Matthew Stanton | Western Sydney University (original) (raw)

Papers by Matthew Stanton

Species distribution models have great potential to efficiently guide management for threatened s... more Species distribution models have great potential to efficiently guide management for threatened species, especially for those that are rare or cryptic. We used MaxEnt to develop a regional-scale model for the koala Phascolarctos cinereus at a resolution (250 m) that could be used to guide management. To ensure the model was fit for purpose, we placed emphasis on validating the model using independently-collected field data. We reduced substantial spatial clustering of records in coastal urban areas using a 2-km spatial filter and by modeling separately two subregions separated by the 500-m elevational contour. A bias file was prepared that accounted for variable survey effort. Frequency of wildfire, soil type, floristics and elevation had the highest relative contribution to the model, while a number of other variables made minor contributions. The model was effective in discriminating different habitat suitability classes when compared with koala records not used in modeling. We validated the MaxEnt model at 65 ground-truth sites using independent data on koala occupancy (acoustic sampling) and habitat quality (browse tree availability). Koala bellows (n = 276) were analyzed in an occupancy modeling framework, while site habitat quality was indexed based on browse trees. Field validation demonstrated a linear increase in koala occupancy with higher modeled habitat suitability at ground-truth sites. Similarly, a site habitat quality index at ground-truth sites was correlated positively with modeled habitat suitability. The MaxEnt model provided a better fit to estimated koala occupancy than the site-based habitat quality index, probably because many variables were considered simultaneously by the model rather than just browse species. The positive relationship of the model with both site occupancy and habitat quality indicates that the model is fit for application at relevant management scales. Field-validated models of similar resolution would assist in guiding management of conservation-dependent species

Ecology and Evolution, 2017

Ecology and conservation of owls, 2002

Fragmentation of native forest and woodland by clearing for agriculture and urban development has... more Fragmentation of native forest and woodland by clearing for agriculture and urban development has had a greater effect than logging on large forest owls and their arboreal marsupial prey. In this study, assessments were made of the contribution of forest and woodland fragments on privately-owned and unprotected lands towards the
regional conservation of these species in southeastern New South Wales. Small (<200 ha) frag- ments did not provide a significant reservoir for populations of the Powerful Owl Ninox strenua, Sooty Owl Tyto tenebricosa and Masked Owl T. novaehollandiae. Virtually all records of these owls in the region were associated with extensively forested areas or occurred within one km of the boundary of these areas (mainly state forests, national parks and nature reserves). The Barking Owl N. connivens, a ‘non-forest’ species, was also rarely recorded in forest/woodland fragments, raising concerns about the conservation status of this species in the region. Several important prey species for the three large forest owls, in particular the Common Ringtail Possum Pseudocheirus peregrinus, Greater Glider Petauroides volans and the Sugar Glider Petaurus breviceps, were either absent or less abundant in small forest fragments. The Barn Owl Tyto alba, a woodland species, and two other nocturnal ‘forest’ birds, the Southern Boobook Ninox novaeseelandiae and Australian Owlet-nightjar Aegotheles cristatus, were common and widespread in small forest and woodland fragments.

Forest Ecology and Management, Jan 1, 2005

The occurrences of 227 vertebrate species recorded on 487 forested sites in north-eastern New Sou... more The occurrences of 227 vertebrate species recorded on 487 forested sites in north-eastern New South Wales were investigated to determine the levels of association among species, the distribution of species by broad functional groups, species relations with key environmental and disturbance variables, the sensitivity of species to intensive logging, and to list species that have potential value as direct indicators of major environmental change and which should be considered as candidates for long-term monitoring.

A total of 40 species appeared to be significantly disadvantaged by logging, another 40 species appeared to be significantly favoured by logging, while the remainder (147 species) appeared to be relatively unaffected. Most species were widely distributed throughout both logged and unlogged landscapes, albeit at possibly quite different abundances. This may be due to the fine-scale complexity of the habitat mosaic for many species, resulting in the continued availability of essential resources within or near logged areas, or the lack of sufficient sensitivity in the analysis due to the need to use presence–absence data. A further 121 species were recorded so rarely in the study (based on 619 sites) that it was not possible to make any assessments of their sensitivity to logging. The study produced several unexpected results, including the failure to identify some species as sensitive to logging that had previously been identified as sensitive in experimental and other retrospective studies.

Species disadvantaged by logging were often either dependent on large old trees and/or tree hollows for nesting, roosting and foraging, such as the Common Brushtail Possum, Red-browed Treecreeper, Satin Flycatcher and Crimson Rosella, or they were species that inhabit open forests and woodland, such as the Rufous Bettong, Buff-rumped Thornbill, White-throated Gerygone, White's Skink and Nobbi, that were probably disadvantaged by the increase in stem density and understorey and mid-canopy cover that usually follows logging. In contrast, species favoured by logging, such as the Eastern Whipbird, Brown Gerygone, Lewin's Honeyeater, Wonga Pigeon and Land Mullet, were often those species preferring wetter forest environments, particularly those with multi-layered foliage and a forest structure that includes fallen logs and a dense ground cover. No species displayed consistently strong preferences for selectively logged forest. Mammals were more likely to be disadvantaged by logging than birds, but reptiles contained the largest proportion of species sensitive to logging (39%). Frogs were inadequately sampled in this study; only nine species were recorded. While each taxonomic group had some species that were apparently sensitive to logging, the time frame to population recovery following logging disturbance may differ widely between groups depending on their habitat requirements. Fire history was also strongly correlated with the occurrences of many species, indicating the need to quantify species sensitivity to regular fuel-reduction burning.

Three main assemblages of species were identified. One group consisted of rainforest or wet forest specialists, another group comprised open-forest generalists, including a number of open-country and woodland species, and a third group was made up of tall, eucalypt forest generalists, including all of the large forest owls and marsupial gliders. Each broad assemblage, and most of their sub-groups, were found to have at least one species that was apparently sensitive to logging. The study was unable to test whether these sensitive species can function effectively as indicators of the changing status of other species within each assemblage because sampling was undertaken only once at each site.

This study has provided one of the few opportunities to date to critically evaluate some aspects of the indicator species concept and its potential as a management shortcut for assessing changes in biodiversity in managed forest landscapes. Qualified support for the concept has been shown, based on the identification of a set of species apparently sensitive to logging and their representation across a range of species assemblages. The choice of candidate species for monitoring will depend on the integration of results from other experimental and retrospective studies, and will be specific to each region depending on the composition of species assemblages and the goals of management.

Australian Field Ornithology 25(1), Mar 2008

We report observations of adult Barking Owls Ninox connivens from different parts of the Australi... more We report observations of adult Barking Owls Ninox connivens from different parts of the Australian continent, involving both subspecies, holding prey at their diurnal roosts. Prey items could be identified in some cases, and prey was partly eaten in most cases. The Owl was a female in all four cases where the sex of the Owl was known.

Revegetation within cleared farming landscapes offers the potential to restore habitat for many w... more Revegetation within cleared farming landscapes offers the potential to restore habitat for many woodland-dependent species that have declined since European settlement. Most species of arboreal marsupials require hollows for breeding and diurnal shelter, a resource that is usually available only in old trees; however, this constraint does not apply to the Koala. In this study, we describe the occupancy and use of young (4- to 7-year old) eucalypt plantations by Koalas in a predominantly cleared landscape used for intensive cropping and grazing. We compare Koala occupancy in 27 eucalypt plantations, 5 paddocks and 11 remnant forest and woodland sites, and we report the relative usage of these three land cover types by two adult male Koalas that were radio-tracked for 5 and 7 months using GPS transmitters. Koalas were recorded using young eucalypt plantations at 7 sites and remnant forest and woodland at 7 sites. Both radio-collared Koalas used eucalypt plantations more than expected based on the availability of this land cover type in their home-ranges. Occupancy of young eucalypt plantations and remnant patches by Koalas was strongly influenced by the proximity of these sites to remnant vegetation.

Wildlife Research, Jan 1, 2007

The koala (Phascolarctos cinereus) is a charismatic, high-profile species whose conservation need... more The koala (Phascolarctos cinereus) is a charismatic, high-profile species whose conservation needs are commonly perceived to be incompatible with logging. However, koala biology and the results of chronosequence studies elsewhere suggest that this species may tolerate a degree of habitat alteration caused by logging. In this study, 30 koalas, five in each of six areas available for logging within a mixed white cypress pine (Callitris glaucophylla)–Eucalyptus forest in north-western New South Wales, were radio-tracked for one year during 1997–1998 to determine their movements, home-range sizes and tree preferences. Five months after the study began, three of these areas were logged selectively for sawlogs and thinnings of the white cypress pine, a tree that is important to koalas for daytime shelter. This removed about one-quarter of the stand basal area, but the eucalypt component was unaffected. The remaining three areas were left undisturbed as controls. Radio-tracking continued in all six areas for another seven months. Koalas continued to occupy all or part of their previous home-ranges after selective logging, and home-range sizes remained similar between logged and unlogged areas. Home-ranges for both sexes overlapped and were ~12 ha for males and 9 ha for females. Koala survival and the proportions of breeding females were similar in logged and unlogged areas. The principal food trees of the koala were red gums, mainly Eucalyptus blakelyi and E. chloroclada, and the pilliga box (E. pilligaensis), none of which were logged in this study. These results suggest that selective logging for white cypress pine does not appear to adversely affect koala populations and that koalas may not be as sensitive to logging as previously thought. Further work is required to determine thresholds in the level of retention of koala food trees in logging operations.

Emu, Jan 1, 2003

Intensive logging for sawlogs and woodchips in south-eastern Australia is known to cause an immed... more Intensive logging for sawlogs and woodchips in south-eastern Australia is known to cause an immediate impact on bird populations. We measured bird abundance and species turnover during the medium term (13 and 22 years) after intensive alternate-coupe logging in a forest area that had been studied previously for short-term (0–4 years) impacts. The study provides a series of snapshots of bird population recovery following one of the most intensive logging operations in the region when, unlike the current practice, no old trees were retained in logged areas and no unlogged forest was retained in strips along all drainage lines. The aim was to assess the longer-term effects of logging on the bird assemblage and to determine whether recovery had occurred. We found that recovery had occurred for a large component of the avifauna within 22 years of intensive logging (as practised in 1976 when few or no old trees were retained in logged areas). Most bird species that forage among canopy foliage, in the air, among the understorey and on the ground had recovered. Five common forest birds were significantly more common on logged coupes than unlogged coupes. The Bell Miner, Manorina melanophrys, a honeyeater often associated with disturbed forest and implicated in eucalypt canopy dieback, had established colonies on some logged coupes 22 years after intensive logging. However, some hollow-nesting bird species (e.g. treecreepers, cockatoos) had still not fully recovered. A number of other common forest birds occurred so rarely in the study area (on both logged and unlogged coupes) that few conclusions could be made about their responses to logging. Evidence suggested that intensive logging may have had an effect at a local landscape (compartment) level as well as at the level of individual logged coupes. This was indicated by changes to the avifauna of the unlogged coupes over time (increasing numbers of species) as the forest regenerated on the adjacent logged coupes.

Austral Ecology, Jan 1, 2007

Most of the original forest and woodland cover on the western slopes of New South Wales and the n... more Most of the original forest and woodland cover on the western slopes of New South Wales and the northern plains of Victoria has been cleared for agriculture (wheat, sheep and cattle) and what remains is highly fragmented and modified by a long history of disturbance. Over the past three decades, native eucalypt trees and shrubs have been planted extensively in a part of this region to provide a range of environmental benefits. Our aim was to determine the extent to which these plantings could improve biological diversity in agricultural landscapes in south-eastern Australia and to identify the variables influencing their effectiveness. We sampled birds at 120 sites encompassing the range of available patch sizes, stand ages, floristic and structural conditions, and habitat attributes for revegetated areas and remnants of native vegetation, and we compared these to nearby paddocks. Eucalypt plantings were found to provide significant improvements in bird population density compared with cleared or sparsely treed paddocks, and mixed eucalypt and shrub plantings had similar bird communities to remnant native forest and woodland in the region. Birds displayed a strong response to patch size, with both larger (≥5–20 ha) eucalypt plantings and larger (≥5–20 ha) remnants having more species and more individuals per unit area than smaller (<5 ha) patches of these vegetation types. Older (10–25 years) plantings had more bird species and individuals than young (<10 years) plantings. The distance from remnant forest and woodland (habitat connectivity) appeared to be an important variable influencing bird species richness in eucalypt plantings. The main differences were due to the greater numbers of species classified as woodland-dependent in the larger-sized patches of plantings and remnants. Eucalypt plantings provided useful habitat for at least 10 declining woodland-dependent species, notably for the Speckled Warbler, Red-capped Robin and Rufous Whistler. The Brown Treecreeper and Dusky Woodswallow appeared to be the species most limited by the extent of remnant forest and woodland in the region. Plantings of all shapes and sizes, especially those larger than 5 ha, have an important role to play in providing habitat for many bird species. Restoration efforts are more likely to be successful if eucalypt plantings are established near existing remnant vegetation.

Australian Zoologist 30(4), Apr 1998

"State Forests of the South Western Slopes of New South Wales were surveyed for ground-dwelling m... more "State Forests of the South Western Slopes of New South Wales were surveyed for ground-dwelling mammals utilising hair traps (42 sites), small mammal traps (11 sites), nocturnal searches (253 sites), scat and animal sign observations, opportunistic observations and records from unpublished studies in the region. A total of 13 indigenous and 10 exotic ground-dwelling mammal species were identified for the region. No species that are recognised as rare, endangered or threatened were encountered during this survey although some evidence suggesting the presence of Long-footed Potoroo Potorous longipes and Smoky Mouse Pseudomys fumeus was found in Bago and Maragle State Forests.
Hair trapping (1170 large traps for an average of 18 nights each and costing 226 person hours) revealed the presence of 9 species/genera. Swamp Wallabies Wallabia bicolor, were the most commonly detected species. Small mammal trapping in swampy environments of Bago and Maragle State Forests (1100 trap nights costing 52 person hours) detected the presence of Bush Rat Rattus fuscipes, Dusky Antechinus Antechinus swainsonii and Brown Antechinus Antechinus stuartii. Rattus fuscipes proved to be abundant in some localities with trapping rates up to 44% on some nights. Records made by nocturnal survey (253 sites costing 306 person hours) revealed 14 ground-dwelling mammal species, notably Long-nosed Bandicoot Perameles nasuta.
All the techniques were judged to be of value for broad forest surveys as no technique covered all species. Hair trapping had the greatest limitations in this study but is still judged to be of value if the method is used as a preliminary technique. Limitations of each technique are discussed."

Australian Zoologist 30(4), Apr 1998

"A regional survey of 253 forest sites on the south-western slopes and adjacent highlands of New ... more "A regional survey of 253 forest sites on the south-western slopes and adjacent highlands of New South Wales recorded a total of 530 animals from five species of nocturnal forest birds and nine species of arboreal marsupials. One additional species, the Squirrel Glider Petaurus norfolcensis, was trapped during supplementary searches. Elevation (climate) and forest type were the major factors accounting for the distribution of this fauna. Minor environmental gradients included logging intensity, the density of hollow-bearing trees, and fire. The patterns of land-use in the region contributed to interactions between elevation, forest type and management history (logging, fire and grazing) which restricted inferences about the habitat requirements of animals.
Forest types occurring at either end of the elevational (wet-dry) gradient (Alpine Ash and Box-Stringybark) were poorer in species than forests occurring at intermediate elevations (Alpine Gums, Gully Gums and Peppermint and Dry Peppermint). The Gum-Peppermint forest type, which was the richest in species of nocturnal animals, has been deliberately selected in the past for conversion to pine plantations. The fauna of the region was generally more sparse and patchy compared to four other regions (north-eastern and south-eastern New South Wales, and north-eastern and Central Highlands of Victoria) where similar studies have been undertaken. Numbers of large forest owls and the Greater Glider Petauroides volans were recorded less commonly than expected and may have declined in the region. The Common Brushtail Possum Trichosurus vulpecula was particularly abundant. This species and the Common Ringtail Possum Pseudocheirus peregrinus appeared to maintain their numbers in small (<2000 ha), isolated forest fragments."

Fishes of Sahul 23/2, Jul 2009

Observations of Eel-tailed Catfish migrating upstream after rain in the Paroo and Bulloo Rivers ... more Observations of Eel-tailed Catfish migrating upstream after rain in the Paroo and Bulloo Rivers
Matthew Stanton and Graham Turner
One place I always wanted to travel in Australia was the Channel Country of South-western Queensland, North-western New South Wales and North-eastern South Australia. I must confess that the main attraction was a suite of unusual birds that inhabit these parts, but fish are always in the back of my mind.
Travel in the area is somewhat restricted. Relentless heat in summer is off putting and drought is the normal situation. When it rains much of the area is off limits because of flooding.
In January and February 2007, a number of rain events in the Bulloo and surrounding catchments caught our attention and we began planning a trip, counting on any floods to have subsided by Autumn. We watched the river height reports through March, following minor flooding events as they moved downstream. By April, most roads were open and we had the opportunity to travel for a couple of weeks. Almost as an afterthought, we threw a small scoop net, a couple of bait traps and a hand line in the back of the car.
Once we arrived in the area, we found it hard to believe that roads that had been closed due to flooding only a week before, now had only a few centimetres of water over the river crossings. It seems that “minor flooding” really meant “flowing rivers”. Roads were closed because they also become rivers when there is nowhere else for the water to disperse in the mostly flat landscape.
Our first proper fish stop was on the Paroo River at Eulo in Southern Queensland. We stopped at a weir, crossed by the main road heading west. The first thing we noticed were numerous water birds lining the waters edge. White-necked Herons (Ardea pacifica) seemed very interested in the area but quickly took flight when we opened the car doors.
The next thing we noticed was that the water was completely turbid and brown. The only way we would see fish here was if they were in our hands.
We tried scooping around the rough cobbled edges of the pool below the weir. In this area we found many Spangled Perch (Leiopotherapon unicolor). They seemed to be congregating around the fast flowing water pouring off the weir. Next we noticed movement in the fast flowing streams themselves. A few quick scoops revealed masses of Hyrtl’s Tandans (Neosilurus hyrtlii), all beating their way into the current. It was quite possible to take a quick scoop and bring up 20 or more, a writhing mass of little catfish. The size of the fish was reasonably consistent being between about 120 and 160 mm total length.
Their swimming ability was most startling. Over a distance of about 20 to 30 metres to the pool below the weir, the river fell about three metres. Any slack water in the otherwise rapidly flowing water was being utilised by catfish to have a bit of a rest on the way up the weir. Many of the catfish seemed to use the shallower and slower flowing bits for moving upstream. Even so, there were a number of tightly constricted places in the stream bed where the water was flowing swiftly and which the fish had to negotiate. They seemed to manage this without trouble. Some of the eddies were so full of catfish taking a breather that the space seemed to be over 50% fish and little eel-tails were constantly flipping out of the water as they struggled against their mates to stay in the calmest spot. These must have made an easy target for the herons.
It was difficult to gauge the exact numbers of fish swimming upstream because of the turbid water. Our general impression was that there were plenty.
Later that day we found the same species at two other localities. The first was a small feeder stream to the Paroo. At this location we could easily see tandans swimming through the shallow water (~20 mm) in a box culvert.
The second location was further south in Currawinya National Park. Here we had our best look at the swimming (crawling) abilities of Hyrtl’s Tandan.
Caiwarro Crossing of the Paroo River, Currawinya National Park, is a concrete weir/causeway with a rubble slope downstream and shallow pools upstream. The area is well vegetated near the river with trees, shrubs, and a fringe of emergent sedges and grasses.
We arrived about an hour before sunset. Once again we noticed the birds first, in this case Whistling Kites (Milvus sphenurus) outnumbering the White-necked Herons. This is a well used tourist camp site and there was a car crossing the causeway every five minutes or so. The birds were used to getting out of the way and we also learnt quickly.
Between vehicles we saw a near continuous stream of little bow waves crossing the smoothly flowing water of the weir. Mostly the water was less than 60 mm deep and fish were swimming upstream in water as shallow as 20 mm deep. The water was almost as turbid as at Eulo yet we could quite clearly see the fish here as their backs were often out of the water. The long anal/caudal fin served the fish well in propelling themselves through the shallows with the fin always being immersed even if the body was not. The head usually stayed under a bow-wave of water.
The occasional catfish was also swimming back down the weir. I guess it is easy to get confused when the only thing a fish could see would be light or shadow. It was easy to see why the birds were gathering at these locations. The fish were easy targets as they crossed the weirs.
Fish passage would improve with greater flow and could easily be improved with minor modification of the causeway.
Later in the trip we visited the Bulloo River where it flows through Thargomindah. Here we encountered a different species of catfish with a different problem. Silver Tandan (Porochilus argenteus) were in abundance at the base of a weir, but seemed to be unable to cross it due to insufficient flow and a very steep drop which terminated in a short free-fall. It would take considerably more than “minor flooding” to remove this barrier to tandan migration. At least these fish were not exposing themselves to the same level of predation that the Paroo River migrants had been suffering. The weir is probably only passable in extreme conditions. Hopefully the Silver Tandans don’t require regular access to upper tributaries in order to breed.

Thesis by Matthew Stanton

"The Barking Owl Ninox connivens population in the Pilliga forests of northern New South Wales is... more "The Barking Owl Ninox connivens population in the Pilliga forests of northern New South Wales is the largest known in southern Australia. Breeding pairs in this population occupy large home-ranges across less than half of the forest. In this thesis, I quantify the diet of Barking Owls in the Pilliga. I consider a number of hypotheses that could explain the species’ large home ranges and restricted distribution, particularly those that are related to prey availability.
This is the first diet study of a Barking Owl population to incorporate data from many territories over several years and all seasons of the year. Radio-tracking of nine owls provided the opportunity to begin a substantial collection of prey remains (regurgitated pellets, food debris and faecal material). Ultimately, the collection period spanned 2003 – 2009, with prey remains from 19 territories in the Pilliga and one territory in a small forest to the south near Dubbo. In total, 1546 regurgitated pellets and 315 faecal samples were collected and examined. Foraging observations improved the understanding of the results.
Barking Owls in the Pilliga forests preyed on most species of diurnal and nocturnal birds, as well as Sugar Gliders, bats and insects, with a few items being taken from the ground. Prey size ranged from 0.3 gram insects to ~800 gram cockatoos and mammals, a similar size to the owls. Most prey were native animals in contrast to some other studies.
The proportions of consumed prey, as determined by pellet analysis, were compared with available prey, as determined by bird counts, spotlight surveys, small mammal trapping, bat surveys and insect netting. Prey items from all prey groups were available from all sampled areas of the Pilliga. Barking Owls distribution was positively associated with prey availability: significantly with the biomass of birds and with flying insect numbers. Mammal groups were not significantly different but showed the same positive trend.
Spatial availability of total prey biomass offered a good explanation for the distribution of Barking Owls within the Pilliga forests. Crucial food resources, particularly available biomass of diurnal birds and nocturnally active prey, may limit the population density and distribution of owls in what appears to be marginal rather than prime habitat. Land cleared for agriculture, because of its higher productivity, may have previously supported higher densities of Barking Owls when wooded."

Reports by Matthew Stanton

Dee Why bat monitoring during culvert repairs, 2016

Project outline As part of the implementation of the Dee Why Town Centre Masterplan, and to manag... more Project outline
As part of the implementation of the Dee Why Town Centre Masterplan, and to manage the potential impacts of flooding, a number of stormwater upgrades are required within the Dee Why Town Centre, Including upgrades to stormwater infrastructure at Oaks Avenue, Dee Why. The primary culvert has previously been identified as a regionally important roost site for the Eastern Bentwing-bat (Miniopterus schreibersii).
In order to mitigate the impacts of the construction works on the bat population, a range of mitigation measure were implemented, including installation of bat protection works and commencement of a monitoring program. Niche Environment and Heritage was engaged to monitor the bat population in the culvert and ensure that the population was not being significantly impacted by the works. The purpose of this report is to present the findings of the monitoring program.
Summary of results
Eastern Bentwing-bats continued to occupy the culvert during the works. The mitigation strategies implemented included the provision of exclusion screens and the installation of additional roost structures. The purpose of these strategies was to influence the roost location of the bats within the culvert, i.e. exclude the bats from the works areas, but maintain use of the culvert by encouraging roosting in other areas. Bats used the alternate roost locations as planned, but chose to use pre-existing roost structures rather than the new structures provided for them. There was a temporary dip in bat numbers that corresponded with the commencement of excavation works. Bats returned by the following week and there were around 100 bats occupying the culvert by the end of the works period. This is broadly consistent with data collected in previous years when that were unaffected by construction disturbance.
Conclusion
Works on the Oak Avenue culvert upgrade may have had a small temporal impact on Eastern Bentwing-bats that occupy the Dee Why trunk culvert. Numbers of bats present by the end of the works were at normal levels for that time of the year. Bat movement within the culvert will continue to be affected while the bat protection barriers remain in place. However, removing the barriers would cause an un-necessary increase in disturbance and is unlikely to increase the occupancy of the culvert given that occupancy already appears to be approximately normal.
It is recommended that the barriers remain in place until November 2016. Monitoring should take place immediately before barrier removal to confirm that bats have vacated the site.
Bat furniture used to augment bat roost habitat in this project, proved to be largely ineffective with the exception of light baffling curtains. The curtains were designed to reduce the light entering the ends of the culvert. This increased the habitat quality of the extremities of the culvert for roosting, thus facilitating a lower disturbance environment for bats to roost in.

Calibration of Songmeters with the bird census results of experienced observers, 2013

Context Niche Environment and Heritage Pty Ltd (Niche) was commissioned by Forests New South Wale... more Context
Niche Environment and Heritage Pty Ltd (Niche) was commissioned by Forests New South Wales to determine the feasibility of using SongMeters (a commercially-available sound recording device) to replace the field deployment of experienced bird observers, and thereby to save costs, in regional forest biodiversity monitoring programs. The setting was the Pilliga forests of north-western New South Wales where a major biodiversity monitoring program is underway.
Aims
The aim of the project was to assess the effectiveness of an autonomous sound recording system (SongMeter 2+, Wildlife Acoustics, Inc. Massachusetts, USA) in recording the presence of forest bird species in comparison with the results of standard, on-ground bird surveys using experienced bird observers. Specifically, we wanted to know which species could be detected reliably, if they were present, using SongMeters.
Methods
Systematic, fixed-time, variable-radius, point-counts of all birds seen and heard by two experienced observers were compared to the species identified from sound recordings made simultaneously at the same sites. Paired comparisons between the two methods were available for 136 ten-minute census periods in November 2012.
Key Results
The two survey methods agreed in their identification of bird species on approximately 50% (range 5-85%) of occasions when common bird species were encountered. This level of agreement declined to about 33% (range 0-100%) for detections of uncommon species. Experienced observers recorded more bird species per ten minute census period during point-counts (mean 12.51 species) than in simultaneous call recordings (mean 8.71 species). 17.5% of species were recorded only by experienced observers in the field.
Conclusions
SongMeters have the potential to match, or exceed, the numbers of bird species recorded by experienced bird observers during standard field surveys or in regional species monitoring programs if they are programmed and set to record autonomously in the field over extended periods. SongMeters provide many advantages, including standardisation of survey procedures and the provision of an archived record of point-count data.
Management Implications
SongMeters have the capacity to replace the field deployment of experienced bird observers, and thereby to save costs, in regional forest biodiversity monitoring programs. However, the collection of such voluminous amounts of data would necessitate the development of an automated process for identifying bird species from digital recordings. This requires the identification and development of a library of “bird call recognisers” which are the unique acoustic diagnostics of the calls for each (or priority) species.

Automated Bird Call Recognition Call recognition for birds in the Pilliga forests, 2014

Context Niche Environment and Heritage Pty Ltd (Niche) was commissioned by the Forestry Corporati... more Context
Niche Environment and Heritage Pty Ltd (Niche) was commissioned by the Forestry Corporation of New South Wales (FCNSW) to investigate the feasibility of developing bird call recognisers for use in software programs to automatically identify the bird species recorded on remote sound recording devices. These cost-effective sampling devices are planned for widespread use in the Pilliga forests regional biodiversity monitoring program.
Aims
We aimed to develop call recognisers for 15 bird species considered to be of interest to forest managers in the Pilliga forests of north-western New South Wales.
Methods
Selections of bird calls recorded during 136 ten-minute census periods at 36 bird survey locations in 2012 were used to develop call recognisers for each species. The entire data set was used to evaluate the effectiveness of each recogniser in identifying the known occurrences of each species. SongScope software (Wildlife Acoustics) was used to develop and test the recognisers. Comparisons were made between the number of occurrences of a species correctly or incorrectly identified in relation to the known occurrences of that species. Species presence was determined for each 10 minute census period from the bird counts made directly by an experienced observer standing at each survey point, from independent species identifications made by listening to the calls collected during simultaneous recording sessions, and by expert assessments of the calls attributed to each species by the SongScope software using our recognisers.
Key results
Call recognisers (25) were produced for 16 species of forest birds. Most species would be surveyed efficiently by the recogniser, given sufficient recording time. Brown Treecreeper, Eastern Yellow Robin and Sacred Kingfisher do not yet have recognisers suitable for use due to variability issues. Some recognisers proved to be more useful for species they were not intended for and, overall, there was a large proportion of “false positive” identifications attributed to most call recognisers. Testing of the recognisers confirmed that some species will always be difficult to detect by automated means.
Conclusions
Recognisers failed to detect as many presence records for each species compared with the results of direct field surveys using experienced observers or the results of experts identifying bird species from simultaneous recorded bird vocalisations. However, when deployed using extended recording periods, SongMeters should prove to be efficient for detecting birds. This technology (i.e. remote sound recording devices combined with the use of “trained” call analysis software) is not yet the most effective way of surveying the full suite of species occurring in an Australian bird community.
Management implications
This study has shown that experienced bird observers are still required to guide the use of recognisers in bird call identification. The study provides a method to assist determination of the likely power (i.e. required sampling effort) of recordings. SongMeters need to be deployed for extended periods to collect sufficient data to compensate for the poor call discrimination ability of most species recognisers. The automated identification of bird species from sound recordings is a new and developing field and managers should remain open to the arrival of new technologies and methods which may improve efficiencies.

Project outline The NSW Roads and Maritime Services recently completed works to widen and upgrade... more Project outline
The NSW Roads and Maritime Services recently completed works to widen and upgrade the existing Nelson Bay Road (MR108) between Cromarty Lane, Bobs Farm and Port Stephens Drive, Anna Bay to a divided dual carriageway. The project was opened to traffic in March 2015.
The Koala (Phascolarctos cinereus) was identified as an issue for the construction and operation of the duplicated and upgraded Nelson Bay Road (Niche 2014). Evidence of Koala use of the area came from Koala road fatality records, and observations of Koalas or Koala signs in adjacent and nearby habitat. The pre- existing two lane road provided no protection for Koalas and therefore additional measures were put in place to minimise Koala/vehicle interactions. These were floppy-top fauna fencing, round-bar grids across driveways and local roads, the construction of two fauna underpasses and ecological monitoring of the Koala within the local landscape.
Monitoring commenced during construction and continued through the first year of road operation. The aims of Koala monitoring work were to:
1. Provide data to help protect Koalas during construction works.
2. Examine habitat use by the Koala in the project area (which covers the Tomaree Peninsula adjacent
to Stage 3) and to identify critical habitat components for these Koalas.
3. Collect data to help determine the population size.
4. Monitor the Koala in the project area to detect any changes in the population that could be attributed to the Nelson Bay Road Upgrade.
5. Determine if the underpasses provided as mitigation for the Nelson Bay Road Upgrade are effective in reducing the mortality of the Koala on Nelson Bay Road.
Three main strategies were employed to achieve these aims:
 Nine Koalas were fitted with GPS/radio collars to provide long term observations of individual Koalas in regards to their use of the environment, including the underpasses.
 The underpasses and adjacent areas were monitored continuously for a year with automated game cameras to record use of these structures by Koalas and any other fauna.
 Koala surveys by spotlight and faecal pellet counts were continued on a subset of sites established for the initial project assessment in order to detect changes in Koala densities.
Summary of Results
Individual tracked Koalas were found to maintain small territories (mean FK95% 1.98 ha). However, most individuals also made occasional longer expeditions into the surrounding areas. With one exception, these expeditions brought the koalas into contact with roads. One tracked Koala made numerous crossings of Nelson Bay Road over a year during the construction phase before being fatally struck by a car three months into the operational phase.
The automated game cameras found no evidence of Koalas using the fauna underpasses. Feral and domestic mammals were the main users of the underpasses (Fox, Cat and Rabbit). However the Short- beaked Echidna, Lace Monitor and an unidentified skink were also observed to transit through the underpasses. Notably, the entrances to the underpasses were also used for non-transitory purposes (e.g. shelter and foraging) by a wide range of birds and mammals.
Koala presence or absence remained largely steady across the monitoring sites. However, some of the sites more distant from the Nelson Bay Road project site showed a decline in Koala presence late in the study.
This study has documented the typical movement and ranging patterns of Koalas in the Bobs Farm to Anna Bay area. Interactions with roads were relatively few within the study period. The study period appears to have been undertaken following a period of decline in the local Koala population.
Conclusion and declaration
Measures to avoid impacting Koalas during the construction phase were successful with no Koala deaths or injuries recorded during that period. The barriers to minimise Koala collisions with cars appear to have had some success with a lower strike rate than for the five years preceding construction. However, Koalas were still falling victim to vehicle strikes during the operational phase. The underpasses were not effective in facilitating Koala passages in the area as the Koala was not recorded using them.
The future of the Koala population in the study area is unknown, however, a continuing decline seems likely based on modelling looking at factors unrelated to roads or the Nelson Bay Road Upgrade Stage 3. Koalas are still breeding successfully in the area and good habitat still contains a moderate density of Koalas.
Although the fauna fencing was installed in accordance with the Review of Environmental Factors Addendum (November 2013), extension of the fauna fencing to the east of Binder Road would improve its effectiveness and reduce the likelihood of future vehicle strikes in this area. Improvements to the fence design, such as removing climb over points and re-working gates to tighter tolerance, would also benefit the operation of the fencing system.
Overall however, the monitoring was unable to detect any changes in the Koala population that could be attributed to the Nelson Bay Road Upgrade Stage 3.

This project documented the capacity of young eucalypt plantations to restore habitat for fauna w... more This project documented the capacity of young eucalypt plantations to restore habitat for fauna within a highly fragmented, and ecologically degraded, agricultural landscape. The study compared fauna occupancy within 4-6 year old eucalypt plantations and paddocks used for grazing and intensive cropping with remnant forest and woodland in the region. The study sought to validate inferences from a previous study in another region about the importance of plantation age and size and proximity to remnant vegetation. It also sought to calibrate forestry-type plantations with eucalypt plantings established primarily for broad environmental benefits. The study assessed opportunities for improving habitat for fauna in agroforestry plantations. It also documented the capacity of eucalypt plantations to provide critical resources for breeding and year-round occupancy for a range of vertebrate species.
We found that the responses varied across the different taxonomic groups of fauna investigated. Young (4-6 year old) eucalypt plantations had a surprisingly similar avifauna to that recorded in remnant forest and woodland in the Liverpool Plains region. Eucalypt plantings clearly lack certain resources and habitat components that are important for many fauna. A rigorous field experiment was established to test the effect of supplementing plantings with nest boxes and ground cover. Initial inspections to document the use of these resources found that the addition of nest boxes provided otherwise missing tree hollow resources for two species of bats, two species of marsupials, two species of parrot and tree frogs. Invertebrates also made extensive use of all kinds of boxes provided. Nest box uptake by some species such as Sugar Gliders was limited by proximity to remnant populations. The addition of cover boards greatly increased the counts of reptiles in plantings compared to those without added cover, further indicating that lack of cover is the main restriction on the use of plantings by reptiles. Both forms of habitat supplementation (nest boxes and ground cover) will require further inspections and surveys in coming years to fully test the hypotheses of the experiment.
Recommendations to improve the biodiversity values of eucalypt plantations need to consider the varying requirements of different fauna groups. Birds benefit from a high stocking rate of trees in plantations, but many bats are unable to forage in dense vegetation. A practical recommendation to balance the needs of these two groups could be to plant at a high density to minimise weeds and optimise tree growth and form, but then to thin some patches non-commercially within the plantation to create gaps. Consideration could also be given to alternate planting of eucalypts and fast-growing acacias, with the latter self-thinning creating gaps in the plantation and increased ground cover with fallen dead wood. The aim would be to maximise structural complexity at the patch scale. Our results suggest the floristic composition (i.e. a few commercial species of eucalypts) compared to richer environmental plantings has less effect on fauna than structural components.
The composition of biodiversity inhabiting eucalypt plantations is influenced differentially by the surrounding landscape. Birds and less mobile terrestrial fauna benefit enormously when plantations are established close to existing remnants. The landscape scale is of less importance to bats, but the retention of remnant trees on farms, even when scattered in the landscape, is vital for bats. Plantation patch area appeared to be less important for birds than proximity to remnants.
Artificial ground cover and nest boxes are predicted to benefit a wide range of fauna, but the extent is yet to be fully revealed.

Extensive areas of trees and shrubs are being planted for land rehabilitation and wood production... more Extensive areas of trees and shrubs are being planted for land rehabilitation and wood production on previously cleared agricultural land in Australia. Environmental benefits are a major incentive for this change in land management, but data concerning the response of biodiversity to new plantings are scarce and are needed to direct restoration efforts and to underpin policy development.
This report outlines a large-scale study to guide future planting schemes for biodiversity recovery in agricultural landscapes. It documents and compares the occurrences of more than 180 species of birds, mammals, reptiles and amphibians sampled at 136 sites representing eucalypt plantings in two broad age-classes, nearby remnants of native forest and woodland, and cleared or sparsely-treed paddocks. The study design also enabled comparison of the occurrences of these species across a range of patch-sizes of both revegetation and existing remnant vegetation. Comparisons were also made of the occupancy of young plantings by birds and bats in two landscape types, which differed mainly in their proportions of retained native vegetation. The role of vegetation type and condition, and management history, in influencing the occurrences of species in plantings and remnants is discussed. The report concludes with recommendations for restoring habitat for wildlife on farms, in the context of improving conservation outcomes from eucalypt plantings that are established for multiple purposes.
This project was funded by the Natural Heritage Trust and the Joint Venture Agroforestry Program (JVAP). JVAP is supported by three R&D Corporations — Rural Industries Research and Development Corporation (RIRDC), Land & Water Australia, and Forest and Wood Products Research and Development Corporation (FWPRDC), together with the Murray-Darling Basin Commission (MDBC). The R&D Corporations are funded principally by the Australian Government. State and Australian Governments contribute funds to the MDBC.

Ecology and Evolution, 2017

Ecology and conservation of owls, 2002

Forest Ecology and Management, Jan 1, 2005

Australian Field Ornithology 25(1), Mar 2008

Wildlife Research, Jan 1, 2007

Emu, Jan 1, 2003

Austral Ecology, Jan 1, 2007

Australian Zoologist 30(4), Apr 1998

Fishes of Sahul 23/2, Jul 2009

Dee Why bat monitoring during culvert repairs, 2016

Calibration of Songmeters with the bird census results of experienced observers, 2013

Automated Bird Call Recognition Call recognition for birds in the Pilliga forests, 2014

Habitat loss and alteration has been ongoing in many landscapes since European settlement. These ... more Habitat loss and alteration has been ongoing in many landscapes since European settlement. These changes in vegetation composition, condition and fragmentation have affected some species more than others.
The objectives of mine-site rehabilitation often don’t acknowledge this broader perspective, instead alternating between attempted restoration of the original plant communities or restoration to enable continued primary production. Within already highly-modifed landscapes grazing or occasionally timber production have been pursued as production goals.
Mine-site rehabilitation provides an opportunity to target habitat restoration, for certain highly threatened species, and these restoration efforts may be simple to implement with a high chance of achieving the desired results within relatively short time-frames.
We provide examples demonstrating that the establishment of young eucalypt plantations in predominantly agricultural landscapes comprising mostly preferred Koala feed trees, can lead to rapid occupation and use by Koalas. Translocation of individuals may be required if there is no local source population. Similar targeted habitat restoration efforts have the potential to benefit many other threatened species such as the Squirrel Glider.

Coastal Conference NSW 2017, 2017

Coastal wildlife will face one of the most certain impacts of global warming; that of changing co... more Coastal wildlife will face one of the most certain impacts of global warming; that of changing coastlines caused by sea-level rise. While not a coastline-dependent species, many of the recognised strongholds of the Koala occur in low elevation coastal areas. The Port Stephens population of the Koala was once considered a stronghold for the species. However, significant population decline has been documented and is projected to continue for the population occurring on the Tilligerry/Tomaree peninsular due to the pressures of wildfire, dogs, increased vehicle traffic and other risks associated with urban expansion and intensification. In this paper, we contend that the preferred habitat for the Koala will be lost or degraded due to the additional pressure of rising sea-level.
Methods
We tracked nine Koalas using GPS data-loggers in the Bobs Farm-Taylors Beach-Anna Bay area to determine their movements, habitat preferences and exposure to urban risks. Koala collars recorded 16,534 reliable track points between February 2014 and June 2016.
We used modelled predictions of sea level rise for 2100 based on a predicted 1.1 metre rise in the Highest Astronomical Tide (HAT), an intermediate forecast by 2017 expectations (Sweet et al. 2017), but a severe prediction by the best estimates made in 2013.
Results
Koalas spent more than 95% of their time in swamp forest or modified swamp habitats with most animals never leaving Broad-leaf Paperbark (Melaleuca quinquenervia)/ Swamp Mahogany (Eucalyptus robusta) dominated forest or variations of this vegetation type which included Forest Red Gum (E. tereticornis), Flooded Gum (E. grandis) or Smooth-barked Angophora (Angophora costata). There were occasional forays into other forest types, particularly those containing Blackbutt (E. pilularis). These primarily were associated with commuting between swamp patches for feeding or breeding activity.
The amount of time spent within the Broad-leaf Paperbark/Swamp Mahogany vegetation type is also shown in their use of the dominant swamp trees (Figure 1). However, Swamp Mahogany is clearly a preferred tree species as it was used over half the time by tracked Koalas, even though it but was not usually the dominant tree. Broad-leaf Paperbark usually was far more abundant. Forest Red Gum was only found at a few locations but it was also clearly a preferred tree for Koalas.
Modelled Highest Astronomical Tides in 2100 will inundate approximately 40% of the study area and around 80% of the present-day home ranges of our study Koalas. Saltwater inundation will prevent the persistence of present-day Koala habitat.
Discussion
In light of predicted sea level rise this century, the freshwater swamp habitat choice by Koalas has strong implications for the Tilligerry/Tomaree peninsular population. Most of the wooded swamp environments selected were at elevations less than five metres above current Highest Astronomical Tide level, and around half the locations recorded were at elevations less than or within the one metre above HAT category (Figure 2). A sea level rise of one metre will cause significant coastline modification and saltwater intrusion of large portions of the remaining Koala habitat on the peninsula. Saltwater intrusion on the estuarine side of the peninsular will shift mangroves, saltmarsh and Swamp She-oak (Casuarina glauca) swamp inland, replacing areas of Broad-leaf Paperbark/Swamp Mahogany freshwater swamps that are currently prime Koala habitat (Figure 3).
Mangrove and saltmarsh communities will be significantly reworked as the sea-level rises rapidly over the remainder of this century. Because there is substantial sediment transport occurring in the tidal estuarine environment, it will be difficult to say precisely which areas will be maintained as land and which will become part of the submerged marine environment. While it is difficult to determine the exact shape of the of the mangrove and saltmarsh habitats, it is easy to see that freshwater swamp forest will be squeezed between the new saltwater intrusions and areas already cleared for human occupation or unsuitable as habitat for Koalas due to well drained dune habitats.
It is much less certain how swamp areas further from the shore line will develop with rising groundwater levels and increases in sediment mobility with climate change. Possibly these Koala habitats will remain Koala friendly, provided that increased environmental pressures on the human inhabitants of the peninsular do not result in more conversion of swamps into urban land. Some wooded swamps are already being destroyed by natural and anthropomorphic factors. The dune field behind Stockton Beach is already smothering Paperbark/Swamp Mahogany swamps. The threats to wetlands in urban areas remain present despite recognition as important Koala habitat.
Conclusion
Managers of the peninsular need to allow space for change with rising sea levels and other, as yet unknown, impacts of human-induced climate change. Rising sea levels will squeeze both human and wildlife populations. Koalas may be one of the significant casualties as they lose a substantial percentage of their current habitat.

Owls 2000, 2000

Fragmentation of native forest and woodland by clearing for agriculture and urban developments ha... more Fragmentation of native forest and woodland by clearing for agriculture and urban developments has had a greater effect than logging on large forest owls and their arboreal marsupial prey. In this study, assessments were made of the contribution of forest and woodland fragments on privately-owned and unprotected lands towards the regional conservation of these species in southeastern New South Wales.
Methods: A total of 120 sites was located evenly throughout the privately-owned or otherwise unprotected lands, including vacant crown land and leasehold land, in southeastern New South Wales. Sites were located on public roads in or near fragments of native forest or woodland. Each site was classified subjectively by the degree of fragmentation in the landscape. A habitat fragmentation score ranging from 1-6 was estimated within a circle of 1 km radius (314 ha) around each site. The classification was based on the assumption that at least some cleared land was present in the landscape surrounding each site.
Results: Small (<200 ha) fragments did not provide a significant reservoir for populations of the Powerful Owl (Ninox strenua), Sooty Owl (Tyto tenebricosa) and Masked Owl (T. novaehollandiae). Virtually all records of these owls in the region were associated with extensively forested areas or occurred within 1 km of the boundary of these areas (mainly state forests, national parks and nature reserves).
Important owl prey, in particular the Common Ringtail Possum (Pseudocheirus peregrinus), Greater Glider (Petauroides volans), Yellow- bellied Glider (Petaurus australis) and the Sugar Glider (Petaurus breviceps), were either absent or less abundant in small forest fragments. The Barn Owl (Tyto alba), a non- forest species, and two other nocturnal “forest” birds, the Southern Boobook (Ninox novaeseelandiae) and Australian Owlet- nightjar (Aegotheles cristatus), were surprisingly common in small forest and woodland fragments.