Oecologia DOI 10.1007/s00442-014-3189-y
COMMUNITY ECOLOGY - ORIGINAL RESEARCH
Effects of reconstruction of a pre-European vertebrate assemblage on ground-dwelling arachnids in arid Australia Colin J. Silvey · Matthew W. Hayward · Heloise Gibb
Received: 12 January 2014 / Accepted: 10 December 2014 © Springer-Verlag Berlin Heidelberg 2015
Abstract Species loss can result in changes in assemblage structure and ecosystem function through ecological cascades. Australian vertebrate assemblages changed significantly following European colonisation, which resulted in the establishment of invasive vertebrates and the loss of native marsupials, many of which consume invertebrates. Conservation focusses on the removal of invasive carnivores and the reintroduction of regionally extinct species to fenced sites, resulting in what could be considered a reconstruction of pre-European vertebrate assemblages. In semi-arid Australian spinifex mallee ecosystems, we asked: (1) what is the effect of reconstructed pre-European vertebrate assemblages on native arachnid assemblages? and (2) what direct or indirect mechanisms (predation, disturbance and/or competition) could plausibly be responsible for these effects? We compared sites with reconstructed vertebrate assemblages
with paired control sites. Arachnids were sampled using pitfall trapping and direct searching. Hypotheses regarding mechanisms were tested using scat analysis (predation) and by comparing burrow depth (disturbance) and scorpion mass (competition) between control and reconstructed sites. The dominant dune scorpion, Urodacus yaschenkoi, was less abundant and a wolf spider (Lycosa gibsoni species group) more abundant in reconstructed sites. Differences in spider assemblage composition were marginally non-significant. Scat analysis confirmed native vertebrate predation on scorpions and we found no evidence that competition or disturbance affected scorpions. We, thus, suggest that changes in spider assemblages may have resulted from ecological cascades via decreases in dune scorpions. The loss of omnivorous mammals and other changes associated with the invasion of carnivores may, therefore, have had broad-reaching consequences for native arachnid assemblages in Australian ecosystems.
Communicated by Sven Bacher.
Keywords Bilby · Bettong · Mesopredator release · Reintroductions · Ecological cascade
Electronic supplementary material The online version of this article (doi:10.1007/s00442-014-3189-y) contains supplementary material, which is available to authorized users. C. J. Silvey · H. Gibb (*) Department of Zoology, La Trobe University, Melbourne, VIC 3068, Australia e-mail: [email protected] C. J. Silvey Museum Victoria, GPO Box 666, Melbourne VIC 3001, Australia M. W. Hayward Australian Wildlife Conservancy, Scotia Sanctuary, CARE P.O., Wentworth, NSW 2648, Australia M. W. Hayward School of Environment, Natural Resources and Geography, Bangor University, Bangor, Gwynedd LL57 2UW, UK
Introduction Biotic interactions play a critical role in the maintenance of biodiversity and ecosystem function (Tylianakis et al. 2008). Colonisation by alien species or the loss of native species can have cascading effects that reshape interaction networks and alter ecosystem function (Jackson et al. 2001; Brown et al. 2001; Bruno and O’Connor 2005). Such ecological cascades include mesopredator release (Soulé et al. 1988). Mesopredator release occurs when an apex predator is removed from an ecosystem, and is functionally replaced by a smaller (meso) predator, whose numbers are normally suppressed through interactions with that top
13
Oecologia
predator (Soulé et al. 1988; Ripple and Beschta 2007; Estes et al. 2009; Ritchie and Johnson 2009). Outbreaks of mesopredators have the potential to lead to extinction of prey, particularly if their prey has a low population growth rate (Courchamp et al. 1999; Prugh et al. 2009). Since European colonisation just over 220 years ago, Australian ecosystems have changed substantially, due mainly to habitat modification and biotic invasions (Burbidge and McKenzie 1989). Substantial losses of vertebrate species are largely attributed to predation by the introduced European fox (Vulpes vulpes) and the feral cat (Felis catus) (Watts 1969; Short 1998). In the past two centuries, Australia has experienced the highest rate of mammal extinctions in the world (Burbidge and McKenzie 1989; Short and Smith 1994; Johnson 2006; Johnson et al. 2007; McKenzie et al. 2007) and mammals within a weight range of 35–5,500 g, known as the critical weight range (CWR), have been particularly affected (Burbidge and McKenzie 1989; Burbidge et al. 2009). Many of the extinct or threatened species in Australia forage by digging and the loss of digging mammals parallels similar losses in ecosystems worldwide (Davidson et al. 2012). The decline and loss of native digging mammals significantly alters both interactions and ecological functions (Davidson et al. 2012). In Australian ecosystems, CWR digging mammals that are now considered ‘ecologically extinct’ (Estes et al. 1989) are thought to have been important ecosystem engineers prior to European colonisation. For example, their feeding behaviour (digging) facilitates the dispersal and germination of plant seeds and fungal spores (Lamont et al. 1985; Murphy et al. 2005; James and Eldridge 2007; James et al. 2010), improves water penetration and mixing of organic materials (Garkaklis et al. 1998, 2004; Martin 2003; James and Eldridge 2007) and alters plant community composition (James and Eldridge 2007; Travers et al. 2012; Verdon et al. in review). However, a significant gap in our understanding of the impact of digging mammal assemblages on ecosystems concerns their interactions with invertebrates (Gibb 2012). Invertebrate assemblages are strongly affected by vertebrate predators (Schoener and Spiller 1987, 1996; Spiller and Schoener 1990, 1998; Dial and Roughgarden 1995) and by habitat disturbance (Gibb and Hochuli 2002; Pearce and Venier 2006), which is often caused by vertebrates. Although omnivorous, the diet of native CWR digging mammals such as bilbies, Macrotis lagotis and bettongs, Bettongia spp., includes significant numbers of invertebrates (Dawson 1989; Seebeck et al. 1989; Gibson 2001; Robley et al. 2001; Bice and Moseby 2008). In addition, invertebrates may be affected by competition for food from vertebrates (e.g. Barton et al. 2011), which might result in lower invertebrate abundance or smaller body size (Polis and McCormick 1986). Disturbance resulting from vertebrate foraging activities might also alter invertebrate assemblages (Davidson and Lightfoot 2007;
13
Davidson et al. 2012; Gibb 2012). Cascading effects of mammal extinctions on invertebrate assemblages are, thus, likely. Recent concurrent reintroductions of CWR mammals and removal of invasive carnivores provide an excellent opportunity to study the effect of a reconstructed pre-European vertebrate assemblage on invertebrate assemblages. Here, we focus on the effects on arachnids, which occur in high densities, are taxonomically diverse and are likely to be key components of arid ecosystems, playing a significant role in the regulation and stabilisation of prey (Polis and McCormick 1986; Polis and Yamashita 1991). Intraguild predation is extremely common in arachnid assemblages (Polis and McCormick 1986; Polis et al. 1989), so it is likely that disruptions to populations of dominant arachnids, such as scorpions, cascade through arachnid assemblages. We hypothesised that reconstructions of the pre-European vertebrate assemblage would cause ecological cascades in arachnid assemblages. Firstly, we predicted that arachnid assemblages would differ between reintroduction/removal sites (referred to here as ‘reintroductions’) and controls, reflected in shifts in habitat use, or changes in abundance or assemblage structure. We hypothesised that changes in scorpion and spider assemblages associated with native vertebrate reintroductions might result from a number of direct mechanisms, including: predation (hypothesis 1), disturbance (hypothesis 2) and competition (hypothesis 3) (Table 1; Fig. 1). Alternatively, spider assemblages may be affected indirectly, through ecological cascades via scorpions (hypothesis 4). If the effect of native vertebrates (mesopredators I) on scorpions (mesopredator II) was due to predation, we would predict that scorpions would constitute a common component of native vertebrate diets. If it was due to competition, we would expect scorpions to have reduced body condition in reintroduction sites due to competition (e.g. Marshall et al. 2000, but see Polis and McCormick 1986). If it was due to disturbance, we would expect scorpions to dig deeper burrows in reintroductions to avoid disturbance, as shown for other taxa (Šumbera et al. 2004). If effects on spiders were due to predation, disturbance or competition from native vertebrates, we would expect declines in the abundance of common species in reintroductions. If effects were indirect, resulting from ecological cascades via declines in populations of scorpions (mesopredator II), which consume spiders (Shorthouse and Marples 1982), we would expect increases in the abundance of common spiders in reintroductions.
Materials and methods Study area This study was conducted within Scotia Sanctuary, a private conservation area run and maintained by the Australian
Oecologia Table 1 Alternative hypotheses examining how introduced mammalian predator removal and critical weight range (CWR) mammal reintroduction may affect: (a) scorpion abundance directly (hypotheses 1, Response
Prediction
2 and 3) and (b) spider assemblages directly (hypotheses 1, 2 and 3) or indirectly (hypothesis 4) through ecological cascades via scorpions (see also Fig. 1)
Outcome
Hypothesis† Direct effects from vertebrates
(a) Scorpions Abundance Habitat shift Component of scats Burrow depth Body mass and condition (b) Spiders
Control > reintroduction Habitat × treatment interaction Scorpions present Control reintroduction
Common species abundances Control
COMMUNITY ECOLOGY - ORIGINAL RESEARCH
Effects of reconstruction of a pre-European vertebrate assemblage on ground-dwelling arachnids in arid Australia Colin J. Silvey · Matthew W. Hayward · Heloise Gibb
Received: 12 January 2014 / Accepted: 10 December 2014 © Springer-Verlag Berlin Heidelberg 2015
Abstract Species loss can result in changes in assemblage structure and ecosystem function through ecological cascades. Australian vertebrate assemblages changed significantly following European colonisation, which resulted in the establishment of invasive vertebrates and the loss of native marsupials, many of which consume invertebrates. Conservation focusses on the removal of invasive carnivores and the reintroduction of regionally extinct species to fenced sites, resulting in what could be considered a reconstruction of pre-European vertebrate assemblages. In semi-arid Australian spinifex mallee ecosystems, we asked: (1) what is the effect of reconstructed pre-European vertebrate assemblages on native arachnid assemblages? and (2) what direct or indirect mechanisms (predation, disturbance and/or competition) could plausibly be responsible for these effects? We compared sites with reconstructed vertebrate assemblages
with paired control sites. Arachnids were sampled using pitfall trapping and direct searching. Hypotheses regarding mechanisms were tested using scat analysis (predation) and by comparing burrow depth (disturbance) and scorpion mass (competition) between control and reconstructed sites. The dominant dune scorpion, Urodacus yaschenkoi, was less abundant and a wolf spider (Lycosa gibsoni species group) more abundant in reconstructed sites. Differences in spider assemblage composition were marginally non-significant. Scat analysis confirmed native vertebrate predation on scorpions and we found no evidence that competition or disturbance affected scorpions. We, thus, suggest that changes in spider assemblages may have resulted from ecological cascades via decreases in dune scorpions. The loss of omnivorous mammals and other changes associated with the invasion of carnivores may, therefore, have had broad-reaching consequences for native arachnid assemblages in Australian ecosystems.
Communicated by Sven Bacher.
Keywords Bilby · Bettong · Mesopredator release · Reintroductions · Ecological cascade
Electronic supplementary material The online version of this article (doi:10.1007/s00442-014-3189-y) contains supplementary material, which is available to authorized users. C. J. Silvey · H. Gibb (*) Department of Zoology, La Trobe University, Melbourne, VIC 3068, Australia e-mail: [email protected] C. J. Silvey Museum Victoria, GPO Box 666, Melbourne VIC 3001, Australia M. W. Hayward Australian Wildlife Conservancy, Scotia Sanctuary, CARE P.O., Wentworth, NSW 2648, Australia M. W. Hayward School of Environment, Natural Resources and Geography, Bangor University, Bangor, Gwynedd LL57 2UW, UK
Introduction Biotic interactions play a critical role in the maintenance of biodiversity and ecosystem function (Tylianakis et al. 2008). Colonisation by alien species or the loss of native species can have cascading effects that reshape interaction networks and alter ecosystem function (Jackson et al. 2001; Brown et al. 2001; Bruno and O’Connor 2005). Such ecological cascades include mesopredator release (Soulé et al. 1988). Mesopredator release occurs when an apex predator is removed from an ecosystem, and is functionally replaced by a smaller (meso) predator, whose numbers are normally suppressed through interactions with that top
13
Oecologia
predator (Soulé et al. 1988; Ripple and Beschta 2007; Estes et al. 2009; Ritchie and Johnson 2009). Outbreaks of mesopredators have the potential to lead to extinction of prey, particularly if their prey has a low population growth rate (Courchamp et al. 1999; Prugh et al. 2009). Since European colonisation just over 220 years ago, Australian ecosystems have changed substantially, due mainly to habitat modification and biotic invasions (Burbidge and McKenzie 1989). Substantial losses of vertebrate species are largely attributed to predation by the introduced European fox (Vulpes vulpes) and the feral cat (Felis catus) (Watts 1969; Short 1998). In the past two centuries, Australia has experienced the highest rate of mammal extinctions in the world (Burbidge and McKenzie 1989; Short and Smith 1994; Johnson 2006; Johnson et al. 2007; McKenzie et al. 2007) and mammals within a weight range of 35–5,500 g, known as the critical weight range (CWR), have been particularly affected (Burbidge and McKenzie 1989; Burbidge et al. 2009). Many of the extinct or threatened species in Australia forage by digging and the loss of digging mammals parallels similar losses in ecosystems worldwide (Davidson et al. 2012). The decline and loss of native digging mammals significantly alters both interactions and ecological functions (Davidson et al. 2012). In Australian ecosystems, CWR digging mammals that are now considered ‘ecologically extinct’ (Estes et al. 1989) are thought to have been important ecosystem engineers prior to European colonisation. For example, their feeding behaviour (digging) facilitates the dispersal and germination of plant seeds and fungal spores (Lamont et al. 1985; Murphy et al. 2005; James and Eldridge 2007; James et al. 2010), improves water penetration and mixing of organic materials (Garkaklis et al. 1998, 2004; Martin 2003; James and Eldridge 2007) and alters plant community composition (James and Eldridge 2007; Travers et al. 2012; Verdon et al. in review). However, a significant gap in our understanding of the impact of digging mammal assemblages on ecosystems concerns their interactions with invertebrates (Gibb 2012). Invertebrate assemblages are strongly affected by vertebrate predators (Schoener and Spiller 1987, 1996; Spiller and Schoener 1990, 1998; Dial and Roughgarden 1995) and by habitat disturbance (Gibb and Hochuli 2002; Pearce and Venier 2006), which is often caused by vertebrates. Although omnivorous, the diet of native CWR digging mammals such as bilbies, Macrotis lagotis and bettongs, Bettongia spp., includes significant numbers of invertebrates (Dawson 1989; Seebeck et al. 1989; Gibson 2001; Robley et al. 2001; Bice and Moseby 2008). In addition, invertebrates may be affected by competition for food from vertebrates (e.g. Barton et al. 2011), which might result in lower invertebrate abundance or smaller body size (Polis and McCormick 1986). Disturbance resulting from vertebrate foraging activities might also alter invertebrate assemblages (Davidson and Lightfoot 2007;
13
Davidson et al. 2012; Gibb 2012). Cascading effects of mammal extinctions on invertebrate assemblages are, thus, likely. Recent concurrent reintroductions of CWR mammals and removal of invasive carnivores provide an excellent opportunity to study the effect of a reconstructed pre-European vertebrate assemblage on invertebrate assemblages. Here, we focus on the effects on arachnids, which occur in high densities, are taxonomically diverse and are likely to be key components of arid ecosystems, playing a significant role in the regulation and stabilisation of prey (Polis and McCormick 1986; Polis and Yamashita 1991). Intraguild predation is extremely common in arachnid assemblages (Polis and McCormick 1986; Polis et al. 1989), so it is likely that disruptions to populations of dominant arachnids, such as scorpions, cascade through arachnid assemblages. We hypothesised that reconstructions of the pre-European vertebrate assemblage would cause ecological cascades in arachnid assemblages. Firstly, we predicted that arachnid assemblages would differ between reintroduction/removal sites (referred to here as ‘reintroductions’) and controls, reflected in shifts in habitat use, or changes in abundance or assemblage structure. We hypothesised that changes in scorpion and spider assemblages associated with native vertebrate reintroductions might result from a number of direct mechanisms, including: predation (hypothesis 1), disturbance (hypothesis 2) and competition (hypothesis 3) (Table 1; Fig. 1). Alternatively, spider assemblages may be affected indirectly, through ecological cascades via scorpions (hypothesis 4). If the effect of native vertebrates (mesopredators I) on scorpions (mesopredator II) was due to predation, we would predict that scorpions would constitute a common component of native vertebrate diets. If it was due to competition, we would expect scorpions to have reduced body condition in reintroduction sites due to competition (e.g. Marshall et al. 2000, but see Polis and McCormick 1986). If it was due to disturbance, we would expect scorpions to dig deeper burrows in reintroductions to avoid disturbance, as shown for other taxa (Šumbera et al. 2004). If effects on spiders were due to predation, disturbance or competition from native vertebrates, we would expect declines in the abundance of common species in reintroductions. If effects were indirect, resulting from ecological cascades via declines in populations of scorpions (mesopredator II), which consume spiders (Shorthouse and Marples 1982), we would expect increases in the abundance of common spiders in reintroductions.
Materials and methods Study area This study was conducted within Scotia Sanctuary, a private conservation area run and maintained by the Australian
Oecologia Table 1 Alternative hypotheses examining how introduced mammalian predator removal and critical weight range (CWR) mammal reintroduction may affect: (a) scorpion abundance directly (hypotheses 1, Response
Prediction
2 and 3) and (b) spider assemblages directly (hypotheses 1, 2 and 3) or indirectly (hypothesis 4) through ecological cascades via scorpions (see also Fig. 1)
Outcome
Hypothesis† Direct effects from vertebrates
(a) Scorpions Abundance Habitat shift Component of scats Burrow depth Body mass and condition (b) Spiders
Control > reintroduction Habitat × treatment interaction Scorpions present Control reintroduction
Common species abundances Control
