Wildlife Roadkill Mitigation Report - Lismore LGA

Wildlife Roadkill Mitigation

8 Report

Lismore Local Government Area

Sandpiper Ecological FINAL – Version 2 1/94 Main Street Alstonville 11 September 2019

sandpipereco.com.au Document Review

Date Version Status Reviewed By Represent Delivered Format Dispatched By 23/7/2018 A Internal draft D. Rohweder Sandpiper MSW B. Taylor 23/7/2018 1 Draft W. Neilan Lismore CC MSW B. Taylor 23/7/2018 1 Draft W. Neilan Lismore CC MSW B. Taylor 5/2/2019 2 Draft W. Neilan Lismore CC MSW B. Taylor

Document Distribution

Date Version Status Sent to Represent Delivered Format Dispatched By 11/9/2019 2 Final W. Neilan Lismore CC Pdf & MSW B. Taylor

Project team: Report prepared for: Dr B. Taylor (field survey, reporting) Lismore City Council Dr D. Rohweder (project management, review, mapping & analysis) Ms. Sandie Manns (data administration) Ms. Nirvarna Makings (data administration, field survey assistance)

PO Box 401 ALSTONVILLE NSW 2477 P 02 6628 3559 | E [email protected]

Cover Photo: Koala roadkill black-spot on Coraki Road, Ruthven (Photo: Sandpiper Ecological).

Disclaimer: This report has been prepared in accordance with the scope of services described in the contract or agreement between Sandpiper Ecological Surveys (ABN 82 084 096 828) and Lismore City Council. The report relies upon data, surveys and measurement obtained at the times and locations specified herein. The report has been prepared solely for Lismore City Council and Sandpiper Ecological Surveys accepts no responsibility for its use by other parties. Sandpiper Ecological Surveys accepts no responsibility or liability for changes in context, meaning, conclusions or omissions caused by cutting, pasting or editing the report. Wildlife Roadkill Mitigation Report - Lismore LGA

Executive Summary

The current report represents Stage 2 of the project. Stage 1 of the project, titled Assessment of Wyrallah Road Koala Roadkill Black-spot and Mitigation Plan, was presented to Lismore City Council in May 2018 (Sandpiper 2018). It included an assessment of the Wyrallah Road koala black-spot and a detailed mitigation plan. Both stages of the project build upon koala black-spot work completed by Biolink in 2017.

Wildlife roadstrike is recognised globally as one of the primary ecological impacts of roads and highways. On a local scale, the Wildlife Information Rescue and Education Service (WIRES) reported 2780 roadstrikes across the Lismore LGA, excluding koalas, for the period 2010-18. At least 72% of roadstrikes resulted in death of the animal. The number of native species struck throughout the LGA included 1221 mammals, 1273 birds and 253 reptiles. Macropods were the most commonly struck fauna group with 663 reported records. Other high incidence groups included possums/gliders (n = 254), parrots (n = 252), water birds (n = 91), raptors (n = 76) and snakes (n = 142).

While numerous wildlife species fall victim of roadstrike, the koala Phascolarctos cinereus, listed as vulnerable both nationally and in NSW, is arguably the species most impacted by roadstrike. Friends of the Koala (FOK) recorded 394 koala roadstrikes across the Lismore LGA for the period 2005-17 (FOK unpub. data) with 74% of roadstrikes reportedly resulting in death. Causes of roadstrike are varied but vehicle speed, visibility and driver vigilance appear to be key factors. Mitigating wildlife roadstrike has largely centered around warning signs, speed reductions and fencing. More recent innovations include vehicle-activated signs (VAS), pavement marking, virtual fencing and break-the-beam warning systems. Some of these recent innovations are included in the toolbox of wildlife road safety measures for the Lismore LGA.

To guide what measures to include in the roadstrike mitigation toolbox, roadstrike black-spots were identified across the Lismore LGA. Wildlife roadstrike data were sourced from FOK, WIRES and NSW Government’s BioNet Atlas. Spatial data were imported into ArcGIS and analysed for spatial clustering. Because of the paucity of spot locations for WIRES data and few BioNET data, the focus of black-spot identification was placed on koala roadstrike records provided by FOK. Through this process, black-spots were rated according to the intensity of koala roadstrikes/km/generation with a generation time being five years. This resulted in four major, four moderate, nine minor and five slight koala roadstrike black-spots being identified, including those identified by Biolink (2017). Black-spot locations were then inspected in the field, and a mitigation plan developed for each site. Implementation of the mitigation plan was prioritised according to the roadstrike intensity rating.

As recommended in Stage 1 of the project, the proposed basis for addressing roadstrike black-spots is to treat the road(s) in question as a management zone within which a variety of management strategies or tools can be applied. Zones are then referred to as ‘koala zone’ (or ‘platypus zone’, etc.) for specific focal species or ‘wildlife zone’ for black-spots featuring a variety of species. It is proposed, cost permitting, that all priority mitigation zones would include green pavement treatment and a low-profile rumble strip marking the entry and exit of the zone and a static ‘Wildlife Zone’ (or ‘Koala Zone’) Trivia Sign featuring a wildlife-related question at the entry point and answer at the exit point.

Sandpiper Ecological Surveys 1

Wildlife Roadkill Mitigation Report - Lismore LGA

With mitigation zones as a basis for black-spot management, a range of other mitigation tools or strategies are proposed and prioritised in accordance with the black-spot priority rating. Highest priority is given to black- spots rated as ‘major’ which occur along the Bruxner Highway at Lindendale and Goonellabah; several minor roads joining the Bruxner Highway at East Lismore; and Bangalow Road at Bexhill. Tools recommended for these zones include speed reductions, Vehicle Activated Signs (VAS), slashing roadside vegetation to improve driver visibility, fence tie-ins to direct fauna to under-road culverts, retrofitting culverts with dry access ledges and extra climbing mesh and jute climbing poles specifically for the retaining wall along the Bruxner Highway near Kadina Street, Goonellabah. A range of mitigation tools are also recommended for moderate-rated black- spots.

To evaluate the effectiveness or otherwise of mitigation measures, it is proposed that the koala be used as a surrogate. As more systematic data collection of other wildlife species occurs and/or datasets with good spatial precision are developed, they should be analysed and become part of the evaluation process and performance indicators developed for them. FOK, WIRES and Northern Rivers Wildlife Carers (NRWC) collect considerable data on roadstrike of koalas and other wildlife. It is proposed that LCC facilitate discussions about data capture, storage and sharing with such organisations. This will enable more robust evaluation and analysis of roadstrike information in the future. Central to this is the need for greater spatial precision of roadstrike locations. Several Web-based and App-based approaches to roadstrike recording have been developed at various locations around the world, including Tasmania. It is recommended that LCC seek funding to develop such a system.

The performance of the mitigation plan should be reviewed annually and assessed against the following four performance Indicators:

1. Increase in proportion of vehicles travelling at or below the signed speed at locations where speed reductions or VAS have been implemented. (Data loggers required to measure speeds at least 1 month prior to installation of mitigation treatment). 2. Reduction in annual number of reported koala roadstrikes/km at roads where mitigation measures implemented compared to mean number of roadstrikes for preceding koala generation (i.e. 5 years). 3. A reduction in annual proportion of koala roadstrikes that result in death on rural roads where mitigation measures have been implemented compared to mean number of roadstrikes for preceding koala generation (i.e. 5 years). 4. A reduction in annual number of reported wildlife roadstrikes/km at locations where mitigation implemented compared to roads that do not receive mitigation. Mitigated (i.e. treatments) and un- mitigated (i.e. controls) road segments should feature equivalent survey effort and adequate ‘before mitigation’ data.

In reviewing the performance of the WRMR, corrective actions will need to be considered at locations where roadstrike has not reduced in response to mitigation efforts. In this regard, the plan should be regarded as adaptive and able to respond to broad changes in road and/or landscape conditions (e.g. bushfire) and the emergence of other black-spot location(s) across the LGA. Moreover, analysis of the most recent data should be conducted prior to expenditure on any new black-spot infrastructure to ensure that locations featuring the highest roadstrike intensity are being prioritised. The plan should also be adaptive to the emergence of alternative mitigation technologies and new information on the performance of road mitigation strategies at other locations.

Sandpiper Ecological Surveys 2

Wildlife Roadkill Mitigation Report - Lismore LGA

Acknowledgements

Angus Underwood and Wendy Neilan from Lismore City Council are thanked for project management support. We would like to thank Ros Irwin from Friends of the Koala (FOK) and Barb Wilkins and other carers from WIRES for generously sourcing and providing roadstrike data. Scott Hetherington from Tweed Council is thanked for sharing his experiences and insights on koala roadkill mitigation strategies in the Tweed LGA. Cheryl Cochrane from Northern Rivers Wildlife Carers also generously provided road mortality records although, due to the timing of receipt of the data, it was not analysed but is included in the report Appendices.

We also wish to thank Cath Dexter (Redlands Council), Jo Green (Byron Council), Matt Woods (Ballina Shire Council), Ian Morphett (Myall Lakes Koala and Environment Group), Rob Appleby (Wildspy), Sean Fitzgibbon (Koala Ecology Group, University of Queensland), Amber Gillett (Australia Zoo), Jane Wickers (Brunswick Valley Landcare) for sharing their insights on wildlife road mortality and its mitigation.

Sandpiper Ecological Surveys 3

Wildlife Roadkill Mitigation Report - Lismore LGA

Table of contents

Executive Summary ...... 1

Acknowledgements ...... 3

1. Introduction ...... 8 1.1 Background ...... 8 1.2 Scope of works and structure of the WRMR ...... 8

2. Wildlife roadstrike and roadkill ...... 9 2.1 Extent of the problem ...... 9 2.2 Contributing factors ...... 11

3. Mitigating wildlife roadkill: the ‘Tool Box’ ...... 12 3.1 Roadkill mitigation tools ...... 12 3.2 Warning signs and speed reductions ...... 12 3.3 Rumble strips ...... 15 3.4 Wildlife threshold pavement marking ...... 16 3.5 Fauna fencing and crossings structures ...... 17 3.6 Virtual fencing ...... 20 3.7 WID-activated signage system ...... 21 3.8 Break-the-beam dynamic wildlife warning system...... 22 3.9 Roadside reflectors and audio deterrents ...... 23 3.10 Summing up – what are effective and feasible mitigation tools for Lismore LGA? ...... 24

4. Identifying roadkill black-spots ...... 26 4.1 Roadkill data sources ...... 26 4.2 Identifying black-spots ...... 27 4.3 Black-spot inspections ...... 31

5. Mitigating black-spots: prioritising and selecting from the tool box ...... 33 5.1 A rationale for black-spot mitigation in the Lismore LGA ...... 33 5.2 Prioritising black-spots ...... 33 5.3 Applying the ‘Mitigation Zone’ to all priority black-spots ...... 34 5.4 Extra tools for priority black-spots ...... 35

6. Identifying future black-spots and evaluating effectiveness ...... 42 6.1 Identifying future black-spots – Roadkill Recording Systems ...... 42 6.2 Evaluating effectiveness ...... 44

7. References ...... 45

Appendix A...... 49

Appendix B ...... 51

Appendix C ...... 52

Appendix D ...... 56

Sandpiper Ecological Surveys 4

Wildlife Roadkill Mitigation Report - Lismore LGA

List of Tables

Table 1: Number of vertebrate roadstrikes reported to WIRES for Lismore LGA for period January 2010 to March 2018 (Source: WIRES Inc. unpub. data)...... 10

Table 2: Localities within Lismore LGA featuring >80 reported roadstrikes (i.e. average of >10 roadstrike reports/year) and their proportion (%)...... 11

Table 3: Advantages, disadvantages and approximate costs of warning sign types...... 15

Table 4: Advantages, disadvantages and approximate costs of low-profile rumble strips...... 16

Table 5: Advantages, disadvantages and approximate costs of pavement marking...... 17

Table 6: Advantages, disadvantages and approximate costs of fencing...... 20

Table 7: Advantages, disadvantages and approximate costs of virtual fencing...... 21

Table 8: Advantages, disadvantages and approximate costs of WID-activated signage system...... 22

Table 9: Advantages, disadvantages and approximate costs of Dynamic Wildlife Warning System...... 23

Table 10: Advantages, disadvantages and approximate costs of roadside reflectors and audio deterrents...... 24

Table 11: Feasibility, cost and proven effectiveness of various roadstrike mitigation technologies. Rating metric: very low (+), low (+ +), moderate (+ + +), high (+ + + +), very high (+ + + + +)...... 25

Table 12: Koala roadstrike black-spot ratings as derived from FOK data...... 28

Table 13: Koala roadstrike black-spot site assessments...... 32

Table 14: Major and moderate koala roadstrike black-spots. X-labelled sites are from the current study and B- labelled sites are from Biolink (2017)...... 34

Table 15: ‘Tools’ proposed for priority black-spots. VAS = vehicle-activated sign...... 41

Table 16: Roadkill recording template (adapted from Californian Roadkill Observation System (CROS) developed by the University of California Road Ecology Center)...... 43

List of Figures

Figure 1: Proportion of koala roadstrikes (n = 394) reported to Friends of the Koala (FOK) for period 2005-2017 according to month of occurrence (Source: FOK unpub. data)...... 12

Figure 2: Virtual fencing - approaching car headlights activate the virtual fencing sensor which emits sound and light to deter animals from crossing the road. (Source: www.wildlifesafetysolutions.com.au )...... 20

Sandpiper Ecological Surveys 5

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 3: WID ear tag (left) and schematic of WID-activated signage system (right). (Source: www.wildspy.com.au)...... 21

Figure 4: Dynamic Wildlife Warning System...... 23

Figure 5: Swareflex roadside reflectors (left; Source: D’Angelo and van der Ree 2015) and the ShuRoo audio deterrent (right; Source: www.shuroo.com.au )...... 24

Figure 6: Wildlife roadstrike locations for Lismore LGA from various data sources. WIRES known roads represent roads identified by WIRES carers as featuring a range of wildlife roadstrike...... 29

Figure 7: Koala roadstrike intensity polylines for Lismore LGA generated from 2005-17 data from FOK. Biolink (2017) intensity polylines were generated from 2005-16 data...... 30

Figure 8: Major koala black-spot (B4) in the Goonellabah area identified by Biolink (2017)...... 36

Figure 9: Major koala black-spot (B6) in the Bexhill area identified from Biolink (2017) data...... 37

Figure 10: Moderate koala black-spots (B1 & B1X) in the south Gundurimba area identified in the current study and by Biolink (2017)...... 38

Figure 11: Moderate koala black-spot (X1) in the Wyrallah area identified in the current study...... 39

Figure 12: Moderate koala black-spot (X2) in the Coraki area identified in the current study...... 39

Figure 13: Moderate koala black-spot (B5) in the Wyrallah area identified by Biolink (2017)...... 40

Figure 14: ‘Roadkill TAS’, one of several smartphone Apps available for recording roadkill...... 43

List of Plates

Plate 1: Examples of static wildlife signs - koala sign on Wyrallah Road adjacent Mathieson Lane (left); Koala Zone speed trial sign in Redlands, Qld (middle; Source: tmr.qld.gov.au); wildlife advisory night speed sign in Tasmania (right; source: efe.com)...... 13

Plate 2: Koala Vehicle-Activated Signs (VAS) manufactured by Pioneerz Safety and installed on Bagotville Road in the Ballina LGA (left). Another version of koala VAS manufactured by HIVIS Group installed at Hawksnest on the mid-north coast (right). Signs are set to activate on a set speed (e.g. 5 km/h below the signed speed limit)...... 13

Plate 3: A constantly scrolling Variable Message Sign (VMS; not vehicle activated) manufactured by HIVIS Group and installed on Clothiers Creek Rd in the Tweed LGA (Source: S. Hetherington)...... 14

Plate 4: Highway trivia question and answer signs (each approximately 5km apart) on the Bruce Highway near Mackay, north Queensland (Source: au.geoview.info)...... 14

Plate 5: Low-profile rumble strips (Source: www.barriergroup.com.au) ...... 16

Sandpiper Ecological Surveys 6

Wildlife Roadkill Mitigation Report - Lismore LGA

Plate 6: ‘Koala Zone’ pavement marking (left) and associated VMS sign (right) on Clothiers Creek Road, Tweed LGA...... 17

Plate 7: Floppy-top exclusion fence on the Pacific Highway, Brunswick Heads (top left); 1500mm-high chain- mesh fence with flashing at Scrub Road, Brisbane (top right); temporary floppy-top on star pickets on the Old Pacific Highway, Urunga (bottom left); and chicken wire on star pickets on the Pacific Highway, Chinderah (bottom right)...... 18

Plate 8: The Pacific Highway upgrade features an array of crossing structures including a box culvert at Glenugie (top left) and a land bridge near Chinderah (top right) targeting terrestrial fauna and a rope bridge (bottom left) and glide poles (bottom right) at the Sapphire to Woolgoolga section targeting arboreal fauna...... 19

Plate 9: Retrofitted ledges under bridges provided dry access for koalas in Brisbane (left; Dexter et al. 2016). Two 1050mm concrete pipes (and frog exclusion fence) along the Pacific Highway at Woolgoolga were used by over 30 species of wildlife (right)...... 19

Sandpiper Ecological Surveys 7

Wildlife Roadkill Mitigation Report - Lismore LGA

1. Introduction

1.1 Background

Sandpiper Ecological Surveys (Sandpiper) was engaged by Lismore City Council (LCC) to develop a Wildlife Roadkill Mitigation Report (WRMR). The need for a WRMR was identified in the Biodiversity Management Strategy (BMS) for the Lismore Local Government Area 2015-2035. The BMS highlighted the need to identify roadkill blackspots in the LGA and to develop an integrated program of works to implement a toolbox of wildlife road-safety measures to mitigate wildlife roadstrike, with a focus on koalas and other threatened fauna. In doing so, the WRMR would build on the Koala Habitat and Population Study (Biolink 2017) which analysed koala roadstrike within the area covered by the Lismore Comprehensive Koala Plan of Management (CKPoM), an area of approximately 20 000 ha in the south of the Lismore Local Government Area (LGA).

The objectives of the WRMR are:

• To provide detailed recommendations and designs for mitigation measures to be implemented at selected roadkill black-spots identified in the Koala Habitat and Population Study (Biolink 2017) • To identify additional wildlife roadkill black-spots across the LGA • To provide a ‘toolbox’ of mitigation measures that LCC can use to reduce wildlife roadkill at identified black-spots • To provide recommendations for ongoing monitoring, evaluation and data collection.

1.2 Scope of works and structure of the WRMR

The scope of works for development of the WRMR features two stages:

• Stage 1: Undertake an assessment of the major koala roadkill black-spot on Wyrallah Road identified by Biolink (2017) and develop a mitigation plan tailored to the specific threat and topographical features of the site. • Stage 2: Develop a Wildlife Roadkill Mitigation Report (WRMR) I. Identify and map additional roadstrike black-spots throughout the LGA. II. Develop a ‘toolbox’ of roadstrike mitigation measures that can be implemented at roadkill black-spots along both existing roads as well as during road construction or upgrade. III. Develop a works program for implementation of mitigation measures. IV. Detail a monitoring program to evaluate the effectiveness of the mitigation measures. V. Provide recommendations for ongoing data collection to monitor wildlife roadkill to enable Council to identify future black-spots.

The final Stage 1 report titled Assessment of Wyrallah Road Koala Roadkill Black-spot and Mitigation Plan was presented to Lismore City Council in May 2018 (Sandpiper 2018). It included an assessment of the Wyrallah Road koala black-spot and a detailed mitigation plan.

The current Stage 2 report, while clearly related to the recommendations of Stage 1, should be regarded as a stand-alone document. It begins by reviewing the scale of wildlife roadstrike and roadkill, particularly locally, and the contributing factors. It then builds upon the koala black-spot work completed by Biolink (2017) and broadens it out to include other fauna groups.

The Stage 2 report then presents a toolbox of roadstrike mitigation measures that may be selectively applied for specific species or applied more broadly for a range of species. The measures are then built in to a works

Sandpiper Ecological Surveys 8

Wildlife Roadkill Mitigation Report - Lismore LGA

program for implementation and a monitoring program to evaluate effectiveness. This is then followed by recommendations for ongoing data collection to enable identification of future black-spots. 2. Wildlife roadstrike and roadkill

2.1 Extent of the problem

Wildlife road mortality (roadkill) is globally recognised as one of the primary ecological impacts of roads and highways (Rytwinski & Fahrig 2015; Bennett 2017). Within Australia, the scale of the problem is difficult to quantify but numerous studies since the 1970s have documented both the range and extent of fauna killed on roads at local scales (e.g. Vestjans 1972; Disney & Fullagar 1978; Coulson 1982; Osawa 1989; Ramp et al. 2005; Hobday & Minstrell 2008). Whereas most studies note that common species are most typically hit, several studies suggest that some threatened species may suffer adverse population effects from road mortality. For example, a road upgrade in central Tasmania was associated with a dramatic decline in the local populations of eastern quolls Dasyurus viverrinus and Tasmanian devils Sarcophilus harrisii (Jones 2000). Road mortality reportedly threatens the persistence of koalas Phascolarctos cinereus in south-east Queensland (Dique et al. 2003) and major roads are described as a threat to the persistence of an endangered population of long-nosed bandicoots Perameles nasuta within Sydney (Scott et al. 1999). On a national scale, it has been suggested that over nine million kangaroo and wallaby road deaths occur each year across Australia’s 810,600 km of sealed and unsealed roads (Burgin and Brainwood 2008).

Within north-east NSW, a study conducted during 2000-01 reported 529 vertebrate roadkills of 53 species along a 100km circuit featuring the Pacific Highway, Bruxner Highway and Bangalow Road (Taylor & Goldingay 2004). Eighty-seven of these records occurred within the Lismore LGA. Bandicoots (long-nosed bandicoot Perameles nasuta & northern brown bandicoot Isoodon macrourus) and brushtail possums (short-eared brushtail possum Trichosurus caninus & common brushtail possum T. vulpecula) were the most commonly recorded species. Interestingly, during the 20 weeks of surveys conducted between July and January, only one koala roadkill was recorded, and it occurred outside of the Lismore LGA (Taylor & Goldingay 2004).

Databases maintained by Wildlife Information Rescue and Education Service (WIRES) and Friends of the Koala (FOK) provide a more indicative temporal picture of the scale of roadstrike within the Lismore LGA. WIRES reported 2780 roadstrikes, excluding koalas, for the period 2010-18 (Table 1). At least 72% of roadstrikes resulted in death of the animal. The number of native species struck throughout the LGA included 1221 mammals, 1273 birds and 253 reptiles. Macropods were the most commonly struck fauna group with 663 reported records. Other high incidence groups included possums/gliders (n = 254), parrots (n = 252), water birds (n = 91), raptors (n = 76) and snakes (n = 142). Only two frogs were reported for the period, highlighting the gross under-reporting that occurs for this fauna group despite hundreds if not thousands being killed in areas where roads traverse high quality frog habitat (e.g. Taylor & Goldingay 2003; Goldingay & Taylor 2006).

Sandpiper Ecological Surveys 9

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 1: Number of vertebrate roadstrikes reported to WIRES for Lismore LGA for period January 2010 to March 2018 (Source: WIRES Inc. unpub. data).

Group Taxa Reported roadstrikes Native Mammals Macropods 663 Other mammals 288 Possums & gliders 254 Fruit bats 11 Rodents 3 Microbats 1 Unknown mammals 1 Feral mammals 13 Native birds Other birds 826 Parrots 252 Water birds 91 Raptors 76 Unknown birds 27 Seabirds 1 Native reptiles Snakes 142 Lizards 77

Turtles 34 Frogs 2 Unclassified 18

Total 2780

When the WIRES data are sorted according to the 66 Lismore LGA localities/suburbs, Lismore, Goonellabah, Nimbin, Rosebank and Dunoon all feature more than 80 reported roadstrikes for the 2010-18 period or the equivalent of >10 roadstrikes/year (Table 2). This is most pronounced for Lismore (n = 574) and Goonellabah (n = 427) which feature very high numbers of reported roadstrikes. Together, the five localities account for 55% of all reported roadstrikes for the period.

Sandpiper Ecological Surveys 10

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 2: Localities within Lismore LGA featuring >80 reported roadstrikes (i.e. average of >10 roadstrike reports/year) and their proportion (%) of 2780 roadstrikes reported for the period January 2010 to March 2018 (Source: WIRES inc. unpub. data).

Locality Reported % of all Lismore roadstrikes LGA roadstrikes Lismore 574 20.6 Goonellabah 427 15.4 Nimbin 286 10.3 Rosebank 152 5.5 Dunoon 87 3.1

There is also a significant number of koala roadstrikes across the LGA. FOK recorded 394 koala roadstrikes across the Lismore LGA for the period 2005-17 (FOK unpub. data). Biolink (2017) analysed the period 2005-16 and reported that 74% of roadstrikes resulted in death of the koala. Moreover, a greater proportion of roadstrikes on rural roads resulted in a confirmed koala mortality (i.e. 83% vs 53%), presumably attributable to higher vehicle speeds on rural roads (Biolink 2017).

2.2 Contributing factors

A variety of factors reportedly contribute to the extent and location of wildlife roadkill (Taylor & Goldingay 2010). Numerous studies have determined that vehicle speed, traffic volume and presence of roadside vegetation cover are associated with higher levels of roadkill (e.g. Osawa 1989; Fahrig et al. 1995; Jones 2000; Freitas et al. 2015). Vehicle speed is a major contributing factor largely because the risk of experiencing a collision increases exponentially with increasing vehicle speed (Huijser et al. 2015). Conversely, reducing vehicle speed lowers collision risk. For example, vehicle activated signs (refer Section 4.1.1) reportedly reduced vehicle speeds by 2-20 km/h and vehicular accidents by 33% on rural roads in the UK (Winnett & Wheeler 2002).

In Australia, reduced roadkill rates of eastern quoll and Tasmanian devil in central Tasmania was attributed to a reduction in vehicle speeds although increased vigilance from signage may have been a contributing factor (Jones 2000). Roadkill hot-spot modelling by Hobday and Minstrell (2008) across numerous other roads in Tasmania showed that vehicle speed reductions would be the most effective method for reducing roadkill numbers. Indeed, the effectiveness of speed reductions is supported by a large international review of mitigation measures for reducing wildlife roadkill (see Glista et al. 2009).

Certain patterns of behaviour and life stages make some species more vulnerable to roadstrike. Elevated risk may be associated with factors such as a preference for roadside habitat, road/roadside scavenging or increased activity during breeding and population dispersal (Mansergh and Scotts 1989; Coulson 1997; Freitas et al. 2015; Madden & Perkins 2017). For example, high numbers of frogs are killed during the breeding season while crossing roads between habitat areas (Fahrig et al. 1995; Goldingay & Taylor 2006) and numerous reptiles are killed in spring/summer while basking on the warm road surface (Vestjens 1972). The habit of eastern grey kangaroos Macropus giganteus to graze on roadside grasses and swamp wallabies Wallabia bicolor to browse on roadside vegetation heighten their risk of being struck by a vehicle (Coulson, 1982; Osawa, 1989). Further, roadkills often attract scavengers such as lace monitors Varanus varius and crows/ravens Corvus spp. which may in turn become roadkill (Vestjans 1972; Schwartz et al. 2018).

Several studies have reported strong temporal trends in roadkill intensity for some species and species groups. A review of various roadkill data for NSW showed incidence peaking during May to July and between the hours

Sandpiper Ecological Surveys 11

Wildlife Roadkill Mitigation Report - Lismore LGA

of 1700 and 2200 (Ramp and Roger 2008). Seasonality is clear for koala roadkills across south-east Qld and north-east NSW with the peak occurring from late winter to mid-summer (Dique et al. 2003; Lassau et al. 2008; Ellis et al. 2016). This is consistent with data for the Lismore LGA where koala roadstrike is especially pronounced between August and November (Figure 1; FOK unpub. data). Indeed, these four months account for 53% of roadstrike records.

16 14 12 10 8 6 4

2 Proportion of roadstrikes (%) 0

Figure 1: Proportion of koala roadstrikes (n = 394) reported to Friends of the Koala (FOK) for period 2005-2017 according to month of occurrence (Source: FOK unpub. data).

3. Mitigating wildlife roadkill: the ‘Tool Box’

3.1 Roadkill mitigation tools

The range of roadkill mitigation tools is wide and varied. To investigate this, several approaches were utilised including key word searches in Google and Google Scholar; review of relevant literature; discussions with other roadkill mitigation researchers, ecologists, FOK representatives, WIRES representative, product suppliers and local council environment officers. A range of feasible tools are detailed below and the advantages, disadvantages and approximate costs (where known) described.

3.2 Warning signs and speed reductions

Wildlife warning signs are amongst the most frequently used roadkill mitigation measure (Huijser et al. 2015). The intent is to alert drivers to the possibility of encountering wildlife on or near the road and thus make the driver more vigilant and reduce speed (Bond & Jones 2013). Standard or static warning signs have typically featured line images of wildlife against a yellow background (Plate 1). At several locations in Australia, signs promoting night/seasonal speed reductions (e.g. Redlands, Queensland) and advisory night speed reductions (e.g. Tasmania) have been used (Plate 1). In more recent years, enhanced or dynamic signs such as vehicle activated signs (VAS) and variable message signs (VMS) have emerged (Plate 2 & 3).

Enhanced warning signs are reportedly more frequently observed and recalled by drivers than standard static signs (Bond & Jones 2013; Huijser et al. 2015). Their effectiveness is reportedly further enhanced if they relate

Sandpiper Ecological Surveys 12

Wildlife Roadkill Mitigation Report - Lismore LGA

to a relatively short section of road to reduce the likelihood that a driver will forget and discontinue their vigilance (Huijser et al. 2015). Similarly, standard signs installed at numerous locations and left indefinitely will become increasingly ignored by drivers over time as they habituate to their presence (Huijser et al. 2015). A recent static sign innovation is the ‘trivia sign’ (Plate 4). The signs feature a question followed by an answer some 5km further along the road. They have been installed at several long, open stretches of the Bruce Highway in central Queensland that feature a high number of accidents attributed to driver fatigue. While there is no reported studies on their effectiveness, the concept may be applicable to wildlife black-spots and include wildlife-related trivia questions albeit their efficacy may be contingent on whether the road is mostly used by tourists or daily commuters.

Sandpiper Ecological Surveys 13

Wildlife Roadkill Mitigation Report - Lismore LGA

Plate 3: A constantly scrolling Variable Message Sign (VMS; not vehicle activated) manufactured by HIVIS Group and installed on Clothiers Creek Rd in the Tweed LGA (Source: S. Hetherington).

Plate 4: Highway trivia question and answer signs (each approximately 5km apart) on the Bruce Highway near Mackay, north Queensland (Source: au.geoview.info).

Other research on warning signs suggests their effectiveness may be improved if they are located at roadkill black-spots and targeted at a specific time of day or season, referred to as ‘temporal signs’ (Huijser et al. 2015). For example, temporal signs reportedly reduced deer roadstrike by 9-45% although their effectiveness appeared to diminish over time (e.g. Found & Boyce 2011; Sullivan et al. 2004). Conversely, a seasonal speed reduction trial using static signs to reduce koala roadstrike in Redlands (Plate 1) reported a marginal though non-significant reduction in vehicle speeds and koala roadstrike and slightly higher survivorship for koalas struck on roads with lower speed limits (Dique et al. 2003).

The combination of a variable message sign and pavement treatment on Clothiers Creek Road in Tweed LGA reportedly reduced the number of drivers exceeding the speed limit by 12% over a two-year period (S. Hetherington pers. comm.). This suggests that vehicle-activated signs and/or variable message signs could assist in reducing vehicle speeds and increasing driver vigilance at koala blackspots in the Lismore LGA and thereby reduce roadstrike and improve survivorship.

The advantages, disadvantages and approximate costs of various warning sign types is described in Table 3.

Sandpiper Ecological Surveys 14

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 3: Advantages, disadvantages and approximate costs of warning sign types.

Advantages Disadvantages Approximate Cost • Relatively low cost. • Effectiveness may • Standard static koala sign (www.hivis.com) = • Low maintenance. be short term, $103 + delivery + installation • Easily relocated. particularly for • Vehicle-activated koala sign (www.hivis.com) • Variable Message Signs static signs. = $10,714 (exc. GST; delivered) + installation may reduce likelihood of • Risk of vandalism. • Vehicle-activated koala sign habituation. (www.pioneerzsafety.com) = approx. • Vehicle-Activated Signs $13,000 (exc. GST; delivered) + installation can record vehicle ($4,000 by supplier) speed, which assists in • Koala Zone Variable Message Sign determining their (www.hivis.com) = approx. $15,000 (exc. effectiveness. GST; delivered) + installation.

3.3 Rumble strips

Rumble strips are raised rubber strips that cause a sound and car vibration when a vehicle passes over them (Plate 5). The intent of the rumble strip is to alert drivers. They are often used along road shoulders to alert errant/drowsy drivers before they run off the road. Rumble strips have been shown to substantially reduce the rate of run-off accidents and may be effective at reducing vehicle speed and thereby reducing wildlife vehicle collisions (Corkle et al. 2001).

The use of rumble strips to increase driver alertness may reduce wildlife vehicle collisions (EDI Environmental Dynamics 2015). A study conducted in a remote part of north-west Tasmania reported an average of 59% reduction in wildlife roadkill on road sections treated with series of rumble strips compared to control sites (Lester 2015). The noise generated by the numerous strips make this design most appropriate for locations away from dwellings. However, the promising results suggest this technology is worthy of further investigation.

A useful application of a scaled-down version of a series of rumble strips is the inclusion of a single strip to reinforce mitigation signage and/or pavement treatments. This approach was recommended for the Wyrallah Road black-spot (Sandpiper 2018).

The advantages, disadvantages and approximate costs of low-profile rumble strips is described in Table 4.

Sandpiper Ecological Surveys 15

Wildlife Roadkill Mitigation Report - Lismore LGA

Plate 5: Low-profile rumble strips (Source: www.barriergroup.com.au)

Table 4: Advantages, disadvantages and approximate costs of low-profile rumble strips.

Advantages Disadvantages Approximate Cost • Reduced incidence of roadkills. • Noise may disturb • $15.40/500mm strip • Low cost. nearby residents. (www.barriergroup.com.au) + • Low maintenance. installation costs.

3.4 Wildlife threshold pavement marking

Wildlife threshold pavement marking/painting is typically used to identify the start and/or end of signed significant wildlife conservation areas (Plate 6). The intent of the threshold treatment is to complement the warning signs into a significant wildlife conservation area/black-spot area and advise drivers they should modify their driving behavior and reduce their speed. The use of a rumble strip may further enhance this message.

The concept has been applied at two koala black spots within the Tweed LGA. At the Clothiers Creek Road site, post-treatment monitoring revealed an 8% reduction in vehicles driving above the speed limit (S. Hetherington, pers. comm.). This rose to an average of 12% over a two-year monitoring period after the addition of variable message signs (Plate 3). Pavement marking was recommended for the Wyrallah Road black-spot (Sandpiper 2018).

The advantages, disadvantages and approximate costs of pavement marking is described in Table 5.

Sandpiper Ecological Surveys 16

Wildlife Roadkill Mitigation Report - Lismore LGA

Plate 6: ‘Koala Zone’ pavement marking (left) and associated VMS sign (right) on Clothiers Creek Road, Tweed LGA.

Table 5: Advantages, disadvantages and approximate costs of pavement marking.

Advantages Disadvantages Approximate Cost • Relatively low cost. • Requires • Approx. $1,800 - • Consistent branding of black-spots. periodic re- $2,250/treatment (i.e. $45/m • Reinforces other treatments such as advisory application. x 40-50m2) + traffic control speed reduction signs, rumble strip.

3.5 Fauna fencing and crossings structures

Well designed and maintained fencing is an effective method for reducing wildlife vehicle strikes and directing animals to under-road crossing structures (Clevenger et al. 2001; van der Ree et al. 2015). Across eastern Australia, floppy top or 1500mm high chain mesh with 500mm wide metal flashing is commonly installed on major road upgrades to prevent wildlife road incursions, particularly by koalas (Plate 7). Within the Lismore LGA, a 2.5 km-long floppy-top exclusion fence was installed on Skyline Road in 2003 complete with an underpass and access road koala grids. However, due to fence breaches and maintenance issues, several koalas have reportedly been killed on the fenced section of road since its installation (A. Brace pers. comm.). Cheaper versions of the steel post fence include floppy-top fence on star pickets and chicken wire suspended on star pickets (Plate 7). Interestingly, chicken wire fencing was reportedly effective in preventing koala roadstrikes along sections of the Pacific Highway at Chinderah (Phillips 2014).

Despite the obvious benefits of fencing in preventing roadstrike, long sections of fence are a barrier to wildlife movement and landscape permeability and should only be installed in conjunction with crossing structures such as land bridges, bridge underpasses and box or pipe culverts for terrestrial fauna, rope bridges targeting arboreal fauna and glide poles specifically targeting gliders (Plate 8). The accumulating evidence confirms that land-bridges and box culverts are used by a range of terrestrial fauna from snakes and rodents up to large kangaroos and koalas (e.g. Taylor & Goldingay 2003; Bond & Jones 2008; Smith et al. 2015); possums, gliders and brush-tailed phascogales Phascogale tapoatafa have been recorded using rope bridges; and several species of glider have been recorded using glide poles (e.g. Sandpiper 2017; Goldingay et al. 2018).

Whereas retrofitting box culverts and/or land bridges into an existing road is prohibitively expensive, several less expensive approaches are feasible. For example, terrestrial fauna may be funneled towards existing road

Sandpiper Ecological Surveys 17

Wildlife Roadkill Mitigation Report - Lismore LGA

underpasses (i.e. box culverts, pipes and bridges) by tying-in short sections of fence. Fences could consist of a standard stock fence with 50mm mesh. Access through wet culverts/pipes may be enhanced by installing raised dry-access ledges (Plate 9). Such an approach was reportedly effective in providing safe passage across busy roads in Brisbane for koalas (Dexter et al. 2016). Moreover, concrete pipes, particularly if only periodically wet, may provide safe under-road crossing for wildlife (Plate 9). Indeed, over 30 species ranging in size from frogs to swamp wallabies regularly used 1050mm diameter dry-access pipes to cross under the Pacific Highway at Woolgoolga (Sandpiper 2018). Ropes (40-60mm diameter) slung between large trees each side of the road may prove effective for possums.

The advantages, disadvantages and approximate costs of fauna fencing and crossing structures is described in Table 6.

Plate 7: Floppy-top exclusion fence on the Pacific Highway, Brunswick Heads (top left); 1500mm-high chain-mesh fence with flashing at Scrub Road, Brisbane (top right); temporary floppy-top on star pickets on the Old Pacific Highway, Urunga (bottom left); and chicken wire on star pickets on the Pacific Highway, Chinderah (bottom right).

Sandpiper Ecological Surveys 18

Wildlife Roadkill Mitigation Report - Lismore LGA

Sandpiper Ecological Surveys 19

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 6: Advantages, disadvantages and approximate costs of fencing.

Advantages Disadvantages Approximate Cost • Well designed and maintained fauna • Fencing is a barrier to • Costs highly fencing is effective in preventing movement/landscape permeability in the variable and wildlife access to roadway. absence of crossing structures. contingent • Fencing combined with crossing • Retrofitting crossing structures into an on method, structures maintains habitat existing road is very expensive. location and connectivity. • Loss of fence integrity if gates left open or materials • Low risk of theft/damage. grids breached. used. • Short sections of fence tie-ins to • Requires ongoing maintenance to repair existing culverts/pipes relatively breaches, clear debris. cheap. • Resistance by local landholders to • Retrofitting a dry ledge into existing fence/gate/grids along property culvert provides dual purpose. boundaries.

3.6 Virtual fencing

Virtual Fencing is promoted as an active electronic protection system that aims to prevent animals from crossing the road when a vehicle is approaching at night. The devices are activated by approaching headlights causing them to emit sound and light stimuli intended to alert, repel and prevent animals from entering the road (Figure 2). The supplier recommends a device every 25m for the length of the road treatment.

Small-scale trials recently completed in western Tasmania reported reductions in small macropod road mortality by 50% (Fox et al. 2019). Considering such results, the devices are worthy of trials at other locations.

The advantages, disadvantages and approximate costs of virtual fencing is described in Table 7.

Sandpiper Ecological Surveys 20

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 7: Advantages, disadvantages and approximate costs of virtual fencing.

Advantages Disadvantages Approximate Cost • Demonstrated • Relatively new technology with limited trials. • $7,800/km (40 reductions in small • Theft of units. units) + macropod roadkill. • Possible habituation by animals residing in the installation + • Low maintenance. road corridor near the units. posts. • Untested on wide range of fauna.

3.7 WID-activated signage system

This signage system is based on catching and placing Wireless Identification (WID) ear tags on koalas residing near a roadkill black-spot. If a tagged individual moves close to the road edge, dataloggers detect their presence and activate nearby warning signs (Figure 3). Warning signs can also feature audio (e.g. an alarm signal). The system is distributed by Brisbane-based company Wildspy (www.wildspy.com.au).

The advantages, disadvantages and approximate costs of WID-activated signage system is described in Table 8.

Figure 3: WID ear tag (left) and schematic of WID-activated signage system (right). (Source: www.wildspy.com.au).

Sandpiper Ecological Surveys 21

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 8: Advantages, disadvantages and approximate costs of WID-activated signage system.

Advantages Disadvantages Approximate Cost • Highly targeted to • Requires capture of local koalas (and • $40,000 – $100,000/km koalas. recapture at end of WID battery life). for signage & • Reduced likelihood • System only applies to tagged individuals. datalogger (depends on of driver habituation • Battery life of WID ear tags ranges between type and frequency of to signage. 0.5 – 2.8 years depending on frequency of signs and frequency of • Field trials transmission. dataloggers). demonstrate • Would require inclusion of accelerometers • Add cost of installation dataloggers effective in ear tags so tagged koalas resting in • Add cost of in detecting WID roadside trees within range of a datalogger capture/tagging of tags. do not continually activate sign. koalas which may be • Activated sign feature of system yet to be substantial depending trialed so effectiveness unknown. on number of • Theft of data loggers. individuals.

3.8 Break-the-beam dynamic wildlife warning system

The technology works on roadside sensors detecting animal movement. Two different options are available:

1) Roadside fencing directs animals to designated crossing locations. Sensors at these crossing locations detect animal movement and activate LED warning signs which alert drivers to animals crossing (Figure 4). 2) In the absence of roadside fauna fencing, a series of sensors are placed along the roadside to create a ‘virtual’ fence. When an animal is detected by a sensor it activates an LED warning sign.

The company promoting the technology claim reductions in wildlife road mortality in the order of 95% at locations in Europe where it has been installed, presumably for option 1. No information was available about option 2 trials or effectiveness. It is yet to be trialed in Australia. The system is distributed by German company Prowild (www.prowild.net).

The advantages, disadvantages and approximate costs of break-the-beam dynamic wildlife warning system is described in Table 9.

Sandpiper Ecological Surveys 22

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 4: Dynamic Wildlife Warning System. Schematic Zone Detection (Option 1 - top) consists of roadside sensors that detect animal movements at designated crossing locations/gaps in fauna fence which in turn activate warning signs alerting drivers. Schematic line detection (Option 2 - bottom) relies on a virtual fence rather than a fauna fence (Source: www.prowild.net ).

Table 9: Advantages, disadvantages and approximate costs of Dynamic Wildlife Warning System.

Advantages Disadvantages Approximate Cost • May be • Requires fencing of road corridor (option 1). • Approx. $90,000/km appropriate for a • Requires ongoing maintenance to retain clear line- (option 2) range of species, of-sight between sensors (option 2). • Uneven and winding including koalas. • Unmaintained roadside vegetation may cause false- roads require a • Capture of koalas triggers. higher frequency of not required. • Not yet trialed in Australia so effectiveness for sensors which koalas unknown. increases costs. • Theft of sensors.

3.9 Roadside reflectors and audio deterrents

Other mitigation technologies include roadside reflectors and car-mounted audio deterrents. Reflectors are designed to deter animals from attempting road crossings by deflecting light from oncoming vehicle headlights into the roadside habitat to provide a visual warning (Figure 5). Wildlife audio deterrents, such as ShuRoo, mount to the front of the vehicle and emit high frequency sound meant to alert and drive off animals on the roadway (Figure 5).

The manufacturers of ShuRoo promote a range of anecdotal evidence of their effectiveness in reducing kangaroo road strike on their website. However, there is no published scientific evidence supporting the effectiveness of audio deterrents or roadside reflectors in reducing wildlife road mortality (D’Angelo and van der Ree 2015). Moreover, we are unaware of any trials of either type of technology specifically targeting koalas.

The advantages, disadvantages and approximate costs of roadside reflectors and audio deterrents is described in Table 10.

Sandpiper Ecological Surveys 23

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 10: Advantages, disadvantages and approximate costs of roadside reflectors and audio deterrents.

Advantages Disadvantages Approximate Cost • Relatively low cost. • No robust evidence supporting their effectiveness. • Approx. $7,600/km • Relatively low • Audio deterrents require uptake by drivers. (Swareflex). maintenance. • Theft of reflectors. • $565/unit (ShuRoo).

Figure 5: Swareflex roadside reflectors (left; Source: D’Angelo and van der Ree 2015) and the ShuRoo audio deterrent (right; Source: www.shuroo.com.au ).

3.10 Summing up – what are effective and feasible mitigation tools for Lismore LGA?

The mitigation technologies described above offer a range of approaches aimed at reducing wildlife road mortality. Their respective feasibility, costs and proven effectiveness vary considerably and are summarised in Table 11. The relative merits of each are discussed according to these three criteria.

Sandpiper Ecological Surveys 24

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 11: Feasibility, cost and proven effectiveness of various roadstrike mitigation technologies. Rating metric: very low (+), low (+ +), moderate (+ + +), high (+ + + +), very high (+ + + + +).

Technology Feasibility Cost Proven Effectiveness Static signs + + + + + + + Trivia signs + + + + + + + Vehicle-activated signs (VAS) + + + + + + + + + + + Variable message signs (VMS) + + + + + + + + + + Rumble strips + + + + + + + + Pavement treatments + + + + + + + + + + WID-activated signs + + + + + + + + + Break-the-beam warning system + + + + + + + + + Virtual fence + + + + + + + + + Fencing & crossing structures + + + + + + + + + + + + Short tie-in fencing to culverts/pipes + + + + + + + + + + + Retro-fit dry ledge in wet culverts/pipes + + + + + + + + + + + + Roadside reflectors + + + + + + + + Audio deterrents + + + + +

Most options rate high on feasibility except WID-activated signs, break-the-beam warning systems, virtual fence, fencing/crossing structures and audio deterrents. Apart from audio deterrents, the technologies that scored low on feasibility present considerable logistical challenges to implement and are relatively high cost. Audio deterrents rate low on feasibility because they are reliant on uptake by motorists.

The least expensive technologies (per unit) include static signs/trivia signs, rumble strips, short tie-in fencing and audio deterrents. VAS and VMS signage, virtual fence and retro-fit ledges are moderate cost options whereas WID-activated signs, break-the-beam systems and fencing/crossing structures score as high to very high cost. It should be noted that cost is largely scale-dependent.

Evidence of effectiveness (proven effectiveness), while critical to informing choice of mitigation technology, is in many instances difficult to quantify. Most mitigation technologies have not been subjected to well-designed trials and much of the evidence supporting their effectiveness or otherwise is either anecdotal, lacking appropriate design, lacking replication or collected over short time periods. Exceptions to this are break-the- beam systems that have undergone extensive trials in Europe although its effectiveness has not been proven in an Australian context.

Well-maintained fauna fencing and crossing structures have been shown at numerous sites to prevent wildlife roadstrike whilst enabling landscape permeability (e.g. Taylor & Goldingay 2003; Dexter et al. 2016). Conversely, fence breaches and inadequate maintenance compromises fencing effectiveness, as demonstrated at Skyline Road. Recent evidence suggests that lower profile and less obtrusive styles of exclusion fence can be effective in guiding fauna to underpasses. Ropes slung between canopy trees on each side of a road may be a cost-effective means of reducing road strike for possums.

There is emerging evidence in support of VAS to reduce both vehicle speed and wildlife road mortality. Further, making the VAS mobile and moving it periodically to different blackspot locations reduce the

Sandpiper Ecological Surveys 25

Wildlife Roadkill Mitigation Report - Lismore LGA

likelihood of habituation. Less has been done on VMS but there is some evidence that they lower vehicle speeds. There is strong evidence of speed reductions in response to rumble strips and they may assist in supporting other technologies such as pavement treatments or signage by providing an auditory cue. The reported reduction in road mortality attributed to rumble strips in NW Tasmanian (Lester 2015) suggests further trials are warranted although, due to their noise, their application in urban areas or near dwellings is constrained.

Evidence emerging for virtual fence suggests this technology may be effective at preventing or reducing road incursions for some wildlife species, particularly small macropods although the technology is relatively expensive. Break-the-beam warning systems and WID-activated signs show much potential but are very expensive technologies to deploy over long road distances. These technologies are most suited to short sections of road where roadstrike is affecting a highly significant species/population.

Static signs, while rated low on effectiveness overall, may be relatively effective for a short period after installation much the same as seasonal signs are relatively effective (Huijser et al. 2015). Indeed, the use of trivia signs that change periodically may improve driver vigilance. This could be applied as a wildlife mitigation tool by posing trivia questions related to the species of concern at a black-spot with the question posed as a motorist enters a black-spot and the answer provided as the motorist exits the black-spot. Costs of changing the question/answer may be markedly reduced by placing/replacing a sticker on the static sign. Lismore CC could even encourage locals to submit trivia questions to post on the signs. 4. Identifying roadkill black-spots

4.1 Roadkill data sources

Roadkill data were accessed from three primary sources – WIRES, FOK and NSW Office of Environment and Heritage (OEH) BioNet Atlas. Northern Rivers Wildlife Carers provided data at the end of the project and were not included in analyses. WIRES data were provided in tabulated summary form according to fauna groups and localities/suburbs (Table 1 & 2). The specific location of roadkills (i.e. grid reference) was not available. To address this and attempt to identify wildlife roadkill black-spots, WIRES carers within the Lismore LGA were asked to nominate what they regarded as the primary roadstrike black-spots within their local area and the fauna most affected. These data were then compiled and sorted into spot locations (specific point locations) and road segment locations (generalized locations referring to a length of road). A grid reference (easting/northing) was generated for spot locations and a poly-line generated for the road segments. This resulted in 9 spot locations and 33 road segment locations. Full details of WIRES carer responses is provided in Appendix 1.

Data for koala roadstrikes for the period 2005 to end of 2017 was provided by FOK. The dataset of 394 records was interrogated for spatial quality. Locations with insufficient spatial information (e.g. only road and/or suburb recorded) were discarded (mostly from the period 2005-2010) and other records converted to a grid reference (easting/northing). This resulted in a total of 321 spot locations.

BioNet Atlas data were sourced through Lismore CC under their license with OEH. All records recorded as ‘roadkill’ for the Lismore LGA were acquired on 26/6/2018. Acquired records (n=135) were sorted by year and records for the period 2005-2017 retained and the rest discarded. From this set, feral species and 12 koala records (to avoid risk of double-counting FOK records) were discarded. This resulted in 26 retained records. Each record included an easting/northing for a total of 26 spot locations. The resultant BioNET and WIRES records did not include any koala records.

Sandpiper Ecological Surveys 26

Wildlife Roadkill Mitigation Report - Lismore LGA

4.2 Identifying black-spots

Identifying roadkill black-spots involved several iterative steps. Firstly, WIRES spot locations and road segments, FOK spot locations and BioNet spot locations were imported into ArcGIS (ArcMap 10.4.1) and overlaid on Lismore LGA roads and vegetation (Figure 6). Because of the paucity of spot locations for WIRES data and BioNet data, the focus of black-spot identification was placed on koala records.

Biolink (2017) analysed FOK koala roadstrike data for the CKPOM area for the period 2011-2016 (and 2005-10 in rural areas) and identified five black-spot road segments although there appeared to be some minor anomalies between roadstrike records and intensity of roadstrike in identified road segments. The current study did not analyse black-spot road segments identified by Biolink (2017).

To identify additional koala black-spots, the mapped 2017 FOK roadstrike locations were interrogated to determine where either 2017 data intensified the rate of roadstrike data for combined 2005-2010 and 2011-16 or new black-spot clusters greater than one roadstrike/km were evident amongst the 2017 locations. Inclusion of both 2005-10 and 2011-16 data (i.e. two koala generations) is consistent with the method applied by Biolink (2016) for rural areas to correct for under-reporting. For the current analysis, 2017 data were effectively treated as a generation despite being a single year. In doing so, we acknowledge that they represent minimum, preliminary values that may or may not increase during the remaining years of the generation period (i.e. 2017-22). However, using 2017 data in this way enables identification of emerging or worsening black-spots. Values for koala roadstrikes/km/generation were derived by averaging FOK 17 koala roadstrikes and sum of FOK 05-10 & 11-16 koala roadstrikes.

The additional koala black-spots were then rated using the same roadstrike intensity rating applied to 2005-16 data (Biolink 2017). That is, black-spots featuring ≥ 3.5 roadstrikes/km/generation (or a small road section within a segment featuring ≥ 9 roadstrikes/km/generation) were designated as ‘major’ black-spots. Biolink (2017) referred to all other black-spots as ‘minor’. To better assist in prioritising black-spot management, an additional category (‘moderate’) was applied to identify black-spots in the order of 2.0-3.0 roadstrikes/km/generation. Black-spots with 1.0-1.9 roadstrikes/km/generation were classified as ‘minor’ and those <1.0 roadstrikes/km/generation were classified as ‘slight’.

The black-spot rating process resulted in two moderate and seven minor black-spots in addition to those identified by Biolink (2017) (Table 12). Both Biolink (2017) black-spots (‘B’ series) and additional or extra black- spots (‘X’ series) were mapped to produce a koala roadstrike intensity map (Figure 7). The additional sites mostly occurred in areas near those identified by Biolink (2017) but also included new areas near Coraki (X2), Dunoon and Rosebank (X3, X5, X9) and Clunes (X7). Notably, 12 of the combined koala black-spots also occurred on roads identified by WIRES carers as wildlife roadstrike black-spots.

Sandpiper Ecological Surveys 27

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 12: Koala roadstrike black-spot ratings as derived from FOK data. ‘Other wildlife records’ sourced from WIRES and BioNET have been included for comparative purposes. WIRES Road refers to whether the subject road(s) is listed as a known WIRES road and WIRES Spot refers to roadstrike spot locations. Koala black-spots were generated from current study (X1-13) and Biolink (2017) (B1-7). Koala RK/km/generation values in the current study were derived by averaging FOK 17 and sum of FOK 05-10 & 11-16. na = not assessed in current study (data previously assessed in Biolink 2017).

Black- Locality Road(s) Length Other wildlife records Koala records Koala RK/ Koala spot (km) BIO WIRES WIRES FOK FOK FOK km/ Black-spot No. NET spot ROAD 05-10 11-16 17 generation Rating B4e Lindendale Bruxner Hwy 3 0 0 YES na na na 4.8 Major Cowlong Rd 1 0 0 NO na na na B6 Bexhill Bangalow Road 1.5 0 0 NO na na na 4.0 Major B2 Tucki Tucki & Tuckurimba Tucki Tucki Rd, Wyrallah Rd 9 0 0 NO na na na 3.9 Major B4w Goonellabah, E Lismore Bruxner Hwy 4 0 0 YES na na na Cynthia Wilson Dv 1 0 0 NO na na na 2.6 Major* Rous Rd, Pleasant St, Invercauld Rd, Kruseana Av 5 0 1 YES na na na B1 South Gundurimba Bruxner Hwy, Tatham Rd 6 1 0 YES na na na 2.8 Moderate X1 Wyrallah Tregeagle Rd 1 0 0 NO 2 1 2 2.5 Moderate B5 Monaltrie, Wyrallah Wyrallah Rd 8 1 0 NO na na na 2.4 Moderate X2 Coraki Coraki Rd 1 0 0 NO 1 3 0 2 Moderate B3 Marom Ck Grennan Rd 2 0 0 NO na na na 1.7 Minor Graham Rd 1 0 0 NO na na na B4c Goonellabah, Tregeagle Taylor-Holland Rd 2 0 0 NO na na na 1.4 Minor Rous Rd, Alphadale Rd, Tregeagle Rd 8 1 0 NO na na na X3 Dunoon, Modanville Dunoon Rd 3 1 0 YES 3 5 0 1.3 Minor X4 Ruthven Coraki Rd 2 0 0 NO 0 3 2 1.3 Minor X5 Rosebank Rosebank Rd 1.5 0 0 YES 0 2 1 1 Minor X6 South Gundurimba Tatham Rd 1 0 0 NO 0 1 1 1 Minor X7 Clunes Bangalow Rd 1 0 0 NO 0 1 1 1 Minor X9 Rosebank Fox Rd 1.5 0 0 YES 0 3 0 1 Minor B1X McKees Hill Bruxner Hwy 2 0 0 YES 0 0 4 1 Minor X8/B7 Numulgi Duncan Rd 2 0 0 NO 1 1 0 0.5 Slight X10 N Lismore, Booerie Ck Nimbin Rd 3 1 0 YES 0 2 1 0.5 Slight X12 Nimbin Nimbin Rd, Blue Knob Rd; Cecil St, Tuntable Falls Rd, West Rd. 9 0 0 YES 1 0 1 0.2 Slight X11 Caniaba Caniaba Rd, Fredericks Rd, Struthers Rd, Rudgely Rd 8 0 0 YES 1 1 0 0.13 Slight X13 Jiggi Jiggi Rd 4 1 1 YES 0 0 0 0.11 Slight

* Rated as ‘Major’ in Biolink (2017) because a segment of the road section featured >9 roadstrikes/km/generation.

Sandpiper Ecological Surveys 28

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 6: Wildlife roadstrike locations for Lismore LGA from various data sources. WIRES known roads represent roads identified by WIRES carers as featuring a range of wildlife roadstrike.

Sandpiper Ecological Surveys 29

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 7: Koala roadstrike intensity polylines for Lismore LGA generated from 2005-17 data from FOK. Biolink (2017) intensity polylines were generated from 2005-16 data.

Sandpiper Ecological Surveys 30

Wildlife Roadkill Mitigation Report - Lismore LGA

4.3 Black-spot inspections

The majority of koala black-spots identified in the current project and in Biolink (2017) were inspected on site between 29 June and 11 July 2018. Inspections were not undertaken at sites B2 (subject of Phase 1 of project), X11, X12 and X13. During inspections the following site attributes were assessed:

• Signed speed limit; • Road view (open or obstructed); • Road alignment (curved or straight); • Adjoining habitat.

Sites varied considerably with regards to road attributes, type and structure of adjoining habitat (Table 13). Signed speed limits ranged between 50 and 100 km/h with rural roads mostly in the range of 80 to 100 km/h. Rural black-spot roads were typically surrounded by paddock and roadsides lined by trees and often featuring forest red gums Eucalyptus tereticornis, a primary koala food tree. By contrast, urban sites (e.g. B4) mostly featured a range of native and exotic street trees and both patches and corridors of native vegetation.

Sandpiper Ecological Surveys 31

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 13: Koala roadstrike black-spot site assessments. Black-spots were generated from current study (X1-13) and Biolink (2017) (B1-7). Pad = paddock; TLR = tree-lined roadside; FRG = forest red gum; Twood = tallow wood; RainF = rainforest; DryF = dry eucalypt forest; Maca = macadamia plantation.

Black- Locality Black-spot Road(s) Road Signed Speed Road alignment & view Adjoining Habitat spot Rating Length No. (km) B4e Lindendale Major Bruxner Hwy 3 80 curved & obstructed Pad; Maca; RainF & DryF TLR inc.Twood & FRG Cowlong Rd 1 60 straight/ curved & obstructed Pad; Maca; RainF & DryF TLR inc.FRG B6 Bexhill Major Bangalow Rd 1.5 80 curved & open Pad; RainF & DryF B2 Tucki Tucki & Major 9 Assessment & mitigation recommendations already completed Tuckurimba (Sandpiper 2018) B4w Goonellabah, Major Bruxner Hwy 4 60 curved & open Urban; street trees; some FRG E Lismore Cynthia Wilson Dv 1 50 straight & open Large adjacent patch of FRG Rous Rd, Pleasant St, Invercauld Rd, 5 50 curved & obstructed RainF; urban; DryF TLR inc. FRG Kruseana Av B1 South Gundurimba Moderate Bruxner Hwy, Tatham Rd 6 100 curved & obstructed Pad; TLR inc. FRG; scattered & clumps of FRG XI Wyrallah Moderate Tregeagle Rd 1 80 open & straight Pad; TLR inc. FRG B5 Monaltrie, Wyrallah Moderate Wyrallah Rd 8 80, 50 mostly open & curved Pad; TLR inc. FRG; scattered & clumps of FRG X2 Coraki Moderate Coraki Rd 1 80 open & curved Pad; scattered FRG; TLR mostly swamp oak B3 Marom Ck Minor Grennan Rd 2 Not signed (80?) curved & obstructed Pad; TLR inc. FRG Graham Rd 1 Not signed (80?) curved & obstructed Pad; scattered FRG B4c Goonellabah, Minor Taylor-Holland Rd 2 50, 60 open & straight TLR inc. FRG Tregeagle Rous Rd, Alphadale Rd, Tregeagle Rd 8 80 curved & obstructed Pad; Maca; TLR inc. FRG X3 Dunoon, Modanville Minor Dunoon Rd 3 80, 60 mostly open & curved Pad; TLR inc. FRG X4 Ruthven Minor Coraki Rd 2 100 open & straight Pad; creek; TLR inc. FRG X5 Rosebank Minor Rosebank Rd 1.5 80, 50 obstructed & curved RainF; Pad; scattered FRG X6 South Gundurimba Minor Tatham Rd 1 80 open & straight Pad; scattered FRG X7 Clunes Minor Bangalow Rd 1 100 mostly open & straight Pad; RainF & DryF TLR inc. Twood X8/B7 Numulgi Minor Duncan Rd 2 80 mostly open & straight Pad; Maca; TLR inc. Twood X9 Rosebank Minor Fox Rd 1.5 Not signed (50-60) curved & obstructed some Pad; RainF & DryF TLR inc. Twood B1X McKees Hill Minor Bruxner Hwy 2 100 curved & obstructed Pad; TLR inc. FRG; scattered & clumps of FRG X10 N Lismore, Slight 50, 80 curved & obstructed Pad; TLR inc. FRG Nimbin Rd 3 Booerie Ck X12 Nimbin Slight Nimbin Rd, Blue Knob Rd; Cecil St, Not inspected 9 Tuntable Falls Rd, West Rd. X11 Caniaba Slight Caniaba Rd, Fredericks Rd, Struthers 80, 60 mostly open & curved Pad; TLR inc. FRG 8 Rd, Rudgely Rd X13 Jiggi Slight Jiggi Rd 4 Not inspected

Sandpiper Ecological Surveys 32

Wildlife Roadkill Mitigation Report - Lismore LGA

5. Mitigating black-spots: prioritising and selecting from the tool box

5.1 A rationale for black-spot mitigation in the Lismore LGA

It is evident from data presented in chapter four that there are numerous roadstrike black-spots across the Lismore LGA. Moreover, it should be apparent from discussions in chapter 3 that short of lining all black-spot road corridors with fauna fencing and retrofitting underpasses at regular intervals, no single technology will eliminate the incidence of wildlife road strike. Permeable landscapes mean that koalas and other wildlife will continue to encounter roads and traffic. Roadkill mitigation efforts, therefore, should aim to increase driver vigilance and reduce vehicle speed to reduce the likelihood that road encounters between wildlife and cars will end in roadkill.

The mitigation report, therefore, proposes an integrated approach featuring several technologies. As such, it is similar to the approach developed by Tweed Shire Council to mitigate koala road strike at black-spots on Clothiers Creek Road, Cabarita and Coronation Avenue, Pottsville. The Tweed strategy has reportedly been successful in reducing vehicle speeds and reducing koala roadstrike although koala road strike before data were incomplete (S. Hetherington pers. comm.). The approach centered on delineating the two black-spots as ‘koala zones’ and utilising a combination of static signs, variable message signs (VMS), slow points, pavement treatments and before/after traffic monitoring. Tweed City Council also promoted the ‘koala zones’ in a ‘slow down for koalas’ marketing strategy focusing on schools and local media.

Rural roads across the Lismore LGA present a familiar and challenging scenario for road managers. How do you prevent or reduce road strike of koalas and other wildlife along long stretches of moderate use road corridors that feature productive roadside habitat within a mostly cleared and modified landscape? The Biolink (2017) study pressed that given the relatively high and stable koala population density within the CKPoM planning area, the emphasis of management should be on ensuring that the landscape remains permeable to koala movement to enable ongoing population processes. The same largely applies to other species of wildlife affected by roadkill. Concomitant with maintaining landscape permeability is managing wildlife encounters with roads and vehicles. For example, roadstrike accounted for 17.5% of known koala mortalities in the Lismore LGA for the 2011-16 period (Biolink 2017). Indeed, koala road mortality has been increasing annually since at least 1998 (Biolink 2017).

5.2 Prioritising black-spots

As described in the previous chapter, black-spots were rated according to their intensity of koala roadstrikes. This was largely due to an absence of sufficient data for other species and because the koala was identified as a priority species. It is envisaged that as systematic, grid-referenced data for other species becomes available black-spot identification at other locations will be possible. In this sense, the WRMR should be adaptable to emerging black-spots. This is discussed further in section 5.4 which covers data collection and identifying future black-spots.

From the combined dataset of the current project and Biolink (2017), four black-spots have been rated as major, four as moderate, nine as minor and five as slight. The following section focuses on mitigation tools for black-spots rated as either major or moderate (Table 14). Locality maps and mitigation suggestions for minor black-spots are provided in Appendix B and C.

Sandpiper Ecological Surveys 33

Wildlife Roadkill Mitigation Report - Lismore LGA

Table 14: Major and moderate koala roadstrike black-spots. X-labelled sites are from the current study and B-labelled sites are from Biolink (2017).

Black- Locality Road(s) Length Koala Black-spot spot (km) RK/km/ Rating No. generation B4 Lindendale Bruxner Hwy 3 4.8 Major east Cowlong Rd 1 B6 Bexhill Bangalow Road 1.5 4.0 Major B2 Tucki Tucki & Tuckurimba Tucki Tucki Rd, Wyrallah Rd 9 3.9 Major# B4 Goonellabah, East Lismore Bruxner Hwy 4 2.6 Major* west Cynthia Wilson Dv 1 Rous Rd, Pleasant St, Invercauld Rd, 5 Kruseana Av B1 South Gundurimba Bruxner Hwy, Tatham Rd 6 2.8 Moderate XI Wyrallah Tregeagle Rd 1 2.5 Moderate B5 Monaltrie, Wyrallah Wyrallah Rd 8 2.4 Moderate X2 Coraki Coraki Rd 1 2.0 Moderate

* Rated as ‘Major’ in Biolink (2017) because a section of the road featured ≥ 9 koala roadstrikes/km/generation. # Mitigation treatments for site B2 was the focus of Stage 1 (Sandpiper 2018).

5.3 Applying the ‘Mitigation Zone’ to all priority black-spots

As recommended in the Phase 1 report for the Wyrallah Road major black-spot (i.e. B2), the proposed basis for addressing roadkill black-spots is to treat the road(s) in question as a management zone within which a variety of management strategies or tools can be applied. Zones are then referred to as ‘koala zone’ (or ‘platypus zone’) for specific focal species or ‘wildlife zone’ for black-spots featuring a variety of species. It is proposed that, cost permitting, all mitigation zones include the following features:

• Green pavement treatment marking zone entry and exit. • Single rumble strip approximately 20m prior to the pavement marking. • Static ‘Wildlife Zone’ (or ‘koala zone’) Trivia Sign with a question at zone entry and answer at zone exit. The ‘Wildlife Zone’ Trivia Sign may be modelled on that used in Queensland (refer to Plate 4). Trivia Sign question should be related to focal species (e.g. koala-related question for koala zone). Question/answer should be periodically changed. To reduce cost, a removable sticker rather than removable panel could be used.

Pavement treatments, rumble strips and static ‘Zone’ signs should be installed concurrently. Design of the ‘Zone’ signs should be discussed with environmental staff and road engineers to ensure specification compliance and branding consistent with LCC policy. Plate 4 shows an example which is useful as a reference point. The Queensland Department of Transport and Main Roads (QDTMR), who developed the specifications for the pavement marking and which was approved for use in Tweed LGA, suggest the associated sign should inform and reinforce the pavement marking (QDTMR 2015). The above recommendations are consistent with this. A single rumble strip should be installed within the lane approaching the ‘Zone’ and approximately 20m prior to the pavement marking. The intention is to provide an auditory stimulus to reinforce the visual sign. Whilst installation of rumble strip/s near residences is problematic this method warrants careful consideration due to its possible benefits.

Sandpiper Ecological Surveys 34

Wildlife Roadkill Mitigation Report - Lismore LGA

The ‘Zone’ branding and trivia questions should be promoted via LCC marketing and local media. It should highlight that these Zones are roadstrike black-spots. Media campaigns may further benefit by having a seasonal focus, e.g. koalas during August to December. Campaigns should be nested within the broader message of drivers to be koala (wildlife) aware on local roads and the link between speed and an increased likelihood of hitting a koala (wildlife). It may also be useful to link the ‘Zone’ branding with that occurring in the region, e.g. Tweed and Ballina LGA’s, to develop a broader, consistent narrative. Tweed City Council environment officers have conducted promotion of ‘Koala Zone’ and ‘Slow down for koalas’ in local schools. A similar strategy should be considered for Lismore LGA schools.

5.4 Extra tools for priority black-spots

In addition to the standard ‘mitigation zone’ treatments described above, other specific mitigation tools are proposed for priority black-spot sites. Details of proposed mitigation for each site is summarised in Table 15 and described in more detail below.

B4-east and B4-west were identified by Biolink (2017) as major koala blacks-spots (Figure 8). B4-east features a 3km section of the Bruxner Highway and 1km of Cowlong Road. Both road sections should feature the standard ‘mitigation zone’ treatments based on ‘koala zone’ signage. It is proposed that a signed speed reduction from 80 km/h to 60 km/h be imposed on the Bruxner Highway ‘koala zone’ section (Table 15). Importantly, Bruxner Highway is a state road and any proposed mitigation will require cooperation and approvals from RMS. A speed reduction is not recommended for Cowlong Road (signed 60 km/h) but the 1km- long strip would benefit from more frequent slashing of the road edge. Thick shrub vegetation up to the road edge obstructs driver vision along this section and reduces the likelihood of spotting wildlife emerging from the roadside.

The Bruxner Highway section of B4-east was also identified by WIRES as a black-spot for possums, bandicoots and echidnas. Further, this section of road was identified in 2003 as featuring high numbers and high diversity of roadkills (Taylor & Goldingay 2003). Biolink (2017) recommended exclusion fencing and strategically placed underpasses would be an effective approach along this stretch. We agree that this would be an effective way to substantially reduce the incidence of roadstrike on this stretch of highway but acknowledge the high cost and logistical challenges associated with such a venture. A scaled-down version of this would be to install fence tie-ins to funnel fauna to existing culverts/pipes located along this section (Table 15). A box culvert in the drainage line approximately 200m west of Mountain Blue Farm would be a prime candidate to begin with. It would require a feasibility assessment and a determination whether dry access was available or if a retro-fitted ledge was required. Ideally, it should include up to 100m of fencing (stock fence with mesh) either side of the culvert to direct fauna to the crossing.

B4-west features a section of the Bruxner Highway and numerous nearby side roads. The disparate nature of the roads involved makes it a challenging zone within which to mitigate roadstrike. It is also a zone where signed speeds are already 50 and 60 km/h and further speed reductions are not feasible. Biolink (2017) identified the retaining wall/median strip between Kadina Street and Rous Road as a barrier to koala movement. Climbable mesh has been installed on the retaining wall at several points along this section. It is recommended that more mesh be installed to fill the larger gaps. Consideration should also be given to attaching jute ‘sausages’ (approx. 100mm-thick jute strips used as a retaining material in bush regeneration and landscaping) at intervals along the wall (Table 15). These do not pose a vehicle hazard and would facilitate a climbing escape. Because of the short lengths of the numerous side roads (i.e. Cynthia Wilson Drive, Rous Road, Invercauld Road, Kruseana Avenue) we recommended that only pavement treatment be applied at start/end to these road sections, particularly as trivia signs are not practical amidst erratic traffic movements.

Sandpiper Ecological Surveys 35

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 8: Major koala black-spot (B4) in the Goonellabah area identified by Biolink (2017).

The major black-spot B6 (Figure 9) identified by Biolink (2017) data featured 4.0 koala roadstrikes/km/generation. The management zone is relatively short (i.e. 1.5 km) and features a relatively open view but curved road orientation. In addition to ‘mitigation zone’ treatment (i.e. ‘koala zone’ signage, pavement treatment and rumble strips), a signed speed reduction from 80 to 60 km/h is recommended for this section (Table 15). A speed reduction is regarded as feasible and reasonable considering the curvature of the road and the high rate of koala roadstrikes. This is a road zone in which installation of VAS should be considered. VAS deployment is discussed further below.

Sandpiper Ecological Surveys 36

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 9: Major koala black-spot (B6) in the Bexhill area identified from Biolink (2017) data.

Black-spot B1 (Figure 10) rated as moderate and includes a section of the Bruxner Highway and Tatham Road identified by Biolink (2017). We have added B1X (rated as minor) identified in the current study to this black- spot segment as it joins the western Bruxner Highway end of B1 and makes logical sense to treat them as a single management zone. Indeed, four koala roadstrikes were recorded during 2017 within the western section of the Bruxner Highway, which highlights this section as an emerging black-spot of concern. In addition to the standard ‘mitigation zone treatments, we recommend reducing the signed speed from 100 km/h to 80 km/h (Table 15). This is also an area that VAS should be considered.

Sandpiper Ecological Surveys 37

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 10: Moderate koala black-spots (B1 & B1X) in the south Gundurimba area identified in the current study and by Biolink (2017).

The moderate black-spots X1 and X2 (Figure 11 & 12) featured 2.5 and 2.0 koala roadstrikes/km/generation, respectively. The management zones of both black-spots are relatively short (0.5 km and 1km, respectively) and feature an open, unobstructed road view. In addition to ‘mitigation zone’ treatment (i.e. ‘koala zone’ signage, pavement treatment and rumble strips), a signed speed reduction from 80 to 60 km/h is recommended for both sites (Table 15). A speed reduction is regarded as feasible and reasonable considering the curvature of the road at both sites. This is particularly so for the X2 site where Coraki Road, between Casey Lane and the bridge, curves considerably. The concentration of roadstrikes within 200m of the bridge at this site suggests that the natural river barrier requires koalas to cross the road if moving east or west along the edge of the river.

Sandpiper Ecological Surveys 38

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 11: Moderate koala black-spot (X1) in the Wyrallah area identified in the current study.

Figure 12: Moderate koala black-spot (X2) in the Coraki area identified in the current study.

Sandpiper Ecological Surveys 39

Wildlife Roadkill Mitigation Report - Lismore LGA

The moderate rated black-spot B5 (Figure 13) identified by Biolink (2017) featured 2.4 koala roadstrikes/km/generation. The management zone for this black-spot is considerably long (i.e. 8 km) and features mostly open, unobstructed road view with both curved and straight sections. In addition to ‘mitigation zone’ treatment (i.e. ‘koala zone’ signage, pavement treatment and rumble strips), a signed speed reduction from 80 to 60 km/h is recommended in the 80 km/h zone sections, particularly the sections featuring 4-6 koala roadstrikes/km/generation (Table 15; Figure 13). These sections may also benefit from regular slashing of the road edge and periodic deployment of VAS.

Figure 13: Moderate koala black-spot (B5) in the Wyrallah area identified by Biolink (2017).

Further to the discussions above on extra tools for priority black-spots, two other tools highlighted in chapter 3 should be considered in the mix of mitigation measures. Vehicle-Activated Signs (VAS) were recommended at two locations at B2 on Wyrallah Road in the Phase 1 report (Sandpiper 2018). VAS should also be considered for locations B6 (Bangalow Road, Figure 9), B1 and B1X (Bruxner Highway, Figure 10) and B5 (Wyrallah Road, Figure 13) to complement speed reductions. Due to the high cost/unit, it may be cost effective to rotate the VAS units between these sites (e.g. six-monthly rotations). Importantly, data loggers should be installed at least one month prior to VAS installation to capture ‘before’ treatment vehicle speeds. VAS records vehicle speeds and thereby logs the level of compliance. The emerging evidence for VAS suggests they are an effective method for reducing vehicle speeds and periodically relocating units would also reduce the likelihood of habituation.

The other tool discussed in chapter three that may be applicable to the current project is Virtual Fence. The east section of the Bruxner Highway (i.e. B4-east, Figure 8), features high numbers of both koala and other wildlife roadstrikes. If fencing of this section of highway is not considered feasible then consideration should be given to trialing Virtual Fence. Thought would need to be given about the design of the trial to ensure

Sandpiper Ecological Surveys 40

Wildlife Roadkill Mitigation Report - Lismore LGA

adequate replication and controls. For example, the road section could be divided up in to 500m sections which are randomly assigned as treatment zones (i.e. virtual fence) or controls (i.e. no virtual fence). Six to 12 months of systematic roadkill surveys would also be required prior to trial commencement. Such a project could be the subject of an NSW Government SOS Grant.

Table 15: ‘Tools’ proposed for priority black-spots. VAS = vehicle-activated sign.

Mitigation Recommendation Black- Black- Signed Extent of spot Locality Road(s) spot Speed Pavement Speed Sign Other Management No. Rating Reduction treatment Zone Slash road edge; tie-in fence up to 100m either side of Koala culvert located Bruxner Hwy 80 80>60 Yes 3km Zone 200m W of Mtn B4e Lindendale Major Blue Farm; retro-fit dry ledge through culvert if required Cowlong Rd 60 No No Yes Slash road edge 1km Koala B6 Bexhill Bangalow Rd Major 80 80>60 Yes VAS 1.5km Zone Tucki Tucki & Tucki Tucki Rd, B2 Major Assessment & recommendations already completed (Sandpiper 2018). Tuckurimba Wyrallah Rd Install extra mesh & climbable jute poles Koala Bruxner Hwy 60 No Yes along retaining wall 4km Zone between Kadina St Goonellabah, & Gordon Blair Dv B4w Major E Lismore Cynthia Wilson Dv 50 No No Yes - 1km Rous Rd, Pleasant St, Invercauld Rd, 50 No No Yes - 5km Kruseana Av Parrots Nest South Bruxner Hwy, Koala Rd to 500m B1 Moderate 100 100>80 Yes VAS Gundurimba Tatham Rd Zone W Tatham Rd (3km) 500m either Koala side of XI Wyrallah Tregeagle Rd Moderate 80 80>60 Yes - Zone Henson Rd (1km) Monaltrie, Koala VAS; slash road B5 Wyrallah Rd Moderate 80, 50 80>60 Yes 8km Wyrallah Zone edge Casey Ln to Koala X2 Coraki Coraki Rd Moderate 80 80>60 Yes _ bridge Zone (0.5km)

Sandpiper Ecological Surveys 41

Wildlife Roadkill Mitigation Report - Lismore LGA

6. Identifying future black-spots and evaluating effectiveness

6.1 Identifying future black-spots – Roadkill Recording Systems

A feature of developing a Wildlife Roadkill Mitigation Report (WRMR) is gathering data from varied sources. In its entirety, it is a considerable and informative dataset. However, differences in recording systems, methods and types of data captured somewhat constrained analyses. Going forward, therefore, we propose that Lismore CC hold discussions with FOK, WIRES and Northern Rivers Wildlife Carers (NRWC) – the primary repositories of roadstrike data – to clarify capture of key data and a means of sharing data. This will also be critical for evaluating the effectiveness of any mitigation measures.

To enable detailed black-spot analyses, the most critical information to capture is date, GPS location and the species encountered (if known). The data generously provided by FOK and WIRES captures date and species but location information often lacks spatial precision (e.g. street name only) which precludes creation of spot locations on data maps. Smart phones now enable accurate and simple recording of spot locations. Presuming location services are enabled on one’s phone, taking a photo of the roadkill creates a geo-reference or spot location which can be later recorded and/or uploaded.

A step-up from this simple approach is web or phone-based reporting of roadkills. Such approaches are a rapidly growing source of data for understanding road impacts and mitigation effectiveness (Shilling et al. 2015). Some systems use smartphone-based applications (Apps) to facilitate data entry from the field and some use social media. Smartphone Apps enable users to photograph/log/describe a roadkill. This information is then uploaded and curated by an administrator, usually a University, conservation organization or within a government environment department. They have been developed in several regions across the globe, including Tasmania. The largest and longest running system that relies on volunteer observers is the ‘Californian Roadkill Observation System’ run by the Road Ecology Center at University of California (Shilling et al. 2015). App examples include ‘Roadkill/SPOTTERON’ run by the University of Natural Resources (BOKU) in Austria and ‘Project Splatter’ operated by Cardiff University in the UK.

‘Roadkill TAS’ (Cloudsource Ltd., Tasmania) is an Australian version that has been recently launched and is receiving much notice on social media. Such an App could be applied to the Lismore LGA (Figure 14). This would require negotiations with the developer and a designated person/position in Lismore CC to administer it. It may be a project that could be developed with assistance of the NSW Government’s SOS program.

Sandpiper Ecological Surveys 42

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure 14: ‘Roadkill TAS’, one of several smartphone Apps available for recording roadkill.

In the absence of an appropriate App, Lismore CC may consider using a standardized roadkill recording sheet such as one shown in Table 16, which was adapted from the Californian Roadkill Observation System (CROS) developed by the University of California Road Ecology Center. The recording sheet essentially captures the ‘what’, ‘where’ and ‘when’ of the roadkill and ‘who’ recorded it. This form could be distributed amongst Council road crews and administered/curated by a designated position within the Environment branch.

Table 16: Roadkill recording template (adapted from Californian Roadkill Observation System (CROS) developed by the University of California Road Ecology Center).

Roadkill Observer Information (Who) Name: Phone: Email: Postcode: Roadkill Observation 1 Roadkill Observation 2 What What Photo: Yes No Photo: Yes No Animal: Frog/toad Reptile Bird Mammal Animal: Frog/toad Reptile Bird Mammal Size: Small Medium Large Size: Small Medium Large Confidence: 100% Fairly sure Best guess Confidence: 100% Fairly sure Best guess Where Where Road name: Road name: Nearest town: Nearest town: Location description: Location description:

GPS location (if known): GPS location (if known): Easting: Easting: Northing: Northing: When When Date & Time of Obs: Date & Time of Obs: Approx. date of impact: Approx. date of impact: Did you witness the event: Yes No Did you witness the event: Yes No Other information/comments: Other information/comments:

Sandpiper Ecological Surveys 43

Wildlife Roadkill Mitigation Report - Lismore LGA

6.2 Evaluating effectiveness

As discussed in section 5.4, FOK, WIRES and NRWC are the primary repositories of roadkill data across the Lismore LGA. Coordinating data sharing and ensuring more precise capture of roadkill locations will be critical to the evaluation process. Whereas FOK data provides a comprehensive ‘before’ dataset of koala roadstrikes across the Lismore LGA, roadstrike data for other species captured by WIRES, while comprehensive in its extent, do not include point locations and therefore limit before/after comparisons. Data from NRWC was received just prior to report submission and not analysed. It appears to feature reasonably good spatial information with which to plot roadkill locations. Future black-spot analyses should include these data (refer Appendix D).

We propose that the koala is the surrogate for evaluating the effectiveness or otherwise of mitigation measures. As more systematic data collection of non-koala species occurs and/or datasets with good spatial precision are developed, they should be analysed and become part of the evaluation process and performance indicators developed for them. Further, those implementing the WRMR should be cognizant of roadstrike data collection survey effort. That is, roadstrike data have largely been generated from incidental observations by members of the public and wildlife carers. Because data have not been systematically collected (e.g. weekly drive survey along a given road) it is difficult to quantify the survey effort. This should be considered in any assessments of data collected after mitigation is implemented – was survey effort similar or has it increased after implementation of mitigation thereby increasing the likelihood of detecting a roadstrike?

As proposed for Phase I of the WRMR, the performance of the mitigation plan should be reviewed annually and assessed against the following four performance Indicators:

Performance indicators

1. Increase in proportion of vehicles travelling at or below the signed speed at locations where speed reductions or VAS have been implemented. (Data loggers required to measure speeds at least 1 month prior to installation of mitigation treatment). 2. Reduction in annual number of reported koala roadstrikes/km at roads where mitigation measures implemented compared to mean number of roadstrikes for preceding koala generation (i.e. 5 years). 3. A reduction in annual proportion of koala roadstrikes that result in death on rural roads where mitigation measures have been implemented compared to mean number of roadstrikes for preceding koala generation (i.e. 5 years). 4. A reduction in annual number of reported wildlife roadstrikes/km at location where mitigation implemented compared to roads that do not receive mitigation. Mitigated (i.e. treatments) and un- mitigated (i.e. controls) road segments should feature equivalent survey effort and adequate ‘before mitigation’ data.

In reviewing the performance of the WRMR, corrective actions will need to be considered in response to poor performance. In so doing, the plan should be regarded as adaptive and able to respond to broad changes in road and/or landscape conditions (e.g. bushfire) and the emergence of other black-spot location(s) across the LGA. Moreover, analysis of the most recent data should be conducted prior to expenditure on any new black- spot infrastructure to ensure that locations featuring the highest roadstrike intensity are being prioritised. The plan should also be adaptive to the emergence of alternative mitigation technologies and new information on the performance of road mitigation strategies at other locations.

Sandpiper Ecological Surveys 44

Wildlife Roadkill Mitigation Report - Lismore LGA

7. References

Bennett, V. (2017). Effects of road density and pattern on the conservation of species and biodiversity. Current Landscape Ecology Reports 2(1): 1-11.

Biolink Ecological Consultants. (2017). Koala Habitat and Population Assessment. Lismore local government area (part). Final Report. Prepared for Lismore City Council.

Bond, A. R., and D. N. Jones. 2008. Temporal trends in use of fauna-friendly underpasses and overpasses. Wildlife Research 35:103-112. Bond, A. and Jones, D. (2013). Wildlife warning signs: public assessment of components, placement and designs to optimize driver response. Animals 3: 1142-1161.

Burgin, S. and Brainwood, M. (2008). Comparison of roadkills in peri-urban and regional areas of NSW Australia and factors influencing deaths. In ‘Too close for comfort: contentious issues in human-wildlife encounters’, pp. 137-144. Edited by D. Lunney, A. Munn & W. Meikle. Royal Zoological Society of NSW, Mosman, NSW, Australia.

Clevenger, A., Chruszcz, B. and Gunson, K. (2001). Highway mitigation fencing reduces wildlife-vehicle collisions. Wildlife Society Bulletin 29(2), 646-53.

Corkle, J., Marti, M. and Montebello, D. (2001). Synthesis on the effectiveness of rumble strips. Technical Report prepared for Minnesota Local Road Research Board, Minnesota.

Coulson, G. (1982). Road kills of macropods on a section of highway in central Victoria. Australian Wildlife Research 9, 21-26.

Coulson, G. (1997). Male bias in road-kills of macropods. Wildlife Research 24, 21-25.

D’Angelo, G. and van der Ree, R. (2015) Use of reflectors and auditory deterrents to prevent wildlife-vehicle collisions. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Department of Transport and Main Roads (DTMR) (2015). Wildlife signage guidelines: Technical Note 110. State of Queensland, Department of Transport and Main Roads.

Dexter, C., Appleby, R., Edgar, J., Scott, J. and Jones, D. (2016). Using complementary remote detection methods for retrofitted eco-passages: a case study for monitoring individual koalas in south-east Queensland. Wildlife Research 45(5): 369-379.

Dique, D., Thompson, J., Preece, H., Penfold, G. de Villers, D. and Leslie, R. (2003). Koala mortality on roads in south-east Queensland: the koala speed zone trial. Wildlife Research 30(4): 419-26.

Disney, H.J. and Fullagar, P.J. (1978). A note on road kills. Corella 2, 89.

EDI Environmental Dynamics Inc. (2015). Large Mammal-Vehicle Collisions: Review of Mitigations and Analysis of Collisions in Yukon. Prepared for the Yukon Government’s Preventing Yukon Wildlife Collisions Interdepartmental Working Group.

Sandpiper Ecological Surveys 45

Wildlife Roadkill Mitigation Report - Lismore LGA

Ellis, W., FitzGibbon, S., Barth, B., Niehaus, A., David, G., Taylor, B., Matsushige, H., Melzer, A., Bercovitch, F., Carrick, F., Jones, D., Dexter, C., Gillett, A., Predavec, M., Lunney, D., Wilson, R. (2016). Daylight saving time can decrease the frequency of wildlife-vehicle collisions. Biology Letters 12(11), 20160632.

Fahrig, L., Pedler, J. H., Pope, S. E., Taylor, P. D. and Wegner, J. F. (1995). Effect of road traffic on amphibian density. Biological Conservation 73, 177-82.

Found, R. and Boyce, M. (2011). Warning signs mitigate deer-vehicle collisions in an urban area. Wildlife Society Bulletin 35: 291-295.

Fox, S., Potts, J., Permberton, D. & Crosswell, D. (2019). Roadkill mitigation: trialing virtual fence devices on the west coast of Tasmania. Australian Mammalogy online early.

Freitas, C., Jusino, C. and Setz, E. (2015). Road-kills of the giant anteater in south-eastern Brazil: 10 years of monitoring spatial and temporal determinants. Wildlife Research 41(8), 673-80.

Glista, D., Devault, T., DeWoody, J. (2009). A review of mitigation measures for reducing wildlife mortality on roads. Landscape and Urban Planning 91(1), 1-7.

Goldingay, R. L., and B. D. Taylor. 2006. How many frogs are killed on a road in north-east New South Wales. Australian Zoologist 33, 332-336.

Goldingay, R. and Taylor, B. (2017). Targeted field testing of wildlife road-crossing structures: koalas and canopy rope bridges. Australian Mammalogy 39(1), 100-104.

Goldingay, R. and Taylor, B. and Parkyn, J. (2018). Use of tall wooden poles by four species of gliding mammal provides further proof of concept for habitat restoration. Australian Mammalogy online early.

Hobday, A. and Minstrell, M. (2008). Distribution and abundance of roadkill on Tasmanian highways: human management options. Wildlife Research 35, 712-26.

Huijser, M.P., Mosler-Berger, C., Olsson, M. and Strein, M. (2015). Wildlife warning signs and animal detection systems aimed at reducing wildlife-vehicle collisions. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Jones, M. (2000). Road upgrade, road mortality and remedial measures: Impacts on a population of eastern quolls and Tasmanian devils. Wildlife Research 27(3): 289-296.

Lassau, S., Ryan, B., Close, R., Moon, C., Geraghty, P., Coyle, A. and Pile, J. (2008). Home range and mortality of a roadside koala population at Bonville, NSW. Pp. 127-36 in ‘Too close for comfort: contentious issues in human-wildlife encounters’. Edited by D. Lunney, A. Munn & W. Meikle. Royal Zoological Society of NSW, Mosman, NSW, Australia.

Lester, D. (2015). Effective wildlife roadkill mitigation. Journal of Traffic and Transportation Engineering 3, 42- 51.

Lismore City Council (2013). Biodiversity Management Strategy for Lismore Local Government Area 2015-2035. Lismore City Council, Lismore, NSW.

Lismore City Council (2015). Comprehensive Koala Plan of Management for south-east Lismore. Lismore City Council, Lismore, NSW.

Sandpiper Ecological Surveys 46

Wildlife Roadkill Mitigation Report - Lismore LGA

Madden, J. and Perkins, S. (2017). Why did the pheasant cross the road? Long-term road mortality patterns in relation to management changes. Royal Society Open Science 4, 170617.

Mansergh, I.M. and Scotts, D.J. (1989). Habitat continuity and social organisation of the mountain pygmy- possum restored by tunnel. Journal of Wildlife Management 53(3), 701-707.

Osawa, R. (1989). Road kills of the Swamp Wallaby, Wallabia bicolor, on North Stradbroke Island, South-east Queensland. Australian Wildlife Research 16, 95-104.

Phillips, S. (2014). Factors contributing to ongoing koala mortalities in areas traversed by recent Pacific Highway upgrades on the far north coast NSW. Proceedings of the Australasian Network for Ecology and Transportation Conference. Coffs Harbour, Australia.

Ramp, D. and Roger, E. (2008). Frequency of animal-vehicle collisions in NSW. In ‘Too close for comfort: contentious issues in human-wildlife encounters’, pp. 118-26. Edited by D. Lunney, A. Munn & W. Meikle. Royal Zoological Society of NSW, Mosman, NSW, Australia.

Ramp, D. and Roger, E. (2008). Frequency of animal-vehicle collisions in NSW. In ‘Too close for comfort: contentious issues in human-wildlife encounters’, pp. 118-126. Edited by D. Lunney, A. Munn & W. Meikle. Royal Zoological Society of NSW, Mosman, NSW, Australia.

Rytwinski, T. and Fahrig, L. (2015). The impacts of roads and traffic on terrestrial animal populations. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Sandpiper Ecological (2017). Pacific Highway Upgrade – Glenugie: Operational phase fauna crossing monitoring program years 1-3. Report prepared for NSW Roads and Maritime Services.

Sandpiper Ecological (2018a). Wildlife Roadkill Mitigation Strategy for Lismore LGA. Stage 1: Assessment of Wyrallah Road koala roadkill black-spot and mitigation plan. Report prepared for Lismore City Council.

Sandpiper Ecological (2018b). Pacific Highway Upgrade – Sapphire to Woolgoolga: Operational phase monitoring Year 3 of Greys Dam frog pipes. Report prepared for NSW Roads and Maritime Services.

Scott, L.K., Hume, I.D., Dickman, C.R. (1999). Ecology and population biology of long-nosed bandicoots (Perameles nasuta) at North Head, Sydney Harbour National Park. Wildlife Research 26, 805-821.

Schwartz, A., Williams, H., Chadwick, E., Thomas, R. and Perkins, S. Roadkill scavenging in an urban environment. Journal of Urban Ecology 4(1), juy 006.

Shilling, F., Perkins, S. and Collinson, W. (2015). Wildlife roadkill observation and reporting systems. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Smith, D., van der Ree, R. and Rosell, C. (2015). Wildlife crossing structures: an effective strategy to restore or maintain connectivity across roads. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Sullivan, T., Williams, A., Messmer, T., Hellinger, L. and Krychenko, S. (2004). Effectiveness of temporary warning signs in reducing deer-vehicle collisions during mule deer migrations. Wildlife Society Bulletin 32, 907- 915.

Sandpiper Ecological Surveys 47

Wildlife Roadkill Mitigation Report - Lismore LGA

Taylor, B. and Goldingay, R. (2003). Cutting the carnage: wildlife usage of road culverts in north-east NSW. Wildlife Research 30 (5), 529-537.

Taylor, B. D. & Goldingay, R. L. (2004). Wildlife road-kills on three major roads in north-eastern NSW. Wildlife Research: 31(1), 83-91.

Taylor, B. D. & Goldingay, R. L. (2010). Roads and wildlife: impacts, mitigation and implications for wildlife management in Australia. Wildlife Research: 37(4), 320-31. van der Ree, R., Gagon, J. and Smith, D. M. (2015). Fencing: A valuable tool for reducing wildlife-vehicle collisions and funneling fauna to crossing structures. In R. van der Ree, D. Smith & C. Grilo (Eds) Handbook of Road Ecology, John Wiley & Sons, Sussex, UK.

Vestjens, W.J. (1972). Wildlife mortality on a road in NSW. Emu 73, 107-112.

Winnett, M.A. and Wheeler, A.H. (2002). Vehicle-activated signs: a large-scale evaluation. Prepared for Road Safety Division, Department of Transport. UK.

Sandpiper Ecological Surveys 48

Wildlife Roadkill Mitigation Report - Lismore LGA

Appendix A

WIRES Inc. Northern Rivers - Lismore LGA wildlife roadkill black-spots (as provided by wildlife carers 26 March 2018)

Nimbin area

• mostly wallabies in these areas but some birds also especially around the Nimbin town area • Stoney Chute Rd - entire length • Cecil St - especially between the preschool and Nimbin Rd, • Gungas Road, entire length • Crofton Rd especially along the straight stretch in front of the Fields place, • Nimbin Road...entire length but especially wherever there are trees around, or big dams etc • Blue Knob Road

Chilcotts Grass area

• Taylor Rd Chilcotts Grass - mainly wallabies, some birds • Rous Rd between Taylor Rd and Skyline Rd. - mainly wallabies

Lismore/Goonellabah area

• Leading into Oliver Ave and Kadina St. intersection, Goonellabah in the area of Birdwing Butterfly Walk - mainly possums, some birds) • Invercauld Road between no 52 and 54 – bushland - mainly wallabies • Cynthia Wilson Drive, midway - wallabies, many birds, occasional other animals • City View Drive (birds, wallabies, occasional other animals e.g. lizards) • Bruxner Hwy from Marom Creek to roundabout at Kadina Street. Many deceased - possums, echidnas, bandicoots removed from hwy. • Nimbin Road – 2-3 km stretch just out of Lismore - wallabies, echidnas, birds

Ducks Crossing signs on

• Dawson St between Zadoc and Uralba St. • Uralba St between Brewster and Dawson Sts. • Ducks also frequently cross in all streets leading unto the Roundabout at the bottom of Cynthia Wilson St. in the SCU area.

Rock Valley

• Rock Valley Road 300m on the Lismore side from approximately near Plim Bridge, and probably all the area up to and around Plim Bridge - mainly wallabies, some bandicoots etc. •

Jiggi/Georgica/Coffee Camp

• mainly wallabies in these areas, but also snakes bandicoots and other animals from time to time • Jiggi school zone, opposite Davis Road

Sandpiper Ecological Surveys 49

Wildlife Roadkill Mitigation Report - Lismore LGA

• Mountain Top Road, near Jacobson Road intersection • From 211 Davis Road heading north into the forest area • Gwynne Road and Oakey Creek Road • Lodge Road – whole length

Rosebank/Durroughby/Tullera

• Rosebank Road between the Mackie Road turnoff and Rosebank school - possums and wallabies • Repentance Creek Road between the Rosebank shop and Fox Road mainly - wallabies • Rosebank Road between Mackie Road turnoff and the Rosebank school - possums and wallabies • All of Mackie Road - possums and wallabies • All of Fox Road - koalas, possums and wallabies • Dunoon Road, especially between the Channon Road turnoff and Lismore - possums and wallabies • All of Arthurs Road Corndale - mainly wallabies • Durroughby Road, especially the Rosebank side - possums and wallabies • Numulgi road stretch to Lismore, - possums and wallabies • All of Emmerson Rd wallabies, - possums

Tuntable

• All of Tuntable Falls Road - Pademelons, wallabies

Caniaba

• Nimoola Road - wallabies • Rudgley Road -wallabies • Entire length of Caniaba Road - wallabies • Gardinia St - wallabies • Fredericks Rd - wallabies • Toohey road - wallabies and eastern grey kangaroos • Struthers Road - wallabies • Dougan road - wallabies, possums

Goolmangar

• Bourganville Road - wallabies • Pinchin road - wallabies

Sandpiper Ecological Surveys 50

Wildlife Roadkill Mitigation Report - Lismore LGA

Appendix B

Table B1. Minor and slight koala roadstrike black-spots and mitigation recommendations. ni = not inspected.

Black- Mitigation Recommendation Black-spot spot Locality Signed Speed Speed Pavement Rating Sign Other Extent of Management Zone No. Reduction treatment Not signed 80>60 Koala Zone No (gravel) Slash road edge 2km (80?) B3 Marom Ck Minor Not signed 80>60 Koala Zone Yes Slash road edge 1km (80?) Tie-in fence to culverts B4 Goonellabah, 50, 60 No Koala Zone Yes 50m either side & retrofit 2km Minor central Tregeagle dry ledge through culverts 80 80>60 Koala Zone Yes Slash road edge 8km Dunoon, X3 Minor 80, 60 80>60 Wildlife Zone Yes Slash road edge The Channon Rd to 3km S (3km) Modanville X4 Ruthven Minor 100 100>80 Koala Zone Yes VAS Flood Reserve Rd to 1km south (1km) Lychee Dr to Byrne Rd & 400m along Eureka Rd X5 Rosebank Minor 80, 50 80>60 Wildlife Zone Yes Slash road edge (1.5km) South X6 Minor 80 80>60 Koala Zone Yes _ 100m E of bend to 900m W of bend (1km) Gundurimba 500m E of Johnston Rd to 2km W of Johnston Rd X7 Clunes Minor 100 100>80 Koala Zone Yes VAS; slash road edge (2.5km) Not signed X9 Rosebank Minor No Wildlife Zone No (gravel) Slash road edge Repentence Ck Rd to Maso Rd (2km) (50-60) N Lismore, X10 Slight 50, 80 80>60 Wildlife Zone Yes Slash road edge Wilson Rd to 4km W (4km) Booerie Ck X8/B7 Numulgi Slight 80 80>60 Koala Zone Yes Slash road edge Dunoon Rd to Hunters Hill Rd (3.5km) X11 Caniaba Slight 80, 60 80>60 Wildlife Zone Yes _ Nimoola Rd to Fredericks Rd & along 5km (6.5km) X12 Nimbin Slight ni ni ni ni ni X13 Jiggi Slight ni ni ni ni ni 250m S Davis Rd to Oakey Ck Rd (3km)

Sandpiper Ecological Surveys 51

Wildlife Roadkill Mitigation Report - Lismore LGA

Appendix C

Figure C1. Minor koala black-spot (B3) in the Tucki Tucki area identified by Biolink (2017).

Figure C2. Minor koala black-spot (B4central) in the Goonellabah area identified in identified by Biolink (2017).

Sandpiper Ecological Surveys 52

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure C3: Minor koala black-spots (X3) in the Dunoon and (X8/B7) in the Rosebank/Numulgi area identified in the current study and by Biolink (2017).

Figure C4: Minor koala black-spot (X4) in the Tucki Tucki area identified in the current study and major koala black-spot (B2) also in the Tucki Tucki area identified by Biolink (2017). B2 was addressed in Phase 1 of the WRMR project.

Sandpiper Ecological Surveys 53

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure C5. Minor koala black-spot (X5, X9) in the Rosebank area identified in the current study.

Figure C6. Minor koala black-spot (X6) in the south Gundurimba area identified in the current study.

Sandpiper Ecological Surveys 54

Wildlife Roadkill Mitigation Report - Lismore LGA

Figure C7. Minor koala black-spot (X7) in the Clunes area identified in the current study

Sandpiper Ecological Surveys 55

Wildlife Roadkill Mitigation Report - Lismore LGA

Appendix D

Table D1. Wildlife roadstrikes in Lismore LGA for 2015-16 as supplied by Northern Rivers Wildlife Carers.

Date Location Suburb Postcode Species Cause Injury type Outcome 2 July 2015 Dunoon Rd Dunoon 2480 Red-necked Collision - Injury to body Died 1oomts from wallaby Motor Channon Rd turn Vehicle 5 July 2015 Corndale Rd Corndale 2480 Bandicoot, Collision - Injury to body Euthanased by Vet Northern brown Motor Vehicle 5 July 2015 275 Rosebank Rd Rosebank 2480 Cockatoo, Collision - Injury to head Euthanased by Sulphur-crested Motor Rehabilitation Vehicle Group 5 July 2015 31 Dalley St Lismore 2480 Pigeon, unknown Collision - Injury to head Euthanased by Vet Motor Vehicle 6 July 2015 At beginning of Tullera 2480 Kookaburra Collision - Injury to Euthanased by Vet 50zone on Motor forelimb/wing Dunnoon rd just Vehicle before Tullera 13 July Wyrallah Rd, just Wyrallah 2480 Echidna Collision - Injury to body Euthanased by Vet 2015 near 50 sign Motor entering Wyrallah Vehicle 27 July Near Recycle Lismore 2480 Noisy minor Collision - Injury to body Euthanased by Vet 2015 Depot at Lismore Motor Square Vehicle 8 August Ballina Rd Lismore 2480 Lorikeet, Collision - Concussed Euthanased by Vet 2015 Rainbow Motor Vehicle 8 August near G'bah Goonellabah 2480 Rosella, Eastern Collision - Concussed Escaped From Care 2015 Primary School, Motor Ballina Rd Vehicle 9 August Channon Market, The 2480 Bandicoot, Collision - Injury to head Dead On Arrival 2015 oval Channon Northern brown Motor Vehicle 10 August Bangalow Rd Bexhill 2480 Magpie Collision - Injury to body 2015 Motor Vehicle 20 August 598 Konorigan rd Konorigan 0 Unknown Collision - Injury to Euthanased by 2015 Motor hindlimb police Vehicle 26 August 1170 Dunoon Rd Dunoon 2480 Unknown Collision - Injury to 2015 macropod Motor hindlimb Vehicle 29 August at Recycling Lismore 2480 Cockatoo, Collision - Injury to 2015 depot opposite Unknown Motor forelimb/wing Lismore Square Vehicle Uralba St 1 23 Phyllis St South 2480 Magpie Collision - Injury to Euthanased by Vet September Lismore Motor forelimb/wing 2015 Vehicle 5 Cullen St Nimbin 2480 Carpet python Collision - Injury to body September Motor 2015 Vehicle 8 Opposite Goonellabah 2480 Red-necked Collision - Injury to Euthanased by September Crematorium wallaby Motor hindlimb Police 2015 Vehicle 9 Beside Cawongla Cawongla 2474 Unknown Collision - Injury to head Euthanased by Vet September Playhouse, Kyogle macropod Motor 2015 Murwullinbah Rd Vehicle 12 rifle Range Rd East Lismore 2480 Bearded dragon, Collision - Injury to body Euthanased by September Eastern Motor Rehabilitation 2015 Vehicle Group 12 10 Cumbrae Ave Lismore 2480 Unknown Collision - Injury to Euthanased by September wallaby Motor hindlimb Police 2015 Vehicle

Sandpiper Ecological Surveys 56

Wildlife Roadkill Mitigation Report - Lismore LGA

Date Location Suburb Postcode Species Cause Injury type Outcome 17 Wyrallah Rd Wyrallah 2480 Barn owl Collision - Concussed Euthanased by Vet September Motor 2015 Vehicle 18 325 Eureka Rd Eureka 2480 Barn owl Collision - Injury to Released September Motor forelimb/wing 2015 Vehicle 24 Invercauld Rd Lismore 2480 Unknown Collision - Injury to Euthanased by September Heights macropod Motor hindlimb Police 2015 Vehicle 28 39 Park St Lismore 2480 Lorikeet, Scaly- Collision - Injury to Euthanased by Vet September brested Motor forelimb/wing 2015 Vehicle 8 October Jiggi Public School Jiggi 0 Rosella, Eastern Collision - Injury to Euthanased by Vet 2015 Motor forelimb/wing Vehicle 8 October Nimbin Rd near Nimbin 2480 Kookaburra Collision - Unknown Resolved By Vets 2015 the Nimbin Motor Neighbourhood Vehicle Ctr 9 October Oliver Ave just Goonellabah 2480 Lorikeet, Collision - Concussed Released 2015 after Kadina St Rainbow Motor Vehicle 18 October Wyrallad Rd, near Lismore 2480 Duck, Wood Collision - Injury to Euthanased by Vet 2015 tip Motor forelimb/wing Vehicle 19 October FOK centre, Rifle Lismore 2480 Lorikeet, Collision - Injury to head Euthanased by Vet 2015 Range Rd Rainbow Motor Vehicle 20 October Billen Rd Georgica 2480 Swamp wallaby Collision - Injury to head Euthanased by Vet 2015 Motor Vehicle 27 October 719 Ballina Rd Goonellabah 2480 Galah Collision - Multiple Dead On Arrival 2015 Motor health issues Vehicle 27 October Just north of 60 Modanville Noisy minor Collision - Concussed Released 2015 signs heading Motor north on Dunoon Vehicle Road 7 Outside southside South 2480 Noisy minor Collision - Concussed Released November pharmacy Lismore Motor 2015 Vehicle 7 Richmond Hill Rd Richmond 2480 Red-necked Collision - Injury to Euthanased by Vet November Hill wallaby Motor hindlimb 2015 Vehicle 13 69 Avondale Ave Lismore 2480 Currawong, Collision - Injury to Euthanased by Vet November Unknown Motor forelimb/wing 2015 Vehicle 14 The Channon Rd The 2480 Lorikeet, Scaly- Collision - Concussed Died November Channon brested Motor 2015 Vehicle 15 next to Lismore Lismore 2480 Lorikeet, Collision - Injury to November Square, Uralba St Unknown Motor forelimb/wing 2015 Vehicle 18 97 Dalley St Lismore 2480 Galah Collision - Injury to November Motor forelimb/wing 2015 Vehicle 20 Caroona Nurding Goonellabah 2480 Lorikeet, Collision - Neurological Euthanased by Vet November home Unknown Motor damage 2015 Vehicle 20 5 Ks from Tuncester Pheasant coucal Collision - Unknown Released November Tuncaster turnoff Motor 2015 on Kyogle road. Vehicle 21 Nimbin Rd Koonorigan 0 Kookaburra Collision - Injury to body Resolved By Vets November Motor 2015 Vehicle 21 108 Wyrallah Rd Wyrallah 2480 Carpet python Collision - Injury to tail November Motor 2015 Vehicle

Sandpiper Ecological Surveys 57

Wildlife Roadkill Mitigation Report - Lismore LGA

Date Location Suburb Postcode Species Cause Injury type Outcome 25 Outside Trinity Lismore 2480 Carpet python Collision - Injury to body Died November college Motor 2015 Vehicle 26 5/18 Oliver St East Lismore 2480 Crow Collision - Injury to Euthanased by Vet November Motor forelimb/wing 2015 Vehicle 26 On road in front Lismore 2480 Figbird Collision - Injury to tail November of City Toyota, Motor 2015 Bruxner Highway Vehicle 29 Nimbin 2480 Kookaburra Collision - Unknown November Motor 2015 Vehicle 5 From Nimbin, Nimbin 2480 Kookaburra Collision - Injury to Resolved By Vets December about 50 meters Motor hindlimb 2015 up hill from flat Vehicle on Mountain Top rd after Lodge. 13 Lot 8 Callistemon Federal 2480 Bandicoot, Collision - Injury to body December Drive Unknown Motor 2015 Vehicle 25 On road in front Lismore 2480 Honeyeater, Collision - Unknown Died December of Carolna Unknown Motor 2015 Nursing home Vehicle 7 January Cox Rd Koonorigan 0 Honeyeater, Collision - Injury to body Euthanased by Vet 2016 Lewin's Motor Vehicle 8 January Near Keen St vets Lismore 2480 Masked lapwing Collision - Injury to Euthanased by 2016 (plover) Motor forelimb/wing Rehabilitation Vehicle Group 9 January 6 Fermony Ave Lismore 2480 Magpie Collision - Injury to Euthanased by Vet 2016 Motor forelimb/wing Vehicle 18 January Sunmmerland Afterlee 2474 Parrot, Unknown Collision - Injury to body Euthanased by Vet 2016 Way near Kyogle Motor turn off Vehicle 18 January 18 Crescent St Lismore 2480 Kingfisher, Collision - Injury to Euthanased by Vet 2016 Unknown Motor forelimb/wing Vehicle 19 January Cnr Oliver Ave, Goonellabah 2480 Tawny Collision - Concussed Unknown And 2016 Rous Rd on frogmouth Motor Cannot Be roundabout Vehicle Determined 28 January 'Blacksheep Farm' Nimbin 2480 Echidna Collision - Resolved By Vets 2016 449A Gungas rd Motor Vehicle 3 February Bruner Highway Lismore 2480 Duck, Unknown Collision - Injury to Died 2016 outside Bunnings Motor hindlimb Vehicle 3 February SCU University East Lismore 2480 Water dragon, Collision - Injury to body Resolved By Vets 2016 Eastern Motor Vehicle 5 February Channon Oval, The 2480 Red-necked Collision - Injury to head Dead On Arrival 2016 The Channon Rd Channon wallaby Motor Vehicle 7 February 560 Mountain Georgica 2480 Tawny Collision - Released 2016 Top Rd frogmouth Motor Vehicle 10 Anderson Rd Nimbin 2480 Eastern long- Collision - Injury to body Died February neck turtle Motor 2016 Vehicle 10 Wyrallah Rd near East Lismore 2480 Ibis, White Collision - Injury to body February Lismore tip Motor 2016 Vehicle 14 Young Rd Tuntable 2480 Red-necked Collision - Injury to body February Creek wallaby Motor 2016 Vehicle 20 Keen St near Lismore 2480 Figbird Collision - Injury to February Gerard St Motor forelimb/wing 2016 Vehicle

Sandpiper Ecological Surveys 58

Wildlife Roadkill Mitigation Report - Lismore LGA

Date Location Suburb Postcode Species Cause Injury type Outcome 22 35 Gwynne Rd Georgica 2480 Tawny Collision - Injury to Released February frogmouth Motor forelimb/wing 2016 Vehicle 23 Basil Rd Nimbin 2480 Red-necked Collision - Orphaned Died February wallaby Motor 2016 Vehicle 29 Emerson Rd, 2nd Rosebank 2480 Red-necked Collision - Other In Care February driveway off pademelon Motor 2016 Repentence Ck Rd Vehicle 29 Emerson Rd, 2nd Rosebank 2480 Red-necked Collision - Other Dead On Arrival February driveway off pademelon Motor 2016 Repentence Ck Rd Vehicle 3 March Emerson rd Rosebank 2480 Tawny Collision - Injury to Euthanased by 2016 frogmouth Motor forelimb/wing Rehabilitation Vehicle Group 4 March 449 Ballina Rd Goonellabah 2480 Dove, Unknown Collision - Unknown Dead On Arrival 2016 Motor Vehicle 7 March 3981 Kyogle Rd Lillian Rock 0 Echidna Collision - Injury to body 2016 Motor Vehicle 11 March 469 Green Pigeon Green 2474 Unknown Collision - Orphaned Transferred to 2016 Road Pigeon macropod Motor Other Animal Vehicle Organisation 24 March Breckenridge Rd Wyrallah 2480 Bearded dragon, Collision - Injury to head Euthanased by Vet 2016 Eastern Motor Vehicle 24 March 2242 Dunoon Rd Dorroughby 2480 Commom ringtail Collision - Injury to Euthanased by 2016 possum Motor hindlimb Rehabilitation Vehicle Group 1 April Corn of Skyline Rd Goonellabah 2480 Kookaburra Collision - Concussed Released 2016 and Rous Rd Motor Vehicle 2 April 83 Dalley St East Lismore 2480 Lorikeet, Collision - Injury to body Died 2016 Rainbow Motor Vehicle 3 April Dibbs Street Lismore 2480 Rosella, Collision - 2016 Unknown Motor Vehicle 7 April 100 mts before Goonellabah 2480 Tawny Collision - Unknown Released 2016 roundabout to frogmouth Motor Coles Vehicle 7 April 398 Rous Rd Tregeagle 2480 Unknown Collision - Orphaned 2016 macropod Motor Vehicle 10 April Tuntable Creek The 2480 Bandicoot, Collision - Orphaned Disappeared 2016 Rd Channon Unknown Motor Before Rescue Vehicle 1 May 2016 on road outside Eureka 2480 Pigeon, White Collision - Injury to Euthanased by Eureka PS headed Motor forelimb/wing Rehabilitation Vehicle Group 1 May 2016 200 River Bank Rd Wyrallah 2480 Tawny Collision - Injury to body Resolved By Vets frogmouth Motor Vehicle 3 May 2016 The Channon Oval The 2480 Kookaburra Collision - Injury to Channon Motor forelimb/wing Vehicle 4 May 2016 146 Gwynne Rd Jiggi 0 Pigeon, unknown Collision - Injury to body Resolved By Vets Motor Vehicle 7 May 2016 614 Dunoon Rd Tullera 2480 Unknown Collision - NA Dead On Arrival macropod Motor Vehicle 8 May 2016 On Rous Rd near Goonellabah 2480 Mountain Collision - Orphaned Died 121 brushtail possum Motor Vehicle 11 May cnr Tullera 2480 Unknown Collision - Injury to Disappeared 2016 Borton,Bentley Rd wallaby Motor hindlimb Before Rescue Vehicle

Sandpiper Ecological Surveys 59

Wildlife Roadkill Mitigation Report - Lismore LGA

Date Location Suburb Postcode Species Cause Injury type Outcome 14 May Near 78 The The 2480 Magpie Collision - Concussed Disappeared 2016 Channon Rd Channon Motor Before Rescue Vehicle 14 May Laural Ave East Lismore 2480 Lorikeet, Collision - Injury to body Euthanased by Vet 2016 Rainbow Motor Vehicle 14 May 43 Tickles Rd Upper 2482 Red-necked Collision - Orphaned Euthanased by 2016 Coopers pademelon Motor Rehabilitation Creek Vehicle Group 15 May 16 Springvale Rd Eureka 2480 Tawny Collision - Unknown Euthanased by Vet 2016 frogmouth Motor Vehicle 17 May 1k west of Eureka 2480 Kookaburra Collision - Concussed released 2016 Springvale rd turn Motor off on Bangalow Vehicle rd 19 May Blue Knob rd Nimbin 2480 Unknown Collision - Injury to Transferred to 2016 wallaby Motor hindlimb Other Animal Vehicle Organisation 24 May 369 Skyline Rd Goonellabah 2480 Unknown Collision - Injury to Disappeared 2016 wallaby Motor hindlimb Before Rescue Vehicle 28 May 54 Invercaul Rd Goonellabah 2480 Unknown Collision - Injury to body 2016 wallaby Motor Vehicle 31 May Crn Dawson & Lismore 2480 Tawny Collision - Concussed Resolved By Vets 2016 Woodlark Sts frogmouth Motor Vehicle 22 June 500mt down Boat 2480 Magpie Collision - Injury to 2016 Boatharbour Rd Harbour Motor forelimb/wing from Bangalow Vehicle Rd end 25 June Goolmangar Shop Goolmangar 2480 Galah Collision - Concussed Escaped From Care 2016 Motor Vehicle

Sandpiper Ecological Surveys 60