Response of frogs to environmental factors at multiple scales in the Lachlan Catchment of New South Wales

Habitat loss and change are major contributors to the global decline of frog populations. In riverine and floodplain environments water diversion and extraction (regulation) can change the character of aquatic habitat and reduce the availability of freshwater. In semi-arid areas these changes to the availability of water can have profound impacts on the persistence of frog communities. This thesis is focused on the frog communities in riverine and floodplain habitats, excluding rain-fed wetlands, of the semi-arid Lachlan Catchment of New South Wales, Australia. It examines the relationship between frog communities and freshwater floodplain and riverine habitat at multiple spatial and temporal scales. Examining these relationships across more than one scale allows for interactions to be identified that are often not all considered when studies are limited to one scale. This can then provide greater insight for the management of floodplain and riverine habitat in semi-arid regions for frogs through the increased understanding of how they interact with factors at multiple scales.
Surveys to examine frog communities in the Lachlan were undertaken across two field seasons at two spatial scales. A broad scale survey (49 sites) was conducted across the mid and lower Lachlan Catchment over four survey periods in September 2012, November 2013, February 2013 and April 2013. Calling activity in response to climate was collected from automated call recorders installed at ten sites, for periods of one and a half to five months between 2010 and 2013. Microhabitat use was investigated at six sites, in September and October 2013.
Limnodynastes fletcheri, Crinia parinsignifera, Litoria peronii, Litoria latopalmata and Litoria rubella were detected throughout the survey area in the mid and lower Lachlan. Habitat occupancy models were developed using the program Presence, considering detection probability. Water was a dominant driver of occupancy, with Lim. fletcheri and Lit. peronii having positive relationships with increasing open water in the top models (within 2 AIC). Litoria latopalmata also occurred at sites with a longer hydroperiod and Lit. rubella had a higher probability of occurring at streams, often perennial due to regulation, in the mid Lachlan. Frogs within the region were also primarily detected when sites were inundated. Upon site drying, sites were surveyed on one more occasion and not re-surveyed unless inundated again. Region, the mid or lower geographic region of the Lachlan Catchment, was present in the top models (within 2 AIC) for Crinia parinsignifera, Lit. latopalmata and Lit. rubella. Occupancy by C. parinsignifera, Lit. peronii, Lit. latopalmata and Lit. rubella all had a negative relationship with vegetation categories.
The probability of calling for four common species, L. tasmaniensis, C. parinsignifera, Limnodynastes interioris and Lit. peronii was estimated from automated call recording data collected between 2010 and 2013 at 10 sites. Rainfall in the last seven days, temperature and decreasing moon illumination featured in top models for L. tasmaniensis and C. parinsignifera, while Lim. interioris calling probability was strongly influenced by rainfall in the last seven days and to a lesser extent decreasing moon illumination. Litoria peronii was strongly driven by increasing temperature with probability of calling not increasing above 0.5 until temperatures reached around 20°C.
The availability of suitable microhabitats such as vegetation, cover/coarse woody debris, soil cracks and holes may support the persistence of frog populations in semi-arid areas, particularly during periods when wetlands are dry. The diurnal, terrestrial microhabitat use by L. tasmaniensis was investigated at six sites between September and October 2013. Frogs were tracked using fluorescent tracking powder and, based on availability, were found to use soil based microhabitat significantly more often than vegetation and coarse woody debris. Soil-based microhabitats (soil cracks/ burrows) were chosen at 64 of 92 locations. Twelve, 36mm and 25mm agar frog models of each size were also placed in open, soil-based, cover and in vegetated microhabitat types in September and October 2013, to assess evaporative water loss conditions over time. The rate of evaporative water loss of agar frog models among microhabitat types was significantly different, with soil cracks having the lowest mean loss.
The four research chapters of this thesis, demonstrate that frog species in the Lachlan Catchment of NSW interacted with environmental factors across a range of physical and temporal scales. This study focused on species that are commonly associated with riverine and wetland habitats, and did not consider species that utilise rain-fed water bodies. Each individual frog species had their own specific relationship with the environmental factors studied. However, in the semi-arid landscape, the availability of water was often the key driver of occupancy, calling and microhabitat use. This thesis
xvi
highlights which of these variables are most important to help inform and improve management of wetland and riverine habitat that these frog species rely on.