Ecological Indicators 85 (2018) 537–547

Contents lists available at ScienceDirect

Ecological Indicators
journal homepage: www.elsevier.com/locate/ecolind

Research paper

Assessing the ecological health of rivers when they are dry MARK
a,b,⁎ a a,b a c
Alisha L. Steward , Peter Negus , Jonathan C. Marshall , Sara E. Clifford , Catherine Dent
a
Water Planning Ecology, Queensland Department of Science, Information Technology and Innovation, Ecosciences Precinct, 41 Boggo Road, Dutton Park, Queensland
4102, Australia
b
Australian Rivers Institute, Griffith University, 170 Kessels Road, Nathan, Queensland 4111, Australia
c
Remote Sensing Centre, Queensland Department of Science, Information Technology and Innovation, Ecosciences Precinct, 41 Boggo Road, Dutton Park, Queensland
4102, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Rivers and streams that dry up are found on every continent, and can form a large proportion of river networks.
Ecological assessment When rivers are dry, traditional indicators of river health – such as aquatic macroinvertebrates, fish or water
Biological monitoring quality – cannot be measured. Aquatic health indicators are widely used to assess wetted habitats, but currently
Temporary river no universally applicable indicators have been developed or applied to assess dry riverbed health. Dry riverbeds
Murray-Darling Basin
are often the ‘typical’ state of many intermittent rivers and streams; however, the ecological health of these
Feral pigs
habitats is rarely, if ever, assessed in monitoring programs. Resource managers have called for indicators of
intermittent river health during the dry phase. The use of terrestrial invertebrate biota (e.g. ants, beetles, and
spiders) as indicators in this study provides a novel solution to assessing rivers when they are dry.
We developed a conceptual model of human-induced stressors (i.e. disturbance by livestock and feral
mammals) on dry riverbed biota, which guided the selection of potential health indicators. Livestock and feral
mammals are one of the most significant stressors on riverine ecosystems in Queensland, and impact riverbeds
by altering the substrate through compaction, rooting and pugging. We trialled the use of metrics of terrestrial
invertebrate assemblages as indicators of dry riverbed health in four Australian dryland catchments: Bulloo,
Paroo, Warrego and Nebine. We used quantile regression and found that terrestrial invertebrate communities
responded negatively (and significantly, p < 0.05) to a gradient of disturbance, defined by on-the-ground field
measurements of livestock and feral mammal impacts. This response to stressors was predicted by the initial
conceptual model.
We conclude that terrestrial invertebrates in this study are suitable indicators of dry riverbed health, as they
are impacted by disturbance from livestock and feral mammals. They can be used in the same way that in-
dicators, such as aquatic macroinvertebrates, are traditionally used to assess river health. We also successfully
combined indicators of wet and dry habitats to provide a holistic assessment of the health of intermittent river
ecosystems incorporating all sections of the river network. We suggest that this approach should be adopted by
other river health monitoring programs in rivers around the world.

1. Introduction climate change, and increased water extraction for human uses (Meehl
et al., 2007; Palmer et al., 2008; Larned et al., 2010). These effects can
1.1. Intermittent rivers are widespread, and there will be more of them in the increase the duration of dry spells in intermittent rivers, and can po-
future tentially convert perennial rivers to intermittent ones.

Rivers that temporarily cease to flow and dry up are a global phe-
nomenon, being found on every continent and nearly every watershed
(Datry et al., 2014). Intermittent rivers have been described as being
more representative of the world’s river systems than those with per-
ennial flows (Williams, 1988). Their spatial extent is likely to further
increase as a result of the combined effects of altered land-use patterns,
1.2. Dry riverbeds are not included in traditional river monitoring programs
Environmental monitoring and assessment of aquatic ecosystems is
undertaken to inform management: either by identifying reductions in
river health in response to anthropogenic stressors, or sometimes to
demonstrate the effectiveness of restoration actions. As such, it is a

⁎
Corresponding author at: Water Planning Ecology, Queensland Department of Science, Information Technology and Innovation, Ecosciences Precinct, 41 Boggo Road, Dutton Park,
Queensland 4102, Australia.
E-mail address: alisha.steward@griffithuni.edu.au (A.L. Steward).

https://doi.org/10.1016/j.ecolind.2017.10.053
Received 8 December 2016; Received in revised form 2 September 2017; Accepted 24 October 2017
1470-160X/ Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved.
A.L. Steward et al.
valuable tool for directing and supporting natural resource manage-
ment (Apitz et al., 2006; Field et al., 2007; Norris et al., 2007). The
unpredictability of flow, and subsequently of surface water presence
and distribution along intermittent rivers have been recognised as
challenges for environmental monitoring (Sheldon, 2005). Despite their
prevalence, intermittent rivers, and in particular dry riverbeds, have
often been neglected and frequently ignored in river management,
policy, and monitoring programs throughout the world (Steward et al.,
2012; Acuña et al., 2014; Mazor et al., 2014). Hence, gaps often exist in
monitoring data sets used to assess river health when sites are dry and
consequently not sampled during particular occasions, seasons or years.
This problem is particularly likely to occur in semi-arid and arid re-
gions, areas with Mediterranean climates, during the dry season in
monsoonal ‘wet-dry’ tropics, and in other regions during drought con-
ditions. Under these circumstances, monitoring and assessment of in-
termittent river ecosystems typically seeks out and considers only the
wet parts of the system (e.g. by using aquatic macroinvertebrates as
biological indicators: Chessman, 1995; Reynoldson et al., 1995), and
does not represent the entire river network. This is an important defi-
ciency because wet parts of an intermittent river network may be un-
representative of the ecological health of the system in totum. Further-
more, even in the absence of surface water, dry riverbed habitats can be
‘healthy’ and can have ecological values that may otherwise be over-
looked, such as unique biodiversity, their use as dispersal corridors for
terrestrial biota, and sites for the storage and processing of organic
matter and nutrients (McClain et al., 2003; Steward et al., 2011, 2012;
Acuña et al., 2014; Sánchez-Montoya et al., 2016). Conversely, dry
riverbed habitats may be degraded by various stressors and therefore
‘unhealthy’ (Chiu et al., 2017; Steward et al., 2017).
There is thus a recognised need to develop indicators of intermittent
river health during the dry phase (Acuña et al., 2014). Several potential
solutions have been proposed, but they rely on targeting specific ha-
bitats that may not be present in all intermittent river systems, or may
be unrepresentative of the overall health of the system being assessed.
For example, Robson et al. (2011) suggested sampling remnant pool
‘drought refuges’ for aquatic invertebrates during the dry phase. These
pools can have sparse, variable and patchy distributions at the catch-
ment scale, so by their very nature are unrepresentative of the overall
intermittent river network where they occur. Some reaches may not
contain any surface water to sample. Variable taxonomic composition
in refuge pools results from stochastic founder effects and strong biotic
interactions (Sheldon et al., 2010). Furthermore, sampling itself may
threaten refuge function and therefore system resilience by depleting
the supply of future colonists utilising refuges, so their use as mon-
itoring habitats may be undesirable. Leigh et al. (2013) partially
overcame these issues with their suggestion to adopt hyporheic in-
vertebrates as indicators of dry river health. However, not every dry
riverbed has a functioning hyporheic zone due to a lack of hyporheic
capacity (e.g. in bedrock or clay substrates), or to temporal absence due
to drying subsurface water during extended periods without flow
(Boulton and Stanley, 1995). To better overcome these issues we de-
monstrate the use of terrestrial riverbed invertebrates (sensu Steward
et al., 2011) as sensitive dry river health indicators, followed by an
approach to integrate these with more traditional aquatic indicators to
assess the entire river network – representing both wet and dry reaches
of an intermittent river system.
1.3. Terrestrial invertebrates as indicators for river health monitoring
For an indicator to be effective for biological monitoring it needs to
be relevant and sensitive (Andersen, 1999; Dobbie et al., 2013). To be
relevant for river health assessment it must be applicable to rivers and
streams within the region being assessed, and to be sensitive it needs to
change in a measureable way along gradients of a stressor. An ideal
indicator is also specific, so that it is responsive to gradients of a single
stressor, allowing for direct diagnosis of changes in river health.
538
Ecological Indicators 85 (2018) 537–547
However, few indicators used for river health monitoring are truly
specific; for example, aquatic macroinvertebrates, which are widely
used as indicators, respond to many varied stressors.
The widespread occurrence of intermittent rivers and streams dis-
cussed above, together with the emerging realisation that terrestrial
invertebrates of dry riverbeds are ubiquitous and contain elements of
specialist fauna for this habitat (Wishart, 2000; Steward et al., 2011;
Corti et al., 2013; Steward et al., 2017), suggest a broad relevance of
these biota as indicators of dry river health. While terrestrial in-
vertebrates have not previously been investigated as river health in-
dicators, they have been successfully utilised for biomonitoring in other
habitats. For example, ants (Formicidae) are routinely used as biolo-
gical indicators of rehabilitated mine sites throughout Australia, in-
cluding forests, semi-arid heathlands, subtropical shrublands and tro-
pical savannah woodlands (Andersen et al., 2004; Andersen and Majer,
2004). Similarly, ground beetles (Carabidae) have been employed as
biological indicators of exposed riverine sediments and riparian zones
in Europe (Boscaini et al., 2000; Eyre and Luff, 2002; Kleinwächter and
Rickfelder, 2007). However, the sensitivity of terrestrial invertebrates
to stressors in dry riverbeds has not yet been investigated.
1.4. A conceptual model of livestock and feral mammal impacts on
intermittent rivers
1.4.1. Natural intermittent rivers
To understand how stressors from livestock and feral mammals
impact intermittent rivers and streams during the dry phase, we de-
veloped a conceptual understanding of their ecological structure and
function (Fig. 1). Natural dry riverbeds can contain interstitial spaces
which are inhabited by terrestrial invertebrates, both when the bed
substrate is coarse (e.g. cobble, pebble) or fine (e.g. sand, silt/clay)
(Steward et al., 2011). A diversity of substrates and other habitat types
supports a comparatively diverse invertebrate fauna. Coarse substrates,
such as cobbles and pebbles, can provide structural complexity and
interstitial spaces for terrestrial invertebrates (Paetzold et al., 2008).
Sandy substrates are used by invertebrates that dig (e.g. Mecynotarsus
spp. (Coleoptera: Anthicidae), Hashimoto and Hayashi, 2012; Steward,
2014). Silt/clay substrates can crack once dry, and produce long fur-
rows in which invertebrates can reside, acting as cool microhabitats.
Catchment vegetation cover minimises runoff, and therefore erosion,
suggesting that more substrate types are available in areas rarely
smothered by sediment. Functioning riparian zones provide buffers,
which minimise nutrient and sediment loads entering waterways, and
inputs of leaf litter and woody debris, which provide habitat and po-
tential food resources for terrestrial invertebrates.
1.4.2. Impacted intermittent rivers
Out of all human-mediated disturbances, land use change has one of
the largest impacts on species richness (Murphy and Romanuk, 2014).
The use of land for livestock grazing (particularly cattle) impacts upon
both aquatic and terrestrial ecosystems (Fleischner, 1994; Agouridis
et al., 2005), with 80% of rivers and riparian zones damaged by grazing
in the United States (Belsky et al., 1999). Cattle grazing for meat pro-
duction is the dominant land use in arid and semi-arid parts of Aus-
tralia, where most of the river network is dry most of the time. More
than 83% of the total area of the Australian state of Queensland, re-
presenting almost 145 million hectares, is managed for beef cattle and
sheep grazing (Barson, 2013).
Dry riverbeds can be subjected to numerous stressors, such as li-
vestock trampling, overgrazing, weed infestation, gravel and sand ex-
traction, wastewater discharge, inundation by dams and weirs, crop-
ping, and their use as roads (Steward et al., 2012). However, stressors
caused by livestock and feral mammals appear to be both widespread
and ecologically important in our study area of Queensland, and are the
focus of this study. Feral mammals are common inhabitants of
Queensland, and include pigs (Sus scrofa), goats (Capra aegagrus hircus),
A.L. Steward et al.
sheep (Ovis aries) and cattle (Bos taurus) (Choquenot et al., 1996;
Robertson and Rowling, 2000; Mitchell et al., 2007a, 2007b; Gentle and
Pople, 2013), and generate similar stressors on river ecosystems as li-
vestock grazing (Figs. 1 and 2, ). There are no native ungulates in
Australia.
The direct impacts of livestock and feral mammals on dry river
ecosystems include a reduction in riparian vegetation cover; the com-
paction, pugging and rooting of soil; and eutrophication from their
urination and defecation (Kauffman et al., 1983a; Belsky et al., 1999;
Figs. 1 and 2). ‘Pugging’ is a term used to describe the physical dis-
turbance of banks and riverbeds by the hoofs of animals, typically when
the substrate is moist. This impact results in characteristic hoof in-
dentations or ‘pug’ marks left in the substrate that can persist long after
it has dried (Fig. 2). ‘Rooting’ is a term used to describe the physical
539
Ecological Indicators 85 (2018) 537–547
Fig. 1. Conceptual model illustrating cause-effect
relationships of livestock and feral mammals on
terrestrial invertebrates in intermittent river ecosys-
tems: upstream of the fence (background) represents
an ungrazed and thus undisturbed dry riverbed,
while downstream of the fence (foreground) is im-
pacted by stressors resulting from livestock and feral
mammals.
disturbance of banks and riverbeds by feral pigs that use their snout and
feet to turn over the substrate in search of food (Fig. 2). Both pugging
and rooting behaviours alter the bed habitat that invertebrates use,
destroying the interstitial spaces that large-bodied invertebrates of dry
riverbeds require. The rooting impact of pigs is somewhat similar to
pugging; however, rooting behaviour can mix and turn over sediments
while pugging compacts sediments.
Livestock and feral mammals are attracted to riparian zones, riv-
erbanks and riverbeds for shade, cooler temperatures, forage, and as
corridors for movement and to access water (Kauffman and Krueger,
1984; Fleischner, 1994; Belsky et al., 1999; Steward et al., 2011). Li-
vestock can form extensive tracks of bare ground in and around inter-
mittent rivers as they tend to repeatedly use the same path (Trimble and
Mendel, 1995; Fig. 2). Grazing can impact the function of riparian
A.L. Steward et al.
vegetation buffers and thus increase nutrient and sediment loads to
adjacent streams (Lowrance et al., 1984; Naiman and Décamps, 1997;
Muenz et al., 2006). Additionally, riparian vegetation has often been
cleared by land managers to promote pasture growth for livestock, and
this has been occurring over the last 200 or more years since grazing
commenced in Australia (NLWA, 2002; Martin and McIntyre, 2007). A
reduction in shade due to a loss of vegetation impacts terrestrial in-
vertebrates, as dry riverbeds can reach high temperatures that exceed
the 60 °C thermal tolerance of most eukaryotic organisms (Tansey and
Brock, 1972; Steward et al., 2017) and can have large diurnal fluc-
tuations, making them stressful environments.
Trampling and compaction by livestock and feral mammals can
compress sediments (Trimble and Mendel, 1995) and decrease soil
porosity (Orr, 1960). This limits the availability of interstitial spaces
and furrows for terrestrial invertebrates to inhabit, and water to in-
filtrate (Kauffman and Krueger, 1984), and in turn, decreases in-
vertebrate abundance and biomass (Schon et al., 2010). Livestock
trampling also disturbs the invertebrates living near the soil surface,
reducing their abundance and species richness (Cluzeau et al., 1992).
Trampling and feeding reduce vegetation cover, and as a result this
reduces the amount and diversity of litter and soil organic matter
content (Naeth et al., 1991; Green and Kauffman, 1995), which are
likely to be important for the provision of habitat and food resources to
terrestrial invertebrate communities. Bromham et al. (1999) found a
more diverse invertebrate fauna in ungrazed woodlands compared to
grazed woodlands, most likely due to a higher diversity of food and
habitat in the areas of lower disturbance. Similarly, Lindsay and
Lindsay and Cunningham, (2009) found that excluding livestock ben-
efitted components of the invertebrate community, vegetation condi-
tion and the process of litter decomposition. Dry riverbeds represent
important movement corridors used by native wildlife (Sánchez-
Montoya et al., 2016), livestock and feral animals to access sparse,
540
Ecological Indicators 85 (2018) 537–547
Fig. 2. The effects of livestock and feral pigs on
riverbeds and banks: a) cattle pugging across the
width of a dry riverbed in the Cooper Creek catch-
ment, Australia; b) riverbed disturbance from pig
rooting near a waterhole in the Diamantina River
catchment, Australia; and c) the difference between
grazed (left) and ungrazed (right) land separated by
a fence in the centre of the image, located in the
Warrego River catchment, Australia. Note the ex-
posed tree roots and bare ground on the left, in
comparison to the increased ground vegetation and
leaf litter, and natural benches up the bank on the
right.
persistent waterholes in intermittent rivers, so are therefore highly
subjected to trampling, pugging and rooting (Steward et al., 2011).
Livestock and feral mammal urine and manure deposited directly
into rivers, or washed in from adjacent areas, can affect water quality
by increasing nitrate and phosphate levels (Schepers and Francis, 1982;
Muenz et al., 2006) and lowering the dissolved oxygen content, causing
fish mortality (Taylor et al., 1989). Feasibly, these added nutrients also
have an impact on terrestrial invertebrate communities once riverbeds
are dry (Fig. 1), potentially disrupting dry riverbed foodwebs. For ex-
ample, added nutrients could foster the growth of cyanobacteria, which
is relatively unpalatable to consumers once stranded as riverbeds dry
up, as opposed to more palatable types of algae which represent higher
quality food (Guo et al., 2017).
Livestock and feral mammals contribute to an increase in bare
ground through the consumption and trampling of vegetation
(Kauffman et al., 1983a; Popolizio et al., 1994; Trimble and Mendel,
1995; Agouridis et al., 2005), and can hinder vegetation succession
(Kauffman et al., 1983b). Barrios-Garcia et al. (2014) found that
feeding and rooting by feral pigs in Hawaii reduced plant biomass by
60% relative to areas where pigs were excluded. In a study in Colorado,
Schulz and Leininger (1990) found that grazed areas had four times
more bare ground than areas that had been ungrazed for 30 years. The
extent to which cattle can generate bare ground is further evident in a
study by Mills et al. (1989) in south-west Queensland, where more than
half of the properties examined had 60% of their land grazed bare,
representing almost two million hectares. Pettit (2002) found that in-
tensive cattle grazing along a river waterhole in the Cooper Creek
catchment in western Queensland led to a significant reduction in ve-
getation cover and an increase in the percentage of bare ground. In-
creases in bare ground can threaten river ecosystems by accelerating
surface water run-off and soil erosion during rainfall events and sub-
sequently increasing fine sediment supply to waterways (Ludwig and
A.L. Steward et al.
Tongway, 2002; NLWA, 2002).
The increased accumulation of deposited sediment in river channels
is well documented to impact aquatic biota (Wood and Armitage, 1997)
via changes to particle size distribution (Schälchli, 1992). We expect
that deposited fine sediments such as silt and clay, which blanket the
channel substrate and fill interstitial spaces (Richards and Bacon, 1994;
Wood and Armitage, 1997; Bartley and Rutherfurd, 2005), will simi-
larly impact dry riverbed terrestrial invertebrates by changing their
habitat and thus their assemblage composition (Fig. 1).
Degradation of dry riverbeds by livestock and feral mammals has
not been studied previously, but given these high exposure rates and
the analogous impacts known from other habitats, we expect that
trampling, compaction, pugging and rooting will impact dry riverbed
invertebrates directly by squashing (and therefore killing them), and
indirectly by decreasing the availability of food and habitat (Fig. 1).
1.5. Research questions
We predicted that terrestrial invertebrate communities of dry riv-
erbeds would respond to stressors related to the impacts of livestock
and feral mammals. In this paper we ask: Do the terrestrial invertebrate
communities of dry riverbeds show an ecological response to measures
of this stressor, in terms of reduced taxon richness or abundance? If
there is a response, can these metrics be used for river health assess-
ment?
2. Methods
This study tested the sensitivity of terrestrial invertebrate richness
and abundance to stressors generated by livestock and feral mammals
in four catchments in southern Queensland, Australia, and then applied
these findings to the monitoring and assessment of river condition.
2.1. Testing the sensitivity of invertebrate richness and abundance to
stressor intensity
Testing of the sensitivity of the richness and abundance of dry riv-
erbed terrestrial invertebrates as indicators of ecosystem response to
disturbance from livestock and feral mammals was conducted in the
Bulloo, Paroo, Warrego and Nebine river catchments (Fig. 3). All four
have semi-arid climates, and contain rivers and streams with inter-
mittent flow. Kennard et al. (2010) classified flow in the region as
highly to extremely intermittent. The Bulloo catchment is an isolated
drainage that flows into several terminal lakes. The Paroo, Warrego and
Nebine catchments are part of the northern most drainages of Aus-
tralia’s largest river system – the Murray-Darling Basin. Most of the area
is used as grazing land for cattle and sheep, with feral pigs and goats
recognised as important pest species (Negus et al., 2013). Sample sites
were selected to be statistically representative of the region using a
Generalised Random Tessellated Stratification (GRTS) design adapted
for use on stream networks (Dobbie and Stevens, 2006; Dobbie and
Negus, 2013). GRTS was used to ensure spatial balance and efficiency
in site selection while accounting for operational difficulties that can
occur during field sampling of specific sites (Dobbie et al., 2013).
Thirty-one sites were sampled: six in the Bulloo catchment, eight in
the Paroo, eleven in the Warrego and six in the Nebine catchment
(Fig. 3, Supplementary Table 1).
2.1.1. Field-based estimates of stressor intensity
Direct measures of stressor intensity were made at each site during
September–November (austral spring) in 2012. Four 200-m transects
were run parallel to the river channel on the bank at each site, with two
transects along each bank where possible. Measurements of riverbed
impact were made in contiguous 10 m x 10 m quadrats along each
transect (20 quadrats per transect). Livestock impacts in each quadrat
were assessed based on the density of pug markings and tracks along
541
Ecological Indicators 85 (2018) 537–547
the ground surface, and feral pig, sheep and goat impacts were assessed
by the density of their tracks and rooting damage. Measurements of the
stressor intensity were categorised as: none (1), light (2), moderate (3),
or severe (4), and were averaged across all 80 quadrats to produce a
score for each site. Subjective but consistent judgements by experienced
field staff were used to describe the categories: light – visible presence
of livestock or feral mammals, such as limited rooted areas or sparse
hoof prints; moderate – obvious damage from pugging and rooting; and
severe – obvious and extensive with pugging and rooting over 50% of
the area covered by the transect (e.g. Fig. 2a, b).
2.1.2. Terrestrial invertebrate sampling
Terrestrial invertebrates were sampled in dry riverbeds using pitfall
traps after Steward et al. (2011). Six replicate pitfall traps were posi-
tioned in the dry channel at each site and set for approximately 24 h.
Traps were positioned in diverse micro-habitats within the dry channel,
such as leaf litter, bare ground and woody debris, and across different
substrate types. The pitfall traps consisted of 250 mL plastic jars, 77 mm
high and 67 mm in diameter, filled to 100 mL with 70% Ethanol, 3%
Glycerol and 27% water as per Steward et al. (2011) and Wishart
(2000). Ethanol was used as a preservative and a killing agent. Glycerol
was used to keep the invertebrate bodies supple for handling and
identification. A drop of detergent was added to break the surface
tension, which prevented captured invertebrates from escaping. A
plastic cover was positioned approximately 100 mm over each trap to
prevent rain, leaf litter and other debris from blocking the trap and
reducing its efficiency (Williams, 1959).
Specimens were identified in the laboratory with a stereo-micro-
scope to order and then family level where possible, using taxonomic
keys (CSIRO Division of Entomology, 1991). Ant subfamilies and
genera, and beetle families were verified by a museum entomologist
(Chris Burwell, Queensland Museum, pers. comm.).
2.1.3. Data analyses
Terrestrial invertebrate taxon richness and abundance were calcu-
lated as the average of the six replicates per site and then log10-trans-
formed for regression analysis. We used quantile regression to test
whether livestock and feral mammal intensity was a limiting factor on
the invertebrate community. Quantile regression is a statistical tech-
nique that can be used to estimate the response effects of limiting
factors (Cade and Noon, 2003; Brooks and Haeusler, 2016). It can de-
termine whether there are relationships between a response and lim-
iting factors that are evident across the distribution of the data and is
especially useful where the data have heterogeneous variation (poten-
tially due to influences of several limiting factors) (Cade and Noon,
2003). Quantile regression determines a linear regression function be-
tween a limiting factor and the proportion of the measured response
data points (equivalent to the quantile) that fall below the regression
line (Brooks and Haeusler, 2016). For our application, a quantile of 0.8
as the maximum limit with the total site number of 31 was selected
based on the formula (maximum quantile < 1–(5/number of sites)) for
determining quantile extremes (Rogers, 1992). We calculated the 0.8
linear quantile regression relationships between the stressor (mean li-
vestock and feral mammal intensity score at each site) and the re-
sponses (mean log10 taxon richness and mean log10 abundance) using
the rq function within the quantreg package (Koenker, 2016) using R
software (R Core Team, 2013).
2.2. Application of dry riverbed invertebrates as indicators for government
reporting of multi-stressor river health assessments
We undertook a multi-stressor assessment as part of the Queensland
Government’s ongoing river health assessment program (Fig. 4).
Stressors (i.e. grazing intensity) were used to measure the threat (li-
vestock and feral mammals), and the biological response (e.g. terrestrial
invertebrate richness) was used to measure dry riverbed condition. In
A.L. Steward et al.
Fig. 3. Map of sites sampled in the Bulloo, Paroo, Warrego and Nebine catchments (closed circles). Bold lines indicate catchment boundaries. The position of the catchments in the
Australian state of Queensland is provided (inset).
this study, we utilised dry riverbed terrestrial invertebrates as one in-
dicator to represent the ecological impacts of livestock and feral
mammals. We used the dry riverbed terrestrial invertebrate richness
and abundance metrics derived from the Bulloo, Paroo, Warrego and
Nebine catchment samples, as described above.
Additional indicators sampled at the same time and within the same
reach as terrestrial invertebrates were used to represent the impacts of
other stressors, and these were brought together into an overall as-
sessment (Negus et al., 2015; Fig. 4). Our multi-stressor assessment
identified priority threats to the health of each river system. Indicators
of river health were chosen based on the threats which were evaluated
to pose the greatest risk to these ecosystems, and were aggregated to
present an ‘all of catchment’ assessment. Conceptual models (e.g. Fig. 1)
were used to link cause to effect, allowing for a risk-based diagnosis of
the most likely causes of any degradation in condition. Dry riverbed
terrestrial invertebrate metrics were used to define river health based
on the stressor of introduced riparian fauna, quantified using the live-
stock and feral mammal damage metrics described above. This was then
combined with assessments of health based on other important stressors
in the catchments (introduced aquatic fauna and deposited sediment in
permanent waterholes) to give an overall picture of river health, in-
corporating both wet and dry parts of the rivers.
For government reporting (Fig. 4), assessments used a reference
condition approach (Hawkins et al., 2010; Dobbie et al., 2013), with
expected values derived from sites formally evaluated to be minimally
disturbed according to set criteria (DSITIA (Department of Science,
Information Technology, Innovation and the Arts) (2013)). Site metrics
were range standardised (Eq. (1)) to theoretically vary between zero
(worst possible health) and one (health equivalent to reference) and
aggregated to provide an assessment of catchment health.
542
Ecological Indicators 85 (2018) 537–547
X − reference
S tan dardised score = 1 −
WCV − reference (1)
Where: X = site value, Reference = reference value, WCV = value
under realistic worst case scenario
Reference values were zero for the stressor metric (the total live-
stock and feral mammal scores from six transects per site), as our
scoring where there are no impacts of livestock or feral mammals would
be zero – an “historical” reference condition using the definitions of
Stoddard et al. (2006). The 20th percentile of mean log10 taxon richness
(5.5) and abundance (15.17) from reference sites was used as the re-
ference values for terrestrial invertebrates. A percentile value is used
rather than the full range to account for anomalous observations in-
fluencing assessments (Dobbie et al., 2013). Worst case values were six
for the stressor metric based on the value that would be obtained if all
quadrats assessed were recorded as having severe stressor intensity, and
zero for the mean sample taxon richness and abundance based on the
hypothetical potential of no terrestrial invertebrates being recorded at a
site.
3. Results
3.1. Stressor intensity
Stressors attributed to livestock and feral mammals along river
channels were assessed in all four catchments, with some sites ap-
pearing to be in natural condition – having no impacts recorded – while
other sites had severe degradation measured in one or more of the as-
sessed quadrats. Quadrats with severe degradation were recorded at
sites in all four catchments, although it was not widespread within the
sites. The average livestock and feral mammal impact score used to
A.L. Steward et al.
Fig. 4. Summary table from a Queensland Government technical report (Negus et al., 2015) on the health of river ecosystems based on the assessment of threats from introduced aquatic
fauna, introduced riparian fauna (i.e. livestock and feral mammals), and deposited sediment. The summary table shows the results of each indicator sampled in each catchment, to
demonstrate how terrestrial invertebrates were incorporated into the assessment of river health. The ‘current threat’ for introduced riparian fauna was a measure of the intensity of
damage from livestock and feral mammals.
represent sites ranged from 1 (no impact) to 2 (light impact) (Table 1).
The Nebine catchment had the lowest impact score (1.08), followed by
the Warrego (1.16), Bulloo (1.27), and Paroo (1.37).
3.2. Terrestrial invertebrates
Ninety invertebrate taxa, represented by 12,979 individuals, were
recorded from the sampling sites (Supplementary Table 2). The fol-
lowing taxonomic groups were represented: Acari, Amphipoda,
Araneae, Blattodea, Chilipoda, Coleoptera, Dermaptera, Diplura,
Diptera, Entomobryomorpha, Hemiptera, Hymenoptera, Isopoda,
Lepidoptera, Mantodea, Neuroptera, Orthoptera, Poduromorpha,
Pseudoscorpionida, Psocoptera, Symphypleona, Thysanoptera, and
Trichoptera.
Ants (Formicidae) and springtails (Collembola) were collected from
every site, with Iridomyrmex ants and Entomobryidae springtails pre-
sent at 29 of the 31 sites (Supplementary Table 2). Flies (Diptera) and
beetles (Coleoptera) were collected from 30 sites each, of which phorid
flies were collected from 28 sites and anthicid beetles collected from 24
sites. Other commonly-encountered taxa included Acari (27 sites),
Apocrita (27 sites), Cecidomyiidae (25 sites), and Lycosidae (22 sites).
Uncommon taxa only found at a single site included Aleyrodidae,
Chilipoda, Chrysopidae, Gryllotalpidae, Hydrophilidae, Mantinae,
Micropholcommatidae, Miridae, Miturgidae, Mycetophagidae,
Nematoda, Oecobiidae, Podomyrma, Pseudomyrmecinae, Reduviidae,
Spercheidae, Tipulidae, and Trichoptera.
The mean taxon richness of terrestrial invertebrates per pitfall trap
at a site varied between sites and catchments, ranging from 3.8 taxa at a
site in the Bulloo, to 13.8 taxa at a site in the Nebine catchment
(Table 1). Mean terrestrial invertebrate abundance per pitfall trap
ranged from 5.5 individuals at a site in the Bulloo, to 385 individuals at
543
Ecological Indicators 85 (2018) 537–547
a site in the Paroo catchment.
3.3. Sensitivity to the stressor
The 0.8 quantile regression models for both invertebrate richness
and abundance were significant (i.e. p < 0.05) (Table 2, Fig. 5). This
indicates that the response of dry riverbed invertebrates is limited by
the intensity of livestock and feral mammal disturbance.
3.4. River health assessment
The intensity of the stressor (livestock and feral mammals) was
assessed as ‘moderate’ in the Warrego, Paroo and Bulloo catchments,
and ‘slight’ in the Nebine (‘current threat’, Fig. 4). The condition of dry
riverbeds was assessed as being ‘slightly disturbed’ in all four catch-
ments, based on the terrestrial invertebrate results. These results con-
tributed to the overall river assessment across multiple stressors in-
dicating the relative risk to the health of rivers in these catchments from
different threatening processes (for more information refer to Negus
et al., 2015).
4. Discussion
4.1. Suitability of terrestrial invertebrates as indicators for river health
monitoring
Several authors have recently recognised the need for indicators of
intermittent river health during the dry phase (Steward et al., 2012;
Acuña et al., 2014; Datry et al., 2014; Steward et al., 2017). This study
fills that need. The terrestrial invertebrates of dry riverbeds meet the
criteria as a suitable indicator because:
A.L. Steward et al. Ecological Indicators 85 (2018) 537–547

Table 1
Terrestrial invertebrate index results for sites in the Bulloo, Nebine, Paroo and Warrego River catchments in Queensland, Australia.

Catchment Site name Site Mean livestock and feral Mean log10 taxon Mean log10
number mammal impact score richness abundance

Bulloo Blackwater Creek at Adavale 1 1.0031 9.2 330.2

Bulloo Bulloo River at Norley Station 2 1.4813 6.5 108.5
Bulloo Bulloo River at Pinketta Station 3 1.0594 3.8 5.5
Bulloo Bulloo River at Quilpie 4 1 6.3 11.5
Bulloo Bulloo River at Thargomindah 5 1.0348 8.2 29.8
Bulloo Fifteen Mile Creek at Hay Paddock Hut Waterhole 6 2.0281 6.3 34.7
Nebine Nebine Creek at Aqua Downs 7 1 13.8 91.7
Nebine Nebine Creek at Murra Murra 8 1.1563 7.5 114.5
Nebine Nebine Creek at Roseleigh Crossing 9 1.1375 5.5 90.3
Nebine Wallum Creek at Bollon Reserve 10 1 13.7 42.2
Nebine Wallum Creek at Homeboin Waterhole 11 1.1563 12.0 47.3
Nebine Wallum Creek at One Mile Waterhole 12 1.0063 11.2 10.5
Paroo Beechal Creek at Boolbury Waterhole 13 1.0438 7.7 32.7
Paroo Gumholes Creek at Bowra 14 1.6781 5.3 15.2
Paroo Moonjaree Creek at Moonjaree 15 1.5906 9.8 21.2
Paroo Paroo River at 14 Mile Waterhole 16 1.0206 3.8 31.5
Paroo Paroo River at Wimmera Waterhole 17 1.9 6.5 7.3
Paroo Pingine Waterhole 18 1.1406 6.8 101.3
Paroo Rolwegan Creek at Unnamed Waterhole 19 1 8.3 5.8
Paroo Yowah Creek at Thandy Waterhole North 20 1.5813 4.7 24.8
Warrego Ambathalla at Ambathalla Creek 21 1.2875 8.2 35.5
Warrego Cuttaburra Creek Tinnenburra Waterhole at Tinnenburra 22 1.5375 10.5 385.0
Warrego Langlo River at Rylestone Waterhole 23 1.2656 3.8 120.7
Warrego Ward River at Bayrick Fish Hole 24 1.0031 6.8 92.2
Warrego Ward River at Binnowee 25 1.1313 4.5 30.0
Warrego Warrego River at Augathella 26 1.1063 11.3 55.3
Warrego Warrego River at Charleville 27 1.0031 9.5 250.0
Warrego Warrego River at Cunnamulla Weir 28 1 12.2 26.3
Warrego Warrego River at Merwah waterhole 29 1.1281 7.8 35.8
Warrego Warrego River at Wallen 30 1.0688 5.7 15.2
Warrego Warrego River at Wyandra 31 1.2188 6.3 17.5

Table 2
Quantile (0.8) regression analysis results for a) mean log10 taxon richness and b) mean
However, we don’t currently know if it is other stressors or natural
log10 abundance. An asterisk indicates significant results (p < 0.05). variation that are the unmeasured influences.

Coefficients Value Standard t value Pr( > |t|) Our results show that terrestrial invertebrates of dry riverbeds meet
Error
Andersen’s (1999) criteria for indicator selection. They are widely
a) Mean log10 taxon richness distributed, abundant and taxa-rich − terrestrial invertebrates were
Intercept 1.35 0.17 7.84 0.00 found at every site sampled in this study, and have been collected in
Livestock and feral mammal −0.27 0.13 −2.12 0.04* high abundance and richness at these sites as well as at sites in many
impact score
other catchments throughout Australia and other parts of the world
b) Mean log10 abundance (Wishart, 2000; Lalley et al., 2006; Larned et al., 2007; Steward et al.,
Intercept 2.63 0.31 8.59 0.00
2011; Corti et al., 2013; Corti and Datry, 2016). Terrestrial in-
Livestock and feral mammal −0.71 0.24 −2.95 0.006*
impact score vertebrates are functionally important in ecosystems − ants in parti-
cular are recognised as the dominant terrestrial invertebrate group in
the Australian environment (Andersen, 1997; Andersen et al., 2004).
1) They are relevant. Terrestrial invertebrates were found in all dry Particular taxonomic and functional groups of terrestrial invertebrates
riverbeds sampled, they were abundant, and were common. They are known to be sensitive and responsive to environmental changes −
have also been recorded as abundant and common elsewhere in for example, beetles from the family Carabidae (Boscaini et al., 2000;
Australia and Europe (Steward et al., 2011, 2017). Terrestrial in- Eyre and Luff, 2002; Bates et al., 2007; Kleinwächter and Rickfelder,
vertebrates are important components of intermittent river ecosys- 2007) and functional groups of ants (King et al., 1998; Hoffmann and
tems (Steward et al., 2017). Andersen, 2003).
2) They are sensitive. Variability in terrestrial invertebrate metrics was Lastly, terrestrial invertebrates are easily sampled (traps are de-
associated with a gradient representing intensity of livestock and ployed one day and retrieved the next) using inexpensive field equip-
feral mammal damage to dry riverbeds, thereby supporting our ment (e.g. Steward et al., 2017), and are easy to sort and identify − the
conceptual model. Terrestrial invertebrates have also responded to laboratory identification effort is comparable to that required for
stressors in other habitats (Andersen et al., 2004; Andersen and aquatic invertebrate samples.
Majer, 2004). We did not address all possible stressors impacting the terrestrial
3) They are specific. We used quantile regression to account for natural invertebrates of dry riverbeds in Queensland in the conceptual model.
variability in substrate composition, hydrology, etc. of the sampled Our model focused on habitat degradation caused by livestock and feral
sites. We included samples from different catchments to broaden the mammals. Of course there may be other relevant stressors that need to
applicability of our results. The quantile regression showed that be included in future conceptual models. Additional stressors to riv-
intensity of livestock and feral mammal damage limits invertebrate erine ecosystems in Queensland that may influence terrestrial in-
richness and abundance, but that other unmeasured influences exist. vertebrates include:

544
A.L. Steward et al.
Fig. 5. Quantile regression results for a) Mean log10 taxon richness and b) Mean log10 abundance in relation to grazing pressure. The 0.8 quantile regressions (solid lines) are both
significant (p < 0.05).
• Acid soil run-off – affecting the pH of riverbed sediments;
• Deposited sediment (resulting from land clearing) – altering dry
riverbed habitats through in-filling of interstitial spaces;
• Flow management – the frequency, timing, and duration of wet/dry
phases affecting the temporal and spatial availability of dry riverbed
habitats;
• Habitat removal or disturbance (e.g. gravel extraction) – causing
physical damage to riverbed habitats;
• Riparian habitat fragmentation (e.g. riparian tree clearing) – chan-
ging allochthonous food supplies;
• Salinity – potential toxic effects from salt in riverbed sediments; and
• Toxicants (e.g. pesticides and heavy metals) – potential toxic effects
from contaminants in riverbed sediments.
To further explore the use of terrestrial invertebrates as biological
indicators, conceptual cause/effect responses should be investigated
with laboratory and/or field experiments (Dobbie et al., 2013). Other
features of terrestrial invertebrate assemblages, such as traits (e.g.
functional feeding groups), could also be examined. Future studies
could use controlled experimental manipulation; for example, sites
could be fenced from livestock and sampled as ground cover improves.
Many studies have advocated the use of invertebrates as biological
indicators in various terrestrial environments, whether it be entire as-
semblages (Obrist and Duelli, 2010) or specific taxonomic groups (ants:
Majer, 1983; Peck et al., 1998; Andersen et al., 2002; Andersen et al.,
2004; Andersen and Majer, 2004; Rainio and Niemelä, 2003; Michaels,
2007; Marc et al., 1999; Greenslade, 2007). It has been suggested that
groups such as ants and beetles are likely to respond to habitat changes
in different ways, and that using both ants and beetles together could
strengthen assessments (Leach et al., 2013).
We now recommend that terrestrial invertebrates be adopted as
health indicators of dry riverbeds in Queensland and suggest they be
tested in other catchments, and in other parts of the world, with both
similar and different stressors to those assessed in this study.
4.2. Response of terrestrial invertebrates to livestock and feral mammal
disturbance
This study found a strong association between the response of ter-
restrial invertebrates that inhabit dry riverbeds and the impacts caused
by livestock and feral mammals. Sites in the Bulloo, Paroo, Warrego and
545
Ecological Indicators 85 (2018) 537–547
Nebine catchments where dry riverbeds were subjected to higher
grazing impacts, as determined through the on-the-ground field mea-
surements of livestock and feral mammal disturbance, were limited
both in terms of terrestrial invertebrate taxon richness and the abun-
dance of individuals.
Our findings of terrestrial invertebrate responses to livestock and
feral mammal impacts are well supported in the literature for other
habitat types. For example, Schon et al. (2010) found high soil porosity
was clearly associated with high abundance and biomass of soil in-
vertebrates, and that these indices were lower in areas grazed by dairy
cows as a result of soil compaction. Similarly, Cluzeau et al. (1992)
found that cattle trampling compacted the soil and affected its struc-
ture, which greatly reduced earthworm density, biomass, and species
richness. Leach et al. (2013) found more species of ants and beetles in
forest than in nearby areas that had been cleared for grazing. Lindsay
and Cunningham (2009) found that woodlands without recent grazing
had a higher abundance of beetles. Bromham et al. (1999) found a more
diverse ground invertebrate fauna in ungrazed than grazed woodlands,
and this was attributed to differences in vegetation and ground litter.
Feral pigs impact terrestrial invertebrates in a similar way: Vtorov
(1993) found a higher abundance and richness of soil invertebrates,
particularly springtails, from rainforests in Hawaii where feral pigs had
been removed. Further research into the responses of terrestrial in-
vertebrates to stressors could involve the collection of dry riverbed
habitat data, such as substrate composition, from natural and degraded
sites. This could determine whether some substrate types are more
prone to degradation than others.
4.3. Inclusion of dry riverbeds in river monitoring programs
We successfully applied the results of our study to a government
river health assessment in four catchments (Negus et al., 2015). We
combined the terrestrial invertebrate results with other indicators for
reporting purposes, illustrating how wet and dry river health indicators
can be integrated into one assessment (Fig. 4). As far as we are aware,
this is the first time that river monitoring programs have combined wet
and dry indicators to represent the health of intermittent river channels.
In the government report we also combined the assessments across
multiple stressors: in summary, the threat intensity was quantified at
reporting scales for each stressor, as was the resulting ecosystem con-
dition based on the ecological response metrics and their deviation
A.L. Steward et al.
from reference expectations. By considering these separately, it is
possible to diagnose the causes of degradation and to prioritise man-
agement actions to conserve or restore ecosystems.
The use of terrestrial invertebrates from dry riverbeds as indicators
solves previous dilemmas over how to assess river health in the absence
of surface water. We have confirmed that the impacts of stressors on the
dry riverbeds of intermittent rivers can be quantified. Terrestrial in-
vertebrates have now been adopted as indicators for government river
health assessments in other catchments in Queensland.
4.4. Future monitoring and assessment of dry riverbeds
It is time to introduce the assessment of dry sites into current and
future river health monitoring and assessment programs (Steward et al.,
2012; Acuña et al., 2014). Dry and wet indicators can then be combined
to report on the health of an entire river network. We have provided an
example of how this can be done. This can easily be adopted by other
programs in intermittent rivers around the world. A key way forward is
to incorporate dry riverbeds into ecosystem monitoring and assessment
programs through government policy and legislation. We need to re-
cognise dry riverbeds as important elements of intermittent rivers – as
habitats in their own right.
Acknowledgements
Bill Senior, Dean Holloway, Delwyn Hansen, Glenn McGregor,
James Fawcett, Jaye Lobegeiger, Joanna Blessing, John Bowlen, Penny
Rogers, Ryan Woods and Tess Mullins from the Queensland
Government and Ric Newson from South West Natural Resource
Management (NRM) Ltd. are acknowledged for field work assistance,
and we also acknowledge the landholders for permission to sample on
their properties. We would like to thank Farah Zavahir for assistance
with laboratory processing of samples, and Chris Burwell from the
Queensland Museum for verifying some of the beetle and ant identifi-
cations. Funding for this project was provided by the Department of
Science, Information Technology and Innovation in the Queensland
Government, Australia.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.ecolind.2017.10.053.
References
Acuña, V., Datry, T., Marshall, J., Barceló, D., Dahm, C.N., Ginebreda, A., McGregor, G.,
Sabater, S., Tockner, K., Palmer, M.A., 2014. Why should we care about temporary
waterways? Science 343, 1080–1081.
Agouridis, C.T., Edwards, D.R., Workman, S.R., Bicudo, J.R., Koostra, B.K., Vanzant, E.S.,
Taraba, J.L., 2005. Streambank erosion associated with grazing practices in the
humid region. Trans. ASAE 48, 181–190.
Andersen, A.N., Majer, J.D., 2004. Ants show the way Down Under: invertebrates as
bioindicators in land management. Front. Ecol. Environ. 2, 291–298.
Andersen, A.N., Hoffmann, B.D., Müller, W.J., Griffiths, A.D., 2002. Using ants as
bioindicators in land management: simplifying assessment of ant community re-
sponses. J. Appl. Ecol. 39, 8–17.
Andersen, A.N., Fisher, A., Hoffmann, B.D., Read, J.L., Richards, R., 2004. Use of ter-
restrial invertebrates for biodiversity monitoring in Australian rangelands, with
particular reference to ants. Aust. Ecol. 29, 87–92.
Andersen, A.N., 1997. Functional groups and patterns of organization in North American
ant communities: a comparison with Australia. J. Biogeogr. 24, 433–460.
Andersen, A.N., 1999. My bioindicator or yours? Making the selection. J. Insect Conserv.
3, 61–64.
Apitz, S.E., Elliott, M., Fountain, M., Galloway, T.S., 2006. European environmental
management: moving to an ecosystem approach. Integr. Environ. Assess. Manag. 2,
80–85.
Barrios-Garcia, M.N., Classen, A.T., Simberloff, D., 2014. Disparate responses of above-
and belowground properties to soil disturbance by an invasive mammal. Ecosphere 5,
1–13.
Barson, M., 2013. Land management practice trends in Queensland’s beef cattle/sheep
industries. Caring for Our Country Sustainable Practices Fact Sheet 17. Department of
Agriculture, Fisheries and Forestry, Canberra (Last Accessed 07 December 2016).
546
Ecological Indicators 85 (2018) 537–547
http://www.agriculture.gov.au/SiteCollectionDocuments/natural-resources/soils/
trends-factsheets/qld-practices-grazing.pdf.
Bartley, R., Rutherfurd, I., 2005. Re-evaluation of the wave model as a tool for quanti-
fying the geomorphic recovery potential of streams disturbed by sediment slugs.
Geomorphology 64, 221–242.
Bates, A.J., Sadler, J.P., Fowles, A.P., 2007. Livestock trampling reduces the conservation
value of beetle communities on high quality exposed riverine sediments. Biodiv.
Conserv. 16, 1491–1509.
Belsky, A.J., Matzke, A., Uselman, S., 1999. Survey of livestock influences on stream and
riparian ecosystems in the western United States. J. Soil Water Conserv. 54, 419–431.
Boscaini, A., Franceschini, A., Maiolini, B., 2000. River ecotones: carabid beetles as a tool
for quality assessment. Hydrobiologia 422, 173–181.
Boulton, A.J., Stanley, E.H., 1995. Hyporheic processes during flooding and drying in a
Sonoran Desert stream: II Faunal dynamics. Arch. fuer Hydrobiol. 134, 27–52.
Bromham, L., Cardillo, M., Bennett, A.F., Elgar, M.A., 1999. Effects of stock grazing on the
ground invertebrate fauna of woodland remnants. Aust. J. Ecol. 24, 199–207.
Brooks, A.J., Haeusler, T., 2016. Invertebrate responses to flow: trait-velocity relation-
ships during low and moderate flows. Hydrobiologia 773, 23–34.
CSIRO Division of Entomology, 1991. 2nd edition. The Insects of Australia: A Textbook
for Students and Research Workers Vols. 1 and 2 Melbourne University Press,
Carlton, Victoria.
Cade, B.S., Noon, B.R., 2003. A gentle introduction to quantile regression for ecologists.
Front. Ecol. Environ. 1, 412–420.
Chessman, B.C., 1995. Rapid assessment of rivers using macroinvertebrates: a procedure
based on habitat-specific sampling, family level identification and a biotic index.
Aust. J. Ecol. 20, 122–129.
Chiu, M.C., Leigh, C., Mazor, R., Cid, N., Resh, V., 2017. Chapter 5.1: Anthropogenic
threats to intermittent rivers and ephemeral streams. In: Datry, T., Bonada, N.,
Boulton, A. (Eds.), Intermittent Rivers and Ephemeral Streams: Ecology and
Management. Elsevier.
Choquenot, D., McIlroy, J., Korn, T., 1996. Managing Vertebrate Pests: Feral Pigs.
Australian Government Printing Service, Canberra.
Cluzeau, D., Binet, F., Vertes, F., Simon, J.C., Riviere, J.M., Trehen, P., 1992. Effects of
intensive cattle trampling on soil-plant-earthworms system in two grassland types.
Soil Biol. Biochem. 24, 1661–1665.
Corti, R., Datry, T., 2016. Terrestrial and aquatic invertebrates in the riverbed of an in-
termittent river: parallels and contrasts in community organisation. Freshw. Biol. 61,
1308–1320.
Corti, R., Larned, S.T., Datry, T., 2013. A comparison of pitfall-trap and quadrat methods
for sampling ground-dwelling invertebrates in dry riverbeds. Hydrobiologia 717,
13–26.
DSITIA (Department of Science, Information Technology, Innovation and the Arts), 2013.
Operations Support Method: Addressing Reference Condition Selection Criteria
(AEMA006). Science Delivery, Queensland Government, Brisbane (version 1.0).
Datry, T., Larned, S.T., Tockner, K., 2014. Intermittent rivers: a challenge for freshwater
ecology. Bioscience 64, 229–235.
Dobbie, M.J., Stevens Jr., D.L., 2006. Data requirements for design of the SEAP central
province pilot survey. CSIRO Mathematical and Information Sciences Commercial-in-
Confidence Technical Report 06/181.
Dobbie, M.J., Negus, P., 2013. Addressing statistical and operational challenges in de-
signing large-scale stream condition surveys. Environ. Monit. Assess. 185,
7231–7243.
Eyre, M.D., Luff, M.L., 2002. The use of ground beetles (Coleoptera: carabidae) in con-
servation assessments of exposed riverine sediment habitats in Scotland and Northern
England. J. Insect Conserv. 6, 25–38.
Field, S.A., O’Connor, P.J., Tyre, A.J., Possingham, H.P., 2007. Making monitoring
meaningful. Aust. Ecol. 32, 485–491.
Fleischner, T.L., 1994. Ecological costs of livestock grazing in western North America.
Conserv. Biol. 8, 629–644.
Gentle, M., Pople, A., 2013. Effectiveness of commercial harvesting in controlling feral-
pig populations. Wildlife Res. 40, 459–469.
Green, D.M., Kauffman, J.B., 1995. Succession and livestock grazing in a northeastern
Oregon riparian ecosystem. J. Range Manage. 48, 307–313.
Greenslade, P., 2007. The potential of Collembola to act as indicators of landscape stress
in Australia. Anim. Prod. Sci. 47, 424–434.
Guo, F., Bunn, S.E., Brett, M.T., Kainz, M.J., 2017. Polyunsaturated fatty acids in stream
food webs–high dissimilarity among producers and consumers. Freshw. Biol. 62,
1325–1334. http://dx.doi.org/10.1111/fwb.12956.
Hashimoto, K., Hayashi, F., 2012. Structure and function of the large pronotal horn of the
sand-living anthicid beetle Mecynotarsus tenuipes. Entomol. Sci. 15, 274–279.
Hawkins, C.P., Olson, J.R., Hill, R.A., 2010. The reference condition: predicting bench-
marks for ecological and water-quality assessments. J. North Am. Benthol. Soc. 29,
312–343.
Hoffmann, B.D., Andersen, A.N., 2003. Responses of ants to disturbance in Australia, with
particular reference to functional groups. Aust. Ecol. 28, 444–464.
Kauffman, J.B., Krueger, W.C., 1984. Livestock impacts on riparian ecosystems and
streamside management implicationsl: a review. J. Range Manag. 430–438.
Kauffman, J.B., Krueger, W.C., Vavra, M., 1983a. Impacts of cattle on streambanks in
northeastern Oregon. J. Range Manag. 683–685.
Kauffman, J.B., Krueger, W.C., Vavra, M., 1983b. Effects of late season cattle grazing on
riparian plant communities. J. Range Manag. 685–691.
Kennard, M.J., Pusey, B.J., Olden, J.D., MacKay, S.J., Stein, J.L., Marsh, N., 2010.
Classification of natural flow regimes in Australia to support environmental flow
management. Freshw. Biol. 55, 171–193.
King, J.R., Andersen, A.N., Cutter, A.D., 1998. Ants as bioindicators of habitat dis-
turbance: validation of the functional group model for Australia’s humid tropics.
A.L. Steward et al.
Biodiv. Conserv. 7, 1627–1638.
Kleinwächter, M., Rickfelder, T., 2007. Habitat models for a riparian carabid beetle: their
validity and applicability in the evaluation of river bank management. Biodiv.
Conserv. 16, 3067–3081.
Koenker, R., 2016. Quantreg: Quantile Regression R Package Version 5.26. (Accessed 3/
11/2016). https://CRAN.R-project.org/package=quantreg.
Lalley, J.S., Viles, H.A., Henschel, J.R., Lalley, V., 2006. Lichen-dominated soil crusts as
arthropod habitat in warm deserts. J. Arid Environ. 67, 579–593.
Larned, S.T., Datry, T., Robinson, C.T., 2007. Invertebrate and microbial responses to
inundation in an ephemeral river reach in New Zealand: effects of preceding dry
periods. Aquat. Sci. 69, 554–567.
Larned, S.T., Datry, T., Arscott, D.B., Tockner, K., 2010. Emerging concepts in temporary-
river ecology. Freshw. Biol. 55, 717–738.
Leach, E., Nakamura, A., Turco, F., Burwell, C.J., Catterall, C.P., Kitching, R.L., 2013.
Potential of ants and beetles as indicators of rainforest restoration: characterising
pasture and rainforest remnants as reference habitats. Ecol. Manag. Restor. 14,
202–209.
Leigh, C., Stubbington, R., Sheldon, F., Boulton, A.J., 2013. Hyporheic invertebrates as
bioindicators of ecological health in temporary rivers: a meta-analysis. Ecol. Indic.
32, 62–73.
Lindsay, E.A., Cunningham, S.A., 2009. Livestock grazing exclusion and microhabitat
variation affect invertebrates and litter decomposition rates in woodland remnants.
For. Ecol. Manag. 258, 178–187.
Lowrance, R., Todd, R., Fail, J., Hendrickson, O., Leonard, R., Asmussen, L., 1984.
Riparian forests as nutrient filters in agricultural watersheds. Bioscience 34, 374–377.
Ludwig, J., Tongway, D., 2002. Clearing savannas for use as rangelands in Queensland:
altered landscapes and water-erosion processes. Rang. J. 24, 83–95.
Majer, J.D., 1983. Ants: bio-indicators of minesite rehabilitation, land-use, and land
conservation. Environ. Manag. 7, 375–383.
Marc, P., Canard, A., Ysnel, F., 1999. Spiders (Araneae) useful for pest limitation and
bioindication. Agriculture. Ecosyst. Environ. 74, 229–273.
Martin, T.G., McIntyre, S., 2007. Impacts of livestock grazing and tree clearing on birds of
woodland and riparian habitats. Conserv. Biol. 21, 504–514.
Mazor, R.D., Stein, E.D., Ode, P.R., Schiff, K., 2014. Integrating intermittent streams into
watershed assessments: applicability of an index of biotic integrity. Freshw. Sci. 33,
459–474.
McClain, M.E., Boyer, E.W., Dent, C.L., Gergel, S.E., Grimm, N.B., Groffman, P.M., Hart,
S.C., Harvey, J.W., Johnston, C.A., Mayorga, E., McDowell, W.H., 2003.
Biogeochemical hot spots and hot moments at the interface of terrestrial and aquatic
ecosystems. Ecosystems 6, 301–312.
Meehl, G.A., Stocker, T.F., Collins, W.D., Friedlingstein, P., Gaye, A.T., Gregory, J.M.,
Kitoh, A., Knutti, R., Murphy, J.M., Noda, A., Raper, S.C.B., Watterson, I.G., Weaver,
A.J., Zhao, Z.-C., 2007. Global climate projections. In: Solomon, S., Qin, D., Manning,
M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller, H.L. (Eds.), Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC).
Cambridge University Press, Cambridge/New York, pp. 747–845.
Michaels, K.F., 2007. Using staphylinid and tenebrionid beetles as indicators of sustain-
able landscape management in Australia: a review. Anim. Prod. Sci. 47, 435–449.
Mills, J.R., Turner, E.J., Caltabiano, T., 1989. Land degradation in south-western
Queensland. Project Report Q0 89008. Queensland Department of Primary
Industries, Brisbane.
Mitchell, J., Dorney, W., Mayer, R., McIlroy, J., 2007a. Ecological impacts of feral pig
diggings in north Queensland rainforests. Wildl. Res. 34, 603–608.
Mitchell, J., Dorney, W., Mayer, R., McIlroy, J., 2007b. Spatial and temporal patterns of
feral pig diggings in rainforests of north Queensland. Wildl. Res. 34, 597–602.
Muenz, T.K., Golladay, S.W., Vellidis, G., Smith, L.L., 2006. Stream buffer effectiveness in
an agriculturally influenced area, southwestern Georgia: responses of water quality,
macroinvertebrates, and amphibians. J. Environ. Qual. 35, 1924–1938.
Murphy, G.E.P., Romanuk, T.N., 2014. A meta-analysis of declines in local species rich-
ness from human disturbances. Ecol. Evol. 4, 91–103.
NLWA (National Land and Water Resources Audit), 2002. Australians and Natural
Resource Management 2002. NLWA, Australian Capital Territory Canberra.
Naeth, M.A., Bailey, A.W., Pluth, D.J., Chanasyk, D., Hardin, R.T., 1991. Grazing impacts
on litter and soil organic matter in mixed prairie and fescue grassland ecosystems of
Alberta. J. Range Manage. 44, 7–12.
Naiman, R.J., Décamps, H., 1997. The ecology of interfaces: Riparian zones. Annu. Rev.
Ecol. Syst. 28, 621–658.
Negus, P., Blessing, J., Clifford, S., Steward, A., 2013. Riverine Assessment in
Queensland's Lake Eyre and Bulloo Catchments: Stream and Estuary Assessment
Program 2012. Department of Science, Information Technology, Innovation and the
Arts Queensland Government, Brisbane.
Negus, P., Blessing, J., Clifford, S., Steward, A., 2015. Riverine Assessment in the
Warrego, Paroo, Bulloo and Nebine Catchments. Q–catchments: Technical Report
2012. Department of Science, Information Technology and Innovation Queensland
Government, Brisbane.
Norris, R.H., Linke, S., Prosser, I., Young, W.J., Liston, P., Bauer, N., Sloane, N., Dyer, F.,
Thoms, M., 2007. Very-broad-scale assessment of human impacts on river condition.
Freshw. Biol. 52, 959–976.
Obrist, M.K., Duelli, P., 2010. Rapid biodiversity assessment of arthropods for monitoring
average local species richness and related ecosystem services. Biodiv. Conserv. 19,
547
Ecological Indicators 85 (2018) 537–547
2201–2220.
Orr, H.K., 1960. Soil porosity and bulk density on grazed and protected Kentucky blue-
grass range in the Black Hills. J. Range Manag. 80–86.
Paetzold, A., Yoshimura, C., Tockner, K., 2008. Riparian arthropod responses to flow
regulation and river channelization. J. Appl. Ecol. 45, 894–903.
Palmer, M.A., Reidy Liermann, C.A., Nilsson, C., Flörke, M., Alcamo, J., Lake, P.S., Bond,
N., 2008. Climate change and the world's river basins: anticipating management
options. Front. Ecol. Environ. 6, 81–89.
Peck, S.L., McQuaid, B., Campbell, C.L., 1998. Using ant species (Hymenoptera: for-
micidae) as a biological indicator of agroecosystem condition. Environ. Entomol. 27,
1102–1110.
Pettit, N.E., 2002. Riparian vegetation of a permanent waterhole on cooper creek,
southwest Queensland. R. Soc. Queensland 110, 15–25.
Popolizio, C.A., Goetz, H., Chapman, P.L., 1994. Short-term response of riparian vege-
tation to 4 grazing treatments. J. Range Manag. 47, 48–53.
R Core Team, 2013. R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing, Vienna, Austria(http://www.R-project.org
/Accessed 3 November 2016).
Rainio, J., Niemelä, J., 2003. Ground beetles (Coleoptera: carabidae) as bioindicators.
Biodiv. Conserv. 12, 487–506.
Reynoldson, T.B., Bailey, R., Day, K., Norris, R., 1995. Biological guidelines for freshwater
sediment based on BEnthic Assessment of SedimenT (the BEAST) using a multivariate
approach for predicting biological state. Aust. J. Ecol. 20, 198–219.
Richards, C., Bacon, K.L., 1994. Influence of fine sediment on macroinvertebrate colo-
nization of surface and hyporheic stream substrates. Western North Am. Natur. 54,
106–113.
Robertson, A.I., Rowling, R.W., 2000. Effects of livestock on riparian zone vegetation in
an Australian dryland river. Regul. Rivers: Res. Manag. 16, 527–541.
Robson, B.J., Chester, E.T., Austin, C.M., 2011. Why life history information matters:
drought refuges and macroinvertebrate persistence in non-perennial streams subject
to a drier climate. Mar. Freshw. Res. 62, 801–810.
Rogers, W., 1992. Quantile regression standard errors. Stata Tech. Bull. 9, 16–19.
Sánchez-Montoya, M.M., Moleón, M., Sánchez‐Zapata, J.A., Tockner, K., 2016. Dry riv-
erbeds: corridors for terrestrial vertebrates. Ecosphere 7. http://dx.doi.org/10.1002/
ecs2.1508.
Schälchli, U., 1992. The clogging of coarse gravel river beds by fine sediment.
Hydrobiologia 235, 189–197.
Schepers, J.S., Francis, D.D., 1982. Chemical water quality of runoff from grazing land in
Nebraska: I : influence of grazing livestock. J. Environ. Qual. 11, 351–354.
Schon, N.L., Mackay, A.D., Yeates, G.W., Minor, M.A., 2010. Separating the effects of
defoliation and dairy cow treading pressure on the abundance and diversity of soil
invertebrates in pastures. Appl. Soil Ecol. 46, 209–221.
Schulz, T.T., Leininger, W.C., 1990. Differences in riparian vegetation structure between
grazed areas and exclosures. J. Range Manage. 43, 295–299.
Sheldon, F., Bunn, S.E., Hughes, J.M., Arthington, A.H., Balcombe, S.R., Fellows, C.S.,
2010. Ecological roles and threats to aquatic refugia in arid landscapes: dryland river
waterholes. Mar. Freshw. Res. 61, 885–895.
Sheldon, F., 2005. Incorporating natural variability into the assessment of ecological
health in Australian dryland rivers. Hydrobiologia 552, 45–56.
Steward, A.L., Marshall, J.C., Sheldon, F., Harch, B., Choy, S., Bunn, S.E., Tockner, K.,
2011. Terrestrial invertebrates of dry river beds are not simply subsets of riparian
assemblages. Aquat. Sci. 73, 551–566.
Steward, A.L., von Schiller, D., Tockner, K., Marshall, J.C., Bunn, S.E., 2012. When the
river runs dry: human and ecological values of dry riverbeds. Front. Ecol. Environ.
10, 202–209.
Steward, A.L., Langhans, S., Corti, R., Datry, T., 2017. Chapter 4.4: The biota of inter-
mittent rivers and ephemeral streams – Terrestrial and semi-aquatic invertebrates. In:
Datry, T., Bonada, N., Boulton, A. (Eds.), Intermittent Rivers and Ephemeral Streams:
Ecology and Management. Elsevier.
Steward, A., 2014. Mecynotarsus: a genus of odd-looking, but interesting beetles. News
Bull. Queensland Entomol. Soc. 41, 177–178.
Stoddard, J.L., Larsen, D.P., Hawkins, C.P., Johnson, R.K., Norris, R.H., 2006. Setting
expectations for the ecological condition of streams: the concept of reference con-
dition. Ecol. Appl. 16, 1267–1276.
Tansey, M.R., Brock, T.D., 1972. The upper temperature limit for eukaryotic organisms.
Proc. Natl. Acad. Sci. USA 69, 2426.
Taylor, F.R., Gillman, L.A., Pedretti, J.W., 1989. Impact of cattle on two isolated fish
populations in Pahranagat Valley Nevada. Great Basin Natur. 49, 491–495.
Trimble, S.W., Mendel, A.C., 1995. The cow as a geomorphic agent—a critical review.
Geomorphology 13, 233–253.
Vtorov, I.P., 1993. Feral pig removal: effects on soil microarthropods in a Hawaiian rain
forest. J. Wildl. Manag. 57, 875–880.
Williams, G., 1959. The seasonal and diurnal activity of the fauna sampled by pitfall traps
in different habitats. J. Anim. Ecol. 28, 1–13.
Williams, W.D., 1988. Limnological imbalances: an antipodean viewpoint. Freshw. Biol.
20, 407–420.
Wishart, M.J., 2000. The terrestrial invertebrate fauna of a temporary stream in southern
Africa. African Zool. 35, 193–200.
Wood, P.J., Armitage, P.D., 1997. Biological effects of fine sediment in the lotic en-
vironment. Environ. Manag. 21, 203–217.