ERMANZ Report: Verified Source: Pestlink Operational Report for Possum Control in the Waitutu

08 Sep 2010 - 18 Oct 2010 6/04/2011 Department of Conservation Murihiku Contents

1. Operation Summary...... 2 2. Introduction...... 3 2.1 TREATMENT AREA...... 3 2.2 MANAGEMENT HISTORY ...... 6 3 Outcomes and Targets...... 6 3.1 CONSERVATION OUTCOMES...... 6 3.2 TARGETS ...... 6 3.2.1 Result Targets ...... 6 3.2.2 Outcome Targets...... 6 4 Consultation, Consents & Notifications...... 7 4.1 CONSULTATION ...... 7 4.2 CONSENTS...... 8 4.3 NOTIFICATION ...... 8 5 Methods...... 10 5.1 TARGET SPECIES ...... 10 5.2 ENVIRONMENTAL EFFECTS ...... 13 5.2.1 Effects on Non-Target Species ...... 13 5.2.2 Effects on Soil and Water Quality...... 17 5.2.3 Effects on Ecosystems...... 18 5.2.4 Effects on Human Health ...... 19 6 Monitoring Results and Outcomes ...... 21 6.1 RESULT MONITORING - TARGET SPECIES...... 21 6.2 RESULT MONITORING - ENVIRONMENTAL EFFECTS...... 22 6.3 OUTCOME MONITORING...... 24

Page 2 of 25

1. Operation Summary

Operation Name Possum Control in Waitutu Operation Date 08 Sep 2010 - 18 Oct 2010 Area Office Murihiku Conservancy: Southland Pestlink Reference 1011MRH03

Treatment Area Size (ha) Waitutu 25450.00 Conservation Unit Name(s) Conservation Unit Number(s) National Park C430001 Waitutu Block Covenant C460015 Waitutu Blk XIII Sec 4 Covenant C460019 Treatment Block Details Treatment Blocks Size (ha) Waitutu aerial 25000.00 Waitutu ground 450.00

Beck Helicopters Ltd (aerial) & Leith Contractors Ltd. Contractor Name (ground)

Treatment Dates Start Completion Waitutu aerial 08 Sep 2010 04 Oct 2010 Waitutu ground 12 Oct 2010 18 Oct 2010 Target Pest Details Treatment Target Pests Control Method Name Blocks Waitutu aerial Possum Pesticide Aerial Pesticide - Aerial in Waitutu-(1) Waitutu ground Possum Trap Trap in Waitutu ground-(1)

Conservation Outcome(s) Protection of the health and integrity of the forest canopy and associated plant communities susceptible to possum browse, but particularly: Mistletoe species Peraxilla colensi, P. tetrapetala, Alepis flavida, and Ileostylus micranthus).

Result Target(s) Treatment Area/Block What we got • less than 3 percent Waitutu aerial 0.06%RTC

RTC

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 3 of 25

• less than 3 percent Waitutu ground 2.2%RTC RTC

Outcome Targets What we got • Possum preferred forest species including mistletoe are protected from adverse effects of

possum browse and recruitment in mistletoe is observed.

2. Introduction

2.1 TREATMENT AREA

Non-target species Common Name Scientific Name Southern falcon Falco novaeseelandiae "southern" Nestor notabilis Fernbird Bowdleria punctata punctata Morepork Ninox novaeseelandiae subsp. novaeseelandiae Petroica australis australis Yellow-breasted Tit Petroica macrocephala macrocephala

Target benefit species Common Name Scientific Name Yellow mistletoe Alepis flavida Scarlet mistletoe Peraxilla colensoi Red mistletoe, pikirangi, Peraxilla tetrapetala pirirangi, pikiraki, pirita White mistletoe, taapia, Tupeia antarctica pirata, tupia South Island Kaka Nestor meridionalis meridionalis - Metrosideros umbellata

Threatened species Common Name Scientific Name - Ourisia modesta - Plagiochila baylisii - Trithuria inconspicua Sea sedge Carex litorosa

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 4 of 25

Shore spurge Euphorbia glauca - Myosotis pygmaea var. pygmaea - Tetrachondra hamiltonii Yellow mistletoe Alepis flavida Scarlet mistletoe Peraxilla colensoi Red mistletoe, pikirangi, Peraxilla tetrapetala pirirangi, pikiraki, pirita White mistletoe, taapia, Tupeia antarctica pirata, tupia - Carex capillacea - Crassula ruamahanga - Gentianella lineata - Grammitis rigida - Hypolepis amaurorachis - Juncus pusillus - Ranunculus ternatifolius Grassland wheatgrass Stenostachys laevis South Island brown teal Anas chlorotis "South Island" Orange-fronted Parakeet Cyanoramphus malherbi Grey Duck Anas superciliosa subsp. superciliosa White Heron Egretta alba subsp. modesta Southern falcon Falco novaeseelandiae "southern" South Island Kaka Nestor meridionalis meridionalis Kea Nestor notabilis Fiordland Crested Penguin Eudyptes pachyrhynchus Blue duck, Whio Hymenolaimus malacorhynchos Yellowhead Mohoua ochrocephala Southern Crested Grebe Podiceps cristatus subsp. australis Reef Heron Egretta sacra sacra Stewart Island Shag Leucocarbo chalconotus Yellow-eyed Penguin Megadyptes antipodes Rock Wren Xenicus gilviventris Anthus novaeseelandiae Pipit novaeseelandiae South Island Fernbird Bowdleria punctata punctata South Island Rifleman Acanthisitta chloris chloris Southern Blue Penguin Eudyptula minor minor Sooty Shearwater Puffinus griseus Variable Oystercatcher Haematopus unicolor Cyanoramphus novaezelandiae Red-crowned Parakeet novaezelandiae Mottled Petrel Pterodroma inexpectata Catharacta antarctica subsp. - lonnbergi

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 5 of 25

Phalacrocorax carbo subsp. Black Shag novaehollandiae

Geographical location The Waitutu is situated 30 km East of .

TREATMENT BLOCK DETAILS: Treatment block Waitutu aerial Hardwood dominant (southern rata, kamahi and broadleaf), podocarp dominant (rimu, miro and Hall's tötara), beech Vegetation type dominant (silver and mountain) and mixed vegetation types. Coastal turf communities, bog vegetation and alpine shrublands and tussocks. Bioclimatic zone coastal lowland semi-coastal sub-montane montane sub-alpine alpine Climate characteristics: Rainfall 3000 mm Temperature: Average Summer 13.0 Average Winter 6.0 Snow level 900 m Altitude 5 - 1395 m

South Island Landless Native Act (SILNA) landowners own approximately 2,500ha of coastal land in the treatment area. The land owned by SILNA landowners, while not necessarily ancestral land, holds great value to the land holders, for Community and cultural, recreation and biodiversity purposes. The area is a Iwi interests popular with hunters and whitebaiters. There are several concessionaires in the area that support these interests. The lies to the east of, and overlooks the treatment area. Within the treatment area the Crombie Stream and Long Point have been identified as important sites for canoe landings, shellfish regeneration sites and short-term campsites (DOC, 2000). The archaeological sites recorded Historic sites within and around the area indicate a range of activities have been undertaken here by early Maori consistent with coastal travel, inland exploration, and the exploitation of local resources (Edgerton pers. Comm., 2009).

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 6 of 25

Treatment block Waitutu ground Hardwood dominant (southern rata, kamahi and broadleaf), Vegetation type podocarp dominant (rimu, miro and Hall's tötara). Coastal turf communities. Bioclimatic zone coastal semi-coastal lowland Climate characteristics: Rainfall 2000 mm Temperature: Average Summer 13.0 Average Winter 6.0 Snow level 900 m Altitude 10 to 60 m

Community and The Waitutu Incorporation, SILNA landowners, own all of Iwi interests this treatment area (450ha). Within this treatment area the Crombie Stream has been Historic sites identified as an important site for canoe landings, shellfish regeneration and short-term campsites (DOC, 2000).

2.2 MANAGEMENT HISTORY

Management history was not chosen to be shown in this operational report. This history is, however, available via Pestlink

3 Outcomes and Targets 3.1 CONSERVATION OUTCOMES Protection of the health and integrity of the forest canopy and associated plant communities susceptible to possum browse, but particularly: Mistletoe species Peraxilla colensi, P. tetrapetala, Alepis flavida, and Ileostylus micranthus).

3.2 TARGETS 3.2.1 Result Targets The result targets for the treatment area were: • less than 3 percent RTC

3.2.2 Outcome Targets

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 7 of 25

The outcome targets for the treatment area were: • Possum preferred forest species including mistletoe are protected from adverse effects of possum browse and recruitment in mistletoe is observed.

4 Consultation, Consents & Notifications 4.1 CONSULTATION Consultation began at least two years prior to the application of bait (Mar 2008 – Oct 2010). South Island Landless Natives Act (SILNA) landowners were the key stakeholders and were consulted extensively as were the Nature Heritage Fund organisation who provided the bulk of the funding for the operation. Every effort was made to run as open and informative a process as possible. Other key stakeholders that were regularly consulted were:

• DOC staff • local iwi through Kaitiaki Roopu • local landowners • local branches of the New Zealand Deer Stalkers Association • Forest and Bird • local communities • DOC concessionaires • WARO operators

Consultation was carried out through regular meetings using PowerPoint presentations, via newsletters, media releases, fact sheets and regular phone and email contact. For further information on consultation refer to:

• Waitutu Communication Plan • Waitutu RMA AEE • Waitutu MOH application • Waitutu Homepage • Southland Conservancy File, NHT-02-16-32-05

Consultation outcomes The operation ran very smoothly, over sixty people were invited to observe at the loading site on the application days, less than a dozen came along. The SILNA owners were initially opposed to 1080 application but after discussing their concerns and explaining the process and pros and cons they agreed and became our staunchest allies, even under pressure from those opposed. Some of those that attended the Resource Consent hearing and spoke in opposition of the application changed their stance or accepted the outcome and supported it. There was a vocal minority of six-seven people who were outspoken in their opposition.

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 8 of 25

Lessons learned Early consultation is essential as people need time to digest the information, consult with others and comeback with more questions. While vocal and persistent there were only six or seven people that were outspoken in opposition while lots of effort was put into consultation with these people they just could not join the dots.

4.2 CONSENTS Consent Consent File Reference Permission ID date DOC permission - Southland NHT-02-16-32-05 04/02/2010 543885 Conservancy or docdm-543885 Ministry of Health Consent - 01/04/2010 NHT-02-16-32-05 10/04/CJM/INVPH Public Health South Resource Consent - 02/02/2010 NHT-02-16-32-05 D036-084 Environment Southland Landowner consent - 08/03/2010 NHT-02-16-32-05 Waitutu Incorporation Landowner consent - Paiki 05/03/2010 NHT-02-16-32-05 Block Landowner consent - 11/06/2010 NHT-02-16-32-05 Waitutu Blk 12, Section 7 Landowner consent - 08/03/2010 NHT-02-16-32-05 Maritime New Zealand

Lessons learned Keep comprehensive records of all correspondence to do with decisions and in particular, amendments. Numerous questions were asked through the Official Information Act process after the operation had been completed about amendments to consents. In this case information was able to be supplied. 4.3 NOTIFICATION The following groups/businesses were notified as per the Resource Consent conditions:

At least seven days prior, but not more than 28 days:

• Environment Southland • Council • MOH • Waitutu Incorporation (SILNA landowners) • a public notice was placed in the local newspaper

A minimum of five working days prior to the operation:

• local concessionaires

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 9 of 25

• local WARO operators

A minimum of 24hrs prior to the operation:

• local medical centres • local veterinarian centres • Invercargill Information centre • local DOC offices • Hump Ridge Track Office • local NZDA branches • National NZDA • local Tramping Clubs • local guided hunting concessionaires

In addition the following were also notified:

• other SILNA landowners • Nature Heritage Fund • DOC Head Office • Maritime NZ • local landowners • Crayfish and päua industry reps • local and national media • local Police • those people that through consultation had requested that they be advised

All the above were also notified when the operation was completed and the caution period ended. For more information refer to:

• Waitutu Resource Consent Decision document • Waitutu Consultation plan

Warning signs notifying public of the operation and risks were placed at entrances or access points to the treatment area. All parts of the Consultation and Notification SOP were followed.

Lessons learned There are a significant number of people that must be contacted, many 24hrs before the drop. It is essential that contact details are confirmed and that prior contact is made with each before the deadline so that any questions can be answered and that best communication methods are confirmed. Enough staff need to be on hand to carryout the final notification at short notice.

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 10 of 25

5 Methods 5.1 TARGET SPECIES

Treatment Block Waitutu aerial Control method Name Target pest species Pesticide - Aerial Pesticide - Aerial in Waitutu-(1) Possum

Target Treatment Block Control Method Name Pest Species Waitutu aerial Pesticide - Aerial Pesticide - Aerial in Possum Waitutu-(1) Trade name of pesticide 0.15% 1080 Pellets RS5 Name of pesticide Sodium fluoroacetate Type of bait Cereal pellet Toxic loading 1.5 g/kg Bait quality sampling Conducted

Bait Details Pre-feed Toxic Bait type Cereal pellet Cereal pellet Lure/mask/deterrent Cinnamon Cinnamon Lure/mask/deterrent 3.00% 3.00% Dye None Green Individual Bait Weight 12.0g 12.0g

Sowing Rate Details Pre-feed Toxic Wind Wind Date Rate(kg/ha) Direction Date Rate(kg/ha) Direction Speed Speed 08/09/2010 1.00 Calm Nil 04/10/2010 2.00 Calm West

Time between pre-feed and toxic 26 End of Caution Period Date 11/03/2011 Aircraft type Iroquois

Hughes 500D or E Number of Aircraft 3

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 11 of 25

Sowing gear details Description Capacity Iroquois used 1,200kg under-slung buckets, Hughes used 0 kg 300kg under-slung buckets all bucket were modified with cowlings to reduce dust escaping during the loading process. UTS AgSys III in Iroquois and Trimble 3 Type of navigational also a DOC handheld garmin 60CSx was

guidance system used placed in each helicopter with the tracking function on. Bait was supplied in 600kg bags which were loaded by hi-ab from the back of the truck. Two 600kg bags were loaded into a Loading Method loading fadge and this was then loaded into the spreading bucket beneath the helicopters.

Complaints and Incidents A small, vocal group of people were opposed to the operation. On the day of the toxic drop one of these people became aggressive and demanding and was asked to leave the area. Some members of the group held public meetings to drum up opposition, anti-1080 DVD's were shown at these meetings. Signatures were collected in support of their campaign at the meetings and at local businesses. Several anti- 1080 placards were erected throughout the district. Numerous OIA's and Ministerials were received from members of this group.

Other Details about this method SILNA landowners, and a few local landowners were staunch in their support of the operation even under pressure from those opposed.

Deviations from planned operation -

Lessons Learned The value of keeping good records of telephone conversations, emails and other correspondence.

Treatment Block Waitutu ground Control method Name Target pest species Trap Trap in Waitutu ground-(1) Possum

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 12 of 25

Target Pest Treatment Block Control Method Name Species Waitutu ground Trap Trap in Waitutu Possum ground-(1) Trap Style Leg hold Trap type Victor #1 steel jaw

Lure Details Pre-feed Set Traps Lure type - Flour and icing sugar

Treatment Details Pre-feed Set Traps Start Date - 12/10/2010 Completion Date - 18/10/2010 Number of lure 0 0 renewals/checks Total number of traps used 872 Pattern of traps lines Grid 200m between lines X 50m Trap spacing between traps Frequency of lure renewal 2 days Frequency of trap checking 1 days Trap placement Natural Trap height 0 mm Type of trap set Single Total trap effort 4360 Trap set density 1.9378

Other Details about this method -

Deviations from planned operation -

Lessons Learned -

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 13 of 25

5.2 ENVIRONMENTAL EFFECTS

5.2.1 Effects on Non-Target Species Effect on native birds. The studies carried out on native and non-native species suggests 1080 is likely to be toxic to most native animals. There is wide variation in sensitivity between taxonomic groups. Mammals are more sensitive than birds and invertebrates on a weight for weight basis. The small size of many native species relative to the target pests means that toxic baits used for pest control are capable of causing harm to almost any animal that eats the bait. Therefore the level of exposure to the bait becomes important in determining the effects on non-target native species in the field. There are records of a range of native bird species found dead after aerial poisoning operations and many of these individuals have contained residues of 1080. However when records are discounted: from operations which did not meet current bait quality standards (e.g. using unscreened, un-dyed carrot bait with berry fruit lures) or from those animals which did not have detectable 1080 residues, the Department of Conservation, Vertebrate Pesticide Residue Database 1994-2008 contains only 31 individuals representing 8 native species across all bait types used in aerial poisoning. No conclusions about population effects can be drawn from this information but it is useful to focus further studies. Eleven species of native birds have been intensively monitored, and several other bird species monitored using less precise techniques. None of these studies have identified a population level mortality that threatens the viability of the species. All 87 kiwi, 73 kaka, 19 blue ducks and 15 kereru monitored through aerial poisoning operations using radio transmitters, survived (Broome et.al, 2009). Limited monitoring of NZ falcon, käkäriki, short tailed bats and Australasian harriers does not indicate detectable mortality due to aerial poisoning. Risks to threatened bird species present in the treatment areas (see section 3.3) are discussed below: A total of 12 radio tagged morepork (Ninox novaeseelandiae) has been exposed to this method and bait type over 4 operations and none have died from poisoning (Table 14). Call count monitoring at Waipoua did not indicate significant 1080 related mortality (Pierce & Montgomery, 1992 cited in Broome, Fairweather & Fisher, 2009). A total of 23 fernbirds (Bowdleria punctata) has been exposed to this method and bait type over 2 operations and 4 have disappeared after poisoning. A total of 40 colour banded robins (Petroica australis) has been exposed to this method and bait type over 2 operations and 10 have disappeared after poisoning. A total of 29 colour banded (Petroica macrocephala) has been exposed to this method and bait type over two operations and 1 has disappeared after poisoning. A monitoring study in Tongariro using distance sampling found no significant difference in the mortality of tomtit between the treatment and non-treatment sites (Westbrooke, Etheridge & Powlesland, 2003 cited in Broome, Fairweather & Fisher 2009). This study was extended with an additional site treated with cereal pellets at 3 kg/ha at Mt Pureora in 2003. Results from all three sites led the authors to conclude that aerial poisoning operations using cereal pellets at low sowing rates causes “…little, if any…” short term impacts on tomtit populations (Westbrooke & Powlesland (in preparation) cited in Broome,

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 14 of 25

Fairweather & Fisher 2009). Blue duck (whio) (Hymenolaimus malacorhynchos) are unlikely to eat cereal pellet baits and their aquatic invertebrate prey are unlikely to be contaminated by 1080. However studies have been done to determine their survival following aerial 1080 operations. There was no reduction in visual counts of blue duck in the Otira valley after application of 0.15% 1080 Pellets at 6 kg/ha in 1989 (Spurr & Powlesland, 1997 cited in Broome, Fairweather & Fisher 2009). Additionally, all 19 radio-tagged blue ducks in Waihaha survived for at least four weeks following aerial application of carrot bait (0.08%) at 15 kg/ha (Greene, 1998 cited in Broome, Fairweather & Fisher 2009). A total of 35 radio tagged kaka (Nestor meridionalis) have been exposed to this method and bait type over 2 operations and none have died from poisoning. Additionally, 38 radio tagged birds have been exposed to 0.08% carrot baits over 2 operations and none have died from poisoning (Greene et al. 1998; Powlesland et al. 2003). Kereru (NZ pigeon/kukupa) (Hemiphaga novaeseelandiae) have not been monitored individually when exposed to this method and bait type. However none of six birds ate non-toxic cereal pellets offered in a trial on Kapiti island (Spurr & Powlesland, 1997). Monitoring of kereru during 5 aerial 1080 operations using cereal pellets did not detect population changes using the five minute count method (Spurr & Powlesland 1997). Additionally, all 15 radio tagged birds exposed to an aerial 1080 operation using carrot bait survived (Powlesland et al. 2003). Karearea (NZ falcon) (Falco novaeseelandiae) have not been monitored individually when exposed to this method and bait type. However falcon territories have remained occupied, presumably by the resident birds, during four aerial 1080 operations using cereal pellets (Pureora 1984, Mapara 1990-92) and one using carrot bait (Waihaha 1994) (Spurr & Powlesland, 1997). The total number of falcon involved in this monitoring is about 13, although the Mapara birds (3 pair) were exposed in three consecutive years (Calder & Deuss, 1985: Bradfield, 1993 cited in Broome, Fairweather & Fisher, 2009); Greene et al. 1998). kakariki (parakeet) (Cyanoramphus spp.) nests have been monitored during two aerial cereal 1080 operations. Fifteen nests were monitored during the October 2007 Hurunui Valley operation and a further seven nests were monitored during a 1080 operation in the Dart Valley. Dead chicks in a failed nest in the Hurunui Valley operation contained 1080 residues and the female was not seen after the nest failed. All the monitored nests in the Dart Valley operation were successful, however two unmonitored kakariki were found dead with 1080 residues in their tissues. The combined estimate of mortality of nesting parakeets from these operations was 2.27% (0.1-12 % 95% CI) (Rhodes, Elliot & Kemp, 2008 cited in Broome, Fairweather & Fisher, 2009). The authors concluded that while some kakariki were killed during the 1080 operations, given the rate of nest predation observed in areas where no predator control was carried out, the net benefit from the 1080 operations was positive. No detectable impact could be determined through five minute bird count monitoring after four aerial 1080 operations using carrot or cereal pellet baits (Spurr & Powlesland et al. 1997). Additionally following an intensively monitored aerial 1080 operation in Waihaha in 1994 using carrot bait, (Greene et al. 1998) observed “…kakariki remained common within the study area...”. kahu (Australasian harrier) (Circus approximans) have not been monitored individually when exposed to this method and bait type. However no detectable impact could be determined through five minute bird count monitoring before and after an aerial 1080 operation using cereal pellets on Rangitoto Island and “the

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 15 of 25

small resident population was still seen…throughout the year following the poisoning” (Miller & Anderson, 1992). Additionally, (Pierce & Maloney, 1989 cited in Broome, Fairweather & Fisher, 2009) found no evidence of dead harriers after aerial 1080 poisoning of rabbits in the McKenzie basin. A total of 27 radio tagged kea (Nestor notabilis) have been exposed to this method and bait type over 2 operations and 7 have died from poisoning. Additionally, 2 radio tagged birds have been exposed to 0.08% carrot baits over 1 operation and none have died from poisoning (Kemp & van Klink, 2008 cited in Broome, Fairweather & Fisher 2009). There is no Lethal Dose (LD)50 data for kea but these birds have recently been found to be more susceptible to aerial poisoning operations than previously thought, under certain conditions. Risk levels to kea will remain uncertain until further monitoring is done. To this end, the Department of Conservation (in conjunction with the Animal Health Board) are undertaking monitored trials of kea through three planned aerial 1080 operations in order to better understand the relationship between specific operational parameters and kea mortality. Monitoring is proposed to test three hypotheses; 1) junk food vulnerability; 2) 1080 dosage rates; and 3) bird repellents. Preliminary findings of these field trials will be available by July 2010 (DOC, 2009). Effect on bats Short-tailed bats (Mystacina tuberculata) have not been individually monitored when exposed to this method and bait type.(Lloyd, 1994) offered non-toxic cereal pellets containing a fluorescent marker to captive bats and hand broadcast baits throughout an area known to be inhabited by bats and concluded “…short-tailed bats are unlikely to eat carrot or grain-based baits…”. However short-tailed bats are possibly vulnerable to secondary poisoning because they are known to feed on arthropods that have been recorded feeding on 1080 baits and residues in these prey can in theory be enough to kill a bat (Lloyd & McQueen, 2000 cited in Broome, Fairweather & Fisher, 2009). In a study in Rangataua Forest where 0.15% 1080 pellets were aerially broadcast (3 – 5 kg /ha) over “…almost the entire winter range…” of the study animals, a total of 269 short- tailed bats were caught at their roost following poisoning and held for 48 hours to determine mortality or signs of poisoning. All animals survived and showed no signs of 1080 poisoning (Lloyd & McQueen et al. 2000). Effect on Invertebrates Invertebrate populations have been monitored in eight aerial poisoning operations and none have shown significant population effects on any species studied, nor is there evidence to suggest poisoned invertebrates are a significant factor in secondary poisoning of other animals. Long term monitoring of native land snails indicates substantial benefits to threatened populations in sites treated with aerial poisoning due to reduced predator populations. An extensive study of forest invertebrates on 1080 baits (Sherley, et al, 1999) found that at any time only a small proportion of baits had invertebrates on them, and the few individuals per bait represented a small section of the fauna present in the litter. The number of invertebrates recorded on baits in treatment grids declined when 0.15% 1080 Pellets were laid at 18 kg/ha, but started to return to original levels (relative to control grids) within 6 days of removal of the toxic baits. This sowing rate is approximately nine times that proposed to be used in the Waitutu. The reduction in invertebrate numbers did not extend further than 20cm around any bait. Another study (Spurr & Berben, 2004 cited in Broome, Fairweather & Fisher, 2009) hand laid 0.15% 1080 cereal pellets at 5 kg/ha to simulate aerial poisoning in Tararua Forest Park in 1999 and monitored the occupancy of artificial refuges by tree weta (Hemideina

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 16 of 25

crassidens) and cave weta (Isoplectron sp.). No significant impact of bait application was found for these species nor was there any effect observed on numbers of slugs, spiders and cockroaches which also commonly used the same refuges. No impact was detected on populations of weta in Waipoua Forest and all cockroaches, centipedes, millipedes, kauri snails and all but one beetle survived in enclosures with 0.08% 1080 pellets (Pierce & Montgomery et al. 1992). Spurr (1994b cited in Broome, Fairweather & Fisher, 2009) found no impacts on populations of amphipods, ants, beetles, collembolans, millipedes, mites, slugs, snails, spiders and cave weta at Puketi Forest or Titirangi Scenic Reserve where 0.08% 1080 pellets were aerially applied at 5 kg/ha. In Mapara where 0.08% 1080 pellets were aerially applied in three consecutive years (1990-92), a comparison of invertebrate fauna showed a greater number of predatory insects in the treatment site, characteristic of a healthy forest, and more fungal eating insects in the non-treatment site, characteristic of unhealthy forest (Bradfield, 1993 cited in Broome, Fairweather & Fisher, 2009). Effect on reptiles Lizards and frogs were not monitored in any 1080 poisoning operations prior to 1994; however, none have been reported killed by 1080. Captive McCann’s skinks (Oligosoma maccanni) ate non-toxic cereal pellets (RS5 and Agtech), especially when the baits were wet, but the level of consumption (0.01-0.02g) was probably insufficient for the animals to have received a lethal dose had the baits been toxic (Freeman, 1997). Effect on fish and other aquatic fauna Significant adverse effects on fish and other aquatic fauna do not occur based on the following data: • Water contamination is rarely found and extremely low level when it has occurred, • In a study conducted by NIWA (Suren & Lambert, et al. 2004) no mortality of fish due to 1080 leaching from baits was observed, • Eels have survived experimental feeding of cereal pellets and possum tissue containing 1080. • No adverse effects on marine mammals have occurred to the applicant’s knowledge. (b) Section 5.3.1 Proposed consent conditions to avoid, remedy or mitigate adverse effects on non-target native species Consent conditions for non- target native species 4. RS5 bait will be used in the operation and bait will be checked for quality to minimise ‘fines’ available / attractive to bird species. 5. Average bait sowing rate will be no greater than 2kg/ha. 6. Baits will be dyed green. 7. Baits will be lured with cinnamon. 8. Baits will have a mean size of 12g or more and 95% of baits should weigh more than 10g. 9. RS5 baits will be used. These baits breakdown quicker in the environment than the alternative (Wanganui #7 baits), therefore provide less risk to all non target species. Section 5.4 Effects of proposed operation on non-target domestic animals Effect on livestock There are no domestic livestock within or adjacent to in the treatment area. Effect on dogs Dogs are especially vulnerable and highly likely to die if they come into contact with 1080 baits or scavenge animals killed by 1080. Carcasses can remain toxic until completely decomposed, which can take several months, depending on weather conditions. (c) Section 5.4.1 Proposed consent conditions to avoid, remedy or mitigate adverse effects on non-target domestic animals Consent conditions to protect dogs 10. Warning signs will remain in place until the caution period is over. 11. Signs at every entry point will clearly state “no dogs”. 12. No permits to take dogs into the treatment areas will be issued until the operational all-clear is given i.e. caution period has ended and warning signs have been removed by DOC staff. 13. All existing dog permit holders and hunting permit holders that would allow dogs within the treatment area will be notified by mail/email or phone call of the

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 17 of 25 proposed bait drop. 14. The SILNA owners caretaker of the Waitutu Lodge will be notified prior to the operation and informed of any risks posed to dogs. A two kilometre buffer zone will be established around the lodge and muzzles provided. Followed Monitored Performance standard(s) ? ? Performance Standard #1 Sodium fluoroacetate 1.5g/kg Yes Yes cereal pellet aerial (0.15% 1080 pellet) Performance standard for trapping with leghold traps Yes Yes Resource Consent conditions regarding non-target deaths and monitoring. refer Southland Conservancy file, NHT-02- Yes Yes 16-32-05 Waitutu dead bird collection protocol Yes Yes Waitutu bird monitoring plan Yes Yes Effectiveness of performance standards All applicable performance standards were complied with.

Bykill of non-target species Annual bird monitoring is carried out within the treatment area over two 150m x 150m grids. This results in staff covering approximately 2,000ha. During this process staff recorded all dead animals encountered.

5.2.2 Effects on Soil and Water Quality Soil & water quality Degradation of 1080 is slow in soil and sediments, taking 1-4 weeks under favourable conditions. The rate of degradation will be influenced by the presence of soil or litter micro-organisms, and temperature, soil moisture and rainfall. Sodium monofluroacetate is highly water soluble so leaching out of soil will occur. In Waitutu, with its high rainfall, 1080 baits are expected to break down very rapidly. While the tested concentration of 1080 in deionised (sterile) water remains relatively constant and independent of temperature, 1080 degradation occurred within 3-4 days in a monitored field environment (Suren & Lambert, 2004). Temperature, and the presence of aquatic plants and microbes all affect 1080 degradation in aquatic environments. Water samples have been collected from streams following numerous pest control operations using 1080. Between 1990 and 2006, 1973 water samples have been collected from streams following aerial 1080 operations throughout New Zealand. 96.3% of these samples contained no residues of 1080. Residues were found in 3.7% (72) water samples but most of these had less than 1 "g/L 1080 (Booth et.al. 2007). Contamination of waterways by 1080 has long been a focus for community concern. Adopting appropriate precautions, as detailed below, will avoid adverse effects on water bodies and ensure the safety of drinking water. The operation will comply with MOH standards and conditions. (a) Section 5.2.1 Proposed consent conditions to avoid, remedy or mitigate adverse effects of proposed operation on soil and water quality Consent conditions for soil and water quality No toxin will be laid within 20m of the Wairaurahiri and Waitutu Rivers or within 50m of the South Coast and within 20m of Lakes Poteriteri and Hauroko. No toxin will be aerially applied within 2kms of the Waitutu Lodge. No toxin will be applied within 20m of any public walking track Water samples are to be taken from

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 18 of 25

11 sites within the treatment area, within 48hrs of the conpletion of the operation by an independant contractor and tested for traces of 1080. Followed Monitored Performance standard(s) ? ? Resource Consent conditions regarding water and soil quality and monitoring. refer Southland Conservancy file, Yes Yes NHT-02-16-32-05 MOH permission conditions regarding water and soil quality and monitoring. refer southland Conservancy File, Yes Yes NHT-02-16-32-05 DOC Performance Standard #1 Sodium fluoroacetate Yes Yes 1.5g/kg cereal pellet aerial (0.15% 1080 pellet) Waitutu Task specification for track clearing Yes Yes Waitutu Task Specification for water sampling Yes Yes Waitutu Task Specification for wash down at loading site Yes Yes Waitutu water sampling contract, Southland Conservancy Yes Yes File, NHT-02-16-32-05 Effectiveness of performance standards All performance standards were followed. We added an extra water sampling site at the loading site and decided to test the soil at the loading site for traces of 1080 after a local resident raised concerns about contamination.

5.2.3 Effects on Ecosystems The proposed operation will have a significant beneficial effect on the health of the indigenous forest ecosystem and animal communities within the treatment area. This effect will become more apparent in the medium to long term. Reinvasion rates of possums is expected to be low (possibly seven to ten years, before reaching pre control levels, due to the existing natural barriers), ongoing maintenance control will be required to retain the low possum numbers achieved as a result of this operation. Studies of native forests and birds have shown the widespread benefits of using 1080 to control possums. Forest response to the effective control of possums will vary between species just as different plant species show different responses to possum browse (Atkinson et. al. 1995). Some forest types in Waitutu are dominated by canopy species such as southern rata which are highly vulnerable to possum browse. These areas are expected to continue to improve gradually over the years following possum control. Monitoring of the vegetation in the Waitutu has been conducted using Foliar Browse Lines, measured in 1990’s, 2007, and 2008. There is clear evidence to show that reduced possum numbers will benefit mistletoe species in this forest (DOC internal reports (unpublished) R. Clayton, 2007 and R Ewans 2008). A significant knockdown of rodents will likely occur in this proposed operation. This is deemed to be an important benefit, particularly as rat numbers are currently increasing due to a beech mast. While it is difficult to quantify the effect of the operation on stoats it is likely that there will be some knockdown of this species. Any reduction of this species will result in positive effects for native fauna. (f) Section 5.7.1 Proposed consent conditions to avoid, remedy or mitigate adverse effects on effects on ecosystems Consent conditions for effects on

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 19 of 25

ecosystems No additional consent conditions are proposed as no significant adverse effects are anticipated; all consent conditions outlined above will mitigate any adverse effects. Followed Monitored Performance standard(s) ? ? DOC Performance Standard #1 Sodium fluoroacetate Yes Yes 1.5g/kg cereal pellet aerial (0.15% 1080 pellet) Effectiveness of performance standards All performance standards were followed and were effective

5.2.4 Effects on Human Health Effects on human health and community well-being. The estimated lethal dose of 1080 in humans lies in the range of 0.7 to 10.0 mg/kg. Sodium monofluoroacetate is absorbed through the gastrointestinal tract or via the lungs if inhaled. It is not readily absorbed through intact skin, but it can be absorbed through cuts and abrasions. The onset of clinical signs usually ranges from 30 minutes to about 2-3 hours. Adults would have to eat relatively large quantities (8 to 12 pellets) of 1080 impregnated baits to be affected. Children, due to their smaller size would require less (2 pellets). 1080 is not a mutagen and is unlikely to be a carcinogen. It has sub- lethal effects on reproduction and is classified as a teratogen. There is no effective antidote for 1080 poisoning in humans and any treatment given is largely symptomatic and supportive (Broome et. al. 2009). Assessment of risk and exposure from this operation Although people may enter the area where toxic baits are laid, actual exposure to the toxin can only occur if they find and handle baits. Exposure of any significance (that which may cause harm) is only likely if baits or contaminated animals are eaten in substantial quantities. Staff working with poisons have potentially the highest risk of exposure (Broome et. Al. 2009). Options to manage risk and/or levels of exposure Public notification: All parties will be advised of the proposed operation through a variety of communication methods, including: direct notification through letters, phone calls and visits; public notification in newspapers, and warning signs within the operational area. This communication would aim to ensure all people living near or visiting the treatment areas, are informed about the operation, the presence of toxic baits, the risks to people and particularly dogs. Certainty about the operation: People require certainty as to the nature, scale and timing of the operation to ensure they stay outside the area during the operation. People also require reassurance it will proceed in a carefully and professionally planned manner, and that the Department undertakes the operation responsibly and does not depart from the standards discussed in this AEE. Overall, the mitigations in this AEE should lessen concerns about the effective management of the operation. Clearly defining the treatment area: The operation can be managed to confine the sowing of baits exclusively to the treatment area. This area is detailed on map1 in Appendix 1. Distinctive baits: All toxic baits are easily recognisable by their standard green colour, cylindrical shape and size. Warning signs installed at all entry points provide a description. These measures reduce the risk of ingestion. Using proven pesticide handling techniques and equipment: The risks to staff involved in the operation can be managed through appropriate hazard planning, training, supervision and

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 20 of 25

adherence to safe handling techniques, and the use of personal protective equipment (PPE) in good condition. Occupational Safety and Health (OSH) have set a “Biological Exposure Index” for 1080 in urine of 15 parts per billion (micrograms per litre) based on studies of the sub-lethal effects of 1080 on animals. This represents a conservative maximum level acceptable for workers frequently exposed to 1080. Twenty-seven workers involved in aerial 1080 pellet operations have so far been tested for exposure to 1080 and all were found to be well below this standard. (d) Section 5.5.1 Proposed consent conditions to avoid, remedy or mitigate adverse effects on human health and community well-being Consent conditions for human health The following measures will be taken to manage the risk to people: 15. The concentration of 1080 baits will be 0.15% weight/weight and shall not exceed 0.17%. 16. Baits will be applied using a helicopter with under-slung bucket over areas as shown in Map 1 Appendix 1, at an application rate up to 2kg/hectare. 17. RS5 baits will be used; these baits breakdown quicker than the alternative (Wanganui #7 baits) in the environment reducing the risk of baits being accessed by people. 18. Toxic bait will not be laid in the following exclusion zones: • Within 2kms of the Waitutu Lodge. • Within 20m of the coast. • Within 20m of the lake edges of and . • Within 20m of the Wairaurahiri and Waitutu Rivers. 19. Aircraft will have a differential global positioning navigational guidance system (DGPS) to ensure baits are distributed appropriately and to reduce the likelihood of baits entering exclusion zones or falling outside the application area. 20. Pilots will upload a digital copy of the treatment boundary and exclusion zones and these boundaries will be physically checked prior to sowing toxic bait. 21. Pilots will receive hard copy maps and briefings from the project manager before laying any toxic bait. 22. Pilots will hold appropriate chemical and agricultural ratings under the Civil Aviation Regulations 1953 to undertake the aerial sowing. 23. Printouts from the DGPS will be checked throughout the day of application for any possibilities of baits falling outside the operational area and any suspected locations will be physically checked for baits. 24. Popular lake shore areas i.e. between the Blackwater Stream and the Brown Burn, Lake Poteriteri, will be physically checked for baits within 48 hours of the operation. All major public walking tracks will be checked directly after bait is applied and any bait found will be removed from the track. 25. No permits to take dogs into the treatment areas will be issued until the caution period is over nad warning signs have been removed by DOC staff. 26. Warning signs will be placed in the following locations a minimum of 24 hours prior to the operation commencement: - Lillburn Valley Road end (Lake Hauroko carpark) - Waitutu Hut - Wairaurahiri Hut - Teal Bay Hut - Waitutu Lodge (with Waitutu Incorporation consent) - Lake Poteriteri Hut - Slaughter Burn Hut - Long Point Hut (with Waitutu Incorporation consent) 27. The following transport providers will be provided with information for clients regarding the operation a minimum of five working days prior to the operation commencement: - South Coast Jet - Wairaurahiri Jet - Hump Ridge Jet - Southwest Helicopters - Southern Lakes Helicopters 28. The following medical facilities will be notified of the operation location, dates and toxin type a minimum of 24 hours prior to the operation commencement: - Nightcaps Medical Centre - Otautau Medical Centre - Riverton Medical Centre - Tuatapere Medical Centre - Public Health South 29. The following information centres will be notified of the operation in writing a minimum of 24 hours prior to the operation commencement: - Invercargill I-Site - Fiordland

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 21 of 25

National Park Visitor Centre () - Hump Ridge Track Office - Department of Conservation Office (Invercargill) 30. The following animal health organisations will be notified of the location, dates and toxin type in writing a minimum of 24 hours prior to the operation commencement: - Otautau Vets - Veterinary Services Invercargill at Waikiwi (Riverton Branch) 31. The following recreation and interest groups will be notified by phone, email or writing of the operation a minimum of 24 hours prior to the operation commencement: - NZDA Western Southland Branch - NZDA Southern Branch - NZDA National - Southland Tramping Club - Hokonui Tramping Club - Fiordland Tramping Club 32. Local Wild Animal Recovery Operation (WARO) concession holders will be notified in writing or email of the operation a minimum of five working days prior to the operation commencement. 33. All guided hunting concessionaires will be notified in writing or email of the operation a minimum of 24 hours prior to the operation commencement. 34. The following organisations will be notified in writing when the caution period is complete: - Otautau Vets - Veterinary Services Invercargill at Waikiwi (Riverton Branch) - Invercargill I-Site - Visitor Centre (Te Anau) - Hump Ridge Track Office - Department of Conservation Office (Invercargill) - South Coast Jet - Wairaurahiri Jet - Hump Ridge Jet - Southwest Helicopters - Southern Lakes Helicopters - NZDA Western Southland Branch - NZDA Southern Branch - NZDA National - WARO concession holders - Guided hunting concession holders Safe Handling of Pesticides Procedures The Department of Conservation has standards for the safe handling of pesticides to ensure pesticides used in animal pest operations are handled, transported and disposed of as safely as possible. These standards also pertain to minimising worker exposure, and they meet the legal requirements relating to the field use of pesticides (DOC, 2009). Followed Monitored Performance standard(s) ? ? DOC Performance Standard #1 Sodium fluoroacetate Yes Yes 1.5g/kg cereal pellet aerial (0.15% 1080 pellet) DOC requirement to carryout urine testing of all those involved with the loading and application of 1080 (with their Yes Yes permission) DOC Warning Sign Standards and Procedure, Consultation Yes Yes and Notification SOP Effectiveness of performance standards All performance standards were followed and were effective.

6 Monitoring Results and Outcomes 6.1 RESULT MONITORING - TARGET SPECIES

Result target(s) less than 3 percentRTC 6.1.1 Target Species Monitoring Method: Residual trap-catch index (RTCI) Species monitored Possum - Trichosurus vulpecula in

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 22 of 25

Waitutu aerial Monitor method details Trap catch monitoring to NPCA protocol. 55 lines for aerial control, six lines for ground control

Deviations None

Target pest result details Pre During/Post Monitoring dates 23/3/09 - 27/3/09 26/11/10 - 11/12/10 Results 13.5%RTC 0.06%RTC

Result target met? Yes Lessons Learned Monitoring operation went well, no incidents were reported

6.1.2 Target Species Monitoring Method: Residual trap-catch index (RTCI) Species monitored Possum - Trichosurus vulpecula in Waitutu ground Monitor method details Trap catch monitoring to NPCA protocol. 55 lines for aerial control, six lines for ground control

Deviations None

Target pest result details Pre During/Post Monitoring dates 15/3/10 - 18/3/10 1/11/10 - 5/11/10 Results 12.8%RTC 2.2%RTC

Result target met? Yes Lessons Learned Monitoring operation went well, no incidents were reported

6.2 RESULT MONITORING - ENVIRONMENTAL EFFECTS 6.2.1 Non Target Species Monitoring of: kaka, ruru, tomtit, , rifleman, mohua Monitor Method details 15 kaka and 11 ruru were radio tracked through the operation. Five study sites (two non-treatment and three treatment sites) were selected for monitoring the smaller passerines refer to full monitoring plan details. Deviations

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 23 of 25

Kereru were also going to be monitored but this proved to problematic i.e. the time of the year the operation was carried out, the best catching time for kereru, kereru feather molt and transmitter viability. Monitoring dates Pre-control transect counts were carried out from 23/7/10 - 30/7/10, post control 14/10/10 - 21/10/1 Results All 15 radio transmitted kaka survived the drop and females successfully breed the following season. One of the 11 radio transmitted ruru died the day after the drop. Its carcass was recovered and sent to Landcare research for testing. No traces of 1080 were found. This bird was found cached under a log so appears to have been predated. Post control transect counts for small passerines showed no change. Lessons Learned Originally we had over 34 ruru radio transmitted to track through this operation however two were killed by predators, one died through entanglement, 18 appear to have died during a extremely heavy snow event in September, just prior to the operation and a few transmitters failed, leaving 11 birds to monitor during the operation.

6.2.2 Soil and Water Quality Monitoring of: Soil and Water Monitor Method details Four soil samples were taken from the loading site area approximately one week after the operation and tested for traces of 1080. Water As a condition of the Resource Consent, water samples were taken from 11 sites within the treatment area and tested for 1080. DOC added an additional sampling site next to the loading site. Deviations There was no requirement to carryout soil sampling, this was done due to a local residents concerns. One extra water sampling site was added next to the loading site. Monitoring dates 5/10/10 and 13/10/10 Results No 1080 residue was detected in any of the samples. Lessons Learned Independent sampling contractors were used for the water sample collection as required by the Resource Consent. This was a good move as those opposed to the operation could not question the results.

6.2.3 Ecosystems Monitoring of: Birds, rodents, stoats, vegetation Monitor Method details In addition to the bird monitoring mentioned in Non-target species monitoring, kaka nesting success was monitored. Rodent and stoats were monitored using

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 24 of 25

tracking tunnels. Mistletoe recruitment plots and FBI lines are being used for vegetation monitoring. Deviations Post control vegetation monitoring has not yet been carried out, due 2014 Monitoring dates Nov 2010, Jan 2011, Feb, 2011, March 2011 Results No rodents have been detected since the operation. One stoat print was identified in the last (Feb 2011) monitor, otherwise stoats are not tracking. Stoats are tracking at 50% just outside the operation boundary. In anticipation of the coming mast event all monitored female kaka are breeding, there has been no sign of predation to date. Lessons Learned Stoat numbers can be reduced significantly through 1080 application even when rodent numbers are very low prior to application. Assume secondary kill from possum carcasses.

6.2.4 Human Health Staff and contractors involved with loading or Monitoring of: application of toxic bait Monitor Method details Urine samples Deviations Not all contractors agreed to provide samples Monitoring dates 4/10/10 and 5/10/10 Results no traces of 1080 were detected Lessons Learned n/a

6.3 OUTCOME MONITORING Outcome targets Possum preferred forest species including mistletoe are protected from adverse effects of possum browse and recruitment in mistletoe is observed.

6.3.1 Outcome monitoring : Metrosideros umbellata Red mistletoe, pikirangi, pirirangi, pikiraki, pirita - Peraxilla tetrapetala Scarlet mistletoe - Peraxilla colensoi Yellow mistletoe - Alepis flavida Monitoring Method(s)

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011 Page 25 of 25

Monitoring information due date 2014 Method details refer to Waitutu result and outcome monitoring plan Monitoring dates Outcome Results No information has been entered into this field Outcome target met? Monitoring incomplete Lessons Learned No information has been entered into this field

Pestlink Ref: 1011MRH03 Date Printed: 07 Apr 2011