Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

Norfolk Island Boobook novaeseelandiae undulata

Key Findings

Norfolk Island Boobook nearly became extinct in the 1980s, following extensive historic clearing of habitat and nesting trees. Very few natural nesting sites remain on the island and the small population of owls is now highly inbred. Conservation efforts have focused on securing nest sites and culling nest competitors, such as introduced rosellas, however there has been no successful breeding observed since 2012. Photo: Amy Svendsen Significant trajectory change from 2005-15 to 2015-18? No, population currently stable, although outlook is poor.

Priority future actions

• Research to determine whether genetic constraints or poor nestbox placement are the primary reasons for the apparent lack of breeding • Intensive and rat control to reduce the risk of predation and competition • Provide and maintain rat proof nest boxes for roosting and breeding both within the national park and community wide where breeding are known to occur

Full assessment information Background information 2018 population trajectory assessment

1. Conservation status and 8. Expert elicitation for population trends 2. Conservation history and prospects 9. Immediate priorities from 2019 3. Past and current trends 10. Contributors 4. Key threats 11. Legislative documents 5. Past and current management 12. References 6. Support from the Australian Government 13. Citation 7. Measuring progress towards conservation

The primary purpose of this scorecard is to assess progress against achieving the year three targets outlined in the Australian Government’s Threatened Species Strategy, including estimating the change in population trajectory of 20 species. It has been prepared by experts from the National Environmental Science Program’s Threatened Species Recovery Hub, with input from a number of taxon experts, a range of stakeholders and staff from the Office of the Threatened Species Commissioner, for the information of the Australian Government and is non-statutory. It has been informed by statutory planning documents that guide recovery of the species, such as Recovery Plans and/or Conservation Advices (see Section 11). The descriptive information in this scorecard is drawn from Threatened Species Scientific Committee (2016) unless otherwise noted by additional citations.

The background information aims to provide context for estimation of progress in research and management (Section 7) and estimation of population size and trajectories (Section 8).

1 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

1. Conservation status and taxonomy

Conservation status 2018 Taxonomy: N. n. albaria of Lord Howe Island is Extinct. Nominate IUCN Critically Endangered subspecies from New Zealand is Least Concern, as is the species. EPBC Endangered

The population on Norfolk Island retains some genes from the island Endemic N. n. undulata and, because of the limited nature of the intentional hybridisation; it is argued that the taxon does not meet the criteria for Extinct.

2. Conservation history and prospects The Norfolk Island Boobook declined to a single female in the 1970s. Before she died in 1996 she had bred with one of two introduced males of the New Zealand subspecies and numbers increased until there were estimated to be as many as 50 individuals (Olsen 1996). However, despite removal of many competitors for the few hollows suitable for breeding, no reproduction has been confirmed since 2012 (M.Wilson pers.comm.) and the population is again in great peril.

3. Past and current trends The rarity of the was first appreciated in the 1960s and from the 1970s just one bird was heard calling. By the 1980s it was realised only a single female survived. Two males of the only other extant subspecies, from New Zealand, were introduced and she mated with one of them (Olsen 1996). The offspring from that mating, which are inevitably highly inbred, were thought to number 45-50 in 2017. However, there appears to have been no successful breeding since 2012, despite the active maintenance of nine nestboxes from 2015-2018 and the active culling of nest competitors. There is no information on why this is so.

Monitoring (existing programs): Population surveys undertaken by Norfolk Island National Park (NINP) staff and community volunteers in 2016 and 2017 used listening locations and call play-back. There is an unpublished report from 2016 held by NINP and one from 2017 in preparation. Nine operational nestboxes placed at 7 m in apparently suitable trees in 2016 are maintained free of pests and checked regularly for activity.

Population trends: Tables 1 and 2 summarise the overall trend and status of the Norfolk Island Boobook. The information provided in these tables is derived from Threatened Species Scientific Committee (2016), with some amendments from contributing experts based on new information.

2 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

Table 1. Summary of the available information on NI Boobook distribution and population size, and (where possible) trend estimates between 2015 and 2018 for each parameter. Published 2015 2018 Population parameters Confidence in estimates baseline Estimate Estimate

Extent of Occurrence 12 km2 35 km2 35 km2 High

Area of Occupancy 8km2 30 km2 30 km2 High Predominantly occurred in the Mt Pitt region of the NINP (460 ha) in 1996 (10 Dates of records and methods females, 7 males of which 3 were juvenile) but birds present over most of island used during community owl surveys and call play-back in 2014, 2016 and 2017, so probably the same in 2015 and 2018.

No. mature individuals 40 45-50 45-50 Low

Any other measure of relative n/a n/a n/a n/a abundance (specify)

No. of subpopulations 1 1 1 High

No. of locations 1 1 1 High

High. Derived from latest global analysis Generation time 4.7 n/a n/a by BirdLife International

Table 2. Estimated recent and current population trend for the NI Boobook.

Est. % of 1988- Confidence 2015- Confidence Sub- populati 2015 in 1988- 2018 in 2015- Details population on trend 2015 trend trend 2018 trend

Numbers were closely monitored from 1988-1999 and all offspring banded. There is high confidence that numbers increased until 1999. The population was perhaps stable between 1999 and 2007. There is low confidence about trends in numbers Whole 100 Very high Low until the surveys in 2016 and 2017 population suggesting numbers were 45-50 at that time with detailed knowledge of the location of clusters of records. A failure to reproduce during this time means it is highly likely that numbers declined if some old birds died but surveys could not detect the change.

KEY: Improving Stable Deteriorating Unknown Confidence Description High Trend documented Medium Trend considered likely based on documentation

? Low Trend suspected but evidence indirect or equivocal

3 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

4. Key threats The threats listed here are derived from Threatened Species Scientific Committee (2016), with some amendments from contributing experts based on new information. Note that this is not a list of all plausible threats, but a subset of the threats that are likely to have the largest impact on populations.

Farming and forestry Extensive clearance of vegetation for agriculture and felling of large trees destroyed most suitable nest sites on Norfolk Island, making competition for remaining sites intense. Thus, although the land clearance has stopped, the legacy continues to have a major impact on the owls’ prospects through lack of hollows (Garnett et al. 2011).

Nest competitors Crimson Rosellas (Platycercus elegans), and to a lesser extent Common Starlings (Sturnus vulgaris) and sometimes feral bees (Apis melifera), all compete for scarce hollows. All are introduced to Norfolk Island. Crimson Rosellas sometimes expel established Boobook pairs from hollows. Common Starlings fill hollows with material that render hollows unsuitable for Boobooks. Bees, rarely, can kill owls if a swarm invades an occupied nest, and thereafter prevent owls from using the hollow (Garnett et al. 2011).

Guava, olives and other feral weeds It has been hypothesised that habitat consisting primarily of a dense understorey created by red guava (Psidium spp.), African olive (Olea europaea), wild tobacco (Solanum mauritianum) and lantana (Lantana camara) is likely to support only half the density of owls when compared to remnant forest (Olsen 1997) It is estimated that the island consists of 190 ha of remnant forest, 200 ha of weed- invaded forest and 1500 ha of cultivated and urban areas (Olsen 1997).

Lack of genetic variability The population of Norfolk Island Boobook Owls is highly inbred and smaller clutch sizes suggest inbreeding depression (Garnett et al. 2011). Potentially this could be behind the apparent lack of breeding in recent years, although one would expect the owls to attempt to breed even if their eggs were infertile, and some birds that successfully bred previously must still be alive. A highly skewed sex ratio towards females, which may also be a problem (Olsen 1997) would also represent a lack of genetic variability.

Climate change, drought and bushfires Some modelling suggests a drying of the island’s climate, perhaps rendering both the eucalypt plantation and the adjacent national park rainforest vulnerable to uncontrolled fire (Garnett and Franklin 2014).

The impacts of the major threats are summarised in Table 3.

4 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

Table 3. The major threats facing the NI Boobook and their associated impact scores.

CURRENT THREAT IMPACT

Threat Timing Extent Severity 1. Farming and forestry Continuing/ongoing >90% of range Not negligible but <20%

2. Nest competitors Continuing/ongoing >90% of range 20-29%

3. Weeds Continuing/ongoing 50-90% of range Not negligible but <20%

4. Climate change, drought and bushfires Distant future >90% of range Not negligible but <20%

5. Lack of genetic variability Continuing/ongoing >90% of range 50-100% Timing: continuing/ongoing; near future: any occurrence probable within one generation (includes former threat no longer causing impact but could readily recur); distant future: any occurrence likely to be further than one generation into the future (includes former threat no longer causing impact and unlikely to recur). Extent: <1% of range; 1-50%; 50-90%; >90%. Severity: (within three generations or 10 years, whichever is longer) Causing no decline; Negligible declines (<1%); Not negligible but <20%; 20-29%; 30-49%; 50-100%; Causing/could cause order of magnitude fluctuations. 5. Past and current management Recent and current management actions thought to have contributed to the conservation of Norfolk Island Boobook Owls are summarised in Table 4. This information is a collation of material provided by experts. A regional Recovery Plan (2010) and Conservation Advice (2016) are in place, guiding recovery action (see Section 11). Table 4. Management actions thought to that have contributed to the conservation of the Norfolk Island Boobook. Est. % Contributors and Action Location Timing pop’n partners

Maintenance of 9 artificial nest boxes, although no successful 2015-2018 NINP 50 Parks breeding attempts have occurred in this timeframe (ongoing)

*Rodent Control: an initial network of 900 stations was NINP 2015-2018 50 Parks Australia / expanded to 1800 stations covering a larger area as a result (ongoing Contractor, with a of funding from OTSC. The funding also enabled the whole with a short- funding 1800 stations to be baited for a period of approximately 12 term contribution from months. Since the funding ceased, the full 1800 station increase in the Department of network is still used, but each station is only baited once effort 2015- the Environment every two months 17) and Energy

Feral bird control: 882 crimson rosellas culled over 3 years 2015-2018 NINP 50 Parks Australia (ongoing)

Habitat rehabilitation and weed control program to improve 2015-2018 NINP 50 Parks Australia habitat quality and extent on Norfolk Island (ongoing)

*The costs and benefits of rat baiting are uncertain: (i) baiting could reduce prey abundance (but some rats may now be entirely arboreal so may avoid baits, which are placed at ground level); (ii) some owls may suffer secondary poisoning by eating poisoned rodents, reducing survival and fecundity (see Stephenson et al. 1999); (iii) abundance of non-rat prey may increase if there are fewer rats; (iv) rat predation of nests, while unlikely given rats are common prey for the owls, cannot be entirely ruled out.

5 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

6. Actions undertaken or supported by the Australian Government resulting from inclusion in the Threatened Species Strategy The activities tabled above have all been supported by the Australian Government (through Parks Australia) and support the Strategy. As part of the Strategy, the Department of the Environment and Energy provided $300,000 to expand the rodent control program in Norfolk Island National Park from 2015 to 2017, which also enabled more strategic management of two threats to the boobook: weeds and crimson rosellas. While this is likely to benefit other threatened species on Norfolk Island (notably the Norfolk Island Green Parrot Cyanoramphus cookii), the impacts on the boobook are uncertain.

7. Measuring progress towards conservation Table 5. Progress towards management understanding and management implementation for each of the major threats affecting the Norfolk Island Boobook in 2015 (i.e. timing of TSS implementation) and 2018, using the progress framework developed by Garnett et al. 2018. PROGRESS IN MANAGING THREATS Threat Year Understanding of how to manage threat Extent to which threat being managed 6. Research complete and being applied OR 4. Solutions are enabling achievement but only with 2015 ongoing research associated with adaptive continued conservation intervention 1. Farming and management of threat forestry 6. Research complete and being applied OR 4. Solutions are enabling achievement but only with 2018 ongoing research associated with adaptive continued conservation intervention management of threat 6. Research complete and being applied OR 4. Solutions are enabling achievement but only with 2015 ongoing research associated with adaptive continued conservation intervention 2. Nest management of threat competitors 6. Research complete and being applied OR 4. Solutions are enabling achievement but only with 2018 ongoing research associated with adaptive continued conservation intervention management of threat 5. Trial management is providing clear evidence 2. Work has been initiated to roll out solutions where 2015 that it can deliver objectives threat applies across the taxon’s range 3. Weeds 5. Trial management is providing clear evidence 2. Work has been initiated to roll out solutions where 2018 that it can deliver objectives threat applies across the taxon’s range 4. Climate 2015 0. No knowledge and no research 0. No management change, drought and bushfires 2018 0. No knowledge and no research 0. No management 1. Research being undertaken or completed but 5. Lack of 2015 0. No management limited understanding on how to manage threat genetic 1. Research being undertaken or completed but variability 2018 0. No management limited understanding on how to manage threat

> Green shading indicates an improvement in our understanding or management of threats between years 2015 and 2018, while red shading indicates deterioration in our understanding or management of threats.

KEY: Score Understanding of how to manage threat Extent to which threat is being managed 0 No knowledge and no research No management Research being undertaken or completed but limited 1 Management limited to trials understanding on how to manage threat Work has been initiated to roll out solutions where threat applies 2 Research has provided strong direction on how to manage threat across the taxon’s range 3 Solutions being trialled but work only initiated recently Solutions have been adopted but too early to demonstrate success Trial management under way but not yet clear evidence that it Solutions are enabling achievement but only with continued 4 can deliver objectives conservation intervention Trial management is providing clear evidence that it can deliver Good evidence available that solutions are enabling achievement with 5 objectives little or no conservation intervention Research complete and being applied OR ongoing research 6 The threat no longer needs management associated with adaptive management of threat

6 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

8. Expert elicitation for population trends An expert elicitation process was undertaken to assess population trends for the period 2005-2015 and post-2015 under the following management scenarios. Please note that differences between Management Scenarios 2 and 3 (Fig. 1) are difficult to attribute, as it can be difficult to determine whether actions undertaken after 2015 were influenced by the Threatened Species Strategy or were independent of it (see Summary Report for details of methods). Management Scenario 1 (red line): no conservation management undertaken since 2015, and no new actions implemented.

• No nestboxes • No control of Crimson Rosellas, rats or weeds Under this scenario there would be very few opportunities for the owls to breed. Management Scenario 2 (blue line): continuation of existing conservation management (i.e. actions undertaken before implementation of the Threatened Species Strategy or independent of the Threatened Species Strategy).

• Nine nestboxes would be maintained • 900 stations baited per month for rats • Some ongoing control of weeds Under this scenario, nestboxes would be available for the owls to breed, although their suitability would not be known and nine boxes may be inadequate. Rat baiting could possibily have unintended negative effects. Management Scenario 3 (green line): continuation of existing management, augmented by support mobilised by the Australian Government under the Threatened Species Strategy.

• An additional 900 bait stations were in operation from 2015 to 2017 Under this scenario there may have been a reduction in pressure from rat predation during 2015- 2017, though this is thought to be very low, and there could also be some unintended negative effects (i.e. reduction in prey availability or indirect poisoning).

Overall estimated population trajectories subject to management scenarios considered The Norfolk Island Boobook is currently being managed under scenario 3 (green line).

7 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

Figure 1. Estimated relative percentage change in population under each of the management scenarios described above. Data derived from 7 expert assessments of Norfolk Island Boobook expected response to management, using four-step elicitation and the IDEA protocol (Hemming et al. 2017), where experts are asked to provide best estimates, lowest and highest plausible estimates, and an associated level of confidence. The dashed line represents the baseline value (i.e. as at 2015, standardised to 100). Values above this line indicate a relative increase in population size, while values below this line indicate a relative decrease in population size. Shading indicates confidence bounds (i.e. the lowest and highest plausible estimates).

Population size projections based on expert elicitation are extended here to 2025, 2035 and 2045 (i.e. 10, 20 and 30 years after the establishment of the Threatened Species Strategy) on the grounds that some priority conservation management actions may take many years to achieve substantial conservation outcomes. However, we note also that there will be greater uncertainty around estimates of population size into the more distant future because, for example, novel threats may affect the species, managers may develop new and more efficient conservation options, and the impacts of climate change may be challenging to predict.

8 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

Improved trajectory (Threatened Species Strategy Year 3 target): The primary purpose of this scorecard is to assess progress against achieving the year three targets outlined in the Australian Government’s Threatened Species Strategy, i.e. a demonstrated improved trajectory for at least half of the priority species (10 birds and 10 mammals). To assess this, we first use the expert-derived trend between 2005-15 (i.e. 10 years prior to implementation of the TSS) as a baseline for assessing whether there has been an improvement in trajectory in the time since implementation of the TSS (i.e. 2015-18). Table 6 below summarises this information, where negative values indicate a declining population, and positive values indicate an increasing population. We used Wilcoxon match-paired tests to compare trajectories for these two periods; a significant result (probability <0.05) indicates that there was a high concordance amongst experts that their trajectory estimates for 2005-15 were different to their estimates for 2015-18.

Table 6. A comparison of the relative annual percentage population change for the periods 2005-2015 and 2015-2018.

Post-TSS Year 3 Pre-TSS trend Significant concordance among trend target (2005-2015) elicitors? (2015-2018) met? Annual The trajectory worsened after 2015, but percentage 0.14 -0.48 without sufficient concordance among population  elicitotrs to be significant. change

The TSS has had limited impact in terms of improving the trajectory of the Norfolk Island Boobook.

Additional actions that could improve trajectory The potential impact of carrying out specific additional conservation measures on the population trajectory of the NI Boobook Owl was also evaluated through expert elicitation. Current management includes maintainign nestboxes, poison-baiting for feral rats, and some weed control.

Additional actions that could further improve the population trajectory include:

• Research would identify whether genetic constraints or poor nestbox placement are the primary reasons for the apparent lack of breeding • If appropriate, nestbox numbers would be increased and their placement optimised • If appropriate, there would be genetic enhancement of the existing population

With thes additional actions, the problems inhibiting breeding by owls would be discerned and solutions put in place to solve them.

9 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

9. Immediate priorities from 2019 The priorities listed here are derived from Threatened Species Scientific Committee (2016) with some amendments made by contributing experts based on new information. Identification of these priorities in this document is for information and is non-statutory. For statutory conservation planning documents, such as Recovery Plans or Conservation Advices, please see Section 11.

Data collection: • Undertake annual community and call-play surveys during breeding season to monitor abundance and occupancy of owls across the island • Monitor artificial nest boxes during the breeding season: monitor breeding attempts and deal with threats • Search during breeding season: locate owls, particularly potential pairs and nest sites • Ask for help from the community to locate owls and nest sites • Monitor during the non-breeding season to locate owls and monitor survival and dispersal • Study the demographic characteristics of the current population (age structure, sex) • Assess the genetic diversity of the current population: determine the genetic diversity of both the species as a whole (i.e. is the owl still the ?) and the population (genetic bottleneck) • Determine home range and location of nests • Radio-track fledglings to gain information on dispersal, survival, age of independence, habitat preferences etc • Study diet (including a measure of secondary poison risk) • Assess reproductive success and band nestlings at 3.5 to 4.5 weeks, weigh and measure them according to guidelines, collect shell fragment and food remains for analysis, assess fledging success • Determine aspects of nest site success to understand where to place artificial nest boxes so they are most preferred Management actions: • Continue intensive crimson rosella, rat control to reduce the risk of predation and competition • Provide and maintain rat proof nest boxes for roosting and breeding both within the national park and community wide where breeding birds are known to occur • A lack of breeding may be indicative of genetic problems that can only be remedied by introduction of more birds from New Zealand

10. Contributors Stephen Garnett, Margarita Goumas, Nigel Greenup, Sarah Legge, Nicholas MacGregor, Oliver Tester, Melinda Wilson, John Woinarski, Hayley Geyle, Peter Menkhorst, Guy Dutson, Richard Loyn.

10 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

11. Legislative documents SPRAT profile: http://www.environment.gov.au/cgi- bin/sprat/public/publicspecies.pl?taxon_id=26188 Threatened Species Scientific Committee (2016). Conservation Advice Ninox novaeseelandiae undulata Norfolk Island boobook owl. Canberra: Department of the Environment. Available from: http://www.environment.gov.au/biodiversity/threatened/species/pubs/26188-conservation-advice- 15072016.pdf. In effect under the EPBC Act from 15-Jul-2016. Director of National Parks (2010). Norfolk Island Region Threatened Species Recovery Plan. Department of the Environment, Water, Heritage and the Arts, Canberra. Available from: http://www.environment.gov.au/biodiversity/threatened/recovery-plans/national-recovery-plan- norfolk-island

12. References Debus, S.J.S. (2012) Norfolk Island Boobook chick deaths. Boobook 30, 6. Garnett, S.T. and Franklin, D.C. (eds.) (2014) Climate Change Adaptation Plan for Australian Birds. CSIRO, Melbourne. Garnett, S., Szabo, J., & Dutson, G. (2011) The Action Plan for Australian Birds 2010. CSIRO Publishing. Garnett S.T., Butchart S.H.M., Baker G.B., Bayraktarov E., Buchanan K.L., Burbidge A.A., Chauvenet A.L.M., Christidis L., Ehmke G., Grace M., Hoccom D.G., Legge S.M., Leiper I., Lindenmayer D.B., Loyn R.H., Maron M., McDonald P., Menkhorst P., Possingham H.P., Radford J., Reside A.E., Watson D.M., Watson J.E.M., Wintle B., Woinarski J.C.Z., and Geyle H.M. (2018) Metrics of progress in the understanding and management of threats to Australian Birds. Conservation Biology https://doi.org/10.1111/cobi.13220. Garnett, S.T., Olsen, P., Butchart, S.H. and Hoffmann, A.A. (2011) Did hybridization save the Norfolk Island boobook owl Ninox novaeseelandiae undulata? Oryx, 45(4), pp.500-504. Hemming, V., Burgman, M.A., Hanea, A.M., McBride, M.F., and Wintle B.C. (2017) A practical guide to structured expert elicitation using the IDEA protocol. Methods in Ecology and Evolution, 9, 169-180. Invasive Species Council (2017) Case Study: Argentine Ant on Norfolk Island. https://invasives.org.au/wp-content/uploads/2018/02/Case-Study-Argentine-Ants.pdf Olsen, P. 1996. Re-establishment of an endangered subspecies: the Norfolk Island Boobook. Ibird Conservation International 6, 63-70. Olsen, P. 1997. Recovery Plan for the Norfolk Island Boobook Owl Ninox novaeseelandiae undulata. Environment Australia, Canberra. Stephenson, B.M., Minot, E.O., and Armstrong, D.P. (1999) Fate of (Ninox novaeseelandiae) during a pest control operation on Mokoia Island, Lake Roturua, North Island, New Zealand. New Zealand Journal of Ecology, 23, 233-240. Threatened Species Scientific Committee (2016). Conservation Advice Ninox novaeseelandiae undulata Norfolk Island boobook owl. Canberra: Department of the Environment.

11 Information current to December 2018 Threatened Species Strategy – Year 3 Priority Species Scorecard (2018)

13. Citation Please cite this document as:

National Environmental Science Program Threatened Species Research Hub (2019) Threatened Species Strategy Year 3 Scorecard – Norfolk Island Boobook Owl. Australian Government, Canberra. Available from: http://www.environment.gov.au/biodiversity/threatened/species/20-birds-by- 2020/norfolk-island-boobook-owl

12 Information current to December 2018