both multiple and single
prey experiments.
Lower tadpole consumption of L. olongburensis and large L. fallax tadpoles by G.
holbrooki and H. galii when compared with the other two, smaller, tadpole categories may
be explained by physical limitations of the predator (i.e. mouth gape limitation) or
unpalatability, which can increase in older tadpoles (Lawler and Hero, 1997). Mouth-gape
limitations by these two predatory species would result in tadpole consumption being
restricted to smaller tadpoles. If mouth-gape limitations were influencing predation rates on
larger tadpoles then G. holbrooki may have used the observed tail-nipping to reduce tadpole
movement in an attempt to kill the tadpole to allow for easier consumption, regardless of
tadpole size (Baber 2001, cited in Baber and Babbitt 2003). Similar tail-nipping attacks
have been observed where G. holbrooki attacked, but did not kill, L. aurea tadpoles (Webb
1994, cited in Morgan and Buttemer 1996; Pyke and White 2000). Furthermore, tadpole
body size of Crinia signifera and L. peronii tadpoles in past experiments did not influence
predation rates of G. holbrooki (Webb 1994, cited on Morgan and Buttemer 1996).
Lower predation rates of small L. peronii tadpoles when compared with small L. fallax
tadpoles by G. holbrooki are unlikely to be due to mouth-gape limitations as the sizes of
tadpoles within these two tadpole groups were similar. Furthermore, it is unlikely that
unpalatability towards fish is the cause of low predation rates, as the other fish species (H.
130
galii) used in the experiment had relatively high predation levels when compared with G.
holbrooki. While not tested, other anti-predator strategies like behavioural avoidance
(Skelly, 1994; Saidapur et al., 2009; Smith and Awan, 2009) or other morphological
adaptations (Hecnar and M'Closkey, 1997; McCollum and Leimberger, 1997; Touchon
Warkentin, 2008) are likely to be the cause of lower predation rates by G. holbrooki on
Lim. peronii tadpoles.
Predator Effectiveness
The primary consumers of tadpoles within single prey experiments were aquatic
invertebrate insects within the Belostomatidae and Aeshnidae families. These results
support past studies that found Belostomatidae and Aeshnidae were effective predators
of tadpoles (reviewed in Wells, 2007), with tadpoles contributing a large portion of
Belostomatidae diet (Ohba and Nakasuji, 2006). Additionally, members within the
Aeshnidae family can exclude or reduce tadpole populations from waterbodies (Hero et
al., 2001; Stav et al., 2007) and have been known to predate on tadpole species that are
unpalatable to fish (Hero et al., 2001).
There was no predation of L. olongburensis tadpoles by individuals in the sub-order
Zygoptera. A literature search on predators of amphibian larvae (conducted by Wells,
2007) showed no publications recording Zygoptera as a predator of tadpoles. These
results further validate these findings and indicate that Zygoptera are ineffective
predators of L. olongburensis tadpoles.
A higher number of tadpoles were consumed by Cherax sp. when compared with other
predators in tadpole experiments. Crayfish have been described as both active (Wells,
2007) and sit and wait (Renai and Gherardi, 2004) predators that have the potential to
influence tadpole assemblages (Dorn and Wojdak, 2004). The high level of tadpole
predation indicates that Cherax sp. have a greater ability to influence the tadpole
community when compared with other predatory species used in predator experiments
(i.e. G. holbrooki, H. galii). Furthermore, crayfish were shown to leave killed
131
bufonid/unpalatable tadpoles (Axelsson et al., 1997), suggesting the differing predation
levels of Cherax sp. between tadpole species may give an indication of tadpole
palatability.
The lower number of L. olongburensis tadpoles consumed by Cherax sp. when compared
with other tadpole categories in multiple prey experiments may be due to significantly
larger Cherax sp. used in L. olongburensis tadpole experiments. Crayfish are polytrophic
omnivores that will consume flora or fauna material (Axelsson et al., 1997; Verhoef et al.,
1998; Furse and Wild, 2004). However, adult crayfish are often found with high levels of
detritus and plants in their gut content while juvenile crayfish feed predominately on
invertebrates (Nyström, 2002). This would explain patterns observed in the current study,
with smaller crayfish being younger than larger crayfish and having higher predation levels
on fauna, possibly due to of higher protein requirements compared with their adult
counterparts.
The number of experiments where Cherax sp. and H. galii consumed tadpoles was lower
in experiments with single L. olongburensis tadpoles than those with multiple L.
olongburensis tadpoles. An increase in prey abundance and longer experimental time for
multiple prey experiments would allow for a higher likelihood of predator-prey
interactions and, possibly, allow for higher predation levels. Alternatively, the starvation
period in predator consumption rate experiments may have contributed to increased
predation levels.
Conservation Implications
This study indicates that G. holbrooki have the potential to be just as effective as tadpole
when compared with native predatory species. When present, G. holbrooki are often the
most abundant fish species (reviewed in Pyke, 2008) and this would therefore increase the
frequency of predator-prey interactions and potentially influence the abundance of
amphibians at the waterbody. This is supported by past studies, that have shown a negative
relationship with amphibian abundance and G. holbrooki presence (Webb and Joss, 1997).
132
Amphibian species richness, however, is unlikely to be influenced by presence of G.
holobrooki, with past field studies finding that presence of L. aurea tadpoles (Hamer et al.,
2002) and adults of other Australian native amphibian species (Reynolds, 2009) were not
influenced by the presence of G. holbrooki. Hence, waterbodies dominated by G. holbrooki
are likely to be influencing tadpole abundances rather than species richness.
Tail-nipping was observed by G. holbrooki during the single L. olongburensis tadpole
experiments. Tail-nipping has been shown to increase the tadpole period of Bombina
orientalis (Parichy and Kaplan, 1992) and similar tail-nipping observed by G. holbrooki
may result in prolonged periods as a tadpole during the tadpole and metamorph lifestages.
Litoria olongburensis (Meyer et al., 2006) and some populations of L. fallax (Anstis, 2002)
and Lim. peronii (Simpkins pers. obs.) are associated with ephemeral and semi-permanent
waterbodies. Therefore, prolonged time as a tadpole, due to predatory attacks, may be lethal
to individual tadpoles of these species as metamorphosis has to occur before waterbody
desiccation. This would likely be the same for any ephemeral breeding amphibian species
that occurs with G. holbrooki.
Surprisingly, there was no significant difference in predator tadpole consumption between
tadpole types for G. holbrooki. There was, however, a significant difference in predator
tadpole consumption between tadpole types for Cherax sp. and H. galii. Therefore, the
ability of the predators used in this study to influence tadpole populations will differ
between tadpole and predator species. Regardless, the introduction of any predator used in
this study into a ‘predator’ free waterbody where tadpoles are naïve to the introduced
predator may significantly impact tadpole assemblages. This needs to be considered in the
construction of permanent waterbodies (i.e. to mitigate against the effects of climate change
(Shoo et al., 2011)) when being built for amphibians that lack adequate anti-predator
strategies.
Results from the no-choice experiments describe the potential for predation and may only
be applicable under specific conditions within the natural environments. Firstly, the
presence of alternative prey sources can influence predation rates in experiments (Pyke and
133
White, 2000; Reynolds, 2009) and levels of predation may vary within the natural
environment where alternative prey sources would be present. Secondly, aquatic refuge
habitat can positively influence survival of tadpoles against predatory fish (Morgan and
Buttemer, 1996) and aquatic invertebrate insects (Babbitt and Tanner, 1998; Tarr and
Babbitt, 2002; Kopp et al., 2006). Therefore, when present in natural waterbodies, both
alternative prey sources and adequate refuge cover can allow for co-existence of anuran
species and G. holbrooki. This may be of particular importance to L. olongburensis, that are
often associated with waterbodies where sedges are dominant (Simpkins unpublished data).
Hence, results from this study may be more applicable to waterbodies where/when refuge
habitat and alternative prey sources are relatively low or absent.
Acknowledgements
Work undertaken for this chapter was performed under approval from Griffith University
Animal Ethics Committee (Permit number: ENV/18/11/AEC), Queensland general fisheries
permit (Permit Number: 90306) and Queensland Department of Environment and
Resource Management (ECOACCESS Permit Numbers: WITK10080611 /
WISP10081411). I would like to thank Clare Morrison, Donna Treby, James Bone, Diana
Virkki, Jean-Marc Hero, Guy Castley and Katrin Lowe for feedback on earlier drafts of this
chapter. I also thank Amanda Winzar, Chays Ogston, Jodie Lee Hills, Chris Dahl, Diana
Virkki, Tempe Parnell and James Bone for assisting in the field. I also thank Alan Kerr
from the Bribie Island Environmental Protection Society provided accommodation during
fieldwork. Special thanks are given to Currumbin Wildlife Sanctuary for providing access
to their land. Funding was provided from the Griffith School of Environment.
134
5.6 References
Álvarez, D., Nicieza, A. G., 2009. Differential success of prey escaping predators: tadpole
vulnerability or predator selection? Copeia 2009, 453-457.
Anstis, M. 2002. Tadpoles of South-eastern Australia: a guide with keys, New Holland
Publishers, Sydney.
Axelsson, E., Nyström, P., Sidenmark, J., Brönmark, C., 1997. Crayfish predation on
amphibian eggs and larvae. Amphibia-Reptilia 18, 217-228.
Azevedo-Ramos, C., Van Sluys, M., Hero, J.-M., Magnusson, W. E., 1992. Influence of
tadpole movement on predation by odonate naiads. Journal of Herpetology 26, 335338.
Babbitt, K. J., Tanner, G. W., 1998. Effects of cover and predator size on survival and
development of Rana utricularia tadpoles. Oecologia 114, 258-262.
Baber, M. J. 2001. Understanding anuran community structure in temporary wetlands: the
interaction and importance of landscape and biotic processes. Dissertation, Florida
International University.
Baber, M. J., Babbitt, K. J., 2003. The relative impacts of native and introduced predatory
fish on a temporary wetland tadpole assemblage. Oecologia 136, 289-295.
Crossland, M. R., 2001. Ability of predatory native Australian fishes to learn to avoid toxic
larvae of the introduced toad Bufo marinus. Journal of Fish Biology 59, 319-329.
Dorn, N. J., Wojdak, J. M., 2004. The role of omnivorous crayfish in littoral communities.
Oecologia 140, 150-159.
Furse, J. M., Wild, C. H., 2004. Laboratory moult increment, frequency, and growth in
Euastacus sulcatus, the Lamington Spiny Crayfish. Freshwater Crayfish 13, 205211.
Gillespie, G., Hero, J.-M. (1999) Potential impacts of introduced fish and fish
translocations on Australian amphibians. In: Campbell A (ed) Declines and
Disappearances of Australian Frogs. Environment Australia, Canberra, pp 131-144
Gregoire, D. R., Gunzburger, M. S., 2008. Effects of predatory fish on survival and
behavior of Larval Gopher Frogs (Rana capito) and Southern Leopard Frogs (Rana
sphenocephala). Journal of Herpetology 42, 97-103.
135
Gunzburger, M. S., Travis, J., 2005. Critical literature review of the evidence of
unpalatability of amphibian eggs and larvae. Journal of Herpetology 39, 547-571.
Hamer, A. J., Lane, S. J., Mahony, M. J., 2002. Management of freshwater wetlands for the
endangered green and golden bell frog (Litoria aurea): roles of habitat determinants
and space. Biological Conservation 106, 413-424.
Hecnar, S. J., M'Closkey, R. T., 1997. The effects of predatory fish on amphibian species
richness and distribution. Biological Conservation 79, 123-131.
Hero, J.-M., Gascon, C., Magnusson, W. E., 1998. Direct and indirect effects of predation
on tadpole community structure in the Amazon rainforest. Australian Journal of
Ecology 23, 474-482.
Hero, J.-M., Magnusson, W. E., Rocha, C. F. D., Catterall, C. P., 2001. Antipredator
defenses influence the distribution of amphibian prey species in the central Amazon
rain forest. Biotropica 33, 131-141.
Heyer, W. R., McDiarmid, R. W., Weigmann, D. L., 1975. Tadpoles, predation and pond
habitats in the tropics. Biotropica 7, 100-111.
Jara, F. G., 2008. Tadpole–odonate larvae interactions: influence of body size and diel
rhythm. Aquatic Ecology 42, 503-509.
Kats, L. B., Ferrer, R. P., 2003. Alien predators and amphibian declines: review of two
decades of science and the transition to conservation. Diversity and Distributions
2003, 99-110.
Kats, L. B., Petranka, J. W., Sih, A., 1988. Antipredator defenses and the persistence of
amphibian larvae with fishes. Ecology 69, 1865-1870.
Kopp, K., Wachlevski, M., Eterovick, P. C., 2006. Environmental complexity reduces
tadpole predation by water bugs. Canadian Journal of Zoology 84, 136-140.
Lawler, K. L., Hero, J.-M., 1997. Palatability of Bufo marinus tadpoles to a predatory fish
decreases with development. Wildlife Research 24, 327-334.
McCollum, S. A., Leimberger, J. D., 1997. Predator-induced morphological changes in an
amphibian: predation by dragonflies affects tadpole shape and color. Oecologia 109,
615-621.
Meyer, E., Hero, J.-M., Shoo, L., Lewis, B. (2006) National recovery plan for the wallum
sedgefrog and other wallum-dependent frog species. Report to the Department of
136
the Environment and Water Resources, Canberra. Queensland Parks and Wildlife
Service, Brisbane,
Morgan, W. A., Buttemer, L. A., 1996. Predation by the non-native fish Gambusia
holbrooki on small Litoria aurea and L. dentata tadpoles. Australian Zoologist 30,
143-149.
Nelson, D. W. M., Crossland, M. R., Shine, R., 2011. Foraging responses of predators to
novel toxic prey: effects of predator learning and relative prey abundance. Oikos
120, 152-158.
Nyström, P. (2002) Ecology. In: Holdich DM (ed) Biology of Freshwater Crayfish.
Blackwell Science, Great Britain, pp 192-235.
Ohba, S.-y., Nakasuji, F., 2006. Dietary items of predacious aquatic bugs (Nepoidea:
Heteroptera) in Japanese wetlands. Limnology 7, 41-43.
Parichy, D. M., Kaplan, R. H., 1992. Developmental consequences of tail injury on larvae
of the Oriental Fire-Bellied Toad, Bombina orientalis. Copeia 1992, 129-137.
Pyke, A. W., White, G. H., 2000. Factors influencing predation on eggs and tadpoles of the
endangered Green and Golden Bell Frog Litoria aurea by the introduced Plague
Minnow Gambusia holbrooki. Australian Zoologist 31, 496-505.
Pyke, G. H., 2008. Plague minnow or mosquito fish? A review of the biology and impacts
of introduced Gambusia species. Annual Review of Ecology, Evolution and
Systematics 39, 171-191.
Renai, B., Gherardi, F., 2004. Predatory efficiency of crayfish: comparison between
indigenous and non-indigenous species. Biological Invasions 6, 89-99.
Reynolds, S. J., 2009. Impact of the Introduced Poeciliid Gambusia holbrooki on
Amphibians in Southwestern Australia. Copeia 2009, 296-302.
Saidapur, S. K., Veeranagoudar, D. K., Hiragond, N. C., Shanbhag, B. A., 2009.
Mechanism of predator–prey detection and behavioral responses in some anuran
tadpoles. Chemoecology 19, 21-28.
Shoo, L. P., Olson, D. H., McMenamin, S. K., Murray, K. A., Van Sluys, M., Donnelly, M.
A., Stratford, D., Terhivuo, J., Merino-Viteri, A., Herbert, S. M., Bishop, P. J.,
Corn, P. S., Dovey, L., Griffiths, R. A., Lowe, K., Mahony, M., McCallum, H.,
Shuker, J. D., Simpkins, C., Skerratt, L. F., Williams, S. E., Hero, J.-M., 2011.
137
Engineering a future for amphibians under climate change. Journal of Applied
Ecology 48, 487-492.
Skelly, D. K., 1994. Activity level and susceptibility of anuran larvae to predation. Animal
Behaviour 47, 465-468.
Smith, A. R., Awan, G. R., 2009. The roles of predator identity and group size in the
antipredator responses of American toad (Bufo americanus) and bullfrog (Rana
catesbeiana) tadpoles. Behaviour 146, 225-243.
Stav, G., Kotler, B. P., Blaustein, L., 2007. Direct and indirect effects of dragonfly (Anax
imperator) nymphs on green toad (Bufo viridis) tadpoles. Hydrobiologia 579, 85-93.
Stoneham, M. 2001. The influence of stream-dwelling predators on the distribution and
density of Mixophyes tadpoles in Southeast Queensland. Unpublished Honours
Thesis, Griffith University.
Tarr, T. L., Babbitt, K. J., 2002. Effects of habitat complexity and predator identity on
predation of Rana clamitans larvae. Amphibia-Reptilia 23, 13-20.
Touchon, J. C., Warkentin, K. M., 2008. Fish and dragonfly nymph predators induce
opposite shifts in color and morphology of tadpoles. Oikos 117, 634-640.
Verhoef, G. D., Jones, P. L., Austin, C. M., 1998. A comparison of natural and artificial
diets for juveniles of the Australian freshwater crayfish Cherax destructor. Journal
of the World Aquaculture Society 29, 243-248.
Vonesh, J. R., Kraus, J. M., Rosenberg, J. S., Chase, J. M., 2009. Predator effects on
aquatic community assembly: disentangling the roles of habitat selection and postcolonization processes. Oikos 118, 1219-1229.
Webb, C., Joss, J., 1997. Does predation by the fish Gambusia holbrooki (Atheriniformes:
Poeciliidae) contribute to declining frog populations? Australian Zoologist 30, 316324.
Webb, C. E. 1994. Does predation by Gambusia holbrooki (Atheriformes: Poeciliidae)
contribute to declining frog populations? Unpublished Honours Thesis, Macquarie
University.
Wells, K. D. 2007. The Ecology and Behavior of Amphibians, The University of Chicago
Press, USA.
138
Woodward, B. D., 1983. Predator-prey interactions and breeding-pond use of temporarypond species in a desert anuran community. Ecology 64, 1549-1555.
139
6.0 General Conclusions
This thesis aimed to determine which environmental variables influenced the adult and
tadpole assemblages of ‘acid’ and ‘non-acid’ frog species within and around wallum
vegetation of eastern Australia, with primary focus on threatened amphibian species
occurring within this habitat. The conclusions are presented from each of four chapters
within this thesis, followed by future research directions and management outcomes
summarised for the acid frog species, Litoria olongburensis and Crinia tinnula.
6.1 Chapter overviews
6.1.1 Chapter 2 - Variables influencing wallum heathland tadpole
assemblages
The tadpole assemblage associated with natural waterbodies in wallum areas included the
species L. olongburensis, C. tinnula, L. fallax, L. gracilenta and L. cooloolensis. The
assemblage was dominated by tadpoles of L. olongburensis and C. tinnula, indicating that
these species are highly adapted to exist within the wallum heathland environment. L.
olongburensis were recorded from 11 survey transects while C. tinnula were recorded from
14 survey transects, with both species dominating the waterbodies surveyed. Tadpoles of
the other ‘acid’ frog species L. cooloolensis were only recorded from two transects. The
absence of this species from other transects are likely to do with its restricted distributional
range. The other ‘acid’ frog species L. freycineti was absent from the tadpole assemblage.
This is likely because of detectability issues with this species.
The distribution and occupancy of L. olongburensis and C. tinnula tadpoles within natural
wallum heathland waterbodies was associated with several environmental variables, with
variables differing between species, distribution and occupancy. The variables influencing
L. olongburensis abundance included pH, depth and turbidity while the variables
influencing L. olongburensis occupancy included pH and depth. Both pH and depth were
140
influencing L. olongburensis tadpole in a unimodel distribution. The variables influencing
C. tinnula abundance were not definitive while the primary variables influencing C. tinnula
occupancy included positive associations with water depth, and turbidity, and a negative
association with predatory fish.
These results confirm that wallum heathland waterbodies are dominated by ‘acid’ frog
species (L. olongburensis and C. tinnula). Furthermore, it confirms hypothesis by Meyer et
al (2006) that indicated that pH is playing a role in L. olongburensis distribution. The lower
relative abundance of L. fallax is likely due to tadpole intolerance to low pH waters (Meyer
2004).
6.1.2 Chapter 3 - Usage of anthropogenic waterbodies, and variables
influencing adult amphibian assemblages
Natural waterbodies contained L. olongburensis, L. fallax, L. tyleri, C. tinnula, L.
gracilenta and Limno. peronii while anthropogenic waterbodies contained L.
olongburensis, L. fallax, L. tyleri, C. tinnula, L. gracilenta, R. marina, L. nasuta, L.
freycineti, a Uperolia sp. and Limno. peronii. The waterbody type (i.e. golf course, road
side ditch) that a species was located in was species specific.
The relative abundance of L. olongburensis within natural and anthropogenic waterbodies
was associated with both both % sedge cover and pH. The result of pH is similar for L.
olongburensis tadpole distribution within Chapter 2 and adult relative abundance in
Chapter 4, indicating that water pH is vital for L. olongburensis population continuance in
wallum heathland waterbodies. Furthermore, L. olongburensis and C. tinnula relative
abundance and L. olongburensis tadpole occupancy was highest in natural waterbodies.
Results from this chapter also showed that an overlap in the ecological niches for L.
olongburensis and L. fallax occur. This indicates co-existence between these two species
within certain waterbodies. However, high pH had a negative influence on L. olongburensis
141
abundance, indicating that tadpole competition between L. olongburensis and L. fallax may
be excluding L. olongburensis within waterbodies where pH is high. Additionally, relative
abundance of L. fallax, while not shown statistically, was lower in waterbodies where pH
was low, suggesting L. fallax tadpoles are either intolerant to low pH or are outcompeted by
L. olongburensis in acidic waters. The intolerance hypothesis is likely correct, as Meyer
(2004) showed eggs and tadpoles of L. fallax fail to metamorphose when exposed to low
pH water (i.e. 3.5.). Furthermore, the absence of adults from ‘unfavourable’ waterbodies
could be explained by adults failing to deposit eggs in waterbodies that are unfavourable for
successful reproduction (Takahashi, 2007; Hamamura, 2008).
6.1.3 Chapter 4 – Compensatory pond usage by wallum heathland amphibians
and variables influencing adult amphibian assemblages
In a similar study of natural and compensatory waterbodies of a four-lane road construction
project, environmental variables associated with the relative abundance of L. olongburensis
were pH, salinity and water depth (similar to the distribution of L. olongburensis tadpoles
in natural wallum habitats (Chapter 2) and disturbed waterbodies (Chapter 3). The result of
water depth is similar for L. olongburensis tadpole distribution within Chapter 2. This gives
a strong indication that both of these variables are important for L. olongburensis
population persistence.
Environmental variables associated with the relative abundance of C. tinnula were salinity,
water depth, minimum hydroperiod and pH. The only variable influencing C. tinnula
tadpole occupancy (Chapter 2) that also influenced the relative abundance of C. tinnula
adults was water depth, indicating that this is an important variable in C. tinnula population
persistence.
The lower relative abundance of ‘acid’ frog species in compensatory habitats is explained
by unfavourable water chemistry variables and the high relative abundance of competitive
amphibian species. These results support the conclusions of Chapter 3, suggesting
142
competition excludes the ‘acid’ frogs from environments where water chemistry is more
favourable for competitive amphibian species.
6.1.4 Chapter 5 - Predation experiments with G. holbrooki and natural
predators
Predation experiments suggest G. holbrooki are either less efficient or equally efficient at
predating on tadpoles when compared with native predatory species. This was dependant
on tadpole species. For L. olongburensis, the most effective predators were Cherax sp. and
Belostomatida, with G. holbrooki predation being low or absent. This trend was similar for
the other two tadpole species (L. fallax and Limno. peronii), with G. holbrooki predation
being low or equal to predation rates by native predatory species.
While these results indicate that G. holbrooki have a lower predation potential when
compared with native predaotrs, itshould be noted that, when present, G. holbrooki are
often the most abundant fish species (reviewed in Pyke, 2008) and would therefore increase
the occurrence of predator-prey interactions and potentially influence the abundance of
amphibians at the waterbody. However, amphibian species richness is unlikely to be
influenced by presence of G. holobrooki, with past field studies finding that presence of L.
aurea tadpoles (Hamer et al., 2002) and adults of other Australian native amphibian species
(Reynolds, 2009) were not influenced by the presence of G. holbrooki.
Results from previous chapters also showed that G. holbrooki presence within natural
wallum waterbodies were low in occupancy (Chapter 2), indicating successful colonization
of G. holbrooki in remote, natural waterbodies is low.
6.2 Management Outcomes
Results from this thesis showed that generalist amphibian species dominate the amphibian
assemblages occurring in anthropogenically disturbed areas surrounding wallum heathland
143
areas (Chapter 3 + 4). The opposite was found in natural waterbodies within wallum
heathland, where a few specialist ‘acid’ frog species dominating the amphibian assemblage
(Chapter 2). For L. olongburensis and C. tinnula, low tadpole occupancy within
anthropogenic waterbodies and high occupancy/relative abundance within natural
waterbodies indicates that ‘acid’ frogs are predominantly restricted to natural waterbodies
for breeding. Therefore, when habitat offsets are considered for L. olongburensis and/or C.
tinnula, natural waterbodies should be targeted for conservation. Furthermore,
compensatory waterbodies should exclude or have low relative abundance of native
predatory fish and ensure optimal water chemistry and vegetation characteristics. These
characteristics should be within the ranges where L. olongburensis and C. tinnula adults
and tadpoles were predominantly encountered during this survey (Chapters 2-4).
The first two years after construction of compensatory ponds are inadequate in providing
habitat for the acid frog species L. olongburensis and C. tinnula. However, compensatory
habitats will provide habitat for non-target competitive species (i.e. L. fallax). As
mentioned previously, failure of compensatory habitats providing adequate habitat for
‘acid’ frogs is probably a result of unfavourable environmental conditions. If future
construction of compensatory habitats is undertaken then managers should target areas
where environmental variables (i.e. pH) would be within the ideal ranges for acid frog
species.
While field surveys indicated that L. olongburensis tadpoles prefer waterbodies where G.
holbrooki are absent, no-choice experiments suggest that native predatory fauna have the
ability to predate on tadpoles of L. olongburensis more than G. holbrooki. Despite this, G.
holbrooki are often high in abundance in waterbodies where L. olongburensis are present
(Pyke, 2008). Hence, G. holbrooki may influence the abundance, rather than the occupancy,
of L. olongburensis. The interactions between G. holbrooki and other predators in these
ecosystems needs to be further examined. Where possible, there should be no introduction
of predatory fish or aquatic tadpole predators into newly constructed waterbodies or natural
wallum heahtland waterbodies as fish are likely to be negatively influencing the abundance
of L. olongburensis tadpoles.
144
6.3 Future priorities for research
Field studies conducted for this thesis were in the summer/autumn period. Hines and Meyer
(2011) recorded over 20 species of amphibians occurring on south-east Queensland dune
islands. These islands contain both wallum and other native vegetation communities. Future
studies need to be conducted during different seasons to identify winter breeding and
potentially missed amphibian species that were not detected during these surveys. This
should be undertaken in both natural and anthropogenic waterbodies to capture a
‘disturbance’ gradient to determine if any missed species are utilizing anthropogenic
waterbodies for breeding.
Additional surveys determining what environmental variables influence the occupancy and
abundance of other acid frog species (i.e. L. freycineti and L. cooloolensis) are needed. This
would not just be targeted to mainland populations, but be extended to the dune islands
what are located within south-east Queensland.
Prolonged field surveys targeting waterbodies that were designed to aid in compensatory
habitats for ‘acid’ frogs are required to determine if compensatory habitats can be
successfully utilized by ‘acid’ frogs after a two year period. Ecological studies are often
confined to relatively short survey periods and these may not provide an accurate
assessment of the long-term patterns or trends within populations. Therefore future
periodic monitoring is required to address research questions related to the long-term
survival of these species. Furthermore, studies should be undertaken at differing latitudes
and environments across the distributional range of both L. olongburensis and C. tinnula.
A thorough examination of the predator – prey interactions among all predators and prey to
determine how fish and other predators interact, and clarify the relationships between acid
frogs and fish, are required. Alternative prey sources and aquatic refuge cover can influence
predation rates on tadpoles by aquatic predators. Therefore, predation experiments using
alternative prey and a varying density of tadpoles and predators need to be conducted. This
145
would allow for results to be further extrapolated to all waterbodies within wallum
heathland environments and not limited to waterbodies where refuge cover and alternative
prey sources are low or absent.
6.4 References
Hamer, A. J., Lane, S. J., Mahony, M. J., 2002. Management of freshwater wetlands for the
endangered green and golden bell frog (Litoria aurea): roles of habitat determinants
and space. Biological Conservation 106, 413-424.
Haramura, T., 2008. Experimental test of spawning site selection by Buergeria japonica
(Anura: Rhacophoridae) in response to salinity levels. Copeia 2008, 64-67.
Hines, H.A., Meyer, E.A., 2011. The frog fauna of Bribie Island: an annotated list and
comparision with other Queensland dune islands. Proceedings of the Royal Society
of Queensland 117, 261-274.
Meyer, E., 2004. Acid adaptation and mechanisms for softwater acid tolerance in larvae of
anuran species native to the “Wallum” of east Australia. PhD Thesis, University of
Queensland.
Pyke, G. H., 2008. Plague minnow or mosquito fish? A review of the biology and impacts
of introduced Gambusia species. Annual Review of Ecology, Evolution and
Systematics 39, 171-191.
Reynolds, S. J., 2009. Impact of the Introduced Poeciliid Gambusia holbrooki on
Amphibians in Southwestern Australia. Copeia 2009, 296-302
Takahashi, M., 2007. Oviposition site selection: pesticide avoidance by gray treefrogs.
Environmental Toxicology and Chemistry 26, 1476-1480.
146
7.0 Appendices: Publications on ‘acid’ frogs published during
candidature
Appendix 1
Simpkins, C.A., Meyer, E., Hero, J.-M., 2011. Long-range movement in the rare
Cooloola sedgefrog Litoria cooloolensis. Australian Zoologist 35:977-978.
Understanding habitat usage is essential for the proper management of rare and threatened
species in the wild. However, current knowledge of habitat usage by many rare and
threatened Australian frog species is inadequate in this regard. Knowledge of non-breeding
habitat usage in Australian amphibian species is particularly poor (Hines et al. 1999), with
current understanding of the habitat requirements of many species based largely on habitat
usage by calling animals during the breeding season, when frogs are more readily
detectable. However, like other fauna, the use of different habitats by amphibians can vary
daily and seasonally as well as between different sexes and life stages (Law and Dickman
1998). Female frogs, for example, are likely to occupy habitat further away from a breeding
water body while males tend to stay near the breeding site (Bartelt et al. 2004; Johnson et
al. 2007; Rittenhouse and Semlitsch 2007). A majority of aquatic breeding amphibians will
also undergo embryonic and larval development within the aquatic environment while
spending their adult and juvenile lives in the terrestrial environment, unlike developing
larvae, post-metamorphic adult and juvenile amphibians are not restricted to the aquatic
body of their birth (Johnson et al 2007) and thus have the ability to disperse or migrate
into adjacent, non-breeding, areas for the purposes of foraging, overwintering (Regosin et
al 2003) or refuge use (Semlitsch and Bodie 2003).
The Cooloola sedgefrog (Litoria cooloolensis) is a rare, largely arboreal species restricted
to coastal wallum (i.e., coastal sand dunes and plains) in south-east Queensland. Though
known to occur 'some distance' away from breeding habitat within the terrestrial landscape
147
(Meyer et al. 2006), details of non-breeding habitat usage in this species (e.g., the type of
habitat utilized by non-breeding animals and movement distances from areas of breeding
habitat) remain poorly documented. Herein, we present detailed observational
records of L. cooloolensis from non-breeding areas in wallum habitat, thus providing
additional insight into non-breeding habitat usage by this species.
On a summer's night in late January-early February 1999, upwards of 50 L. cooloolensis
were heard calling from trees along a walking track through rainforest near Lake Poona,
within the Cooloola section of the Great Sandy Region National Park, south east
Queensland. The majority of animals heard calling at this time appeared to be calling high
up in the forest canopy, producing the same loud 'kik' call as that made by male L.
cooloolensis heard calling later at Lake Poona and elsewhere (EM, pers. obs.; Naturesound,
1998). While most animals were heard calling from forest habitat close to Lake Poona (i.e.,
within 100 metres of suitable breeding habitat), between 30-50 animals were heard calling
much further (> 300 m) from water. This includes 5-10 animals heard calling from
rainforest at Bymien Picnic Area (-25.95447 153.1045°) approximately 900 m from Lake
Poona, the nearest known breeding site for this species. On the same night, numerous L.
cooloolensis (upwards of 200 animals) were heard and/or seen calling from vegetation
fringing at Lake Poona (-25.9637° / 153.1104°). Male calling activity within forest habitat
surrounding Lake Poona, including rainforest at Bymien Picnic Area, continued well past
midnight.
Small numbers of L. cooloolensis (between 10 and 20 animals) were again heard calling
from rainforest along the same walking track to Lake Poona on the evening of January 29,
2000. As before, animals were heard calling from Bymien Picnic Area at the start of the
walking track to Lake Poona, 900 m from the lake itself Large numbers of L. cooloolensis
(upwards of 200 animals) were again heard and/or seen calling at Lake Poona. Male calling
activity within forest habitat away from Lake Poona (as far away as Bymien Picnic Area)
continued through to dawn. Diurnal calling of L. cooloolensis was also opportunistically
recorded in the southern Cooloola section of the Great Sandy Region National Park (26.2389° / 153.0684°) on the 29th April, 9th July and the 13th July of 2010 in open
148
Eucalyptus forest trees at an elevation of approximately 95 metres above sea level. The
calls made by animals at this time were the same as described above. In contrast with
the above observations, calling activity was short-lived, lasting no longer then 30 seconds.
The maximum number of calling individuals at this site was two (recorded on the
29th April 2010). The nearest known breeding habitat for L. cooloolensis is a
wetland/coastal swamp located approximately 1.3 kilometres from the location of the
calling individuals (-26.2351° / 153.0556°). No other suitable breeding habitat was located
near this site during surveys conducted along a 5km east-west transect or on examination of
Google Earth satellite images.
The presence of L. cooloolensis up to 1.3 kilometres from suitable breeding habitat
demonstrates this species is capable of moving widely across forested terrain. The large
numbers of animals heard calling from rainforest near Lake Poona, moreover, suggests
forest surrounding breeding areas can, at times, support large numbers of this
rare species. The loss or degradation of forest habitat in wallum areas occupied by L.
cooloolensis (e.g., as a result of intense wildfire) may therefore have a significant impact
on numbers of this species and, potentially, the ability of animals to move within the
broader landscape.
The ability of L. cooloolensis to move large distances is a trait shared with the closelyrelated common sedgefrog (Litoria fallax), a species often encountered in dry eucalypt
forest hundreds of metres from suitable breeding habitat (CAS, EM, J-MH pers. obs). Like
L. cooloolensis, L. fallax can often be heard calling from the canopy of forest trees,
occasionally on dry ridges many hundreds of metres from water (EM pers. obs.). Why these
small, diminutive (SVL <45 mm) frogs should travel so widely is presently unclear. In the
case of L. cooloolensis, movement of animals away from breeding areas could occur in
response to: (1) increased competition amongst males at breeding sites during times of
heightened calling activity; or (2) limited availability of food in breeding areas (due to
increased competition with congeners at breeding sites during peaks in breeding activity).
As well as helping animals avoid competition, large scale movements of L. cooloolensis
could potentially promote gene flow between isolated populations, as well as facilitate
149
establishment of new breeding populations (Krebs 2009). Further data on competition,
movement and gene flow in L. cooloolensis are needed to determine which, if any, of these
scenarios is correct.
Though male frogs may call for a number of reasons, most call either to attract a mate (i.e.,
for breeding purposes) or to advertise their presence to other males (i.e., for territorial
purposes) (Zug et al. 2001; Wells 2007). In the case of L. cooloolensis, calling at distance
from breeding sites is unlikely to facilitate breeding, as females mating with males remote
from areas of breeding habitat would have to move many hundreds of metres in order to
spawn. Observations of more than one calling L. cooloolensis in the described areas
suggests observed calling activity away from breeding sites could instead be a territorial
response towards other male L. cooloolensis in the area. Whether this calling behaviour
represents practiced territorial calling in response to calling by conspecific males or calling
for some other purpose (e.g., defence or delineation of foraging territories) (see e.g., Ryan
2001) is presently unclear as visual observations of calling individuals, remote from
breeding areas, were not made. More detailed study of the calling and social behaviour
of this species are needed to ascertain the functional significance of calling by animals in
non-breeding habitat.
References
Bartelt, P.E., Peterson, C.R. and Klaver, R.W, 2004. Sexual differences in tbe postbreeding movements and habitats selected by westem toads {Bufo bóreas) in southeastern
Idaho. Herpetologica 60(4): 455-467.
Hines, H.B., Mahony, M. and McDonald, K., 1999. An assessment of frog declines in wet
subtropical Australia. Pp 44-63 in Campbell, A. (ed.) Declines and Disappearances of
Australian Frogs. Environment Australia, Canberra.
Johnson, J.R., Knouft, J.H. and Semlitsch, R.D., 2007. Sex and seasonal differences in tbe
spatial terrestrial distribution of gray treefrog (Hyla versicolor) populations. Biological
Conservation 140: 250-258.
150
Krebs, C.J., 2009. Ecology: The Experimental Analysis of Distribution and Abundance:
Intemal Edition. Pearson Education, United States of America.
Law, B.S. and Dickman, C.R., 1998. The use of habitat mosaics by terrestrial vertebrate
fauna: implications for conservation and management. Biodiversity and Conservation 7:
323-333.
Meyer, E., Hero, J.-M., Shoo, L. and Lewis, B., 2006. Recovery plan for tbe wallum
sedgefrog and otber wallum- dependent frog species 2005-2009. Report to Department of
Environment and Heritage, Canberra. Queensland Parks and Wildlife Service, Brisbane.
Regosin, J. V, Windmiller, B.S. and Reed, J.M., 2003. Terrestrial Habitat Use and Winter
Densities of tbe Wood Frog (Rana sylvatica). Joumal of Herpetology 37(2): 390-394.
Rittenhouse, T.A.G. and Semlitsch, R.D., 2007. Distribution of amphibians in terrestrial
habitat surrounding wetlands. Wetlands 27(1): 153-161.
Ryan, M., 2001. Anuran Communication. Smithsonian Institute Press, Washington.
Semlitsch, R.D. and Bodie, J.R., 2003. Biological Criteria for Buffer Zones around
Wetlands and Riparian Habitats for Amphibians and Reptiles. Conservation Biology 17(5):
1219-1228.
Wells, K.D., 2007. The Ecology and Behavior of Amphibians. The University of Chicago
Press, Chicago.
Zug, G.R., Vitt, L.J. and Caldwell, J.P, 2001. Herpetology: An Introductory Biology of
Amphibians and Reptiles, Second Edition. Academic Press, San Diego.
151