Introduction of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

Biodiversity in Queensland, Australia is being lost at a rapid rate which is putting significant pressure on species, habitats and ecosystems (Australia Wildlife Conservancy 2017). It is important to Understand only a small part of the biodiversity puzzle that is the Mulga (Acacia aneura) Bioregion in Bowra, south west Queensland will aid in denoting reasoning for the preservation of landscapes with visible bat functioning from being cleared. The Bowra Wildlife Sanctuary (‘Bowra’) is a refuge for native flora and fauna (Australia Wildlife Conservancy 2017). Prior to Bowra becoming a protected area, the study site was used for five generations for cattle grazing and thus subject to a mosaic of different ages of vegetation re-growth caused by the prior land clearing.

 The aged large old trees dispersed across paddock in landscapes and they are widely appreciated for their aesthetic appeal. Over the years they are also being recognised for their economic benefits, such as lowering the risks of dry land salinity, reducing erosion and providing shade and shelter for livestock (Bennett & Lumsden 2003). Insectivorous bats are found to commonly use paddock trees. Bats are valuable mammals that regularly go unnoticed due to their small size, their silence (to human ears), and the fact that they are hidden during daylight and only feed in the evenings. Bats species consume a range of invertebrates, predominantly moths, beetles, mosquitoes and bugs. A single bat can consume up to half of its body weight in insects each night. It has been recorded that some bat species have been seen to catching an average of 600 mosquitoes within an hour (Bennett & Lumsden 2003).

The role that bats play in reducing insects is likely to be particularly important around sparsely scattered trees, because they are few insectivorous birds that occur in these situations. We assessed environmental factors at two levels, first at the landscape scale and second at a local scale where information was obtained regarding the physical environment at each of the study sites chosen. Initially we predicted that older and density rich landscapes will result in a greater variety of bat species richness when in comparison to open spaces with minimal vegetation cover.

 1. Methods and Study Area and Site Selectionof the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

The study area was situated at Bowra Wildlife Sanctuary located northwest of Cunnamulla in Queensland, Australia (AWC, 2017).  Currently Bowra is a conservation site owned by the Australian Wildlife Conservancy (‘AWC’) a hotspot for inland Australian species. However, formerly the site was a cattle station for five generations (Department of Environment and Energy, 2018). During that period of time, the region experienced clearing and thinning for pastoral fodder (cattle grazing) which has caused a mosaic of regrowth of different ages. Currently this site is home to woodlands, shrublands, riparian zones, grasslands, and vast regions of mulga. The study area has an elevation range of 190m to 240m above sea level. During the month of March 2018, the site has an average maximum temperature of 32.7^oC and the average minimum temperature of 19.3^oC, with a mean number of rain days being 3.8 (Farmonline Weather 2018).

 A total of five monitoring sites were used in the study, which were selected based on three varying landscapes of Mulga classes: regrowth mulga (‘RM’) denoting recently cleared (<15 years cleared), intermediate mulga (‘IM’) (>15 years) and old/remnant sites (‘OM’) (>30 years).  The study sites were chosen using aerial photography and QLD Herbarium Regional Ecosystem vegetation mapping and only areas with vegetation patches greater than 150m were selected.

 The Sites were selected to be similar with regards to geographical location as this alters the types of bat species present, with the main difference across sites. This is being the management history which has effects on vegetation, type, composition and cover. It will demonstrate whether management/clearing history has an impact on bat species richness.

 2 Acoustic Sensor Deployment and Calibration of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 One SM3BAT (Wildlife Acoustics 2016) ultrasonic acoustic recorder was deployed at each site. Sensors were evenly distributed along all sites; one sensor was secured on a tree approximately 1.5 m above ground at each site. In particular, sensors were placed in forks and away from branches to avoid any disturbances. Microphones were attached to 2m leads and secured to branches away from sensor to avoid echo/reflection of bat calls of recording device. The selected trees in which the sensors were deployed were positioned as close to the centre of the vegetation patch as possible to reduce edge effects. Recording occurred from 18^th March 2018 until 24^th March 2018. The sensors were set to record data of bat species consecutively between the hours of 6pm and 9pm (hours after sunset) where the probability of bat activity is at its peak (Thomas & West, 1989). In addition, a trigger recording was set to provide for the range of species that fly outside this bracket gap (Szewack, 2004). This setting is of particular importance as it minimised the chances of reaching storage capacity and time spent thereafter sifting through large data sets with only two researchers. All recorders were set to their default configuration and on zero-crossing recording was selected to maximise recording time. Files were then transferred onto an external hard-drive and manually analysed [1].

 Overall, there were 7 nights of recordings at each of the 5 sites, but only those collected on 3 nights (18^th of March 2018 to the 20^th of March 2018) were utilised for the purposes of this study.

 The recorded files obtained were analysed using the program Kaleidoscope (Wildlife acoustics 2016) and species identification were undertaken using two keys (Reinhold et at2001; Pennay et al2004). Furthermore, the AWC’s survey listing bat species within Bowra was used to seldom identify species known to be located within the region (Table 1). This ensured that wrongful species identification is reduced based on previous geographical bat analysis data. Consequently, reference calls were collected from all species known from the region. Each call was examined, and parameters were extracted from the search phase pulses, however pulse type were also recoded such, as search phase or attack phase. All reference calls could be identified for some species ( Saccolaimus Flaviventris, Chalinobolus  picatus, Nyctophilus geoffreyi and Scotorepens greyii), whereas identification rates for species with overlapping characteristics were pooled ( Chalinobolus gouldii, Mormopterus sp “3” & “4” and Scotorepens balstoni grouped as “CMB” and Vespadelus baverstocki, Vespadelus troughtoni and Vespadelus Vultrunus grouped as Vespadelus). A minimum number of four good quality pulses were required from a call sequence (i.e. a pass) for an identification to be attempted [2].

Table 1. Table 1 Bowra Wildlife Sanctuary Bat Species List (November 2014) Retrieved from http://www.australianwildlife.org/media/180976/Bowra-Species-List-November-2014-Mammals.pdf

2.1 Data analysis of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

The data summaries were generated for total bat activity (i.e the number of times a bat of any species made a pass (Law et al. 1998) and activity type (i.e. the number of times a bat of any species was identified in a search or attack phase). The Boxplots, histograms and pie charts were also created to visually assess the relationship between bat activity and activity type against the landscape, date and site. Additionally, the use of linear models and linear mixed-effects models are to be used to quantify the relationship between regrowth stage, bat activity and vegetation. All analysis was undertaken in R statistical software (R Core Team, 2018).

 3. Results and Summary of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

The Bats were detected at every study site, and of the 721 files recorded 514 individual bats were successfully identified of 6 species: Vespadelus,CMB, S. greyii, N. geoffroyi and S. flaviventris). However, despite our initial assumptions of bat species richness being greater in older sites in comparison to that of younger sites, our findings have shown that bat species richness was consistent across the three tested sites.

 Table 2. Overall number of files analysed

 Table 3 Study site vegetation data

Table 4 Summary of data result analysis

 3.1 Results from each site of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 From the study it was evident that some species groups were formed as an alternative to identifying individual species (Table 5; Fig. 2-6). This was done to avoid the misidentification of bat species and ultimately mislead our findings. For example, bat species grouped under CMB were grouped together because it is well known these species have similar calls. These species are almost impossible to distinguish from each other using acoustics alone and they would need to be captured on site to be confident in its identification. By grouping similar bat calls together, it avoids potentially misidentifying one of these bats as another species. It also provides the next best alternative other than individual species identification, because it is as close as possible to the correct specie. Other studies have grouped by Family or Taxon even though the calls are very distinguishable from one another, resulting in very general findings. As opposed to this study, where the researches in this study were confident in identifying C. picatus as it has a unique call which alternates in frequency (Pennay et al2004). Rather than grouping C. picatus with other species of the Vespertiliondae family, it was left as individual specie [3].

 Using bats calls to assess species diversity and activity is an effective way to analyse large data sets whilst working remotely (away from the test site). However, it also means that it is difficult maintain precision and accuracy throughout data collection, when many factors impact a bat call and they are easily confused with other species.

 Table 5 Displays the individual species comprising of the ‘pooled’ species groups and explains their overlapping call characteristics

3.2 Statistical Analysis of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 Vegetation data (Table 2) and bat data from (Table 3) were used to undertake linear regression tests for three sets of response variables: species richness; average passes per night; and attack calls.

 The regressions conducted denote low R-squared values for each test suggesting that there is a very low likelihood that the vegetation characteristics can cause any response in species richness. The high P-values (>0.05) confirms that changes in the tree height, tree cover, DBH and shrub height are not associated with changes species richness (Table 6). Similarly, the low R-squared values and corresponding high P-values (>0.05) suggest that there is no linear relationship between vegetation characteristics and average passes per night (Table 7). In contrast, R-square values between vegetation characteristics and average passes per night (Table 8) show strong linear relationships to exist between some vegetation characteristics (namely tree height and tree cover) and the number of attack calls emitted. Contritely, these regression tests also confirm that DBH and shrub height have little effect on number of attack calls. This is shown in the low R-squared values (0.01426 and 0.3313 respectively) and the corresponding high p-values (0.9483 and 0.3099) (Table 8).

 A strong linear relationship is present between tree height and number of attack calls emitted (Fig. 5). This was evident by the high R-squared value generated (0.9165). This means that change in tree height accounts for approximately 92% of the variation in the attack calls (Table 8). The low p-value of 0.0105 means that the null hypothesis being that change in tree height will show no response in attack calls can be rejected (Table 8). In addition, here the line of best fit clearly depicts that as tree height increases so does attack calls. Similarly to tree height a strong linear relationship is also present between tree cover and number of attack calls emitted (Fig. 5). This regression too produced a high R-squared value (0.7837). This means that change in tree cover accounts for approximately 78% of the variation in the response variable (attack calls). The low (but significant) p-value (0.04584) means that it is safe to reject the null hypothesis that tree cover will show no response in attack calls. Additionally, from the line of best fit it is evident that as tree cover increases so do attack calls.

 Table 6Regressions showing the effect vegetation characteristics have against species richness

Table 7Regressions showing effect vegetation characteristics has against average passes per night

Table 8Regressions showing effect vegetation characteristics has against attack calls

Figure 6 Linear regression denoting the relationship between tree cover and number of attack calls emitted

3.3 Bat Activity of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

It was clearly evident that bat activity (attacks) increased within our data set the older the land site was. This also correlates with the findings that at older sites there are functioning’s that a bat can carry out thus heightening the level of activity in comparison to recently cleared sites with little vegetation and biodiversity.

4. Discussion of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 Overall, this study has illustrated that bat species richness across the three varied sites did not differ (Refer to figure 8). However, across the various landscape assessed in this study, bat activity (nature of call) was highest at the OM site, medium in the intermediate sites and the lowest in the regrowth sites. Although the vegetation analysis was not the predominant focus of this study, the data gathered are indicative of activity levels per site rather than the particular activity around specific vegetation, the high activity of insectivorous bats found in old regrowth suggest that richer vegetation landscapes produce higher biological benefits. These key findings, discussed, have implications for the conservation of beneficial biodiversity through strategic vegetation restoration and future land clearing designs in vegetation dense landscapes.

 4.1 Regrowth Mulga Sites of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 The two study sites RMA and RMB represent the newly cleared (>15 years) landscapes found within the study region. Within the RM sites a high presence of the pooled group of Vespadelus was found. A large presence of Vespadelus was found at both sites A and B in comparison to their presence in either the intermediate or old Mulga sites. Reasoning for such occurrences is due to the characteristics of Vespadelus being medium fast aerial feeders whom forage below, alongside, and inside canopy and sub-canopy cover (Kutt 1995: Dwyer 1965). Given the high shrub height and cover recorded at RMA (3.1m and 33% respectively) this has allowed Vespadelus species the opportunity to forage in its ideal habitat. Vespadelus species are ones which thrive in open spaces rather than highly dense and cluttered spaces. This also explains this species absence in the OM site as there is no understory for it to forage in its ideal habitat. Given OM densely covered undergrowth (shrub cover = 31.51%) and dense tree canopy (tree cover = 61.8%), this site is thought to be too dense for species belonging to the Vespadelus group and may explain the absence of Vespadelus [4].

 Additionally, a high presence of CMB group was present in all sites but more so in RMA and RMB. Within CMB, Morm sp “3” and “4” are known to be fast aerial movers with low manoeuvrability (Kutt, 1995). Morm sp “4” are often the more common species at a site and like to share roosts with S. balstoni (Susan’s book, pg 497). Similarly to Morm sp “4”, Morm sp “3” fly over canopy, over water-holes or along the borders of tree-lined creeks, they generally don’t fly through dense forest because of limited manoeuvrability (Susan’s book, pg 494). Thus, it can be assumed that these species enjoy open spaces similar to the ones of RM. Species such as Morm sp 3 will find insects on the ground or over the trunk of a tree and scurry towards it on its hindfeet (Susan’s book, 494). In a study conducted by Kutt (1995), identified that the occurrence of Morm sp in thinned forests means that these species enjoy accessible, open spaces where they can fly under the canopy, just as they would fly above dense canopies (as they are also known to do). This versatility denoted by this species ability to adapt to different age classes of Mulga goes forth to explain the large presence of CMB percentages across all classes and sites [5].

 Furthermore, a presence of Nyctophilus geoffroyi at A (3%) was found in RMA. N. geoffroyi are known to roost in dead trees, under exfoliating barks, or in hollows (Susan’s mammal book, pg 520). They also move roosts every fews days. N. Geoffroyi was ID’d the most at RMA (5 times), this may be due to land clearing as RMA has more fallen and dead trees present which add to the complexity of its vegetation profile which in turn allows for more suitable roost sites for N. Geoffroyi.It is important to note that Geoffroyi was consistently found at IMA and OM (although less times) and never at RMB or IMB. This is evidence that there may be a high chance of observation bias for this particular species. Meaning, the researcher that identified calls from RMA and IMA was more inclined to ID N.geoffroyi than the researcher who ID’d calls from RMB and IMB [6].

 4.2 Intermediate Mulga Sites of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 In terms of species richness, as previously stated both IMA and IMB results were in line with both regrowth sites and the old site. However, previous studies have shown the opposite with intermediate regrowth sites sharing similar characteristic as old sites in comparison to older sites rather than with regrowth sites and should exhibit changes in species richness or presence between both IM and OM sites in comparison to RM sites.

In IMA, CMB accounted for 71% percent whereas in IMB it accounts for 47%. S. Balstoni is likely causing increases in CMB here as IMA is close to a creek (100m away) and this specie enjoys habitats that are close to waterways (susan’s book). Given that IMA has relatively sparse tree cover (32.2%) and shrub cover (12.2%)  the vegetation profile at IMA seems to have satisfied S. Balstoni’s preference for open woodlands and shrublands and ultimately caused an increase in presence of CMB as opposed to IMB (susan’s book).

 4.3 Old Mulga Sites of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 Larger evidence of CMB species are found in this study region and especially S. greyii in comparison with other sites, this is because denser vegetation and canopy suit these species groups.  Within CMB, S. Gouldii is known to have strong correlations with tree canopy cover and local roost potential (Milne, 2005). Within CMB, as explained above details that  Morm sp “4” are fast, adaptable aerial feeders and like to fly above dense canopy. This is another reason for a large percentage of CMB species found at the Old Mulga site. S. Greyii and S. Flaventris are known to exhibit the same correlations between local roost potential and tree canopy cover as Gouldii.  With tree height, cover and DBH being highest at OM (12m, 61.8% and 19.45cm respectively) it is likely that there the dense tree coverage and potential to roost in larger trees have attracted more S. Greyii and CMB to the site [7].

 4.4 Chalinolobus Picatus of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

 C. Picatus is found everywhere and evenly across age classes (10% at RM, IM and OM). C. Picatus are agile, fast bats whom forage along woodland canopies, taking insects’ mid-flight (susan’s book, pg 540). This species roosts in hollow trees, with an apparent preference for dead trees (susan’s book, pg 540). This preference of dead trees is explanatory to the findings of this species in the RM studies sites. Furthermore, it is known that C. Picatus will travel long distances to forage and drink where small pools of water is present (susan p540). This trait within C. Picatus therefore also is explanatory for its occurrence in IM site as the study site was located in close proximity to a river. Finally, this species is known to enjoy roosting in mulga - in fact it’s been recorded to travel 17km every night to return to the same mulga roost in Bourke, NSW. Their greatest relative abundance in mixed woodlands and mulga - see Australian Bats, pg 122 (book by Sue Churchill, 2009, available on line from QUT). They are known to fly close to vegetation and glean the top of the canopy, or among the foliage and will swoop down to 2-4m above the ground (likes all types of vegetation). C. Picatus predominantly eats moths, as found in an analysis. Species decline because of habitat loss in the eastern parts of the species range (which is Bowra) loss of roost and disturbance.

 Anatomy of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

The lower limb consists primarily of the shin, the fibia being undeveloped and amalgamate to the shin. The limb is turned through 180°, thus once walking knees purpose ventrally. The complete limb is capable of a large angle rotation, permitting an entire 360° flip once hanging. The toes of this limb have claws that are extraordinarily robust and laterally compressed. A connective tissue that runs through rubbery rings connected to the phalange enable associate automatic lockup system. Its bone structure is analogous to several different kooky, with minor variations that outline the species. Its os is just tiny, with a os breadth of solely seven millimetre. The cavity swellings aren't pronounced and there's no median crest on the brain case. the full os length around is eleven.7 mm. very little varicolored kooky have seven cervical vertebrae, eleven body part vertebrae, four body part vertebrae and are thought to own three caudal vertebrae that form up the little tail structure. The pelvis bones (ilium, ischial bone and pubis) are powerfully amalgamate, additional thus than in different mammals [6].

Habitat of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

It is best-known to vary from North Western and South Western New South Wales but they're solely in an exceedingly few massive remnant of environment that stay during this space. Some specific places the small varicolored bat is found include; Willandra lakes government agency, Idalia parkland QLD and Sturt parkland government agency. It usually roosts in tree hollows of the assorted bushland trees of government agency and QLD like semi-arid tall shrublands and vascular plant forests, however, are usually found in eucalyptus and tree open woodlands.

 4.5 Inland Forest Bat of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

The inland Forest Bat is one amongst variety of tiny (3 to seven grams) myrmecophagous wacky within the genus Vespadelus. it's usually sandy-brown on top of, with the underparts being paler (cream to pale brown). Identification is troublesome, with overlap in size and fur colouration with some species occurring within the same space, significantly Southern Forest Bat V. regulus and tiny Forest Bat V. vulturnus.

Habitat & Ecology of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland

Roosts in tree hollows and abandoned buildings. famous to roost in terribly tiny hollows in inferior trees solely some metres high. The environment necessities of this species are poorly famous however it's been recorded from a spread of earth formations, as well as eucalyptus, Mulga and stream Red Gum. Most records are from drier earth habitats with bank areas peopled by the miscroscopic Forest Bat. However, different habitats could also be used for forage and/or drinking. Colony size ranges from some people to quite sixty. Females congregate to lift young in November and Dec, with young carried for the primary week following birth. Young are freelance by January. These wacky fly quickly and canopy an intensive forage space and are plausible to kill flying insects

6 Limitations and  Conclusion of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queenslan

 In this study, the focus was to identify that what effects land can have on the assembling of bats in particular sites. There are certainly few environmental factors, which play their part in shaping the richness of bat species. The assumption for the study was made that bat species richness will be on great level especially in older as well as density rich landscapes, whereas open spaces will result in minimal vegetation cover for bat species. The study used overall five monitoring sites, where results were obtained to see whether the assumption made by the study is correct or not. The total bat activities along with their activity types were done. The results of the study showed that Bats presence was evident on each selected for the study. The six species of bats were also identified with the help of this study. The assumption made earlier in the study proved incorrect regarding bat species richness on older sites would be greater as compared to younger sites. On all sites, the richness of bat species was consistent and there was no significant difference between them.

 In terms of species richness, as previously stated both IMA and IMB results were in line with both regrowth sites and the old site. However, previous studies have shown the opposite with intermediate regrowth sites sharing similar characteristic as old sites in comparison to older sites rather than with regrowth sites and should exhibit changes in species richness or presence between both IM and OM sites in comparison to RM sites.The inland Forest Bat is one amongst variety of tiny (3 to seven grams) myrmecophagous wacky within the genus Vespadelus. It’s usually sandy-brown on top of, with the underparts being paler (cream to pale brown). Roosts in tree hollows and abandoned buildings. Famous to roost in terribly tiny hollows in inferior trees solely some metres high. The environment necessities of this species are poorly famous however it's been recorded from a spread of earth formations, as well as eucalyptus, Mulga and stream Red Gum

Overall, this study has illustrated that bat species richness across the three varied sites did not differ. However, across the various landscape assessed in this study, bat activity (nature of call) was highest at the OM site, medium in the intermediate sites and the lowest in the regrowth sites. Although the vegetation analysis was not the predominant focus of this study, the data gathered are indicative of activity levels per site rather than the particular activity around specific vegetation, the high activity of insectivorous bats found in old regrowth suggest that richer vegetation landscapes produce higher biological benefits.

 7. References of the effects of bat assemblages in different age classes with Bowra Wildlife Conservancy Queensland