James I. D. Smith – fauNature Pty Ltd


A Street Tree Renewal Program required the removal of 27 English Elms, which provided habitat for various wildlife species. Looking to minimise the impact to these fauna, theCouncil responsible (City of Unley) had a pre-removal assessment undertaken and management plans developed to deal with the resident wildlife.  Thirteen Common Brushtail and Common Ringtail Possums were trapped, assessed and dealt with during tree removal.  Planning investigations for the project revealed that habitat structures such as hollow bearing trees used by arboreal fauna are not adequately protected under South Australian legislation.  Additionally, the Australian arboricultural industry would benefit from developing guidelines on the management and handling of native fauna encountered during tree pruning or removal activities, as no formalised approach exists.


Pruning and removal are essential elements of tree management across the urban landscape.  Arboreal wildlife species are likely to be impacted by such tree management practices. Over 300 wildlife species utilise tree cavities or hollows across Australia (Gibbons &Lindenmayer, 2002), with potentially dozens of vertebrates encountered at any given locality.  The range of species utilising tree hollows may also change depending on the time of year tree management is undertaken.  There are currently no industry guidelines in Australia which detail inspection measures or handling practices forwildlife, during pruning and tree felling operations.

In the United Kingdom (UK) legislation covers the protection of native wildlife species, as well as the structures or places that the wild animals occupy (Wildlife and Countryside Act 1981).  The interpretation of this legislation and how it translates into practice is illustrated across the industry.  A typical example from an arborist’s website states “it is illegal to remove or destroy a nest while it is being built or used…” (Norfolk Trees, 2014).  The legislation places the responsibility for wildlife, whether killed, injured or interfered with (intentionally or recklessly) on the individual who caused the offence.

In South Australia legal protection for native fauna is mostly dealt with through two Acts:

  • National Parks and Wildlife Act 1972
  • Animal Welfare Act 1985.

Relevant objectives of the National Parks and Wildlife Act (NPW Act) set out protections for individual native species and some rules around how humans interact with native wildlife.  An omission of the NPW Act, as it was drawn up 40 years ago, was to not protect the habitat of wildlife specifically (e.g. hollow bearing trees), but rather to focus on the animals (and plants) themselves.  Protection of individual animals is extended further when considering the Animal Welfare Act (AW Act), which guides the humane treatment of allanimals in the state; whether wild, domestic, feral, native or introduced.  Under the AW Act it is an offence to knowingly or recklessly harm, injure, kill or cause suffering to an animal.  If a person could have reasonably expected to find an animal when felling a tree they must take steps to mitigate the harm.  Where an animal is present and likely to be harmed by the proposed activity

  • The animal can be moved if it is in immediate danger and such actions are in the animal’s best interest,
  • A permit is required to trap and release or kill the animal.

Two other Acts potentially affect tree removal activities, the Native Vegetation Act 1981 (NV Act) and the Development Act 1983 (DVP Act).  The NV Act protects native vegetation but is rarely used in urban situations and does not apply in the case of non-native trees.  The DVP Act covers “damaging activities” to significant and regulated trees. This Act also recognises significant trees as important for their habitat value, however it does not explicitly safeguard wildlife dwelling in a tree or indicate the methods by which animals are to be extracted or removed.  Hence the current legislation in South Australia is open to interpretation and wildlife may be accidentally injured or killed as a consequence.

This paper outlines a case study of the procedures used by The City of Unley, Adelaide (South Australia), to manage resident wildlife affected during the removal of a stand of mature English Elms (Ulmusprocera) growing along a suburban street in Unley.

Native wildlife was reported to use the Elms.  The trees were in poor health, despite management over a number of years, and presented a risk to life and property.  A Street Tree Renewal Project incorporating water sensitive urban design had been scoped and the project planning was advanced when concerns arose for the resident wildlife.  These concerns were driven by council arboricultural staff dealing with the project and the public’s perception about the removal of the trees and the potential effect on the trees’ resident wildlife.  Wildlife habitation had not previously been considered, however The City of Unley was looking to manage the animals likely to be affected, in a responsible manner and suggested the UK model could positively influence their own approach.


The survey/tree management location was situated approximately 5km south of the Adelaide business district.  The 33 elms were reasonably evenly spaced along a 350m north-south oriented street; electricity infrastructure ran along the northern side of the street, and the trees grew within the southern nature strip.

Various methods for assessing use of hollows by native fauna have been established, including techniques such as “stag watching” (Smith et al., 1989) and physical inspection.  An arboreal inspection was likely to be most accurate, particularly given that Common Ringtail Possums (RTP; Pseudocheirusperegrinus) are typically missed using the Stag Watching technique, as this species tend to emerge after dark (Lindenmayer 1991b).

Wildlife Survey

The City of Unley provided an arboricultural report on the state of the elms (Thornton, C. S, 2012), as background.  An aerial analysis of the surrounding streetscape was also undertaken (Google Maps) before visiting the site.  The last phase of the pre-arboreal inspection involved an on ground evaluation, which enabled distances, infrastructure, connectivity and non-target vegetation to be assessed and recorded.

An elevated work platform (EWP) was employed to provide access to the tree hollows.  The equipment used for inspecting the hollows included a tape measure, camera and torch.  The author identified wildlife; confirmation was provided by the South Australian Museum (P. Horton pers. comm.) for eggs recorded in two of the nests.

The EWP operator was briefed and each tree was first reviewed from the ground to identify hollows and then aerial investigations were undertaken.  The route taken to each hollow was left to the EWP operator’s discretion.  The information recorded included the tree number and for each hollow: entrance size, hollow depth (where possible), signs of usage, resident wildlife and associated photographs.  Following completion of the first tree, each tree moving up the street was assessed in a similar manner as time and access allowed.

Tree Removal and wildlife management

In moving onto the second phase of the project, a practical guide was developed to manage any wildlife encountered during the felling process.

Two Department for Environment, Water and Natural Resources (DEWNR) permits were sought for possums, the taxa considered to most likely require management during the tree removal process.  One Permit to Trap and Release (Possum: D29653; Appendix A) approved the release of four Common Brushtail Possums (BPT; Trichosurusvulpecula) and six RTP.  Approval was also received for a Trap and Destroy permit (Possum: D29654; Appendix B), for up to eight BTP and five RTP.   An additional Trap and Destroy permit was approved following the commencement of the project.

In addition to the EWP there was some specific equipment of assistance in locating and securing the wildlife, which included:

  • Torch
  • Sounding Hammer
  • Welding Glover
  • ·Towel/rags (various sizes)
  • ·Catching Net
  • Holding Bags/Containers
  • Compact Camera
  • Duck/Gaffer Tape
  • Tough Clear Plastic Bag

The compact camera was used to record wildlife present, including in difficult to view hollows. The torch was useful to gauge the depth of hollows and helped determine whether any resident wildlife were present.

A council arboricultural team (AbT) undertook the tree removals.  The original wildlife survey was provided and the author gave an on-site briefing.  A visual inspection was initially undertaken from the ground to identify any hollows that may contain vertebrate wildlife.  The approach to inspecting an individual tree was left to the arborist; generally for smaller trees all the hollows were inspected, while larger trees were divided into workable sections, before chainsawing/felling started.  All hollows with entrance diameters over 15-20mm were considered as potential wildlife (e.g. microbats) refuges, although hollows with a diameter greater than 45mm were recognised as more likely.  In limbs/trunks where there were no hollows or where it could be established there was no resident wildlife – pruning/felling took place in the conventional manner.

The majority of the wildlife encountered during the second phase of this project were possums, the focal species for this aspect of the paper.  Most possums retreated into thedepths of their hollow, when inspection or felling operations commenced.  Once a possum was identified within a hollow, the following five tasks were carried out:

The position of the possum and depth of the hollow was determined.

The possum was secured within the hollow. Initially this was achieved by blocking the hollow with a towel or rag wedged into the entrance. For shallow hollows (<500mm depth), this isolation procedure was all that was required.  In deep hollows the towel/rag was progressively moved into the cavity, thereby exposing excess timber; the excess was then removed, while ensuring the safety of the animal.  By progressing in this way, and cutting small sections from the end of the limb/trunk, the cavity was reduced to 400-500mm in length, complete with a possum safely retained within it.

Table 1: Number and type of hollows recorded during the initial survey (21/10/2013).

Tree No. All Hollows Vertebrate Hollows Active Hollows
4 4 4 2
5 9 7 5
6 9 2 2
7 6 5 4
8 8 5 4
10 9 6 3
11 4 2 2
12 3 1 1
15 1 0 0
16 1 1 0
17 6 6 3
18 1 1 1
19 3 3 3
20 2 2 1
21 3 3 2
24 4 0 0
25 4 2 0
26 3 2 1
27 7 4 3
28 3 0 1
29 4 4 3
30 5 2 2
31 7 6 5
32 10 6 3
33 8 2 2
No. of Hollows 124 76 53
Average/Tree 5.0 3.0 2.1

With the possum confined in the cavity, a solid wooden section below the hollow was identified (using a sounding hammer). The limb was sawn through, ensuring there was at least 150-200mm of solid wood between the possum cavity and the cut.

Once the hollow section was cut, it was steadily lowered to the ground. On the ground, the wedged towel/rag was removed from the hollow, then a layer of thick cloth was laid over the entrance and duck taped into position. This was to aid air circulation, while securing the possum/animal within the cavity.

Finally the hollow was set aside in a protected, shady location for the possums’ subsequent assessment.

Possums that escaped ahead of felling activities were let go.  Where a possum took refuge in the top of the tree that was being felled, activities ceased.  If the escapee moved to the ground or a different tree, felling operations continued, however the possum’s location was noted, as reference for future activities.

Prior to commencing the felling operation, contact was made with several wildlife carers and a veterinary practice recognised for dealing with native fauna.  The project was explained and preparations were made to care for captured wildlife or euthanase animals, if release was not an option on welfare grounds (including under the NPW Act).


Wildlife Survey

The arboriculture report stated the majority of the trees were in poor condition with a life expectancy of less than 10 years (Thornton, C. S., 2012).  The aerial review and on-site ground assessment indicated that there was limited arboreal connectivity, other than that offered by the elms themselves.  The only exceptions to this were provided by good shrub/tree cover at the eastern end of the street (trees 1-6) and between trees 11-15.  Further west down the street there was insufficient vegetation to potentially provide for possums’ habitat needs. The wildlife survey was conducted on the 21 October 2013.

Considerable variation was seen in hollow development of the elms along the residential street.  Due to time, access and funding constraints, 25 of the 33 elms were surveyed (75%).  The assessed trees were considered representative of the elms along the length of the street.   The surveyed elms contained 124 hollows (“All Hollows” Table 1).  Some cavities were likely to have been missed due to challenges with access, however it was estimated that >90% were recorded.  At least one hollow was recorded in each tree; tree 32 exhibited the highest number with 10 cavities; an average of 5 hollows per tree were seen.

Seventy-six “Vertebrate Hollows” (cavities large enough to accommodate the anticipated vertebrate species e.g. possums and parrots) were recorded across the surveyed trees, at an average of 3.0 per elm.  “Active Hollows” were identified as those where recent animal activity had been noted (chewing, feathers, trapped fur, etc.); these accounted for 54 of the Vertebrate Hollows (70%).  Hollows occupied by bees, covered in spider webs or showing no signs of use were considered inactive.

Twenty cavities (26%) of the identified Vertebrate Hollows were occupied during the survey, by five different wildlife species (Graph 1).  These animals were spread across 15 (60%) of the elms surveyed.  Individual BTP were recorded in seven (7) different hollows, spread the length of the avenue.  RTP were observed in two trees; one included a mother and its joey, the second a single adult.

Four active Rainbow lorikeet (Trichoglossushaematodus) nests were observed, including two with eggs/young.  Two nests containing eggs and/or chicks were also recorded for Adelaide Rosellas (Platycercuselegansadelaidae) and Wood Ducks (Chenonettajubata). Three cavities were occupied by European honey bees (Apismellifera), and a fourth was uninhabitable due to the presence of an abandoned hive.

Tree Removal and Wildlife Management

Approval of the DEWNR Trap and Release and Trap and Destroy permits enabled the felling procedure and management of associated wildlife to commence.  The City of Unley had agreed to retain six elms, with the remaining 27 trees scheduled for removal. Tree removal commenced on the 11 March 2014 and was completed on the 7 April.

A total of 13 possums were removed, from 10/27 elms (37%).  The captured animals were spread along the length of the street, across the available habitat.  The majority of possums were caught in substantial, better quality elms with good connectivity, however some animals were recovered from poor quality, isolated trees.

The trapped possums were released or euthanased depending on their state of health and proximity to remaining habitat.  Five BTP and two RTP were captured and subsequently released, based on the Permit to Trap and Release (Table 2)

The hollow sections containing the possums were secured in a shady location until nightfall.  The animals were usually released within 60 minutes of sunset, at the base of a retained elm.  Released possums climbed into trees and all appeared in good condition.  Six BTP were euthanased by a veterinary surgeon under the Permit to Trap and Destroy (Table 2).  No Ringtails required euthanasia.

During tree removal, four additional BTP possums were observed.  Two retreated through the canopy ahead of the felling process and the AbT disposed of two recently deceased animals (pre-dating the tree removal process).

No nesting birds or roosting bats were observed using hollows during the felling process.  The only additional wildlife specimen recovered, was one abandoned Wood Duck egg.

A report on the possums released and those destroyed under the three NPW Act permits were completed and returned to DEWNR in accordance with the permit requirements.


The management of resident wildlife affected by pruning or felling of trees is conducted with varying degrees of rigour depending on the arborist, client, reason for removal, legislation and the public profile of the project.  The described Street Tree Renewal Project was delivered under considerable public scrutiny.  The AbT and The City of Unley complied with all legal requirements and extended their approach to ensure the welfare of the affected animals was addressed.

Native species had been observed using the elms, however no quantitative data were available.  The arboreal survey confirmed the City of Unley’s impression regarding hollow usage and identified the number and potential value of cavities found across the stand.  Seventy-six of the 124 recorded tree hollows potentially provided roosting and denning sites for native vertebrate species.  Of those hollows, 53 (60%) showed signs of recent activity and a total of 20 cavities were occupied by six different taxa: BTP, RTP, rainbow lorikeets, Adelaide rosellas, wood ducks and honey bees.  Three feral honey bee hives were destroyed in preparation for tree removal.  BTP are considered rare under the NPW Act in natural woodlands across much of South Australia, though they can be quite common in Adelaide’s suburbs.

A plan was developed to minimise the impact on this animal, together with the four other protected species likely to be affected.

The broader landscape was also assessed as part of the survey.  An aerial review indicated that once the elms had been removed, the suburban vegetation likely to provide suitable habitat for possums was mainly towards the eastern end of the street.  On ground observations reinforced the aerial assessment and confirmed there was no appropriate habitat, nor the necessary arboreal connectivity required in which to release possums, towards the western end of this residential street.

The various findings of the survey led to three key recommendations:

  • Remove the elms between March and June. This period is outside the breeding season for most bird species.  By February microbats have typically completed their breeding, so maternity roosts are unlikely to be affected and all individuals are capable of flight.  Minimising confrontations with wildlife has two major benefits; it reduces the impacts on wildlife and the time taken to deal with animals present, thereby reducing project costs.
  • Stage the tree removal, with 6-8 elms being retained, if possible. Trees selected for retention were to be based on tree health, groupings of appropriate specimens, arboreal connectivity and the practical scope of the project.
  • Produce a wildlife handling document and use this as the basis for managing any animals encountered during the felling process.

The City of Unley accepted all the recommendations and approved the retention of three elms on the central-eastern and far eastern end of the residential street.

The NPWA and the AW Act guide the manner in which wildlife is dealt with in South Australia.  The NPWA was activated when the City of Unley became aware wildlife were using the elms and these animals were likely to be affected by the planned removal.  The AW Act is triggered when an animal is encountered and needs to be managed.  South Australian legislation states “Possums must be released within 24 hours of capture (at sunset on the day of capture) and released on the same property within 50 metres of the capture site.”  If release is not an option either because the animal is sick, injured or because the carrying capacity of the area cannot sustain the animal(s), it needs to be euthanased.  It is illegal for a possum(s) to be released elsewhere.  For the western end of the residential street, release was not possible and the two remaining elm stands were insufficient to support all the possums living along the street.

The numbers of possums likely to be trapped during the tree removal phase were estimated and application for Trap and Release (four BTP and six RTP) and Trap and Destroy (eight BTP and five RTP) permits was made. Discussion with a DEWNR senior ecologist were initiated and following a site visit the plan and associated permits were approved.

The City of Unley also undertook pre-removal tree poisoning.  This was not initially recommended within the plan, but was instigated by the council arborist to prevent the elms suckering.  A potential benefit of this measure was that resident RTP moved to more favourable locations once the trees had died, but before they were removed.  Poisoning trees targeted for removal will be recommended as an integral part of the process in future.

The AbT were briefed within a matter of days of the permit approval.  The author had personal responsibility for the NPW Act permits, so undertook an initial practical session with the AbT to ensure they understood the animal handling procedures and could fulfil the permit conditions satisfactorily.  During the initial session four possums were encountered (three live and one dead) and the recommended animal handling procedure proved effective.  Having demonstrated their competence, the AbT were allowed to continue with independent removal.  Daily updates were scheduled, so any issues could be addressed and trapped animals could be dealt with promptly.

The additional equipment required for handling the wildlife encountered during tree felling is modest.  The storage bags were not used as the removed hollow tree section method easily accommodated each captured possum, however they provided a useful fall-back option.  The camera and torch were extensively used in determining the location of animals and the extent of associated hollows.  Additionally, a borescope (optical device consisting of a flexible tube with objective lens on one end, a small screen on the other linked by an optical relay) could be used – it has the benefit of longer reach and being able to see around corners; however is of limited value for large or particularly deep hollows and was not used on this project.

Wildlife carer assistance was not required during the project, however veterinary assistance was required on several occasions.  On the first day three possums were taken to a local clinic to be euthanased on welfare grounds.  The limb sections in which the possums were isolated and transported were between 1.2-1.7m in length.  This size proved unwieldy to manage and also increased the challenge of extracting the animal.  The AbT leader was briefed on this issue and the hollow log section containing the possum, were subsequently reduced to between 400-650mm total length.  This simplified handling of the section, together with possum management during the remaining tree removals.

All possums caught in the central and eastern sections of the residential street were released into the retained elms.  This included two RTP and five BTP; which exceeded the number of BTP approved for release under the permit, however subsequent discussions with a DEWNR representative confirmed this variance was acceptable given the scope of the project.  Overall the arboreal survey provided a reasonable indication of the possums that would require management during the tree removal phase.

Nesting birds were not encountered during the tree removal phase of the project.  This was considered primarily to be as a result of the season in which the tree removal took place; by March most hollow nesting birds in urban Adelaide have completed their breeding for the year.  While not considered in the original management plan, killing the elms also reduced the likelihood of small passerine bird species nesting and none were encountered during the tree removal process.  Microbats are an important component of the urban tree fauna, however none were recorded at any stage of the project.  Handling guidelines were also developed for birds and bats as part of the recommendations, however these remain untested.

If a tree is known to contain wildlife, then provisions of the NPW Act and the AW Act are triggered.  Under South Australian legislation an individual must take steps to mitigate harm if they could reasonably expect to find an animal when felling a tree.  Unfortunately, the word reasonable is open to interpretation.  This study highlights the importance of protecting habitat structures used by wildlife, such as hollow bearing trees.  The UK Wildlife and Countryside Act 1981 could be used as a model to develop new regulations or re-interpretexisting legislation regarding habitat structures.

This project found 60% of the elms assessed in the October 2013 survey contained resident wildlife and possums (13) were removed from ten (37%) of the trees felled in March-April, 2014.  Given these figures the author suggests it is reasonable to expect that for hollow bearing trees, with high connectivity, in areas known to support possums and other native fauna there is a high likelihood that wildlife could be impacted by pruning or felling activities.  Under such circumstances it is incumbent upon attending arborist to obtain a DEWNR permit(s) to “trap” such an animal(s) and they are obliged by law to treat it in a humane manner.

Wildlife enhance the value of a tree rather than being an inconvenience or a nuisance.  Integrating wildlife management considerations into a tree felling operation will modify the approach taken.  Depending on work practices, these changes may be slight or considerable.  Currently, wildlife species are accidentally injured or killed if trees scheduled for removal are not adequately inspected or appropriate measures to manage the animals that dwell therein are ignored.  The arboricultural industry is encouraged to develop guidelines so its members can plan for, assess and handle wildlife likely to be encountered when pruning or removing a hollow bearing tree.  In doing so the industry will be brought in line with existing wildlife protection provisions and be well placed to adapt to future legislative changes.


  • Gibbons, P. &Lindenmayer, D. (2002) Tree Hollows and Wildlife Conservation in Australia.
  • Lindenmayer, D. B., Cunningham, R. B., Tanton, M. T. and Nix, H. A. (1991)  Aspects of the use of den trees by arboreal and scansorial marsupials inhabiting montane ash forests in Victoria.  Australian Journal of Zoology, 39: 57-65.
  • Norfolk Trees. Wildlife Management and Protected Species. : http://www.norfolk-trees.co.uk/tree-surgeons 12 August 2014.
  • Smith, A. P., Lindenmayer, D. B., Begg. R. J., MacFarlane, M. A., Seebeck, J. H. and Suckling, G C. (1989).  Evaluation of the stagwatching technique for census of possums and gliders in tall open forests.  Australian Wildlife Research, 16:575-580.
  • Thornton, C. S. Tree Audit/Condition Assessment – Randolph Ave, FULLARTON  SA    (2012)  Treevolutions.  Adelaide


Leave a Comment