Paekakariki-Pukerua Bay Escarpment Lizard Protection Trial Lizard on the escarpment: Peter Kentish

Paekakariki-Pukerua Bay Escarpment Lizard Protection Trial Lizard on the escarpment: Peter Kentish Prepared by Paul Callister April 2017

Vision By 2030 lizards will become abundant on the Paekakariki-Pukerua Bay escarpment. Goals To gain an understanding of lizard diversity and abundance on a small site on the Paekakariki- Pukerua Bay escarpment through lizard surveys and monitoring programmes. On this small site to improve animal pest control regimes so they protect existing lizard populations and allow these populations to recover over a five to ten year period. If pest control is successful, to explore options for translocation from local lizard breeding programmes. To develop a template to guide other community groups. To help increase community awareness of the key role lizards can play in New Zealand s ecosystems. Acknowledgements Many people contributed to this draft and most are mentioned in the body of the report. However, I would like to particularly thank a number of people. Angus Hulme-Moir first suggested setting up the lizard protection trial and then provided ongoing support and feedback as the plan was developed. Trent Bell and Murray Williams also provided useful comment on early drafts. Dalice Sim offered helpful statistical advice in the early planning stages. Working in the field with Sarah Herbert helped me gain some confidence in surveying lizard populations. Many people volunteered in setting up the trial and the success of it will depend on a continuing high level of support over its lifetime. The project would not have been possible with the support of the Ministry of the Environment funded Kapiti Biodiversity project, a range of specialists within the Department of Conservation and Nga Uruora.

Contents Introduction... 4 Mice control as a key element of lizard protection... 6 The site and the trial design... 8 A control site... 15 Setting up the trial area... 16 Pest monitoring and trap management on the trial site... 18 The wider pest control regime... 19 Lizard monitoring... 21 Analysing the data... 23 Risks including health and safety... 23 Publicity for the project... 24 Timeline... 25 Options if the trial is not successful... 26 References... 26 Appendix 1 Data sheets (based on an Ecogecko template)... 27 Appendix 2 Lizard monitoring H & S plan... 29

Introduction Lizards are New Zealand s largest terrestrial vertebrate group with more than 100 species. Historically, they would have occupied almost all available ecosystems from coastal shores to mountain peaks. Lizards play an important role in ecosystem processes and function as predators, pollinators, frugivores and seed dispersers. Lizards are emerging as iconic flagship and indicator species in conservation and ecological restoration. Despite 85 percent of this fauna being threatened or at-risk, lizards can be exceptionally abundant when released from mammalian predation pressure. Nga Uruora began animal pest control in 1997. In the early days of pest control on the Paekakariki- Pukerua Bay escarpment most attention focussed on making the area stock proof and targeting possums in our Kohekohe remnants. 1 A key aim was to protect the existing forest and assist in natural revegetation. Over time, the pest control effort changed. The area trapped has expanded. Mustelids and, to a lesser degree, rats have been targeted. The aims were to protect both forest areas and to support bird populations. Nga Uruora has a vision to bring the birdsong back from Kapiti Island and the protection of birds has been a key driver in pest control efforts. In this early phase of pest control, mice were not seen as a key threat to birdlife. In 2012 the Wellington Regional Lizard Network published a lizard strategy for the Wellington region (Romijn et al. 2012). This document brought together the views of lizard experts (Herpetologists) along with a variety of other stakeholders, including the Department of Conservation and local authorities. This report painted a picture of significant lizard diversity in the Wellington region. However, it also identified a range of threats to these populations. Balancing these factors, the study also identified opportunities to support gecko and skink populations in the Wellington region. Lizards should be abundant in our area. But through predation and habitat change they are under extreme pressure. Predators include rats, mice, hedgehogs and mustelids. Habitat change can include areas of former scree grassing over. 1 Macrons are not used in this report.

In June 2015, it was announced that the government was providing a $294,000 Community Environment Fund grant for a project aimed at protecting and restoring biodiversity on the southern part of the Kapiti Coast. The funding has been aimed at helping local communities to support bird populations, improve habitats for lizards and weta, propagate and plant rare dune plants, increase fish and insect populations in streams, and facilitate an increase in local pest control. 2 Aside from the fact that there have historically been few sightings of lizards on the escarpment, little was known about the local lizard populations. 3 The MfE project gave not only Nga Uruora a chance to study lizard abundance but also allowed our neighbours at Queen Elizabeth Park and Whareroa Farm Reserve to explore their own lizard populations. It was also hoped the research would give Nga Uruora and the other groups some guidance as to how to improve pest control to support lizard populations. In 2015, at the start of the MfE project, two pest control reports were prepared to guide the increased pest control on the Paekakariki-Pukerua Bay escarpment. There was a report setting out how there would be the creation of a wider Kapiti Mainland Island 4. This was followed by a more detailed operational report for Nga Uruora. 5 A lizard strategy report was also prepared to help guide the lizard research. 6 At this point there was a lack of information on the best ways to support local lizard populations aside from the standard ideas of increasing overall predator control. In summer 2016, Ecogecko and volunteers undertook surveys at both Queen Elizabeth Park and Whareroa farm. Ecogecko also undertook two small surveys on the Paekakariki escarpment. A report summarising the findings was then prepared. This set out a list of lizards that might be expected to be found in the area, what was found and some recommendations for further study. 7 It also gave some ideas for enhancing pest control, especially in relation to mice on the escarpment. In both Queen Elizabeth Park and Whareroa farm very few lizards were found. While more lizards were found on the escarpment, it was clear from the survey work mice were common and likely to 2 http://www.naturespace.org.nz/groups/kapiti-coast-biodiversity-project 3 In the summer of 2016/2017 lizards were starting to be seen regularly at the site known at the quarry. 4 http://www.naturespace.org.nz/sites/default/files/kapiti_biodiversity_pest_control_2015.pdf 5 http://www.naturespace.org.nz/sites/default/files/pest_control_strategy_nga_uruora_2015-2018_0.pdf 6 http://www.naturespace.org.nz/sites/default/files/lizard_strategy_kapiti_biodiversity_project_december_2015.pdf 7 http://www.naturespace.org.nz/sites/default/files/ecogecko%20lizard%20report%20for%20kapiti%20biodivesity%20project%20year%20one%20.p df

be impacting lizard populations. It has also become clear from autopsy work carried out by Sue Blaikie on locally trapped rats and mustelids that lizards formed part of their diet. In mid-2016 a pest control workshop was held bringing together the local pest control community. A key theme was how to support lizard populations with a particular emphasis on controlling mice. At this workshop, Ecogecko set out their ideas for mice control. 8 In addition, Angus Hulme-Moir explained what the Friends of Whitireia Park were doing with their lizard protection trial. 9 Following the workshop, Nga Uruora began exploring setting up a lizard protection trial on the escarpment. There were a number of reasons for choosing the escarpment. These were that it had suitable habitat to support lizard populations, that survey lizard populations were higher than in neighbouring areas, that existing non-mouse predator control was well developed and that toxins were able to be used on the site. Angus Hulme-Moir helped facilitate this process both as a volunteer at Whitireia Park and as an employee of the Department of Conservation. The Department of Conservation provided advice on both pest control and lizard monitoring. Assistance was also provided by researchers from Victoria University (Nicola Nelson and Sarah Herbert) and Otago University (Dalice Sim). As a result of this advice, but also consideration of what work volunteers could undertake, a lizard protection the trial plan was devised for the escarpment. The idea was to set up some demonstration projects which would provide guidelines for community groups to support lizard populations on small sites. This project is led by the Kapiti Biodiversity project (represented by Nga Uruora). The Department of Conservation has provided, and will continue to provide, technical support facilitated by a Memorandum of Understanding with Nga Uruora (signed in April 2017). The project has the support of Ngati Toa. Ecogecko has also provided technical and practical support. Mice control as a key element of lizard protection This section draws heavily on a 2016 paper prepared by James Reardon of the Department of Conservation. There has only been minor editing to make it more relevant to the Paekakariki- Pukerua Bay escarpment. 8 http://www.naturespace.org.nz/sites/default/files/ecogecko.pdf 9 http://www.naturespace.org.nz/sites/default/files/angus_lizard%20conservation%20whitireia.pdf

It has long been known that mice are a direct predator and competitor of lizards in New Zealand (Newman 1993). More recently, researchers have observed their predatory behaviour during post release monitoring of some of our largest lizard species suggesting their impacts are not limited to small lizard species (Norbury et al. 2014). Therefore a key aim of a lizard protection trial is to maintain mice to sustainably low indexes of abundance throughout all seasons. The most cost effective control of mice has traditionally involved toxins, notably the second generation anticoagulant brodifacoum. However many scientists suggest that due to the bioaccumulative and persistent nature of this toxicant that strategy is not viable long-term. Ideally, other methods of mouse control will be found. To date there is no evidence of trapping programmes achieving sustained suppression to appropriate levels. Recent results from community conservation efforts to suppress mice at Whitireia Park using the A24 self-resetting trap demonstrated good levels of suppression at high density deployment but unfortunately these outcomes reversed as productivity increased mouse populations which then quickly overwhelmed the trapping effort. The strategy in the lizard protection trial will be to supplement a trapping network with triggered poisoning of mice to maintain indexed abundance of mice to below 5%. Given the promising results from the Whitireia Park management areas (Angus Hulme-Moir pers comm.), together with strong pressure to employ self-resetting traps, the trial will follow recommendations of employing a network of A24s at 20m spacings supplemented by bait stations. As well as covering the area intended for lizard recovery the grids will extend to buffer lines surrounding this area. According to the Department of Conservation, the Saddle island study has provided much information about mouse home range changes with density and their initial estimations Densities on the island ranged from 8.8-19.2 mice/ha, with home ranges varying from 0.15-0.48 ha (MacKay et al. 2011). Taking a mid-point of that range would require a buffer of at least 40m to account for extrapolated home range diameter. Therefore, to protect 1ha of habitat would require an area of 180m x 180m filled with traps and bait stations at 20m spacings. 10 The Department of Conservation note that it is vital that close scrutiny of detection and bait take data is maintained to help adaptively manage an expansion of this buffer if data suggests it is inadequate. For the first phase of this management (year 1 &2) it has been recommended that the project record tracking rates (using tracking tunnels) monthly as standard procedure and the frequency is elevated to fortnightly when tracking is rebounding between 3-5%. Once tracking rates hit 3%, fortnightly 10 For the mouse eradication program on nearby Mana Island the spacings of bait stations was 50 metres http://www.manaisland.org.nz/mana-island- 2/history/

filling of bait stations it is recommended begin filling bait stations. For practical reasons (number of tracking tunnels) only 5% tracking bands are able to be used in this trial. The most effective toxicant, brodifacoum, will initially be used in the trial. Due to concerns about its long life in the environment those running the trial will continue to seek safer toxins. In terms of safety of brodifacoum for lizards, a 2016 study gives some comfort (Weir, et al. 2016). The Department of Conservation note that the rebaiting strategy is critical to maintain lethal dose availability at all times during the toxin knockdown. The duration of the toxicant application will be informed by peripheral and control rodent abundance indexing. As soon as tracking rates in the core area decline to acceptable levels the toxicant can be withdrawn and the hope is that the A24 self-resetting traps can maintain suppression. A risk in this design is that the toxicant dominates the suppression mechanism and the expensive investment in A24 traps plays a negligible role in suppression. This risk further emphasises the need to maintain monitoring data and to adaptively review the programme. The site and the trial design The lizard protection site is located on the Paekakariki-Pukerua Bay escarpment. The Department of Conservation recommended an overall site 180 by 180 metres for pest control. Within this was recommended a 100 x 100 lizard monitoring area. Finding the best site on the escarpment has its challenges. Ideally the site would have easy access, be either fully above or below Te Araroa trail, be surrounded by intensive pest control, have suitable lizard habitat plus not be directly above the railway tracks as there is a need to keep away from this area with a danger of debris falling on tracks. Trevor Thompson, Field Officer for the QE2 Trust, has noted that the best sites for lizards in the Wellington Region are those without long grass and also very dry (personal communication, 2017). This site does have long grass but is dry with no visible watercourses in the area. After a number of investigations, including a visit from Les Moran and Lynn Adams from the Department of Conservation Technical Advisory Group, it was decided to use an open site at the centre of the escarpment. 11 While it does not fully meet the ideal site criteria it seems the best option. There is regenerating forests on the margins of this site, but the central part is mainly covered in grass 11 The site chosen is directly below a memorial seat for ecologist Geoff Park. This seems very appropriate as it was Geoff s book Nga Uruora that the group looking after the escarpment is named after.

and muehlenbekia with occasional propinqua and flax plants. This area is adjacent to an area known as the Ecosite which contains mature Kohekohe forest. This site has had some form of pest control since 1997 and in 2015 became one of Nga Uruora s two rat free sites. The Ecosite has an extensive network or traps and bait stations and has a set of tracking tunnels within it. In the first tracking tunnel exercise carried out in November 2015 before the expanded pest control network was put in tracking tunnel rates for rats were 33%. In May 2016 there were 5% and in November 2016 a zero result was achieved. There was only a small amount of mice and hedgehog presence recorded. To reach both the Ecosite and the proposed lizard protection site, a zig zag track from the current State Highway 1 (known as Bob s track) can be utilised. 12 Bob s track was the site of one of Ecogecko s surveys in summer 2016 and a small number of Northern grass skinks and Raukawa geckos were detected. But the survey also showed the area had a high number of mice at that time of the year. Figure 1 shows Bob s track, a 180 by 180 metre square (yellow) and the 100 by 100 metre square (red). Figure 1 While both the 180 metre square and the 100 metre are mostly below Te Araroa some parts are above the track. However, all of the 50 metre square is below it. In addition, almost all the 50 metre square 12 This track was built by Nga Uruora to allow easy access to the Ecosite before Te Araroa was built.

contains mainly grass and muehlenbekia whereas outside this there are forested areas. There are also some small areas of flax. The site is very steep especially below the 50 metre square. There is also some weed invasion starting to occur within the 50 metre square, primarily fennel. It is proposed the fennel would be controlled. The steep exposed nature of the site provides a risk to the success of the project as there will be challenges accessing parts of it, especially in poor weather conditions. Figure 2 The Department of Conservation recommended a 180 metre square with a 100 metre square within it (Figure 3). Within this 180 metre square would be traps and/or bait stations at 20 metre intervals. Figure 3

In the Paekakariki-Pukerua Bay site it was decided to put in 40 Goodnature A24 traps with 75 bait stations to cover the full 180 x 180 metre grid. The traps/bait stations are 20 metres apart in the core area and, initially in the first year, in a 30 x 20 metre grid outside of this. 13 The bait stations are Philproof economy rodent bait stations. 14 While it was recommended that the lizard monitoring area be 100 x 100 metres, site considerations as well as concerns about finding enough long term volunteer input meant that initially it was decided to initially focus on a 50 x 50 metre lizard monitoring site. However, during the installation phase in late December 2016 it was decided to increase this to 60 x 50 metres. Within this 60 x 50 metre lizard monitoring site, onduline and pitfall traps (4 litre) were installed in December 2016at 10 metre spacings. This provides 42 sites for monitoring lizards. Within this monitoring area and slightly beyond tracking tunnels were installed in late December 2016/early January 2017 at 20 metre spacings. There are 20 tracking tunnels. This means tracking indexes have to be calculated in 5% bands so the trigger of 3% for putting out poison bait cannot be used. Thus if any one (or more) of the tracking tunnels shows mice activity it will trigger a need to fill the bait stations. In this core site 20 Goodnature traps were installed in April 2017. These also have a bait station attached. Ten of the Goodnature traps have counters attached to give another monitor of catch activity. In addition, a further ten A24s will have remote sensors attached. 13 If a 20 x 20 metre spacing is needed across to whole site to keep mouse numbers down to acceptable levels then additional bait stations will be installed. 14 http://www.philproof.co.nz/rat_showproduct.php?cat=3&sub=4

Figure 4 Surrounding the core of the lizard monitoring site out to the full 180 metres either way are another 20 Goodnature traps with bait stations and then bait stations only around the edge (a total of 40 Goodnature traps). 15 On the outer edge the traps/bait stations are at 30 metre spacings across the slope by 20 metres down. If this proves inadequate to control mice then this will be reduced to 20 x 20 metre spacings. 15 The additional 20 A24s have been lent to the project by Friends of Whitireia Park.

Figure 5 In addition, included within the lizard site trap network as well as nearby to it will be up to 13 remote sensed traps. Nine of these traps will be placed within the 180 x 180 metre lizard control trial at approximately 90 metre spacings. These will be placed on site in April 2017. These will be targeting primarily mustelids, but also potentially catch any hedgehogs or rats that have avoided the A24s and, in the case of rats, the bait stations. There are two reasons including the remote sensed traps. First, the remote sensing allows a quicker response to keeping the network fully covering the area. But more importantly in relation to the lizard trial it will allow fresh specimens to be quickly collected and autopsied to assess what they are eating. 16 This will be another way of monitoring lizard populations. 16 To date only rats and mustelids have been autopsied but not hedgehogs.

Figure 6 The remote sensing experiment is a collaboration between Econode, Nga Uruora (on behalf of the Kapiti Biodiversity Project) and Groundtruth. 17 In the first stage of the trial, the setup the aerial for radio coverage did not cover the lizard protection site. As a result an additional aerial was set up on Tokomapuna/Airplane Island on the east side of Kapiti Island in April 2017. 17 http://www.naturespace.org.nz/sites/default/files/remote%20sensing%20trial.pdf

A control site It was discussed in early stages of planning that it would be ideal to have a control site on the escarpment for lizard monitoring. For a variety of reasons it was decided this was impractical. As a rough guide to what would be happening without pest control two similar sites on the escarpment were chosen and a small network of 12 onduline squares set up at 10 metre distances. These will be checked at the same time as the main monitoring. It is understood that while this will not provide valid scientific data it may some idea of how background conditions (season etc) might be affecting lizard numbers. In the first lizard monitoring this site will not have pitfalls but these will be installed for further monitoring. The edge of the control sites are at least 75 metres away from the edge of the intense mouse control, well outside of mice home ranges. They are also at the top of Bob s track to allow easy checking. Figure 7 Potentially Mana Island, given its proximity to the trial area and that it is a pest free environment, should provide some measure of expected lizard abundance in a low pest environment. It may be that if regular surveys were carried out at Mana then seasonal variations in lizard abundance could be assessed giving some idea of whether changes in lizard numbers on the escarpment were due to pest control or other factors such as weather conditions. This will need to be explored further.

Setting up the trial area As discussed, the onduline, pitfalls and tracking tunnels were laid out in later December 2016/early January 2017. A number of working bees were held and following people assisted: Jim Hammond, Chris Keating, Jean Fleming, Tony Older, Sue Boyde, Glenda Robb, Liz Johns, Andy McKay, Paul Callister, Peter McLaughlin, David McKay, Michael Bennett and Vicky Griffin. Figure 8

Much of the work setting ups the pest control network has been carried out by Peter McLaughlin as a contractor to the Kapiti Biodiversity Project. Pest monitoring and trap management on the trial site Pest numbers within the trial area will be monitored in a number of ways. These are: Tracking tunnels Catch data from the remote sensed DOC200s Catch data from 10 remote sensed A24s. Data recorded on ten of the A24s with counters. Motion cameras placed in and around the area from time to time Tracking tunnels will be the main tool used to determine how the pest control regime is working and what responses are needed. However, the other tools will give us some additional information. The first full tracking tunnel monitor will be carried out in mid-april 2017. This will determine mice presence just after the first lizard monitor is carried out. Tracking tunnel monitors will then be carried out monthly until May. While the recommendation is two weekly monitors in peak mouse season are recommended it is considered not possible using volunteer labour. Monitors will then be carried out every two months. Monthly tracking tunnel monitors will resume again in spring (September) and the same regime will be carried out each year. Once the first lizard monitor is carried out, the 35 outer bait stations without A24s will be filled with brodificoum blocks. Throughout the year, these will be checked once a month and filled when empty. The bait stations with the A24s will only be filled when the A24s no longer keep mice numbers down to 5% or less on tracking cards (ie having mouse footprints on just one of the 20 tracking cards).

The wider pest control regime The lizard trial is embedded within the wider pest control on the escarpment. The Department of Conservation provides an ideal of how this wider pest control would operate (see earlier Figure 4). On the escarpment such a broad pattern of pest control does occur in a North and South direction. However, topography and other factors mean the spacing is not as exact as the diagram. On the site chosen for the lizard protection trial directly above it is an area with intensive pest control. Next to this in farmland. Below the trial is State Highway 1 then the sea. Figure 9 Figure 10 uses December 2016 Trap.nz data to show bait stations and traps surrounding the lizard protection trial.

Figure 10 Currently the traps used around the lizard protection site are DOC200s with a small number of Goodnature A24s. The bait stations are Philproof minis. There are also a small number of Timms traps deployed targeting possums and cats depending on what bait is used. This existing trap network relies volunteers checking them at least monthly. The bait stations are filled twice a year, in spring with brodificoum and in autumn with diphacinone. The success (or otherwise) of this wider trapping effort is measured in a number of ways. First, there is the data on catch rates. However, this does not indicate what is left in the environment only what is caught. Nga Uruora has also begun using motion cameras to help assess pest numbers. In addition, tracking tunnels are monitored on two sites on the escarpment twice a year, once in spring and once in autumn. One of the monitoring sites is directly above the trial site with 12 tracking tunnels deployed at approximately 120 metre spacing (Figure 11). These are mostly in the forest area but some are near to open grassland. Therefore there are a total of 32 tracking tunnels either within or next to the lizard protection trial.

Figure 11 Lizard monitoring Ecogecko suggests that the escarpment should support populations of Ngahere gecko, Northern grass skinks, Barking geckos, Ornate skinks, Copper skinks, Brown skinks, Spotted skinks and Raukawa gecko. Not long ago, there were probably Whitakers skinks. A key goal of this experiment is to see lizard numbers, and hopefully variety of lizards, increase over the trial period. However, given the slow rate of breeding the change in numbers over five years is not likely to be dramatic. For example for geckos the rate of increase is intrinsically very low (3-4 years to reach maturity, 2 offspring per year). Copper skinks also have only 2-3 offspring per year, but they do mature in 2

years. Given changes are likely to be small, it is particularly important to collect high quality reliable data. Expert statistical assistance will also be needed in interpreting lizard data collected. The Department of Conservation advise that there is no magic number or configuration of pitfalls or days checked. However, they stress that there is a need to do exactly the same thing consistently over the duration of the study. They note that the number of checks per year can vary somewhat because they will be converted to captures per trap per day. However, if a check is carried out, all the pitfalls must be checked that day, and the lay-out, set-up, baiting etc. of traps cannot change over the course of the study. The Department of Conservation recommend only checking during mild weather thus avoiding extremes of cold, heat and heavy rain. On such an exposed site checking in poor weather conditions would be difficult anyway. They also recommend avoiding checking during prolonged droughts when animals may be stressed by water loss and water provided in pitfalls may evaporate quickly. While the volunteers involved in monitoring will have all been trained in lizard handling there is no provision for on-going expert support of monitoring. Therefore only basic data will be collected. This will be species (if known) and size (an estimate if they are a juvenile or adult). The identification guides produced by Greater Wellington Regional Council will be made available to those carrying out the surveys. 18 If expert help does become available during any of the monitors then additional data can be collected. Some additional background data, including weather conditions, will be collected during each monitor. The project will use datasheets provided by Ecogecko (see appendix 1) The first lizard monitor will be carried out in late /early April 2017. By this time the onduline squares and pitfalls will have been in place for three months. The monitor will take place over four days over two weekends (a total of 8 days). The pitfalls will be opened on Thursday evening and baited with tinned pear. The control onduline site will also be baited with pears. The pitfalls and onduline will then be checked on Friday, Saturday and Sunday. The pitfalls will be closed after checking on the Sunday. In the first monitor, Trent Bell from Ecogecko has agreed to help on the first few days. 18 http://www.naturespace.org.nz/sites/default/files/u4/geckos%20of%20the%20wellington%202012.pdf http://www.naturespace.org.nz/sites/default/files/u4/skinks%20of%20the%20wellington%202012.pdf

This pattern of checking will be repeated each year. The tracking tunnels will be also used from time to time to see if lizards are present on the site. Instead of being baited with peanut butter honey and/or pear will be used. Tracking tunnels deployed elsewhere on the escarpment have proved useful in demonstrating lizard presence. Figure 12 Analysing the data We will seek professional advice with regards to analysing the data being produced by the trial. Risks including health and safety There are a number of risks within this project. One is a financial risk. It may not be able to attract sufficient funding to allow the project to continue over its expected life of 5-10 years. Another risk relates to volunteer input. There will need to be two teams of volunteers. One team will be responsible for predator control and tracking tunnel monitoring the other team for lizard monitoring. Both teams need to be able to work on steep terrain often in strong winds. It may not be able to attract sufficient volunteers to allow the project to continue over its expected life of 5-10 years. Given the steep nature of the site there is a potential that volunteers may be injured in their work. The

Health and Safety protocols used on the Whitireia lizard monitoring site have been adapted for use on the escarpment (Appendix 2) Handling of toxins will be required in this work. All Nga Uruora pest control volunteers work under GWRC Health and Safety protocols so this will help minimise risk. Toxin warning signs are already on entrances to the escarpment given the current use of toxins. If few lizards are found in initial years volunteers can lose interest and forget skills. As this project is reliant on volunteers, ways need to be found to keep them interested Wasps are a risk on the escarpment. If wasps are detected, it is proposed to deploy Vespex on this site to protect it. Nga Uruora has a licensed Vespex operator as part of its committee. There is a potential for extreme weather events or an earthquake to damage the area. There is a small risk that boulders dislodged could reach State Highway 1. As the area is above rail tunnels there is little risk to the railway. There is some small potential for vandalism in relation to the project. There is an overall risk that the project might not be successful and that lizard numbers may not increase. Publicity for the project Publicity would be given to the project through the use of various social media including newsletters. Given the popularity of Te Araroa (with potentially over 20,000 walkers per year), a sign providing information about lizards will be placed above the trial site. This will have general information about lizards and some information about the trial. It will be of the same style of other signs on the escarpment designed by Isobel Gabites. These are high quality vandal proof signs based on Izone panels. Help will be sought from lizard specialists in designing this sign.

Figure 12 There is also the potential to create short low cost publicity video. 19 Timeline It is proposed that this project will run for at least five years and hopefully ten. The first stage of the project began in December 2016/January 2017 with the laying out of the onduline, pitfalls and tracking tunnels. This does not make a commitment to continue the project. The following will take place in the first six months. 1. The 42 onduline and pitfall sites have installed by the end of December 2016. The small comparison sites of 12 ondulines was installed at the same time. 2. The 20 tracking tunnels have been installed by the end of January 2017. The first monitor will take place in mid-april 2017. 3. The first lizard monitor is scheduled to place in late /early April 2017 (three months after the onduline squares have been installed). Ecogecko will lead this first monitor. 19 As an example see https://www.youtube.com/watch?v=aebn1m9su0c

4. The main pest control network will be installed between December and April and will go live in April 2017. Options if the trial is not successful If there is a lack of long term support for the project, or very few lizards are found, this site could simply be used as a way of setting up lizard surveys for a couple of years and then project abandoned. However, if this occurred an option is to proceed with revegetating the site while leaving the onduline and pitfalls in place. Once the long grass was replaced new surveys could be carried to see the effect of habitat change. References Romijn, R., Adams, L., Hitchmough, R. 2012. Lizard strategy for the Wellington region 2012-20, Wellington Regional Lizard Network. MacKay, J.W.B. et al. 2011. A successful mouse eradication explained by site-specific population data. pp. 198 203 in Veitch, C.R.; Clout, M.N.; Towns, D.R. (Eds): Island invasives: eradication and management. IUCN (International Union for Conservation of Nature), Gland, Switzerland. Newman, D.G. 1994. Effects of a mouse, Mus musculus, eradication programme and habitat change on lizard populations of Mana Island, New Zealand, with special reference to McGregors skink, Cyclodina macgregori. New Zealand Journal of Zoology 21(4): 443-456. Norbury G, et al. 2014. Impacts of invasive house mice on post-release survival of translocated lizards. New Zealand Journal of Ecology 38: 322-327. Weir, S.M. et al. 2016. Acute toxicity and risk to lizards of rodenticides and herbicides commonly used in New Zealand. New Zealand Journal of Ecology 40(3): 342-350

Appendix 1 Data sheets (based on an Ecogecko template) Escarpment lizard monitoring data collection sheet: NB: Use one row for each animal caught (e.g. use 2 rows for 2 lizards caught in the same pitfall) Team leader s name: Site name: Date: Start time: Finish time: Time Location (trap number) Trap type 11.05am e.g Trap 1.1 e.g. pitfall or under onduline Species abbreviation Temperature ( C): Rain past 24 hrs?: Wind (Beaufort scale): Cloud cover (in eighths): Whose camera: Sex M=male F=female J=juvenile SVL (mm) (length from snout to vent) TL/BL (mm) Photo number and/or photo of site tag Notes e.g. OP M 64mm 79mm e.g. 5024-5028 Scar on right side

Trap # Done? Notes Maintenance needed (cross out when fixed) 0 1 2 3 4 5 6 7 8 9 C1 C2 C3 etc

Appendix 2 Lizard monitoring H & S plan PAEKAKARIKI-PUKERUA BAY HAZARD REGISTER FOR LIZARD SURVEY Slips, Trips & Falls Significant Hazards Eliminate, Isolate or Minimise? Hazard Controls Comments Last review Next review Uneven tracks and surfaces on the escarpment, steep site with chance of falls, and chances of falling rocks from above Carrying loads, tools etc increases the risk of taking a fall. Remoteness Significant Hazards Distance from help and poor communications in places M Stout footwear with good grip required. Recommend walking stick for those with weak ankles Induction for new participants identifies particular hazardous areas such as areas with steep drops, slippery grass and sharp rocks. Take care with dislodging rocks if working above others. Group support while working on the lizard protection site. The area to be checked preferably in pairs to ensure group support First aid to be carried for each separate group Cell phone carried for each separate group Personal locator beacon to be carried for each separate group M Minimise carrying loads to bare essentials (notebook, cell phone, camera, food, water, essential clothes & first aid) Organise additional volunteer help to minimise need to carry heavy loads Eliminate, Isolate or Minimise? 2017 2017 Hazard Controls Comments Last review M Checks to be carried out preferably in pairs First aid carried between pairs Cell phone carried Personal locator beacon carried if working alone. Bright clothing or hi viz jackets to be worn to assist with a rescue if needed. 2017 2018 2018 Next review 2018

If a person works alone they either a) register this with a contact person who will report them missing if they do not return home b) they check in and out via txt with project coordinator (Paul Callister 0220862405) Weather Significant Hazards Eliminate, Isolate or Minimise? Hazard Controls Comments Last review Cold windy conditions M Check forecast before leaving to survey Clothes appropriate for the conditions windjacket and jersey Carry water and snacks Hot sunny conditions M Check forecast before leaving to survey Sun hat and sunscreen Carry water and snacks Wet conditions M Check forecast before leaving to survey Clothes appropriate for the conditions raincoat Consider abandoning survey as conditions are not ideal for lizard survey and grass becomes slippery. In relation to all weather hazards, decision to go ahead to be made by project co-ordinator (Paul Callister) or, if unavailable, Peter McLaughlin. 2017 2017 2017 Next review 2018 2018 2018 Health Significant Hazards Salmonella poisoning from lizards Eliminate, Isolate or Minimise? Hazard Controls Comments Last review M Each group to carry hand sanitizer as well as hand washing water and soap. Either is to be used at the end of the checks Do not eat food during checks Wasp stings M & E Take care if seeing wasps flying Specific prescribed anti-histamines to be carried by individuals with a known history of allergic reaction 2017 2017 Next review 2018 2018

Traffic Significant Hazards Eliminate, Isolate or Minimise? When concentrations of wasps are detected install Vespex bait stations at appropriate time Hazard Controls Comments Last review Parking next to SH1 M Be aware when entering and exiting vehicles at the bottom of Bob's Track on SH1 and be careful pulling out onto the road 2017 Next review 2018 Reviewed by Peter Kentish, Nga Uruora Health & Safety co-ordinator