Fauna Spotters logo

Sandhill Dunnart (Sminthopsis psammophila)

The Sandhill dunnart is a small, nocturnal, ground-dwelling marsupial of sandy arid and semi-arid habitats in Western Australia and South Australia, with historical records from the Northern Territory. It is Endangered nationally and may occur on project sites containing long-unburnt spinifex hummocks, sandy dunes and mallee or shrubland, even where detection rates are very low.

Conservation status

Commonwealth, EPBC Act 1999: Endangered.

South Australia, National Parks and Wildlife Act 1972: Vulnerable.

Western Australia, Biodiversity Conservation Act 2016: Endangered.

Northern Territory, Territory Parks and Wildlife Conservation Act 1976: Endangered.

Breeding season

Timing varies with rainfall, food availability and local conditions. The pattern below reflects the best documented field observations and should be treated as a planning guide rather than a substitute for site-specific survey results.

Avoid clearing occupied or potentially occupied hummocks, burrows, logs and dense spinifex during the breeding and dependent-young period wherever practicable. Survey timing and mitigation must follow project approvals, permits and the advice of the relevant state or territory authority.

  • Breeding: Mating has been recorded in August and September. Captive husbandry information identifies the mating build-up from June.
  • Young: Pouch young have been recorded in September and October. Juveniles emerge in December and January.
  • Dispersal and wider movement: The species is resident or mobile across large home ranges rather than seasonally migratory. Mean home range was 70 ha, with reproductive males averaging 148 ha.
  • Nocturnal foraging: Adults are primarily nocturnal and may forage throughout the night, returning to daytime shelters before dawn.
  • Highest clearing risk: Clearing is most likely to harm the species from June to October because mating, pouch young and dependent young overlap this period. December and January also require care because juveniles are emerging and may still depend on nearby shelter habitat.

Identification

The Sandhill dunnart is one of the larger Sminthopsis species. It is elusive, mostly nocturnal and difficult to detect, so identification should combine the animal's appearance with habitat, tracks and other evidence.

Adults have grey-buff, brindled fur, dark eye rings, large ears, a dark forehead, white feet and a white or pale underside.

The tail is tapering, pale grey above and darker grey below, with a distinctive crest of stiff black or dark grey hairs on the underside of the distal tail. This tail crest is the most useful field character.

Head and body: 85 to 114 mm

Tail: 107 to 128 mm

Foot length: Approximately 22 to 26 mm

Average body weight: 33 g for females and 36 g for males

Identification: Similar species

The species overlaps with several other dunnarts, including the Ooldea, striped-faced, hairy-footed, little long-tailed, grey-bellied, long-tailed, Gilbert's, lesser hairy-footed and fat-tailed dunnarts.

The Sandhill dunnart is generally separated by its larger size and the dark crest on the underside of the tail. The Julia Creek dunnart is also large, but the two species do not overlap in distribution.

  • Do not rely on size alone where several Sminthopsis species occur.
  • Check the underside of the terminal quarter of the tail for the vertical crest of dark hairs.
  • Use photographs, measurements and expert confirmation where the animal is seen briefly or the tail is not clearly visible.

Identification: Field signs and detection

Tracks are quadrupedal, with a reported gait of 60 to 80 mm and a foot length of approximately 22 to 26 mm. Tracks may be found on open sandy substrates, especially near suitable spinifex habitat.

The species uses nests and shelters in spinifex hummocks, burrows, hollow logs, bark and other animal burrows. It may construct a nest chamber from spinifex needles inside a hummock.

Scats, predator scats containing remains, owl pellets and camera images can assist detection. No species-specific call or eyeshine description is provided in the supplied documents.

  • Search sandy substrates for tracks during daytime habitat inspections, taking care to separate them from tracks of other small mammals.
  • Inspect mid-aged, dense spinifex hummocks, burrows, hollow logs and fallen timber for shelter resources without dismantling potential nests.
  • Use camera traps in suitable habitat because the species is cryptic and occurs at low density.
  • For targeted surveys, use deep pitfall traps, Elliott traps, hair sampling devices and searches of predator scats or owl casts in accordance with permit and animal welfare requirements.

Distribution: States and bioregions

The current known distribution is fragmented across parts of the Great Victoria Desert and Eyre Peninsula. The species was collected historically near Lake Amadeus in the Northern Territory, but there have been no recent live records from that state.

In Western Australia, records are concentrated in the Great Victoria Desert, including the Queen Victoria Spring area, Mulga Rock and the south-western edge of the Great Victoria Desert.

In South Australia, records are from the Yellabinna sand dune country, including Yarle Lakes, Ooldea and Mount Christie, and from several localities on the Eyre Peninsula, including areas near Whyalla and Pinkawillinie.

The Northern Territory record is from the Lake Amadeus region in 1894. Remains have also been identified from owl pellets at Uluru and Kata Tjuta, but there are no recent confirmed live records.

Distribution: Known populations and strongholds

Three extant population areas are recognised in the supplied Western Australian guidelines: the south-western Great Victoria Desert in Western Australia, Yellabinna Regional Reserve in South Australia, and the Eyre Peninsula in South Australia.

The three populations are genetically similar, suggesting that they were connected historically. Records are scattered and the species can remain undetected despite surveys in apparently suitable habitat.

Distribution: Population information

A current national population estimate is not provided. Western Australian assessments cited in the survey guidelines estimated fewer than 2,500 mature individuals, fewer than six locations and continuing decline, although later information recorded the species at 15 locations across approximately 4,674 km2.

Western Australian historical assessment: Fewer than 2,500 mature individuals

Earlier estimated area of occupancy: Less than 500 km2

Earlier estimated extent of occurrence: Less than 5,000 km2

Later reported records: 15 locations across approximately 4,674 km2, based on information available to March 2016

Habitat: Vegetation communities

Sandhill dunnarts occur in sandy dune and interdune systems with spinifex hummock grassland. Suitable sites usually contain mallee, eucalypt, Callitris or mixed shrubland overstorey and a structurally dense ground layer.

Recorded habitats include open mallee, eucalypt and Callitris woodland, desert oak or Allocasuarina woodland, and mixed shrubland over Triodia or Plectrachne hummock grass.

Common associated plants include Eucalyptus gongylocarpa, Eucalyptus oleosa, Eucalyptus socialis, Eucalyptus gracilis, Callitris verrucosa, Acacia and Hakea species.

The substrate is commonly yellow or yellow-orange sand, often on dunes 5 to 30 m high or on sandy plains.

Habitat: Key habitat structure

Large, dense and relatively unbroken Triodia hummocks provide cover, nesting material and protection from predators. Detection and breeding have been associated with Triodia cover greater than 25% and a 90th percentile hummock height above 40 cm.

Radio-tracked animals preferred dense, long-unburnt foraging habitat, particularly dune slopes. In one Western Australian study, habitats in the 32 years or more post-fire stage were preferred and recently burned habitat was generally avoided.

Logs, vertical habitat complexity, mixed-sized spinifex hummocks and a diverse shrub layer may also increase habitat suitability.

Triodia cover associated with detection: More than 25%

90th percentile hummock height associated with abundance and breeding: More than 40 cm

Reported dune height: 5 to 30 m

Preferred fire age in one Western Australian survey: Most captures were in habitat 17 to 26 years after fire

  • Map dune crests, slopes, swales and sand plains separately because habitat use can vary between these landforms.
  • Record Triodia cover, mean and 90th percentile hummock height, hummock spacing, litter, logs, shrub richness and fire age during habitat assessment.
  • Retain connected patches of dense spinifex and avoid isolating suitable hummock habitat with linear infrastructure or cleared corridors.

Habitat: Climate and elevation

The species is associated with arid and semi-arid inland Australia. The supplied documents do not provide a specific elevation range or an upper temperature limit.

Foraging habitat: Diet and feeding sites

The Sandhill dunnart is primarily an insectivore that forages on the ground at night. It searches structurally dense sandy habitat and may travel well beyond the immediate shelter hummock during a nightly foraging period.

Fecal analysis from Western Australia found that ants, particularly Formicidae, made up a significant part of the diet. Other recorded prey included beetles, Hemiptera, centipedes and spiders.

Earlier records also include grasshoppers, termites, wasps and small vertebrates such as reptiles and probably small rodents. The species may take a broad range of invertebrates, but ants can form a large dietary component.

Most frequently identified prey group in the 2022 study: Hymenoptera, mainly ants

Other recorded prey: Beetles, Hemiptera, centipedes, spiders, grasshoppers, termites and wasps

  • Retain mature spinifex, shrub cover, litter and logs that support invertebrate prey and reduce exposure while the animal forages.
  • Avoid assuming that recently burned open ground is suitable simply because sandy substrate remains, as burned habitat was generally avoided in the Western Australian tracking study.

Foraging habitat: Foraging range

The species forages throughout the night and can return to a shelter up to 1 km away. Dense, long-unburnt habitats, especially dune slopes, were preferred over open or recently burned areas.

Mean home range was 70 ha in a study of 24 individuals. Reproductive males had the largest mean home range, at 148 ha, compared with 16 ha for reproductive females.

Mean home range: 70 ha

Reproductive male mean home range: 148 ha

Maximum recorded shelter return distance in the tracking study: Up to 1 km

Breeding habitat: Breeding biology

Breeding occurs in suitable spinifex habitat where females can maintain a protected nest or burrow and access nearby prey. Breeding timing may respond to rainfall and food availability.

Field observations recorded mating in August and September, pouch young in September and October, and juveniles emerging in December and January. Females usually produce one litter each year, but a second litter may occur in good seasons.

Females have been recorded with up to eight pouch young. Young may be deposited in a nest after leaving the pouch, although this has not been demonstrated in the wild.

Mating: August to September

Pouch young: September to October

Juveniles emerging: December to January

Litter size: Up to 8 pouch young recorded

Gestation in captivity: 16 to 19 days

Time in pouch: Approximately 45 days before weaning begins

Age at sexual maturity: 8 to 11 months

Recorded lifespan: Males recorded living for more than 1 year; a reliable maximum lifespan is not provided

Breeding habitat: Breeding habitat features

Breeding and the presence of subadults have been associated with Triodia cover greater than 25% and large hummocks, with a 90th percentile hummock height above 40 cm providing a useful habitat indicator.

Females may need secure nest chambers within spinifex hummocks or burrows, together with nearby dense cover and invertebrate prey. Avoid disturbing occupied hummocks, burrows, logs or dense ground cover during the breeding period.

  • Treat dense, long-unburnt Triodia on sandy dunes and interdunes as potential breeding habitat.
  • Retain shelter habitat around any detected female, pouch young, juvenile, nest or burrow and seek species-specific advice before clearing.
  • Schedule unavoidable clearing outside the main breeding and dependent-young period where practicable, subject to project approvals and survey findings.

Sheltering habitat: Shelter types

Daytime shelter is usually within or beneath dense ground cover. Shelters provide protection from predators and extreme desert conditions and may also support nesting and care of young.

Recorded shelters include Triodia hummocks, burrows, hollow logs, bark and burrows made by other animals. Animals have also been observed forming a nest chamber from spinifex needles in the centre of a stage 3 hummock.

In Western Australia, radio-tracked animals commonly sheltered in burrows under stage 3 to stage 5 Triodia hummocks. In South Australia, mid-aged stage 3 hummocks were preferentially selected for daytime shelter.

Triodia growth stages reported as shelters in Western Australia: Stages 3 to 5

Stage selected for daytime shelter in an Eyre Peninsula study: Stage 3

Return distance from foraging area to shelter: Up to 1 km

  • Search for potential shelter resources during the day, when animals are more likely to be concealed.
  • Do not remove, roll, cut or collapse spinifex hummocks, hollow logs or burrows during clearing or construction unless authorised and managed under the project fauna procedure.
  • Retain shelter patches within and beside cleared areas where safe, and maintain links between hummock patches.

Sheltering habitat: Shelter structure

The documents do not provide a standard nest, burrow or hollow-log entrance size. Shelter quality is described mainly by dense foliage, hummock height, foliage density, logs and surrounding structural complexity.

Shelters moderate temperature and humidity compared with exposed conditions and may reduce predation risk.

  • Record hummock height, width, density and separation, along with the presence of logs, litter and visible burrow entrances.
  • Use the reported 40 cm 90th percentile hummock height and more than 25% Triodia cover as screening indicators, not as proof that a site is occupied.

Threats: Clearing and fragmentation

The main project-related risks are loss of sandy spinifex habitat, removal of shelter hummocks and logs, fragmentation, altered fire regimes and predation by introduced cats and foxes. Low detection rates mean that an absence from a short survey should not be treated as evidence that the species is absent.

Clearing of dunes, interdunes, mallee, shrubland, spinifex hummocks and fallen timber can remove feeding, breeding and daytime shelter resources.

Roads, pipelines, tracks, pads and other linear infrastructure can divide habitat and increase the amount of exposed ground. The supplied documents identify habitat loss and fragmentation as threats, but do not quantify a road avoidance distance.

  • Avoid clearing long-unburnt Triodia on sandy dunes and interdunes wherever practicable.
  • Keep construction footprints, access tracks, laydown areas and drainage works out of dense hummock patches and known records.
  • Retain connected habitat between suitable patches and prevent unnecessary widening of existing clearings.
  • Use a fauna spotter catcher during vegetation removal where required by the approval or fauna management plan.

Threats: Fire

Large or frequent fires can remove the long-unburnt, structurally dense Triodia preferred by the species. Recently burned areas were generally avoided in the Western Australian tracking study.

Fire effects are not uniform. Some records occur in areas with different fire ages, and the supplied studies identify uncertainty about the relationship between fire age, hummock condition and abundance.

  • Protect long-unburnt hummock patches from construction ignition, escaped burns and unnecessary prescribed burning.
  • Map fire age and retain unburnt refuges when fire management or hazard reduction is proposed.
  • Do not assume that a post-fire site is unsuitable without considering surviving hummocks, logs and connected refuges.

Threats: Introduced predators

Feral cats and foxes are recognised threats. Sandhill dunnart remains have been found in feral cat scats in the Great Victoria Desert.

A 2022 survey found fewer fox observations in a fox-managed national park but more cat observations there than in unmanaged Crown Lands, showing that predator responses may differ between sites.

  • Do not attract cats or foxes with unsecured food, waste, carcasses or poorly managed temporary facilities.
  • Coordinate predator control, monitoring and construction hygiene with the land manager where the project is within or near known habitat.
  • Record cat and fox signs, camera detections and carcasses during project monitoring.

Threats: Other pressures and survey limitations

Introduced flora, introduced herbivores and altered vegetation structure are identified threats. The supplied documents do not identify a disease, hydrology or vehicle-strike threat specific to the Sandhill dunnart.

Detection is difficult. In Western Australia, historical capture rates were approximately 0.22 captures per 100 pitfall trap nights, 0.03 per 100 Elliott trap nights and 0.25 per 100 camera trap nights.

Western Australian pitfall capture rate: 0.22 captures per 100 trap nights

Western Australian Elliott trap capture rate: 0.03 captures per 100 trap nights

Western Australian camera capture rate: 0.25 captures per 100 trap nights

  • Control weeds and introduced herbivores where they reduce spinifex, shrub cover or habitat connectivity.
  • Use more than one detection method in suitable habitat, particularly deep pitfall traps and camera traps.
  • Treat a negative result cautiously and document habitat suitability, survey effort, fire history and weather.

Survey methodology: 1. Habitat and sign searches

Survey for Sandhill Dunnart must account for its low density, nocturnal activity and strong association with sandy, spinifex-dominated habitat. Use several complementary methods because a failure to capture the species does not demonstrate absence.

Search the full project area and representative sampling sites during daylight, focusing on sandy dunes and interdunes, open mallee or woodland, mixed shrubland and dense Triodia hummocks.

Map mid-aged to old, dense and unbroken spinifex, especially hummocks about 40 cm or taller, spinifex cover above 25%, logs, burrows and other potential shelter sites.

Search sandy substrates for tracks, nests, burrows and scats, and take care when separating Sandhill Dunnart tracks from those of other small mammals.

Use vegetation maps, aerial photographs, topographic maps and current satellite imagery to plan the search.

For small sites, use systematic transects spaced 50 to 100 m apart, or quadrats in representative habitat. For areas larger than 5 ha, replicate sampling across habitat types and plant communities.

Activity period: Primarily nocturnal

Indicative Triodia habitat: Hummocks about 40 cm or taller and cover above 25% are useful indicators of suitable habitat

Recommended small-site search layout: Transects at 50 to 100 m intervals, or quadrats in representative habitats

  • Record fire age, Triodia species, hummock height and cover, shrub density, logs, bare sand, tracks, burrows and other shelter features at each search location.
  • Give particular attention to long-unburnt dune slopes and dense lower vegetation, which are preferred foraging or shelter habitat.

Survey methodology: 2. Pitfall trapping

Use pitfall trapping as a primary detection method where access, approvals and animal welfare arrangements allow.

For a standard trapping site, use two parallel lines of six pitfall traps, with the lines 100 m apart and traps about 10 m apart along each line.

Use pits about 22.5 cm wide and 60 cm deep, with a 30 cm high and 60 m long flywire drift fence along each line. Extend the fence about 5 m beyond the end traps.

Operate the array for seven consecutive sampling nights unless the approved survey design or regulator specifies otherwise.

Check traps within three hours of dawn, provide shelter such as a toilet roll and a small amount of sand, and protect captured animals from heat, ants and predation.

Use at least 60 cm deep pits. Wider, deep pits have produced higher capture success than narrower or shallower pits.

Typical array: 12 pits, in two lines of six

Line spacing: 100 m

Trap spacing: About 10 m

Pit size: About 22.5 cm wide by 60 cm deep

Recommended duration: Seven consecutive nights

Historical Western Australian detection rate: 0.22 captures per 100 pitfall trap nights, including recaptures

  • Place trap lines across suitable habitat so that they intercept potential home ranges rather than concentrating only at a single hummock.
  • Release animals near their capture point after identification and processing, unless an approved fauna management plan requires another action.
  • Use a trained and authorised team, and comply with animal ethics, state permits and project conditions before trapping.

Survey methodology: 3. Elliott trapping and hair sampling

Use Elliott A or E/A traps as a supplementary method, especially in winter, and follow the effort and placement specified in the approved survey design.

Place Elliott traps through suitable sandy spinifex and mallee habitat, rather than relying on cleared tracks or disturbed margins.

Use hair sampling devices as a supplementary non-capture method because Sandhill Dunnart hair can be distinguished from hair samples when a reasonable sample is obtained.

Leave hair devices for several consecutive days, protect adhesive inserts from dust and weather, and send samples to an appropriately experienced specialist for identification.

Historical Western Australian Elliott detection rate: 0.03 captures per 100 Elliott trap nights, including recaptures

Identification from hair: Possible when a reasonable sample is available

  • Do not treat a negative Elliott trapping result as evidence that the species is absent, because capture rates are low.
  • Record the trap type, bait, trap nights, weather, habitat, captures and non-target interactions for every site.
  • Use the species-specific identification features, including its relatively large size and the dark hair crest on the underside of the distal tail, when checking captured animals or photographs.

Survey methodology: 4. Cameras, predator remains and local knowledge

Deploy camera traps in suitable habitat, particularly where the species may occur at low density or avoid traps.

Aim cameras at ground level near dense Triodia, sandy runways, burrows, shelter sites or other likely movement points, and use an appropriate lure under the approved survey design.

Collect predator scats, owl casts and remains from predator nests or dens where this can be done safely and lawfully, because remains may confirm presence when live capture is unsuccessful.

Use daytime searches for tracks on sandy substrates and consult local people, including Traditional Owners, about records and habitat knowledge.

Species activity: Nocturnal, with activity generally continuing through the night

Camera suitability: Useful for cryptic species occurring at low densities

Recorded home range: Mean 70 ha in a 2015 to 2018 radio-tracking study, with reproductive males averaging 148 ha

  • Check camera images for the Sandhill Dunnart's relatively large body size, dark eye rings and distinctive tail crest, while having uncertain records reviewed by a qualified zoologist.
  • Record camera placement, operating dates, trap nights, lure, image identification and independent detections.
  • Do not disturb predator dens, nests or cultural sites without the required authority.

Survey methodology: Minimum effort and reporting

  • Use a stratified design that covers all potentially suitable habitat and repeat sampling across each relevant habitat type or plant community.
  • Combine habitat searches with at least one trapping method and camera trapping where the species is reasonably likely to occur or where detection is important for a clearing decision.
  • Document the area surveyed, dates, nights, trap numbers, camera effort, search distances, weather, fire age, habitat measurements, personnel and animal welfare controls.
  • Treat the species as present if a reliable record, identifiable remains, hair sample or diagnostic camera image is obtained.
  • Submit confirmed records, negative survey results and the full survey effort to the relevant state or territory wildlife database, and provide records required by the project approval and the Australian Government species information systems.
  • Retain photographs, coordinates, specimens or samples, chain-of-custody information and expert identification notes with the project records.

Before habitat disturbance: Avoid and minimise habitat loss

Treat suitable Sandhill Dunnart habitat as a potential occurrence area before clearing or construction. The species is endangered nationally and has a small, fragmented known distribution, so avoid removing dense spinifex habitat wherever practicable.

  • Avoid clearing sandy dunes, interdunes and long-unburnt Triodia hummock grassland where practicable.
  • Retain large, dense and unbroken hummocks, logs, burrows, diverse shrub cover and connected areas of mallee or woodland.
  • Keep retained habitat connected across the project area and avoid isolating small patches between roads, compounds, cleared corridors and firebreaks.
  • Use site-specific exclusion areas around occupied habitat and important shelter features, with the extent set by the ecological assessment, approval conditions and construction method.
  • Avoid new fragmentation, night lighting and unnecessary vehicle movement through retained sandy spinifex habitat.

Before habitat disturbance: Approvals and work planning

National status: Endangered under the EPBC Act

State listings provided: Endangered in the Northern Territory, Vulnerable in South Australia and Endangered in Western Australia

Recorded breeding period: Spring to early autumn, with mating recorded in August and September

  • Check whether the action requires referral under the Environment Protection and Biodiversity Conservation Act 1999 because the Sandhill Dunnart is nationally listed as Endangered.
  • Check state or territory threatened species legislation, fauna survey licences, scientific collection permits, animal ethics requirements, land access permissions and Aboriginal land approvals before fieldwork.
  • Prepare a fauna management plan that identifies suitable habitat, survey results, exclusion areas, clearing limits, authorised fauna handlers, stop-work triggers, release sites, injury response and reporting requirements.
  • Stage clearing so the smallest practicable area is opened at a time and retained habitat remains available if an animal is found.
  • Plan works to minimise disturbance during the breeding period. Field records include mating in August and September, pouch young in September and October, and juveniles emerging in December and January.

Before habitat disturbance: Pre-clearance preparation

  • Complete the approved pre-clearance survey before vegetation removal, using the species-specific methods and effort described in the survey methodology.
  • Mark all suitable Triodia hummocks, logs, burrows, tracks, camera detections, trap captures and other habitat features in the field and on construction plans.
  • Install exclusion fencing around retained habitat and confirmed shelter areas where it will prevent machinery entry without directing animals into the clearing zone.
  • Brief operators, supervisors and fauna spotter catchers on the species' appearance, nocturnal activity, shelter habitat and stop-work procedure.
  • Complete pre-start checks for open trenches, pits, pipes and other artificial traps, and cap or fence them when unattended.

During habitat disturbance: Daily controls and machinery coordination

Use a controlled clearing process with an authorised fauna spotter catcher present where required by approval or the fauna management plan. The aim is to prevent injury, retain shelter and respond quickly to any detection.

  • Hold a daily pre-start briefing covering the approved clearing boundary, retained habitat, active fauna controls, weather and the location of the fauna spotter catcher.
  • Inspect the work area, exclusion fencing, trenches, pits, pipes, stockpiles and machinery before work starts and at the end of each shift.
  • Clear vegetation in stages towards retained habitat, where the approved method and site conditions allow, so animals have an escape path away from machinery.
  • Keep the fauna spotter catcher in direct communication with the operator and stop machinery immediately when the spotter gives the agreed stop signal.
  • Do not allow machinery to enter marked habitat, exclusion areas or areas awaiting fauna inspection.

During habitat disturbance: If a Sandhill Dunnart is found

Activity: Nocturnal, so daytime clearing may encounter animals resting in hummocks, burrows, logs or other shelters

Handling risk: The species is fast-moving, easily stressed and capable of delivering a sharp bite

Preferred release setting: Nearby retained sandy habitat with suitable spinifex cover and shelter

  • Stop work within the nominated exclusion area and keep people, vehicles and dogs away from the animal.
  • Allow the fauna spotter catcher to assess the animal from a distance and minimise handling because the species is easily stressed and can bite.
  • Only an appropriately trained and authorised person may capture, handle, mark, sample or relocate the animal.
  • Release an uninjured animal into nearby retained habitat that matches the occupied sandy spinifex conditions, at a location approved in the fauna management plan and away from active machinery.
  • Do not move a female with pouch young unless a veterinarian or the relevant wildlife authority directs the action, because handling stress may cause abandonment or loss of young.
  • Place an injured or heat-stressed animal in a secure, quiet container and contact the nominated wildlife veterinarian, wildlife carer or government wildlife authority immediately.
  • Do not resume work in the exclusion area until the fauna spotter catcher and environmental officer confirm that the animal has left or that an approved response has been completed.

During habitat disturbance: Habitat features, stop-work and records

  • Treat dense Triodia hummocks, logs, burrows, bark, fallen timber and other identified shelter sites as fauna habitat features during clearing.
  • Inspect hollow logs, timber piles, pipes, trenches, rock piles and burrows before they are moved, crushed, buried or filled.
  • Stop work for any Sandhill Dunnart detection, injured native animal, occupied burrow, unexpected habitat feature, escaped animal, damaged exclusion fence or breach of an approval condition.
  • Record the date, time, coordinates, habitat, fire age, detection method, animal condition, sex and age class where known, handling, release or treatment, photographs, personnel and machinery involved.
  • Report any death, injury, unauthorised clearing or other incident to the project environmental officer and the relevant regulator within the time required by the approval or management plan.

After habitat disturbance: Post-clearance checks

Post-clearance work must confirm that no animals remain in the disturbance footprint, maintain retained habitat controls and test whether the project has affected nearby habitat or records.

  • Complete a final inspection of the cleared footprint, trenches, spoil areas, material stockpiles and retained habitat edges before demobilising fauna controls.
  • Search for tracks, burrows, nests, scats, remains and trapped animals in the footprint and report any sign that was not documented during clearing.
  • Check that all exclusion fencing, trench covers, fauna ramps and other controls remain intact until the area is safe and the environmental officer authorises removal.
  • Confirm that no animal has been isolated by temporary fencing, access roads, spoil placement or changes to drainage.

After habitat disturbance: Monitoring and reporting

Monitoring context: The species can have a mean home range of about 70 ha, so monitoring should not be limited to the cleared footprint

Important habitat condition: Dense lower vegetation and long-unburnt habitat are associated with foraging and shelter

  • Monitor retained habitat and any known detection sites using cameras, track searches or the approved trapping method when required by the fauna management plan or approval.
  • Repeat monitoring after major fire, construction disturbance or rehabilitation works, because the species selects dense, long-unburnt foraging habitat and can have large home ranges.
  • Compare monitoring effort and detection rates with the pre-disturbance survey, including trap nights, camera nights, search distance, habitat condition and fire age.
  • Report all confirmed records, mortalities, injuries, relocations, control failures and monitoring outcomes to the relevant state or territory wildlife database and approval holder.
  • Review the fauna management plan if monitoring shows loss of spinifex cover, increased predator activity, isolation of retained habitat or repeated animal interactions with the works.

After habitat disturbance: Offsets and residual impacts

  • Calculate residual impacts after avoidance, minimisation, rehabilitation and monitoring measures have been applied.
  • Apply any offset, conservation payment or other compensatory measure required by the EPBC Act approval, state approval or project conditions.
  • Direct residual-impact measures towards protecting and managing suitable sandy Triodia habitat, predator control, fire management and connectivity between known or potential populations where this is required by the approval.

Rehabilitation actions: Rebuild habitat structure

Rehabilitation should recreate the sandy, structurally complex habitat used for shelter and foraging. It should be planned for long-term habitat development, not only for short-term ground cover.

Indicative shelter feature: Triodia hummocks with a 90th percentile height above 40 cm have been associated with breeding and detection in South Australian studies

Habitat structure: Dense lower vegetation, shrub diversity, logs and vertical complexity provide shelter and may support invertebrate prey

  • Reinstate sandy soil profiles and dunes where excavation or vehicle movement has removed or compacted them.
  • Re-establish locally appropriate Triodia species and retain natural regeneration of mallee, eucalypt, Callitris, Acacia and other native shrubs recorded at the site.
  • Aim to develop dense, connected hummock cover, with monitoring against the site-specific reference condition rather than relying on planted stem numbers alone.
  • Retain and reinstate suitable fallen logs, woody debris and other natural shelter features where they are free of weeds, pathogens and contaminants.
  • Do not clear or burn recovering hummocks to make the site appear tidy.

Rehabilitation actions: Fire, weeds and predators

  • Protect rehabilitating Triodia from unplanned fire and use a fire regime that maintains a mosaic of long-unburnt, dense habitat rather than creating large recently burnt areas.
  • Obtain ecological and cultural fire advice before any prescribed burn near known or potential Sandhill Dunnart habitat.
  • Control weeds that alter the spinifex, shrub and mallee structure, using methods that do not damage retained hummocks or contaminate fauna shelter.
  • Manage feral cats and foxes in accordance with the approved threat management programme, and monitor both predator activity and Sandhill Dunnart response.
  • Assess rabbit, goat and other introduced herbivore impacts where grazing is preventing Triodia and shrub recovery.

Rehabilitation actions: Monitoring and success measures

Useful habitat benchmark: Triodia cover above 25% and 90th percentile hummock height above 40 cm are reported indicators of suitable habitat, but local reference conditions must also be considered

Monitoring limitation: Low detection rates mean that non-detection alone should not be used as the only rehabilitation success measure

  • Inspect rehabilitation at least during establishment and after major fire or rainfall events, with the frequency set by the approval and rehabilitation plan.
  • Measure Triodia species, hummock height, hummock cover, separation, shrub diversity, log availability, bare ground, weed cover, fire age and predator activity.
  • Use cameras, track searches and, where authorised, pitfall or Elliott trapping to test whether rehabilitated habitat is being used by Sandhill Dunnarts and other native fauna.
  • Set success measures that include connected sandy habitat, increasing dense Triodia cover, stable or increasing shelter features, low weed and predator impacts, and confirmed use by native fauna.
  • Continue management until the environmental officer and regulator accept that the rehabilitation meets the approved completion criteria.

Sources