{"id":"F2007B00386","name":"National Recovery Plan for the Christmas Island Goshawk Accipiter fasciatus natalis","slug":"national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis","collection":"legislative_instrument","jurisdiction":"commonwealth","status":"in_force","isInForce":true,"actNumber":null,"makingDate":null,"administeringDepartment":null,"currentVersion":{"id":62196,"registerId":"commonwealth-F2007B00386-current","compilationNumber":null,"startDate":"2026-04-02","status":"InForce","reasons":null,"registeredAt":null},"sections":[{"sectionNumber":"NATIONAL RECOVERY PLAN","sectionType":"part","heading":"NATIONAL RECOVERY PLAN","content":"# NATIONAL RECOVERY PLAN\n\nFOR THE","sortOrder":0},{"sectionNumber":"CHRISTMAS ISLAND GOSHAWK","sectionType":"part","heading":"CHRISTMAS ISLAND GOSHAWK","content":"# CHRISTMAS ISLAND GOSHAWK\n\nAccipiter fasciatus natalis\n\nRichard Hill\n\nBirds Australia\n\n  \nPrepared by Richard Hill, Birds Australia, for the Australian Government Department of the Environment and Heritage.\n\nPublished by the Commonwealth of Australia.\n\nMade under the Environment Protection and Biodiversity Conservation Act 1999: August 2004\n\nISBN 0 642 55009 3\n\n© Commonwealth of Australia\n\nThis publication is copyright. Apart from any use permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Commonwealth. Requests and inquiries regarding reproduction should be addressed to:\n\nAssistant Secretary\n\nNatural Resource Management Policy Branch\n\nDepartment of the Environment and Heritage\n\nGPO Box 787\n\nCANBERRA ACT 2601\n\nThis plan should be cited as follows:\n\nHill, R. 2004. National Recovery Plan for the Christmas Island Goshawk Accipiter fasciatus natalis. Commonwealth of Australia, Canberra.\n\nDisclaimer:\n\nThis recovery plan sets out the actions necessary to stop the decline of, and support the recovery of, the listed threatened species or ecological community. The Australian Government is committed to acting in accordance with the plan and to implementing the plan as it applies to Commonwealth areas.\n\nThe plan has been developed with the involvement and cooperation of a broad range of stakeholders, but the making of this plan does not necessarily indicate the commitment of individual stakeholders to undertaking any specific actions. The attainment of objectives and the provision of funds may be subject to budgetary and other constraints affecting the parties involved. Proposed actions may be subject to modification over the life of the plan due to changes in knowledge.\n\nCopies available from: [http://www.deh.gov.au/biodiversity/threatened/recovery/list-common.html](http://www.deh.gov.au/biodiversity/threatened/recovery/list-common.html)","sortOrder":1},{"sectionNumber":"Executive Summary","sectionType":"part","heading":"Executive Summary","content":"## Executive Summary\n\nThe Christmas Island Goshawk is currently listed as Endangered under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). It is currently considered a subspecies of Accipiter fasciatus, but its taxonomy needs resolution. It is considered to be the rarest endemic bird on Christmas Island, where it occurs in all habitats from primary and marginal rainforests to suitable areas of secondary regrowth vegetation. The total population size is thought to be very small, perhaps as few as 100 adults, and is probably limited by the availability of suitable rainforest habitat. Crazy Ants pose an unknown but potentially critical threat to the survival of this bird. The recovery plan objective is to downgrade the Christmas Island Goshawk from Endangered to Conservation Dependent, primarily through successful implementation of the Invasive Ants on Christmas Island Action Plan and protection of habitat critical to the survival of the species from clearance. An assessment of goshawk population dynamics is the most essential requirement of this recovery plan, and community awareness and participation in the conservation of this endemic raptor are also important actions.","sortOrder":2},{"sectionNumber":"Background Information","sectionType":"part","heading":"Background Information","content":"## Executive Summary\n\nThe Christmas Island Goshawk is currently listed as Endangered under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). It is currently considered a subspecies of Accipiter fasciatus, but its taxonomy needs resolution. It is considered to be the rarest endemic bird on Christmas Island, where it occurs in all habitats from primary and marginal rainforests to suitable areas of secondary regrowth vegetation. The total population size is thought to be very small, perhaps as few as 100 adults, and is probably limited by the availability of suitable rainforest habitat. Crazy Ants pose an unknown but potentially critical threat to the survival of this bird. The recovery plan objective is to downgrade the Christmas Island Goshawk from Endangered to Conservation Dependent, primarily through successful implementation of the Invasive Ants on Christmas Island Action Plan and protection of habitat critical to the survival of the species from clearance. An assessment of goshawk population dynamics is the most essential requirement of this recovery plan, and community awareness and participation in the conservation of this endemic raptor are also important actions.\n\n##   \nBackground Information\n\n### Conservation Status\n\nThe Christmas Island Goshawk (Accipiter fasciatus natalis) is currently listed as Endangered (EPBC Act), however, according to Garnett & Crowley (2000) it now fulfils the criteria for Critically Endangered.\n\n### Taxonomic Status\n\nSince the Christmas Island Goshawk was described by Lister in 1888 as a full species Accipiter natalis, debate has continued over whether it is more closely allied to the Brown Goshawk Accipiter fasciatus, or to the Grey Goshawk A. novaehollandiae. Lister (1888) thought the Christmas Island Goshawk was allied to the Grey Goshawk from the Maluku A. n. griseogularis. Chasen (1933) disagreed, thinking it a member of the A. fasciatus group. The Christmas Island Goshawk may have been isolated from populations of either of these species for a long period of time, and Carter (1994) raised the possibility of resurrecting the Christmas Island subspecies natalis to a full species.\n\n### International Obligations\n\nThe Christmas Island Goshawk is not listed under international agreements.\n\n### Distribution and location\n\nThe Christmas Island Goshawk is restricted to Christmas Island, a volcanic island in the Indian Ocean (10o0'S; 105o40' E), approximately 1400km northwest of Australia. The island sits on the northernmost edge of the Australasian continental plate, and immediately north the ocean floor drops into the Java Trench and depths of up to 6000m. The nearest land is Java in the Republic of Indonesia, 360km to the north. Christmas Island is truly oceanic and all its native biota has colonised by sea or by air (Gray 1981). The island is 135km2 and 75% is covered with original vegetation (Environment Australia 2002).\n\nGibson-Hill (1947) reported that Christmas Island Goshawks seemed to prefer areas of ‘slightly thinner growth on the edge of thick jungle or the borders of clearings’. During 1994 and 1995 goshawks were observed in all major habitats on the island (Hill unpubl. data,). Goshawks were regularly seen hunting in regrowth vegetation along roadsides. This may be merely because they were much more visible in this habitat compared to intact forests. Parks Australia North (PAN) data from a systematic survey over all habitat of the island (over 1000 survey points) suggest that more goshawks are located within rainforest (95% of sightings) than in cleared areas. The Christmas Island Goshawk is the least frequently encountered diurnal forest bird and might best be described as ‘widespread and uncommon’.\n\nThere are no data on adult or juvenile movements. Adult Christmas Island Goshawks are probably resident on their territories year-round. Grey Goshawks defend a territory around the nest site which appears to be significantly smaller than the home range in which they feed (Burton 1991).\n\nThere are no detailed population estimates for Christmas Island Goshawks. New Caledonian Brown Goshawks, a similar-sized closely related bird, occurred at a density of 0.89 - 0.96 adults/km2 in rainforest (Thiollay 1993). There is approximately 101 km2 of primary forest on Christmas Island which suggests that it is unlikely that there are many more than 50 pairs or 100 mature Christmas Island Goshawks on Christmas Island.\n\n###   \n\n### ![](image.002.png)\n\n###   \nBiology\n\nThe Christmas Island Goshawk has been recorded eating a wide range of vertebrates and invertebrates including birds, small mammals and insects. Prey includes Christmas Island White-eye Zosterops natalis, Island Thrush Turdus poliocephalus, Emerald Dove Chalcophaps indica, Java Sparrow Lonchura oryzivora, domestic chickens Gallus gallus, (Lister 1888, Gibson-Hill 1947, Hill, unpubl. data); as well as nestling Red-footed Boobies Sula sula (Hill, unpubl. data) and an adult Golden Bosun Phaethon rubricauda westralis (J. Tranter pers. comm.). It has also been recorded catching rats Rattus rattus (Gibson-Hill 1947) and ‘lizards’ (Lister 1888). It is frequently seen catching grasshoppers (probably Valanga sp.) which are common in secondary vegetation around the island (Gibson-Hill 1947, Hill unpubl. data). Gibson-Hill (1947) also recorded it feeding on the large centipede Scolopendra morsitans, which is widespread on the island, and on beetles (Coleoptera) and mantids (Mantidae).\n\nThe Christmas Island Goshawk captures its prey off the ground or in the air by seizing it in flight with its long, strong legs. Like other goshawks, it will hunt from a perch where it may sit for some time waiting for suitable prey. Christmas Island Goshawks will chase small flying birds through the forest as well as pounce on birds foraging on the ground (Hill unpubl. data). Hill observed goshawks ‘short-stay perch-hunting’ (after Newton 1986) flying along a track, perching for a few minutes then flying 50-100m further and landing again, as well as pursuing insects on the ground and snatching insects off foliage. There is likely to be some sex differences in hunting and in diet, as is found in both Brown and Grey Goshawks (Marchant & Higgins 1993). Given the large sexual dimorphism (similar to Grey Goshawk, Hill unpubl. data) and the opportunity for niche expansion on the island in the absence of other diurnal raptors, this difference may be quite pronounced.\n\nGibson-Hill (1947) described three Christmas Island Goshawk nests all of which were in near-horizontal forks of forest trees, one 10m high and two 15-20m in tall forest trees. Hill (unpubl. data) observed four nests: two in Syzygium nervosum in Primary rainforest; and two in Terminalia catappa in Marginal rainforest. All nests were 25-35m above the ground, in the fork of two or more branches, three towards the edge of the crown, and one in the middle of the tree. One nest (probably not used) was 30-40cm in diameter, two others were 50-80cm in diameter, and the fourth was under construction. Nest size and location within the tree share more similarities with Grey than Brown Goshawks (J. Young pers. comm.; see Burton et al. 1994).\n\n### Habitat\n\nApproximately 75% of Christmas Island is still covered with natural vegetation and 84% of this (or 63% of the island) is protected within National Park (Figure 1). The remaining Primary rainforest is currently protected from clearance by a Federal Government moratorium, in conjunction with the requirements for approval under the EPBC Act and Environment Protection and Biodiversity Conservation Regulations 2000 (EPBC Regulations). Marginal rainforest and Second-growth rainforest or Secondary Vegetation outside the National Park may only be cleared if approval is given under the EPBC Act and Regulations.\n\nSchedule 12 of the EPBC Regulations use Du Puy’s classification to describe the Christmas Island forests. Primary Rainforest is restricted to the central plateau and is evergreen closed forest 30-40m high with emergent trees up to 45m. Common canopy trees include Syzygium nervosum, Planchonella nitida, and Hernandia ovigera. The canopy formed by these trees is irregular, especially on the western side of the island where it is sheltered from strong southeast trade winds which blow throughout the dry season. The understorey comprises dense thickets of Pandanus elatus up to 4m tall, and more open areas which commonly include the plants Arenga listeri, Leea angulata, Ochrosia ackeringae, Pisonia umbellifera and Aidia aff. racemosa (Environment Australia 1994).\n\nMarginal Rainforests grow on generally shallower soils on the coastal terraces and scree slopes surrounding the island. Many Marginal Rainforest trees lose their leaves in the dry season. Marginal rainforest is generally lower than Primary Rainforest, often less than 30m, and may include many vines. The canopy height and shape are variable, determined primarily by the degree of exposure to the southeast trade winds. Marginal rainforest facing south and east has a smooth, wind-pruned canopy and tends to increase in height with increasing distance from the sea cliff. The inland cliffs and scree slopes may have no vegetation or carry a closed forest, depending on the degree of the slope. Trees such as Ficus microcarpa and Dendrocnide sinuata are common there.\n\nOld stockpiles and cleared areas that have not been mined may support low second-growth forest of colonising trees such as Macaranga tanarius and Claoxylon indicum and an introduced tree Leucaena leucocephala generally less than 10m high. Previously mined areas tend to have very little remaining soil and on them grow dense herblands of a fern Nephrolepis multiflora to 2m high along with introduced scramblers and occasional low trees.\n\n### Habitat Critical to Survival\n\nIdentification of habitat critical to survival of the Christmas Island Goshawk is difficult due to the lack of information on specific habitat requirements. Although Christmas Island Goshawks may be generalists which forage in most habitats, it is highly likely that they require rainforest to breed as these habitat contain suitable trees. Applying the precautionary principle and using the criteria provided by the EPBC Act, habitat critical to survival of the Christmas Island Goshawks is defined as all Primary Rainforest, Marginal Rainforest (Du Puy 1993) and possibly second-growth forest suitable for nesting. Primary Rainforest and Marginal Rainforest has been identified and mapped (Figure 1). Dexter (2000) recommended mapping the ecological quality of second-growth forests outside the national park. As this has not occurred, and there is insufficient information on the habitat requirements of the Christmas Island Goshawks, suitable second-growth forest cannot be identified at this time.\n\n### Affected Interests\n\nCommonwealth Parks Australia North (PAN), Shire of Christmas Island, Christmas Island Phosphates, Union of Christmas Island Workers, the Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services (DOTARS), and the Department of Immigration, Multicultural and Indigenous Affairs.\n\n### Role and interests of indigenous people\n\nNot applicable. Christmas Island does not have an indigenous population.\n\n### Social and economic impacts\n\nThe actions in this plan may have positive and negative social and economic impacts. Positive social impacts will arise from community education actions that will increase Christmas Islanders knowledge and interest in their own environment. The rainforest rehabilitation program provides on-island jobs, as will the goshawk monitoring program. Christmas Island endemic birds attract specialist bird watching groups each year which is high value, low-impact tourism.\n\nNegative social and economic impacts arising from implementation of the plan could include greater restrictions due to review of the quarantine barrier, and potential economic effects of preventing illegal destruction of Christmas Island goshawks by residents. The EPBC Act already provides a regulatory framework for the protection of rainforest on Christmas Island, and one element of this is assessing potential impacts of proposed developments on listed threatened species. These provisions have the potential to impact on economic activity, for example by adding additional obligations for industry and other development on the island in order to minimise impacts on listed species. This arises from the listing of the species under the EPBC Act invoking a range of protective provisions and offences where a population is to be affected. The magnitude of this potential impact is unknown, as it will vary with the location, size and extent of the activity, proposed or current. However, the plan will aid in determining what could be significant impacts on the Christmas Island Goshawk (through defining habitat critical to survival and developing a management plan outside the national park).\n\n### Biodiversity Benefits:\n\nProtection of the habitat of the Christmas Island goshawk provides protection for numerous other listed species (Table 1). Community education targeted at this species will promote awareness of all the endemic forest birds and their conservation needs.\n\nTable 1: Native species listed under the EPBC Act that will benefit from recovery actions listed in this plan (after Environment Australia 2002).\n\n| Taxon                                                     | National Status |\n| --------------------------------------------------------- | --------------- |\n| Tectaria devexa var. minor                                | E               |\n| Carmona retusa                                            | V               |\n| Christmas Island Pipistrelle (e) Pipistrellus murrayi     | E               |\n| Christmas Island Shrew 1 (e) Crocidura attenuata trichura | E               |\n| Christmas Island Blind Snake (e) Ramphotyphlops exocoeti  | V               |\n| Christmas Island Gecko Lepidodactylus listeri             | V               |\n| Christmas Island Hawk-owl (e) Ninox natalis               | VJ              |\n| Abbott’s Booby (e) Papasula abbotti                       | E M S J         |\n| Red-footed Booby Sula sula rubripes                       | M S C J         |\n| Christmas Island Frigatebird (e) Fregata andrewsi         | V M S C J       |\n| Great Frigatebird Fregata minor minor                     | M S C J         |\n\nNotes:\n\n(e) species and subspecies endemic to the island E Listed under the EPBC Act as Endangered\n\nV Listed under the EPBC Act as Vulnerable M Listed under the EPBC Act as a Migratory species\n\nMF Within a family listed under the EPBC Act as Migratory S Listed Marine species under the EPBC Act\n\nC Listed under China-Australia Migratory Bird Agreement\n\nJ Listed under Japan-Australia Migratory Bird Agreement\n\n### Relationship to other plans\n\nThe Christmas Island National Park Plan of Management is the strategic nature conservation document for the island. This recovery plan makes numerous recommendations in common with other recovery plans for Christmas Island threatened taxa. Opportunities for sharing resources and points shared in common with other recovery plans are identified in the implementation section of this plan.\n\n### Recovery Team\n\nA Recovery Team for the Christmas Island Goshawk has not yet been formed. The team should comprise representatives of PAN, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate.\n\n##   \nThreats to the species\n\n### Birds on Islands\n\nIsland birds are particularly vulnerable to extinction and a very high number of island birds relative to continental species have become extinct in the past century (Diamond 1985, King 1985). This is because (i) island birds tend to have smaller population sizes than continental species and small populations are more vulnerable to extinction (Lacy 1987); (ii) island birds have often evolved in the absence of many predators, diseases and competitors and they can be catastrophically affected by the introduction of one of these (King 1985); (iii) there are often no opportunities for dispersal and thus recolonisation on islands, and this factor also predisposes a population to extinction (Lacy 1987); and (iv) there is often less opportunity to make reserves on islands because land area is limited (Diamond 1985). Predators in general are often naturally rare because their prey, other animals, are less abundant than other food resources. Thus island raptors are a particularly vulnerable group of island birds.\n\n### Crazy Ants\n\nThe exotic invasive yellow crazy ant (Anoplolepis gracilipes) arrived on Christmas Island more than 70 years ago, and is now widespread throughout rainforest (Orchard et al. 2002). The ants can form multi-queened 'super-colonies', in which they occur at very high densities. This has apparently been a relatively recent phenomenon; with the first supercolony discovered in 1989, with further dramatic increases probably beginning around the mid-1990s.\n\nAt supercolony densities, the ant is having a devastating impact on the island's ecosystems. Red crabs, robber crabs, blue crabs and most other ground dwelling animals such as reptiles, have been and are continuing to be severely impacted, sometimes to the point of local extinction in heavily infested areas. By eliminating local populations of red crabs, the ants are also having a marked effect on forest composition and structure, and litter dynamics in infested areas. The feeding activities of the ants and their mutualistic scale insects can stress large trees to the point of death, and appear to be causing canopy dieback in areas of supercolony formation (Orchard et al. 2002). In addition groundings of birds in crazy ant supercolonies would most probably result in the death of the bird. Although research has shown that crazy ants have little impact on canopy insects, it is possible that the ants may reduce insect populations. In the recent Risk Watch List compiled for the Christmas Island National Park and Conservancy, the crazy ant invasion was rated as an Extreme Risk to biodiversity and conservation values, with catastrophic consequences of failure to implement effective control measures.\n\nPAN field workers and assistants conducted an island-wide ant survey during May-August, 2001. The survey was designed by CAMBI (Centre for the Analysis and Management of Biological Invasions, Monash University). Of 972 sites surveyed; 741 surveyed points fell in natural forest. Crazy ants were recorded at 46.7% of sites in natural forest. Super-colony densities were recorded at 22.7% of sites in natural forest. Using these survey points as a representative sample of the forest, it was estimated that 2,379 ha of the estimated 10,492 ha of forest on the island was infested with crazy ants at supercolony densities (Orchard et al. 2002). Crazy ants are not evenly distributed throughout natural forest but are more commonly found on terrace forests and less commonly found in deep-soil tall-closed forest, which make up most of the natural forests on the island plateau. Deep-soil tall-closed forest made up 31% of census sites in natural forest but only 6.5% of supercolony records (Orchard et al. 2002).\n\nIn September 2002 an aerial baiting program was undertaken, and all known supercolonies treated with insecticide. Results so far indicate the program was successful in controlling supercolonies over 2500ha of Christmas Island. Crazy ants are still present in low densities, however, further high densities of crazy ants may establish in the terrace forests without warning. PAN staff will continue to monitor any new supercolony formation and treat by hand baiting over the next few years.\n\n### Disease\n\nA serious threat to all island birds is the introduction of new diseases. Island birds have often evolved in the absence of diseases common in continental bird faunas and the introduction of such diseases to island birds can be disastrous. An example of this is the introduction of avian malaria to Hawaii, which caused the extinction of almost the entire endemic bird fauna from below 600 m altitude, and was probably the main cause of the total extinction of several bird species (Hay 1986). The range of many surviving species was severely reduced and fragmented which in turn markedly increased their chances of extinction. Avian malaria arrived with the accidental introduction of a new species of mosquito (Hay 1986). Christmas Island has been very vulnerable to the accidental introduction of new disease. In 1994 a quarantine barrier was established between the island and Indonesia and Australia and this has reduced that risk.\n\n### Habitat Loss\n\nApproximately 25% of the island’s original forests have been cleared and replaced by shrublands of ferns on minefields, Secondary Vegetation on stockpiles, and roads and housing (Environment Australia 1994). This has meant a loss of available forested habitat, although secondary vegetation growing along roadsides and on stockpiles contributes to the feeding habitat for Christmas Island Goshawks at least at some times of the year. Nonetheless, based on the decrease in available habitat, the total Christmas Island Goshawk population has probably decreased by at least 25% since settlement (Stokes 1988).\n\nPrimary forests on Christmas Island have been fragmented in places by clearing and most Primary Rainforest is dissected by roads and old mining 'grid lines' The grid line system, a series of parallel lines 120 m apart and a bulldozer blade in width was installed in the 1970's to explore the island's phosphate resources. After 25 years these are mostly overgrown with dense stands of Secondary Vegetation.\n\nThere are some major development proposals and associated infrastructure in particular the Immigration Centre, Satellite launching station, airport upgrade, and expansion of phosphate mining that have the potential to reduce the area of goshawk habitat. One development condition for the satellite launching facility is to develop and implement an environment management plan, which is currently underway. The EMP will address threatened fauna and their habitat on the sites. The immigration centre is exempt from assessment under the EPBC Act, however, the DEH is being consulted on environmental aspects of the development, including monitoring of habitat for potential impacts. The airport upgrade and phosphate mining expansion proposals are undergoing assessment for environmental impact under the EPBC Act. Survey and monitoring of the Christmas Island Goshawk will aid in determining potential impacts of these and future developments on goshawk habitat.\n\nThe Christmas Island Rainforest Rehabilitation Program (CIRRP) is a long term program to revegetate old minefields. The CIRRP is funded by a conservation levy that is collected from Phosphate Resources Limited at an indexed rate per tonne. Rehabilitation priorities are primarily determined to protect Abbott’s Booby sites, although other threatened species are also considered in determining the priority of sites. A significant additional result of CIRRP will be the reestablishment of habitat that supports other species such as the goshawk.\n\n###   \nNatural Catastrophes\n\nAlthough goshawks, as generalist raptors, are likely to cope well with habitat modifying natural catastrophes, the effect of events such as cyclones are more severe on small populations. There is a possibility of an increase in severe storms and cyclones occurring as a result of Greenhouse gas-induced climate change. It would be useful to understand the effects of cyclones and severe storms on goshawk populations to help predict the impact of any increases.\n\n### Small population size\n\nThe risk of extinction varies inversely with population size (Lacy 1987) due primarily to demographic and environmental variability the effects of which can become marked in small populations. Demographic variability is the individual variation in reproductive success, which is masked when population size is large. Very small populations (< 30 individuals) can easily become extinct as a result of random demographic variation between individuals (Caughley & Sinclair 1994).\n\nThe reproductive success of a population is also likely to vary with fluctuations in the environment. The degree of this fluctuation is determined by how much seasonal and annual variation in habitat quality there is. The influence of this on the risk of extinction increases markedly with decreasing population size. For example, a cyclone, an extreme example of environmental variation, causing the death of half the population of 10000 animals has an insignificant affect on the probability of extinction of that population. However, if half a population of 200 animals dies, the risk of extinction of that population increases significantly (Caughley & Sinclair 1994).\n\n### Inbreeding Depression\n\nThe risks of inbreeding depression increases with decreasing population size due to the increase in frequency of homozygous alleles. Populations at risk of inbreeding depression are probably a couple of dozen individuals or smaller in size which have been at that size for several generations, as a population contraction must last for several generations to lead to a significant loss of heterozygosity (Caughley & Sinclair 1994). Any action on the island which might permanently, rather than temporarily, reduce the number of mature individual Christmas Island Goshawks (such as permanent forest clearance rather than temporary storm damage) would raise the possibility of inbreeding depression as a concern in their conservation.\n\n### Other Potential Threats\n\nKilling of Christmas Island Goshawks, especially by poultry owners, has been considered to be a significant threat to goshawks in the past (Stokes 1988). Given the small population size, any killing of either juveniles or adult Christmas Island Goshawks would be considered a significant threat to this taxon. Another possible threat is posed by cats, especially near settled areas.\n\nRoad killed Christmas Island Goshawks have been reported. A substantial increase in vehicular traffic will be associated with the proposed satellite launching station and the new immigration, reception and processing centre (IRPC). This will likely increase the number of roadkills in high traffic areas. Thus it is possible that collisions with cars will become an issue for the conservation of populations in high traffic areas.\n\nWeeds, especially newly introduced invasive species, could potentially impact on Christmas Island Goshawk nest sites, for example by forming vine towers over nesting trees.\n\n##   \nRecovery Objectives and Timelines\n\n### Overall objective:\n\nTo shift the status of the Christmas Island Goshawk from Endangered to Conservation Dependent within 10 years.\n\n### Specific objectives:\n\nTo reduce the risk of extinction and improve the conservation status of the taxon by:\n\n1.  Determine taxonomic status\n2.  Determine and maximise total population size and area of occupancy\n3.  Implement threat abatement strategies\n4.  Increase community involvement and awareness\n5.  Implement the Recovery Plan through a Recovery Team\n\n### Performance Criteria\n\n-   Taxonomic position of Christmas Island Goshawk clarified.\n\n-   Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring.\n-   Density of Christmas Island Goshawks increased as demonstrated by population monitoring\n\n-   Crazy Ants having an insignificant impact on Christmas Island Goshawks.\n-   Protection of all habitat critical outside the national park on Christmas Island.\n-   A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n-   Maintenance of effective quarantine against the introduction of all avian diseases\n-   Demonstrated increase in community awareness and support for habitat protection\n-   Demonstrated successful operation of the Recovery Team over five years.\n\n## Actions\n\nSpecific objective 1. Determine taxonomic status\n\nPerformance Criteria Taxonomic position of Christmas Island Goshawk clarified.\n\nAction 1\\. Investigate the taxonomic status of the Christmas Island Goshawk.\n\nThis project is designed to describe the relationships of the Christmas Island Goshawk to the Brown and Grey Goshawks by comparing DNA sequences of selected mitochondrial genes. Mitochondrial DNA may be collected from feathers allowing specimens already in collections to be used for analysis. The Museum of Victoria has material from Christmas Island Goshawks.\n\n  \n\nSpecific objective 2\\. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring\n\nDensity of Christmas Island Goshawks increased as demonstrated by population monitoring\n\nAction 2\\. Conduct detailed population survey\n\nThe recovery team will obtain estimates of population size, distribution, age structure, fledging success, and survival rates. This is required to assess the status of this taxon. Data on age structure and fledging success will be used to monitor the population, and will, over time, indicate whether or not any significant change in population size is occurring.\n\nTrap and colour band the Christmas Island Goshawk population across the island with the aim of individually marking the majority of the population. A majority of the population will remain individually marked by periodic supplementary trapping after this initial program. This action will provide data on the age-structure of the population, distribution on Christmas Island (including some information on habitat), ranging behaviour and spacing of individuals from resightings during the 18 month study, and thus some data on population size.\n\nThe members of the Australasian Raptor Association (ARA), part of Birds Australia, include many of the Australian professional and amateur experts on birds of prey. They may be able to provide unpaid assistance in trapping and marking birds, and in training of local staff in trapping techniques. The cost of their involvement would be airfares from Australia and accommodation and transport on Christmas Island. With the involvement of the ARA this project lends itself to being done by a post-graduate student. If ARA involvement was not possible dedicated researchers would re required, and the cost would be substantially higher.\n\nAction 3\\. On-going monitoring of the population\n\nIt is essential to formally and regularly monitor the size of the population, and breeding success. The recovery team will monitor the total population size at regular intervals (initially annually) such that, if a decline takes place, it provides sufficient warning to managers to allow remedial management and research to take place. The monitoring program shall be based on observations of the individually marked population of Christmas Island Goshawks. PAN Christmas Island staff, tour operators, and the community in general will be encouraged to record all observations of goshawks, and to report their locations accurately. This will provide an ongoing data set on the proportion of resightings of marked birds, and on the habitats in which birds are seen. Once every five years, volunteers from the ARA will visit the island and trap and mark additional birds to maintain a majority of marked birds in the population.\n\nIn addition a sample of goshawk nests should be monitored annually. Monitoring of ten nests over a number of years will provide a considerable amount of information on breeding success, diet, and habitat requirements, and would be relatively inexpensive. Breeding success data will allow evaluation of the species’ status and population viability. Knowledge of diet is essential to evaluate the effects of habitat change, and predict the likely impact of any proposed changes in land management on the island. Information on diet and breeding requirements, combined with data from the population survey on distribution, ranging and spacing, will enable refinement of the definition of habitat critical to the survival of Christmas Island Goshawks, and determination of at what stage/s they utilise rehabilitated areas.\n\nTen nests will be monitored for signs of breeding, and once breeding has commenced monitored regularly (fortnightly) while the breeding attempt continues. In addition roost sites of parents should be located and pellets collected for dietary analysis. The pellets should be relatively straightforward to analyse due to the limited number of potential vertebrate prey. Insects should be identified as accurately as possible. Monitoring of nesting and collection and analysis of pellets could be part of the general duties of the resource management staff of PANCI.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Protection of all habitat critical outside the national park.\n\nAction 4\\. Develop and implement wildlife management plan outside the national park\n\nAction 5\\. Ensure protection of habitat critical outside the national park.\n\nMost forested land outside the National Park is publicly owned, by the Commonwealth. PAN staff are routinely involved with the other stakeholders in negotiations over development proposals. This is ongoing. These proposals can occur unexpectedly and as there is no overall wildlife management plan outside the park, are assessed on an ad hoc basic. While consultation with other agencies and proponents on projects will be essential and ongoing (under the EPBC Act referrals process), achievement of long-term protection of threatened species outside the park requires a more comprehensive and considered approach through the development of a wildlife management program. The EPBC Regulations include that actions under an approved management plan are not considered offences in relation to protected species, and that a management plan must not, or not be likely to, adversely affect the conservation status of a protected species or a population of a protected species. The Goshawk meets the definition of a protected species.\n\nThe plan should allow for the input of relevant landholders and decision makers, and provide more strategic and comprehensive protection of flora and fauna, while facilitating greater certainty for environmentally sensitive developments. The plan should also allow for adaptive management as better information becomes available on the Christmas Island Hawk-Owl and other threatened species, in particular greater definition of the areas of habitat which are critical to survival.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n\nAction 6\\. Continue effective and long-term rainforest rehabilitation program managed by DOTARS & supported by PAN and the mining company.\n\nThe Christmas Island Rainforest Rehabilitation Program has been occurring over the last three decades. Currently, the main aim is to revegetate mined areas adjacent to Abbott's Booby nesting habitat to reduce wind turbulence caused by the clearings and implicated in reduced nesting success of the booby (Reville et al. 1990). There are approx. 70 clearings covering 3200ha or 24% of the island area (Carew-Reid 1987). It is not known how Christmas Island Goshawks use these areas and at what stage of rehabilitation this habitat become important for foraging and for nesting – this will be investigated as part of the monitoring program.\n\nIn 2002, the Government decided to redirect the payment of the conservation levy to the portfolio agency with responsibility for the mine lease (DOTARS). A MOU between PAN and DOTARS was signed in February 2004 to allow PAN to continue the CIRRP within Christmas Island National Park. Rehabilitation is critical in terms of increasing rainforest habitat to further protect a range of species including the goshawk. The CIRRP is funded through a conservation levy that forms part of Christmas Island Phosphate’s lease. Effective management of the lease by DOTARS will ensure optimal outcomes for rainforest rehabilitation.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Crazy Ants have a negligible impact on Christmas Island Goshawk\n\nAction 7\\. Implement the Invasive Ants on Christmas Island Action Plan.\n\nPAN is currently implementing an Invasive Ants Action Plan, which aims to control and minimise the impact of Crazy Ants across the island. This program will require ongoing monitoring of Crazy Ant numbers and control of populations as detected. The recovery team must ensure that the control program does not have any negative impacts on Christmas Island Goshawks.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Maintenance of effective quarantine against the introduction of all avian diseases\n\nAction 8. Maintenance \\[and regular review\\] of a quarantine barrier between Christmas Island and all other lands which minimises the risks of new avian diseases establishing on Christmas Island.\n\nAs part of the regular review process of the Christmas Island Quarantine Service, the effectiveness of the quarantine barrier preventing the man-made introduction of avian diseases onto Christmas Island should be assessed. Review of the Christmas Island Quarantine Service is undertaken by the Australian Quarantine Inspection Service, with input from PAN and in consultation with DOTARS. There is no additional cost for this action.\n\nSpecific objective 4. Increase community involvement and awareness\n\nPerformance criteria Demonstrated increase in community awareness and support for habitat protection\n\nAction 9. A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks.\n\nAs part of the Christmas Island Education and Interpretation Plan, which has a key objective of increasing knowledge of and local support for habitat and species protection, develop a multi-lingual pamphlet on Christmas Island Goshawk stressing their uniqueness and how they are part of Christmas Island ecosystems. This pamphlet would encourage the community to report sightings of colour-banded goshawks. It should also discuss managing problem goshawks around poultry, emphasising that goshawks are protected, and offering solutions to dealing with problem birds.\n\nThis should be further supported by a conservation related curriculum unit for both primary and secondary school students which raises awareness of the Christmas Island habitats and species generally.\n\n  \n\nSpecific objective 5\\. Implement the Recovery Plan through a Recovery Team\n\nPerformance criteria Demonstrated successful operation of the Recovery Team over five years\n\nActions 10. Establish a recovery team which meets regularly\n\n11\\. Carry out a major review of the recovery plan\n\nA recovery team shall be formed to implement this recovery plan. The team should comprise representatives of PANCI, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate eg. other affected Christmas Island interests, and other specialists where necessary. Progress will be monitored and evaluated yearly by members of the recovery team through an annual review. The monitoring process will include: compiling information from resighting of banded birds, success of monitored nests, and assessing progress of all actions against the criteria and objectives of the recovery plan (data comparisons performed by participating biologist). A major review of this performance must be conducted at the end of the first five years of implementation, by an independent reviewer.","sortOrder":3},{"sectionNumber":"Conservation Status","sectionType":"division","heading":"Conservation Status","content":"### Conservation Status\n\nThe Christmas Island Goshawk (Accipiter fasciatus natalis) is currently listed as Endangered (EPBC Act), however, according to Garnett & Crowley (2000) it now fulfils the criteria for Critically Endangered.","sortOrder":4},{"sectionNumber":"Taxonomic Status","sectionType":"division","heading":"Taxonomic Status","content":"### Taxonomic Status\n\nSince the Christmas Island Goshawk was described by Lister in 1888 as a full species Accipiter natalis, debate has continued over whether it is more closely allied to the Brown Goshawk Accipiter fasciatus, or to the Grey Goshawk A. novaehollandiae. Lister (1888) thought the Christmas Island Goshawk was allied to the Grey Goshawk from the Maluku A. n. griseogularis. Chasen (1933) disagreed, thinking it a member of the A. fasciatus group. The Christmas Island Goshawk may have been isolated from populations of either of these species for a long period of time, and Carter (1994) raised the possibility of resurrecting the Christmas Island subspecies natalis to a full species.","sortOrder":5},{"sectionNumber":"International Obligations","sectionType":"division","heading":"International Obligations","content":"### International Obligations\n\nThe Christmas Island Goshawk is not listed under international agreements.","sortOrder":6},{"sectionNumber":"Distribution and location","sectionType":"division","heading":"Distribution and location","content":"### Distribution and location\n\nThe Christmas Island Goshawk is restricted to Christmas Island, a volcanic island in the Indian Ocean (10o0'S; 105o40' E), approximately 1400km northwest of Australia. The island sits on the northernmost edge of the Australasian continental plate, and immediately north the ocean floor drops into the Java Trench and depths of up to 6000m. The nearest land is Java in the Republic of Indonesia, 360km to the north. Christmas Island is truly oceanic and all its native biota has colonised by sea or by air (Gray 1981). The island is 135km2 and 75% is covered with original vegetation (Environment Australia 2002).\n\nGibson-Hill (1947) reported that Christmas Island Goshawks seemed to prefer areas of ‘slightly thinner growth on the edge of thick jungle or the borders of clearings’. During 1994 and 1995 goshawks were observed in all major habitats on the island (Hill unpubl. data,). Goshawks were regularly seen hunting in regrowth vegetation along roadsides. This may be merely because they were much more visible in this habitat compared to intact forests. Parks Australia North (PAN) data from a systematic survey over all habitat of the island (over 1000 survey points) suggest that more goshawks are located within rainforest (95% of sightings) than in cleared areas. The Christmas Island Goshawk is the least frequently encountered diurnal forest bird and might best be described as ‘widespread and uncommon’.\n\nThere are no data on adult or juvenile movements. Adult Christmas Island Goshawks are probably resident on their territories year-round. Grey Goshawks defend a territory around the nest site which appears to be significantly smaller than the home range in which they feed (Burton 1991).\n\nThere are no detailed population estimates for Christmas Island Goshawks. New Caledonian Brown Goshawks, a similar-sized closely related bird, occurred at a density of 0.89 - 0.96 adults/km2 in rainforest (Thiollay 1993). There is approximately 101 km2 of primary forest on Christmas Island which suggests that it is unlikely that there are many more than 50 pairs or 100 mature Christmas Island Goshawks on Christmas Island.\n\n###   \n\n### ![](image.002.png)","sortOrder":7},{"sectionNumber":"Biology","sectionType":"division","heading":"Biology","content":"###   \nBiology\n\nThe Christmas Island Goshawk has been recorded eating a wide range of vertebrates and invertebrates including birds, small mammals and insects. Prey includes Christmas Island White-eye Zosterops natalis, Island Thrush Turdus poliocephalus, Emerald Dove Chalcophaps indica, Java Sparrow Lonchura oryzivora, domestic chickens Gallus gallus, (Lister 1888, Gibson-Hill 1947, Hill, unpubl. data); as well as nestling Red-footed Boobies Sula sula (Hill, unpubl. data) and an adult Golden Bosun Phaethon rubricauda westralis (J. Tranter pers. comm.). It has also been recorded catching rats Rattus rattus (Gibson-Hill 1947) and ‘lizards’ (Lister 1888). It is frequently seen catching grasshoppers (probably Valanga sp.) which are common in secondary vegetation around the island (Gibson-Hill 1947, Hill unpubl. data). Gibson-Hill (1947) also recorded it feeding on the large centipede Scolopendra morsitans, which is widespread on the island, and on beetles (Coleoptera) and mantids (Mantidae).\n\nThe Christmas Island Goshawk captures its prey off the ground or in the air by seizing it in flight with its long, strong legs. Like other goshawks, it will hunt from a perch where it may sit for some time waiting for suitable prey. Christmas Island Goshawks will chase small flying birds through the forest as well as pounce on birds foraging on the ground (Hill unpubl. data). Hill observed goshawks ‘short-stay perch-hunting’ (after Newton 1986) flying along a track, perching for a few minutes then flying 50-100m further and landing again, as well as pursuing insects on the ground and snatching insects off foliage. There is likely to be some sex differences in hunting and in diet, as is found in both Brown and Grey Goshawks (Marchant & Higgins 1993). Given the large sexual dimorphism (similar to Grey Goshawk, Hill unpubl. data) and the opportunity for niche expansion on the island in the absence of other diurnal raptors, this difference may be quite pronounced.\n\nGibson-Hill (1947) described three Christmas Island Goshawk nests all of which were in near-horizontal forks of forest trees, one 10m high and two 15-20m in tall forest trees. Hill (unpubl. data) observed four nests: two in Syzygium nervosum in Primary rainforest; and two in Terminalia catappa in Marginal rainforest. All nests were 25-35m above the ground, in the fork of two or more branches, three towards the edge of the crown, and one in the middle of the tree. One nest (probably not used) was 30-40cm in diameter, two others were 50-80cm in diameter, and the fourth was under construction. Nest size and location within the tree share more similarities with Grey than Brown Goshawks (J. Young pers. comm.; see Burton et al. 1994).","sortOrder":8},{"sectionNumber":"Habitat","sectionType":"division","heading":"Habitat","content":"### Habitat\n\nApproximately 75% of Christmas Island is still covered with natural vegetation and 84% of this (or 63% of the island) is protected within National Park (Figure 1). The remaining Primary rainforest is currently protected from clearance by a Federal Government moratorium, in conjunction with the requirements for approval under the EPBC Act and Environment Protection and Biodiversity Conservation Regulations 2000 (EPBC Regulations). Marginal rainforest and Second-growth rainforest or Secondary Vegetation outside the National Park may only be cleared if approval is given under the EPBC Act and Regulations.\n\nSchedule 12 of the EPBC Regulations use Du Puy’s classification to describe the Christmas Island forests. Primary Rainforest is restricted to the central plateau and is evergreen closed forest 30-40m high with emergent trees up to 45m. Common canopy trees include Syzygium nervosum, Planchonella nitida, and Hernandia ovigera. The canopy formed by these trees is irregular, especially on the western side of the island where it is sheltered from strong southeast trade winds which blow throughout the dry season. The understorey comprises dense thickets of Pandanus elatus up to 4m tall, and more open areas which commonly include the plants Arenga listeri, Leea angulata, Ochrosia ackeringae, Pisonia umbellifera and Aidia aff. racemosa (Environment Australia 1994).\n\nMarginal Rainforests grow on generally shallower soils on the coastal terraces and scree slopes surrounding the island. Many Marginal Rainforest trees lose their leaves in the dry season. Marginal rainforest is generally lower than Primary Rainforest, often less than 30m, and may include many vines. The canopy height and shape are variable, determined primarily by the degree of exposure to the southeast trade winds. Marginal rainforest facing south and east has a smooth, wind-pruned canopy and tends to increase in height with increasing distance from the sea cliff. The inland cliffs and scree slopes may have no vegetation or carry a closed forest, depending on the degree of the slope. Trees such as Ficus microcarpa and Dendrocnide sinuata are common there.\n\nOld stockpiles and cleared areas that have not been mined may support low second-growth forest of colonising trees such as Macaranga tanarius and Claoxylon indicum and an introduced tree Leucaena leucocephala generally less than 10m high. Previously mined areas tend to have very little remaining soil and on them grow dense herblands of a fern Nephrolepis multiflora to 2m high along with introduced scramblers and occasional low trees.","sortOrder":9},{"sectionNumber":"Habitat Critical to Survival","sectionType":"division","heading":"Habitat Critical to Survival","content":"### Habitat Critical to Survival\n\nIdentification of habitat critical to survival of the Christmas Island Goshawk is difficult due to the lack of information on specific habitat requirements. Although Christmas Island Goshawks may be generalists which forage in most habitats, it is highly likely that they require rainforest to breed as these habitat contain suitable trees. Applying the precautionary principle and using the criteria provided by the EPBC Act, habitat critical to survival of the Christmas Island Goshawks is defined as all Primary Rainforest, Marginal Rainforest (Du Puy 1993) and possibly second-growth forest suitable for nesting. Primary Rainforest and Marginal Rainforest has been identified and mapped (Figure 1). Dexter (2000) recommended mapping the ecological quality of second-growth forests outside the national park. As this has not occurred, and there is insufficient information on the habitat requirements of the Christmas Island Goshawks, suitable second-growth forest cannot be identified at this time.","sortOrder":10},{"sectionNumber":"Affected Interests","sectionType":"division","heading":"Affected Interests","content":"### Affected Interests\n\nCommonwealth Parks Australia North (PAN), Shire of Christmas Island, Christmas Island Phosphates, Union of Christmas Island Workers, the Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services (DOTARS), and the Department of Immigration, Multicultural and Indigenous Affairs.","sortOrder":11},{"sectionNumber":"Role and interests of indigenous people","sectionType":"division","heading":"Role and interests of indigenous people","content":"### Role and interests of indigenous people\n\nNot applicable. Christmas Island does not have an indigenous population.","sortOrder":12},{"sectionNumber":"Social and economic impacts","sectionType":"division","heading":"Social and economic impacts","content":"### Social and economic impacts\n\nThe actions in this plan may have positive and negative social and economic impacts. Positive social impacts will arise from community education actions that will increase Christmas Islanders knowledge and interest in their own environment. The rainforest rehabilitation program provides on-island jobs, as will the goshawk monitoring program. Christmas Island endemic birds attract specialist bird watching groups each year which is high value, low-impact tourism.\n\nNegative social and economic impacts arising from implementation of the plan could include greater restrictions due to review of the quarantine barrier, and potential economic effects of preventing illegal destruction of Christmas Island goshawks by residents. The EPBC Act already provides a regulatory framework for the protection of rainforest on Christmas Island, and one element of this is assessing potential impacts of proposed developments on listed threatened species. These provisions have the potential to impact on economic activity, for example by adding additional obligations for industry and other development on the island in order to minimise impacts on listed species. This arises from the listing of the species under the EPBC Act invoking a range of protective provisions and offences where a population is to be affected. The magnitude of this potential impact is unknown, as it will vary with the location, size and extent of the activity, proposed or current. However, the plan will aid in determining what could be significant impacts on the Christmas Island Goshawk (through defining habitat critical to survival and developing a management plan outside the national park).","sortOrder":13},{"sectionNumber":"Biodiversity Benefits:","sectionType":"division","heading":"Biodiversity Benefits:","content":"### Biodiversity Benefits:\n\nProtection of the habitat of the Christmas Island goshawk provides protection for numerous other listed species (Table 1). Community education targeted at this species will promote awareness of all the endemic forest birds and their conservation needs.\n\nTable 1: Native species listed under the EPBC Act that will benefit from recovery actions listed in this plan (after Environment Australia 2002).\n\n| Taxon                                                     | National Status |\n| --------------------------------------------------------- | --------------- |\n| Tectaria devexa var. minor                                | E               |\n| Carmona retusa                                            | V               |\n| Christmas Island Pipistrelle (e) Pipistrellus murrayi     | E               |\n| Christmas Island Shrew 1 (e) Crocidura attenuata trichura | E               |\n| Christmas Island Blind Snake (e) Ramphotyphlops exocoeti  | V               |\n| Christmas Island Gecko Lepidodactylus listeri             | V               |\n| Christmas Island Hawk-owl (e) Ninox natalis               | VJ              |\n| Abbott’s Booby (e) Papasula abbotti                       | E M S J         |\n| Red-footed Booby Sula sula rubripes                       | M S C J         |\n| Christmas Island Frigatebird (e) Fregata andrewsi         | V M S C J       |\n| Great Frigatebird Fregata minor minor                     | M S C J         |\n\nNotes:\n\n(e) species and subspecies endemic to the island E Listed under the EPBC Act as Endangered\n\nV Listed under the EPBC Act as Vulnerable M Listed under the EPBC Act as a Migratory species\n\nMF Within a family listed under the EPBC Act as Migratory S Listed Marine species under the EPBC Act\n\nC Listed under China-Australia Migratory Bird Agreement\n\nJ Listed under Japan-Australia Migratory Bird Agreement","sortOrder":14},{"sectionNumber":"Relationship to other plans","sectionType":"division","heading":"Relationship to other plans","content":"### Relationship to other plans\n\nThe Christmas Island National Park Plan of Management is the strategic nature conservation document for the island. This recovery plan makes numerous recommendations in common with other recovery plans for Christmas Island threatened taxa. Opportunities for sharing resources and points shared in common with other recovery plans are identified in the implementation section of this plan.","sortOrder":15},{"sectionNumber":"Recovery Team","sectionType":"division","heading":"Recovery Team","content":"### Recovery Team\n\nA Recovery Team for the Christmas Island Goshawk has not yet been formed. The team should comprise representatives of PAN, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate.","sortOrder":16},{"sectionNumber":"Threats to the species","sectionType":"part","heading":"Threats to the species","content":"## Executive Summary\n\nThe Christmas Island Goshawk is currently listed as Endangered under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). It is currently considered a subspecies of Accipiter fasciatus, but its taxonomy needs resolution. It is considered to be the rarest endemic bird on Christmas Island, where it occurs in all habitats from primary and marginal rainforests to suitable areas of secondary regrowth vegetation. The total population size is thought to be very small, perhaps as few as 100 adults, and is probably limited by the availability of suitable rainforest habitat. Crazy Ants pose an unknown but potentially critical threat to the survival of this bird. The recovery plan objective is to downgrade the Christmas Island Goshawk from Endangered to Conservation Dependent, primarily through successful implementation of the Invasive Ants on Christmas Island Action Plan and protection of habitat critical to the survival of the species from clearance. An assessment of goshawk population dynamics is the most essential requirement of this recovery plan, and community awareness and participation in the conservation of this endemic raptor are also important actions.\n\n##   \nBackground Information\n\n### Conservation Status\n\nThe Christmas Island Goshawk (Accipiter fasciatus natalis) is currently listed as Endangered (EPBC Act), however, according to Garnett & Crowley (2000) it now fulfils the criteria for Critically Endangered.\n\n### Taxonomic Status\n\nSince the Christmas Island Goshawk was described by Lister in 1888 as a full species Accipiter natalis, debate has continued over whether it is more closely allied to the Brown Goshawk Accipiter fasciatus, or to the Grey Goshawk A. novaehollandiae. Lister (1888) thought the Christmas Island Goshawk was allied to the Grey Goshawk from the Maluku A. n. griseogularis. Chasen (1933) disagreed, thinking it a member of the A. fasciatus group. The Christmas Island Goshawk may have been isolated from populations of either of these species for a long period of time, and Carter (1994) raised the possibility of resurrecting the Christmas Island subspecies natalis to a full species.\n\n### International Obligations\n\nThe Christmas Island Goshawk is not listed under international agreements.\n\n### Distribution and location\n\nThe Christmas Island Goshawk is restricted to Christmas Island, a volcanic island in the Indian Ocean (10o0'S; 105o40' E), approximately 1400km northwest of Australia. The island sits on the northernmost edge of the Australasian continental plate, and immediately north the ocean floor drops into the Java Trench and depths of up to 6000m. The nearest land is Java in the Republic of Indonesia, 360km to the north. Christmas Island is truly oceanic and all its native biota has colonised by sea or by air (Gray 1981). The island is 135km2 and 75% is covered with original vegetation (Environment Australia 2002).\n\nGibson-Hill (1947) reported that Christmas Island Goshawks seemed to prefer areas of ‘slightly thinner growth on the edge of thick jungle or the borders of clearings’. During 1994 and 1995 goshawks were observed in all major habitats on the island (Hill unpubl. data,). Goshawks were regularly seen hunting in regrowth vegetation along roadsides. This may be merely because they were much more visible in this habitat compared to intact forests. Parks Australia North (PAN) data from a systematic survey over all habitat of the island (over 1000 survey points) suggest that more goshawks are located within rainforest (95% of sightings) than in cleared areas. The Christmas Island Goshawk is the least frequently encountered diurnal forest bird and might best be described as ‘widespread and uncommon’.\n\nThere are no data on adult or juvenile movements. Adult Christmas Island Goshawks are probably resident on their territories year-round. Grey Goshawks defend a territory around the nest site which appears to be significantly smaller than the home range in which they feed (Burton 1991).\n\nThere are no detailed population estimates for Christmas Island Goshawks. New Caledonian Brown Goshawks, a similar-sized closely related bird, occurred at a density of 0.89 - 0.96 adults/km2 in rainforest (Thiollay 1993). There is approximately 101 km2 of primary forest on Christmas Island which suggests that it is unlikely that there are many more than 50 pairs or 100 mature Christmas Island Goshawks on Christmas Island.\n\n###   \n\n### ![](image.002.png)\n\n###   \nBiology\n\nThe Christmas Island Goshawk has been recorded eating a wide range of vertebrates and invertebrates including birds, small mammals and insects. Prey includes Christmas Island White-eye Zosterops natalis, Island Thrush Turdus poliocephalus, Emerald Dove Chalcophaps indica, Java Sparrow Lonchura oryzivora, domestic chickens Gallus gallus, (Lister 1888, Gibson-Hill 1947, Hill, unpubl. data); as well as nestling Red-footed Boobies Sula sula (Hill, unpubl. data) and an adult Golden Bosun Phaethon rubricauda westralis (J. Tranter pers. comm.). It has also been recorded catching rats Rattus rattus (Gibson-Hill 1947) and ‘lizards’ (Lister 1888). It is frequently seen catching grasshoppers (probably Valanga sp.) which are common in secondary vegetation around the island (Gibson-Hill 1947, Hill unpubl. data). Gibson-Hill (1947) also recorded it feeding on the large centipede Scolopendra morsitans, which is widespread on the island, and on beetles (Coleoptera) and mantids (Mantidae).\n\nThe Christmas Island Goshawk captures its prey off the ground or in the air by seizing it in flight with its long, strong legs. Like other goshawks, it will hunt from a perch where it may sit for some time waiting for suitable prey. Christmas Island Goshawks will chase small flying birds through the forest as well as pounce on birds foraging on the ground (Hill unpubl. data). Hill observed goshawks ‘short-stay perch-hunting’ (after Newton 1986) flying along a track, perching for a few minutes then flying 50-100m further and landing again, as well as pursuing insects on the ground and snatching insects off foliage. There is likely to be some sex differences in hunting and in diet, as is found in both Brown and Grey Goshawks (Marchant & Higgins 1993). Given the large sexual dimorphism (similar to Grey Goshawk, Hill unpubl. data) and the opportunity for niche expansion on the island in the absence of other diurnal raptors, this difference may be quite pronounced.\n\nGibson-Hill (1947) described three Christmas Island Goshawk nests all of which were in near-horizontal forks of forest trees, one 10m high and two 15-20m in tall forest trees. Hill (unpubl. data) observed four nests: two in Syzygium nervosum in Primary rainforest; and two in Terminalia catappa in Marginal rainforest. All nests were 25-35m above the ground, in the fork of two or more branches, three towards the edge of the crown, and one in the middle of the tree. One nest (probably not used) was 30-40cm in diameter, two others were 50-80cm in diameter, and the fourth was under construction. Nest size and location within the tree share more similarities with Grey than Brown Goshawks (J. Young pers. comm.; see Burton et al. 1994).\n\n### Habitat\n\nApproximately 75% of Christmas Island is still covered with natural vegetation and 84% of this (or 63% of the island) is protected within National Park (Figure 1). The remaining Primary rainforest is currently protected from clearance by a Federal Government moratorium, in conjunction with the requirements for approval under the EPBC Act and Environment Protection and Biodiversity Conservation Regulations 2000 (EPBC Regulations). Marginal rainforest and Second-growth rainforest or Secondary Vegetation outside the National Park may only be cleared if approval is given under the EPBC Act and Regulations.\n\nSchedule 12 of the EPBC Regulations use Du Puy’s classification to describe the Christmas Island forests. Primary Rainforest is restricted to the central plateau and is evergreen closed forest 30-40m high with emergent trees up to 45m. Common canopy trees include Syzygium nervosum, Planchonella nitida, and Hernandia ovigera. The canopy formed by these trees is irregular, especially on the western side of the island where it is sheltered from strong southeast trade winds which blow throughout the dry season. The understorey comprises dense thickets of Pandanus elatus up to 4m tall, and more open areas which commonly include the plants Arenga listeri, Leea angulata, Ochrosia ackeringae, Pisonia umbellifera and Aidia aff. racemosa (Environment Australia 1994).\n\nMarginal Rainforests grow on generally shallower soils on the coastal terraces and scree slopes surrounding the island. Many Marginal Rainforest trees lose their leaves in the dry season. Marginal rainforest is generally lower than Primary Rainforest, often less than 30m, and may include many vines. The canopy height and shape are variable, determined primarily by the degree of exposure to the southeast trade winds. Marginal rainforest facing south and east has a smooth, wind-pruned canopy and tends to increase in height with increasing distance from the sea cliff. The inland cliffs and scree slopes may have no vegetation or carry a closed forest, depending on the degree of the slope. Trees such as Ficus microcarpa and Dendrocnide sinuata are common there.\n\nOld stockpiles and cleared areas that have not been mined may support low second-growth forest of colonising trees such as Macaranga tanarius and Claoxylon indicum and an introduced tree Leucaena leucocephala generally less than 10m high. Previously mined areas tend to have very little remaining soil and on them grow dense herblands of a fern Nephrolepis multiflora to 2m high along with introduced scramblers and occasional low trees.\n\n### Habitat Critical to Survival\n\nIdentification of habitat critical to survival of the Christmas Island Goshawk is difficult due to the lack of information on specific habitat requirements. Although Christmas Island Goshawks may be generalists which forage in most habitats, it is highly likely that they require rainforest to breed as these habitat contain suitable trees. Applying the precautionary principle and using the criteria provided by the EPBC Act, habitat critical to survival of the Christmas Island Goshawks is defined as all Primary Rainforest, Marginal Rainforest (Du Puy 1993) and possibly second-growth forest suitable for nesting. Primary Rainforest and Marginal Rainforest has been identified and mapped (Figure 1). Dexter (2000) recommended mapping the ecological quality of second-growth forests outside the national park. As this has not occurred, and there is insufficient information on the habitat requirements of the Christmas Island Goshawks, suitable second-growth forest cannot be identified at this time.\n\n### Affected Interests\n\nCommonwealth Parks Australia North (PAN), Shire of Christmas Island, Christmas Island Phosphates, Union of Christmas Island Workers, the Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services (DOTARS), and the Department of Immigration, Multicultural and Indigenous Affairs.\n\n### Role and interests of indigenous people\n\nNot applicable. Christmas Island does not have an indigenous population.\n\n### Social and economic impacts\n\nThe actions in this plan may have positive and negative social and economic impacts. Positive social impacts will arise from community education actions that will increase Christmas Islanders knowledge and interest in their own environment. The rainforest rehabilitation program provides on-island jobs, as will the goshawk monitoring program. Christmas Island endemic birds attract specialist bird watching groups each year which is high value, low-impact tourism.\n\nNegative social and economic impacts arising from implementation of the plan could include greater restrictions due to review of the quarantine barrier, and potential economic effects of preventing illegal destruction of Christmas Island goshawks by residents. The EPBC Act already provides a regulatory framework for the protection of rainforest on Christmas Island, and one element of this is assessing potential impacts of proposed developments on listed threatened species. These provisions have the potential to impact on economic activity, for example by adding additional obligations for industry and other development on the island in order to minimise impacts on listed species. This arises from the listing of the species under the EPBC Act invoking a range of protective provisions and offences where a population is to be affected. The magnitude of this potential impact is unknown, as it will vary with the location, size and extent of the activity, proposed or current. However, the plan will aid in determining what could be significant impacts on the Christmas Island Goshawk (through defining habitat critical to survival and developing a management plan outside the national park).\n\n### Biodiversity Benefits:\n\nProtection of the habitat of the Christmas Island goshawk provides protection for numerous other listed species (Table 1). Community education targeted at this species will promote awareness of all the endemic forest birds and their conservation needs.\n\nTable 1: Native species listed under the EPBC Act that will benefit from recovery actions listed in this plan (after Environment Australia 2002).\n\n| Taxon                                                     | National Status |\n| --------------------------------------------------------- | --------------- |\n| Tectaria devexa var. minor                                | E               |\n| Carmona retusa                                            | V               |\n| Christmas Island Pipistrelle (e) Pipistrellus murrayi     | E               |\n| Christmas Island Shrew 1 (e) Crocidura attenuata trichura | E               |\n| Christmas Island Blind Snake (e) Ramphotyphlops exocoeti  | V               |\n| Christmas Island Gecko Lepidodactylus listeri             | V               |\n| Christmas Island Hawk-owl (e) Ninox natalis               | VJ              |\n| Abbott’s Booby (e) Papasula abbotti                       | E M S J         |\n| Red-footed Booby Sula sula rubripes                       | M S C J         |\n| Christmas Island Frigatebird (e) Fregata andrewsi         | V M S C J       |\n| Great Frigatebird Fregata minor minor                     | M S C J         |\n\nNotes:\n\n(e) species and subspecies endemic to the island E Listed under the EPBC Act as Endangered\n\nV Listed under the EPBC Act as Vulnerable M Listed under the EPBC Act as a Migratory species\n\nMF Within a family listed under the EPBC Act as Migratory S Listed Marine species under the EPBC Act\n\nC Listed under China-Australia Migratory Bird Agreement\n\nJ Listed under Japan-Australia Migratory Bird Agreement\n\n### Relationship to other plans\n\nThe Christmas Island National Park Plan of Management is the strategic nature conservation document for the island. This recovery plan makes numerous recommendations in common with other recovery plans for Christmas Island threatened taxa. Opportunities for sharing resources and points shared in common with other recovery plans are identified in the implementation section of this plan.\n\n### Recovery Team\n\nA Recovery Team for the Christmas Island Goshawk has not yet been formed. The team should comprise representatives of PAN, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate.\n\n##   \nThreats to the species\n\n### Birds on Islands\n\nIsland birds are particularly vulnerable to extinction and a very high number of island birds relative to continental species have become extinct in the past century (Diamond 1985, King 1985). This is because (i) island birds tend to have smaller population sizes than continental species and small populations are more vulnerable to extinction (Lacy 1987); (ii) island birds have often evolved in the absence of many predators, diseases and competitors and they can be catastrophically affected by the introduction of one of these (King 1985); (iii) there are often no opportunities for dispersal and thus recolonisation on islands, and this factor also predisposes a population to extinction (Lacy 1987); and (iv) there is often less opportunity to make reserves on islands because land area is limited (Diamond 1985). Predators in general are often naturally rare because their prey, other animals, are less abundant than other food resources. Thus island raptors are a particularly vulnerable group of island birds.\n\n### Crazy Ants\n\nThe exotic invasive yellow crazy ant (Anoplolepis gracilipes) arrived on Christmas Island more than 70 years ago, and is now widespread throughout rainforest (Orchard et al. 2002). The ants can form multi-queened 'super-colonies', in which they occur at very high densities. This has apparently been a relatively recent phenomenon; with the first supercolony discovered in 1989, with further dramatic increases probably beginning around the mid-1990s.\n\nAt supercolony densities, the ant is having a devastating impact on the island's ecosystems. Red crabs, robber crabs, blue crabs and most other ground dwelling animals such as reptiles, have been and are continuing to be severely impacted, sometimes to the point of local extinction in heavily infested areas. By eliminating local populations of red crabs, the ants are also having a marked effect on forest composition and structure, and litter dynamics in infested areas. The feeding activities of the ants and their mutualistic scale insects can stress large trees to the point of death, and appear to be causing canopy dieback in areas of supercolony formation (Orchard et al. 2002). In addition groundings of birds in crazy ant supercolonies would most probably result in the death of the bird. Although research has shown that crazy ants have little impact on canopy insects, it is possible that the ants may reduce insect populations. In the recent Risk Watch List compiled for the Christmas Island National Park and Conservancy, the crazy ant invasion was rated as an Extreme Risk to biodiversity and conservation values, with catastrophic consequences of failure to implement effective control measures.\n\nPAN field workers and assistants conducted an island-wide ant survey during May-August, 2001. The survey was designed by CAMBI (Centre for the Analysis and Management of Biological Invasions, Monash University). Of 972 sites surveyed; 741 surveyed points fell in natural forest. Crazy ants were recorded at 46.7% of sites in natural forest. Super-colony densities were recorded at 22.7% of sites in natural forest. Using these survey points as a representative sample of the forest, it was estimated that 2,379 ha of the estimated 10,492 ha of forest on the island was infested with crazy ants at supercolony densities (Orchard et al. 2002). Crazy ants are not evenly distributed throughout natural forest but are more commonly found on terrace forests and less commonly found in deep-soil tall-closed forest, which make up most of the natural forests on the island plateau. Deep-soil tall-closed forest made up 31% of census sites in natural forest but only 6.5% of supercolony records (Orchard et al. 2002).\n\nIn September 2002 an aerial baiting program was undertaken, and all known supercolonies treated with insecticide. Results so far indicate the program was successful in controlling supercolonies over 2500ha of Christmas Island. Crazy ants are still present in low densities, however, further high densities of crazy ants may establish in the terrace forests without warning. PAN staff will continue to monitor any new supercolony formation and treat by hand baiting over the next few years.\n\n### Disease\n\nA serious threat to all island birds is the introduction of new diseases. Island birds have often evolved in the absence of diseases common in continental bird faunas and the introduction of such diseases to island birds can be disastrous. An example of this is the introduction of avian malaria to Hawaii, which caused the extinction of almost the entire endemic bird fauna from below 600 m altitude, and was probably the main cause of the total extinction of several bird species (Hay 1986). The range of many surviving species was severely reduced and fragmented which in turn markedly increased their chances of extinction. Avian malaria arrived with the accidental introduction of a new species of mosquito (Hay 1986). Christmas Island has been very vulnerable to the accidental introduction of new disease. In 1994 a quarantine barrier was established between the island and Indonesia and Australia and this has reduced that risk.\n\n### Habitat Loss\n\nApproximately 25% of the island’s original forests have been cleared and replaced by shrublands of ferns on minefields, Secondary Vegetation on stockpiles, and roads and housing (Environment Australia 1994). This has meant a loss of available forested habitat, although secondary vegetation growing along roadsides and on stockpiles contributes to the feeding habitat for Christmas Island Goshawks at least at some times of the year. Nonetheless, based on the decrease in available habitat, the total Christmas Island Goshawk population has probably decreased by at least 25% since settlement (Stokes 1988).\n\nPrimary forests on Christmas Island have been fragmented in places by clearing and most Primary Rainforest is dissected by roads and old mining 'grid lines' The grid line system, a series of parallel lines 120 m apart and a bulldozer blade in width was installed in the 1970's to explore the island's phosphate resources. After 25 years these are mostly overgrown with dense stands of Secondary Vegetation.\n\nThere are some major development proposals and associated infrastructure in particular the Immigration Centre, Satellite launching station, airport upgrade, and expansion of phosphate mining that have the potential to reduce the area of goshawk habitat. One development condition for the satellite launching facility is to develop and implement an environment management plan, which is currently underway. The EMP will address threatened fauna and their habitat on the sites. The immigration centre is exempt from assessment under the EPBC Act, however, the DEH is being consulted on environmental aspects of the development, including monitoring of habitat for potential impacts. The airport upgrade and phosphate mining expansion proposals are undergoing assessment for environmental impact under the EPBC Act. Survey and monitoring of the Christmas Island Goshawk will aid in determining potential impacts of these and future developments on goshawk habitat.\n\nThe Christmas Island Rainforest Rehabilitation Program (CIRRP) is a long term program to revegetate old minefields. The CIRRP is funded by a conservation levy that is collected from Phosphate Resources Limited at an indexed rate per tonne. Rehabilitation priorities are primarily determined to protect Abbott’s Booby sites, although other threatened species are also considered in determining the priority of sites. A significant additional result of CIRRP will be the reestablishment of habitat that supports other species such as the goshawk.\n\n###   \nNatural Catastrophes\n\nAlthough goshawks, as generalist raptors, are likely to cope well with habitat modifying natural catastrophes, the effect of events such as cyclones are more severe on small populations. There is a possibility of an increase in severe storms and cyclones occurring as a result of Greenhouse gas-induced climate change. It would be useful to understand the effects of cyclones and severe storms on goshawk populations to help predict the impact of any increases.\n\n### Small population size\n\nThe risk of extinction varies inversely with population size (Lacy 1987) due primarily to demographic and environmental variability the effects of which can become marked in small populations. Demographic variability is the individual variation in reproductive success, which is masked when population size is large. Very small populations (< 30 individuals) can easily become extinct as a result of random demographic variation between individuals (Caughley & Sinclair 1994).\n\nThe reproductive success of a population is also likely to vary with fluctuations in the environment. The degree of this fluctuation is determined by how much seasonal and annual variation in habitat quality there is. The influence of this on the risk of extinction increases markedly with decreasing population size. For example, a cyclone, an extreme example of environmental variation, causing the death of half the population of 10000 animals has an insignificant affect on the probability of extinction of that population. However, if half a population of 200 animals dies, the risk of extinction of that population increases significantly (Caughley & Sinclair 1994).\n\n### Inbreeding Depression\n\nThe risks of inbreeding depression increases with decreasing population size due to the increase in frequency of homozygous alleles. Populations at risk of inbreeding depression are probably a couple of dozen individuals or smaller in size which have been at that size for several generations, as a population contraction must last for several generations to lead to a significant loss of heterozygosity (Caughley & Sinclair 1994). Any action on the island which might permanently, rather than temporarily, reduce the number of mature individual Christmas Island Goshawks (such as permanent forest clearance rather than temporary storm damage) would raise the possibility of inbreeding depression as a concern in their conservation.\n\n### Other Potential Threats\n\nKilling of Christmas Island Goshawks, especially by poultry owners, has been considered to be a significant threat to goshawks in the past (Stokes 1988). Given the small population size, any killing of either juveniles or adult Christmas Island Goshawks would be considered a significant threat to this taxon. Another possible threat is posed by cats, especially near settled areas.\n\nRoad killed Christmas Island Goshawks have been reported. A substantial increase in vehicular traffic will be associated with the proposed satellite launching station and the new immigration, reception and processing centre (IRPC). This will likely increase the number of roadkills in high traffic areas. Thus it is possible that collisions with cars will become an issue for the conservation of populations in high traffic areas.\n\nWeeds, especially newly introduced invasive species, could potentially impact on Christmas Island Goshawk nest sites, for example by forming vine towers over nesting trees.\n\n##   \nRecovery Objectives and Timelines\n\n### Overall objective:\n\nTo shift the status of the Christmas Island Goshawk from Endangered to Conservation Dependent within 10 years.\n\n### Specific objectives:\n\nTo reduce the risk of extinction and improve the conservation status of the taxon by:\n\n1.  Determine taxonomic status\n2.  Determine and maximise total population size and area of occupancy\n3.  Implement threat abatement strategies\n4.  Increase community involvement and awareness\n5.  Implement the Recovery Plan through a Recovery Team\n\n### Performance Criteria\n\n-   Taxonomic position of Christmas Island Goshawk clarified.\n\n-   Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring.\n-   Density of Christmas Island Goshawks increased as demonstrated by population monitoring\n\n-   Crazy Ants having an insignificant impact on Christmas Island Goshawks.\n-   Protection of all habitat critical outside the national park on Christmas Island.\n-   A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n-   Maintenance of effective quarantine against the introduction of all avian diseases\n-   Demonstrated increase in community awareness and support for habitat protection\n-   Demonstrated successful operation of the Recovery Team over five years.\n\n## Actions\n\nSpecific objective 1. Determine taxonomic status\n\nPerformance Criteria Taxonomic position of Christmas Island Goshawk clarified.\n\nAction 1\\. Investigate the taxonomic status of the Christmas Island Goshawk.\n\nThis project is designed to describe the relationships of the Christmas Island Goshawk to the Brown and Grey Goshawks by comparing DNA sequences of selected mitochondrial genes. Mitochondrial DNA may be collected from feathers allowing specimens already in collections to be used for analysis. The Museum of Victoria has material from Christmas Island Goshawks.\n\n  \n\nSpecific objective 2\\. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring\n\nDensity of Christmas Island Goshawks increased as demonstrated by population monitoring\n\nAction 2\\. Conduct detailed population survey\n\nThe recovery team will obtain estimates of population size, distribution, age structure, fledging success, and survival rates. This is required to assess the status of this taxon. Data on age structure and fledging success will be used to monitor the population, and will, over time, indicate whether or not any significant change in population size is occurring.\n\nTrap and colour band the Christmas Island Goshawk population across the island with the aim of individually marking the majority of the population. A majority of the population will remain individually marked by periodic supplementary trapping after this initial program. This action will provide data on the age-structure of the population, distribution on Christmas Island (including some information on habitat), ranging behaviour and spacing of individuals from resightings during the 18 month study, and thus some data on population size.\n\nThe members of the Australasian Raptor Association (ARA), part of Birds Australia, include many of the Australian professional and amateur experts on birds of prey. They may be able to provide unpaid assistance in trapping and marking birds, and in training of local staff in trapping techniques. The cost of their involvement would be airfares from Australia and accommodation and transport on Christmas Island. With the involvement of the ARA this project lends itself to being done by a post-graduate student. If ARA involvement was not possible dedicated researchers would re required, and the cost would be substantially higher.\n\nAction 3\\. On-going monitoring of the population\n\nIt is essential to formally and regularly monitor the size of the population, and breeding success. The recovery team will monitor the total population size at regular intervals (initially annually) such that, if a decline takes place, it provides sufficient warning to managers to allow remedial management and research to take place. The monitoring program shall be based on observations of the individually marked population of Christmas Island Goshawks. PAN Christmas Island staff, tour operators, and the community in general will be encouraged to record all observations of goshawks, and to report their locations accurately. This will provide an ongoing data set on the proportion of resightings of marked birds, and on the habitats in which birds are seen. Once every five years, volunteers from the ARA will visit the island and trap and mark additional birds to maintain a majority of marked birds in the population.\n\nIn addition a sample of goshawk nests should be monitored annually. Monitoring of ten nests over a number of years will provide a considerable amount of information on breeding success, diet, and habitat requirements, and would be relatively inexpensive. Breeding success data will allow evaluation of the species’ status and population viability. Knowledge of diet is essential to evaluate the effects of habitat change, and predict the likely impact of any proposed changes in land management on the island. Information on diet and breeding requirements, combined with data from the population survey on distribution, ranging and spacing, will enable refinement of the definition of habitat critical to the survival of Christmas Island Goshawks, and determination of at what stage/s they utilise rehabilitated areas.\n\nTen nests will be monitored for signs of breeding, and once breeding has commenced monitored regularly (fortnightly) while the breeding attempt continues. In addition roost sites of parents should be located and pellets collected for dietary analysis. The pellets should be relatively straightforward to analyse due to the limited number of potential vertebrate prey. Insects should be identified as accurately as possible. Monitoring of nesting and collection and analysis of pellets could be part of the general duties of the resource management staff of PANCI.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Protection of all habitat critical outside the national park.\n\nAction 4\\. Develop and implement wildlife management plan outside the national park\n\nAction 5\\. Ensure protection of habitat critical outside the national park.\n\nMost forested land outside the National Park is publicly owned, by the Commonwealth. PAN staff are routinely involved with the other stakeholders in negotiations over development proposals. This is ongoing. These proposals can occur unexpectedly and as there is no overall wildlife management plan outside the park, are assessed on an ad hoc basic. While consultation with other agencies and proponents on projects will be essential and ongoing (under the EPBC Act referrals process), achievement of long-term protection of threatened species outside the park requires a more comprehensive and considered approach through the development of a wildlife management program. The EPBC Regulations include that actions under an approved management plan are not considered offences in relation to protected species, and that a management plan must not, or not be likely to, adversely affect the conservation status of a protected species or a population of a protected species. The Goshawk meets the definition of a protected species.\n\nThe plan should allow for the input of relevant landholders and decision makers, and provide more strategic and comprehensive protection of flora and fauna, while facilitating greater certainty for environmentally sensitive developments. The plan should also allow for adaptive management as better information becomes available on the Christmas Island Hawk-Owl and other threatened species, in particular greater definition of the areas of habitat which are critical to survival.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n\nAction 6\\. Continue effective and long-term rainforest rehabilitation program managed by DOTARS & supported by PAN and the mining company.\n\nThe Christmas Island Rainforest Rehabilitation Program has been occurring over the last three decades. Currently, the main aim is to revegetate mined areas adjacent to Abbott's Booby nesting habitat to reduce wind turbulence caused by the clearings and implicated in reduced nesting success of the booby (Reville et al. 1990). There are approx. 70 clearings covering 3200ha or 24% of the island area (Carew-Reid 1987). It is not known how Christmas Island Goshawks use these areas and at what stage of rehabilitation this habitat become important for foraging and for nesting – this will be investigated as part of the monitoring program.\n\nIn 2002, the Government decided to redirect the payment of the conservation levy to the portfolio agency with responsibility for the mine lease (DOTARS). A MOU between PAN and DOTARS was signed in February 2004 to allow PAN to continue the CIRRP within Christmas Island National Park. Rehabilitation is critical in terms of increasing rainforest habitat to further protect a range of species including the goshawk. The CIRRP is funded through a conservation levy that forms part of Christmas Island Phosphate’s lease. Effective management of the lease by DOTARS will ensure optimal outcomes for rainforest rehabilitation.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Crazy Ants have a negligible impact on Christmas Island Goshawk\n\nAction 7\\. Implement the Invasive Ants on Christmas Island Action Plan.\n\nPAN is currently implementing an Invasive Ants Action Plan, which aims to control and minimise the impact of Crazy Ants across the island. This program will require ongoing monitoring of Crazy Ant numbers and control of populations as detected. The recovery team must ensure that the control program does not have any negative impacts on Christmas Island Goshawks.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Maintenance of effective quarantine against the introduction of all avian diseases\n\nAction 8. Maintenance \\[and regular review\\] of a quarantine barrier between Christmas Island and all other lands which minimises the risks of new avian diseases establishing on Christmas Island.\n\nAs part of the regular review process of the Christmas Island Quarantine Service, the effectiveness of the quarantine barrier preventing the man-made introduction of avian diseases onto Christmas Island should be assessed. Review of the Christmas Island Quarantine Service is undertaken by the Australian Quarantine Inspection Service, with input from PAN and in consultation with DOTARS. There is no additional cost for this action.\n\nSpecific objective 4. Increase community involvement and awareness\n\nPerformance criteria Demonstrated increase in community awareness and support for habitat protection\n\nAction 9. A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks.\n\nAs part of the Christmas Island Education and Interpretation Plan, which has a key objective of increasing knowledge of and local support for habitat and species protection, develop a multi-lingual pamphlet on Christmas Island Goshawk stressing their uniqueness and how they are part of Christmas Island ecosystems. This pamphlet would encourage the community to report sightings of colour-banded goshawks. It should also discuss managing problem goshawks around poultry, emphasising that goshawks are protected, and offering solutions to dealing with problem birds.\n\nThis should be further supported by a conservation related curriculum unit for both primary and secondary school students which raises awareness of the Christmas Island habitats and species generally.\n\n  \n\nSpecific objective 5\\. Implement the Recovery Plan through a Recovery Team\n\nPerformance criteria Demonstrated successful operation of the Recovery Team over five years\n\nActions 10. Establish a recovery team which meets regularly\n\n11\\. Carry out a major review of the recovery plan\n\nA recovery team shall be formed to implement this recovery plan. The team should comprise representatives of PANCI, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate eg. other affected Christmas Island interests, and other specialists where necessary. Progress will be monitored and evaluated yearly by members of the recovery team through an annual review. The monitoring process will include: compiling information from resighting of banded birds, success of monitored nests, and assessing progress of all actions against the criteria and objectives of the recovery plan (data comparisons performed by participating biologist). A major review of this performance must be conducted at the end of the first five years of implementation, by an independent reviewer.","sortOrder":17},{"sectionNumber":"Birds on Islands","sectionType":"division","heading":"Birds on Islands","content":"### Birds on Islands\n\nIsland birds are particularly vulnerable to extinction and a very high number of island birds relative to continental species have become extinct in the past century (Diamond 1985, King 1985). This is because (i) island birds tend to have smaller population sizes than continental species and small populations are more vulnerable to extinction (Lacy 1987); (ii) island birds have often evolved in the absence of many predators, diseases and competitors and they can be catastrophically affected by the introduction of one of these (King 1985); (iii) there are often no opportunities for dispersal and thus recolonisation on islands, and this factor also predisposes a population to extinction (Lacy 1987); and (iv) there is often less opportunity to make reserves on islands because land area is limited (Diamond 1985). Predators in general are often naturally rare because their prey, other animals, are less abundant than other food resources. Thus island raptors are a particularly vulnerable group of island birds.","sortOrder":18},{"sectionNumber":"Crazy Ants","sectionType":"division","heading":"Crazy Ants","content":"### Crazy Ants\n\nThe exotic invasive yellow crazy ant (Anoplolepis gracilipes) arrived on Christmas Island more than 70 years ago, and is now widespread throughout rainforest (Orchard et al. 2002). The ants can form multi-queened 'super-colonies', in which they occur at very high densities. This has apparently been a relatively recent phenomenon; with the first supercolony discovered in 1989, with further dramatic increases probably beginning around the mid-1990s.\n\nAt supercolony densities, the ant is having a devastating impact on the island's ecosystems. Red crabs, robber crabs, blue crabs and most other ground dwelling animals such as reptiles, have been and are continuing to be severely impacted, sometimes to the point of local extinction in heavily infested areas. By eliminating local populations of red crabs, the ants are also having a marked effect on forest composition and structure, and litter dynamics in infested areas. The feeding activities of the ants and their mutualistic scale insects can stress large trees to the point of death, and appear to be causing canopy dieback in areas of supercolony formation (Orchard et al. 2002). In addition groundings of birds in crazy ant supercolonies would most probably result in the death of the bird. Although research has shown that crazy ants have little impact on canopy insects, it is possible that the ants may reduce insect populations. In the recent Risk Watch List compiled for the Christmas Island National Park and Conservancy, the crazy ant invasion was rated as an Extreme Risk to biodiversity and conservation values, with catastrophic consequences of failure to implement effective control measures.\n\nPAN field workers and assistants conducted an island-wide ant survey during May-August, 2001. The survey was designed by CAMBI (Centre for the Analysis and Management of Biological Invasions, Monash University). Of 972 sites surveyed; 741 surveyed points fell in natural forest. Crazy ants were recorded at 46.7% of sites in natural forest. Super-colony densities were recorded at 22.7% of sites in natural forest. Using these survey points as a representative sample of the forest, it was estimated that 2,379 ha of the estimated 10,492 ha of forest on the island was infested with crazy ants at supercolony densities (Orchard et al. 2002). Crazy ants are not evenly distributed throughout natural forest but are more commonly found on terrace forests and less commonly found in deep-soil tall-closed forest, which make up most of the natural forests on the island plateau. Deep-soil tall-closed forest made up 31% of census sites in natural forest but only 6.5% of supercolony records (Orchard et al. 2002).\n\nIn September 2002 an aerial baiting program was undertaken, and all known supercolonies treated with insecticide. Results so far indicate the program was successful in controlling supercolonies over 2500ha of Christmas Island. Crazy ants are still present in low densities, however, further high densities of crazy ants may establish in the terrace forests without warning. PAN staff will continue to monitor any new supercolony formation and treat by hand baiting over the next few years.","sortOrder":19},{"sectionNumber":"Disease","sectionType":"division","heading":"Disease","content":"### Disease\n\nA serious threat to all island birds is the introduction of new diseases. Island birds have often evolved in the absence of diseases common in continental bird faunas and the introduction of such diseases to island birds can be disastrous. An example of this is the introduction of avian malaria to Hawaii, which caused the extinction of almost the entire endemic bird fauna from below 600 m altitude, and was probably the main cause of the total extinction of several bird species (Hay 1986). The range of many surviving species was severely reduced and fragmented which in turn markedly increased their chances of extinction. Avian malaria arrived with the accidental introduction of a new species of mosquito (Hay 1986). Christmas Island has been very vulnerable to the accidental introduction of new disease. In 1994 a quarantine barrier was established between the island and Indonesia and Australia and this has reduced that risk.","sortOrder":20},{"sectionNumber":"Habitat Loss","sectionType":"division","heading":"Habitat Loss","content":"### Habitat Loss\n\nApproximately 25% of the island’s original forests have been cleared and replaced by shrublands of ferns on minefields, Secondary Vegetation on stockpiles, and roads and housing (Environment Australia 1994). This has meant a loss of available forested habitat, although secondary vegetation growing along roadsides and on stockpiles contributes to the feeding habitat for Christmas Island Goshawks at least at some times of the year. Nonetheless, based on the decrease in available habitat, the total Christmas Island Goshawk population has probably decreased by at least 25% since settlement (Stokes 1988).\n\nPrimary forests on Christmas Island have been fragmented in places by clearing and most Primary Rainforest is dissected by roads and old mining 'grid lines' The grid line system, a series of parallel lines 120 m apart and a bulldozer blade in width was installed in the 1970's to explore the island's phosphate resources. After 25 years these are mostly overgrown with dense stands of Secondary Vegetation.\n\nThere are some major development proposals and associated infrastructure in particular the Immigration Centre, Satellite launching station, airport upgrade, and expansion of phosphate mining that have the potential to reduce the area of goshawk habitat. One development condition for the satellite launching facility is to develop and implement an environment management plan, which is currently underway. The EMP will address threatened fauna and their habitat on the sites. The immigration centre is exempt from assessment under the EPBC Act, however, the DEH is being consulted on environmental aspects of the development, including monitoring of habitat for potential impacts. The airport upgrade and phosphate mining expansion proposals are undergoing assessment for environmental impact under the EPBC Act. Survey and monitoring of the Christmas Island Goshawk will aid in determining potential impacts of these and future developments on goshawk habitat.\n\nThe Christmas Island Rainforest Rehabilitation Program (CIRRP) is a long term program to revegetate old minefields. The CIRRP is funded by a conservation levy that is collected from Phosphate Resources Limited at an indexed rate per tonne. Rehabilitation priorities are primarily determined to protect Abbott’s Booby sites, although other threatened species are also considered in determining the priority of sites. A significant additional result of CIRRP will be the reestablishment of habitat that supports other species such as the goshawk.","sortOrder":21},{"sectionNumber":"Natural Catastrophes","sectionType":"division","heading":"Natural Catastrophes","content":"###   \nNatural Catastrophes\n\nAlthough goshawks, as generalist raptors, are likely to cope well with habitat modifying natural catastrophes, the effect of events such as cyclones are more severe on small populations. There is a possibility of an increase in severe storms and cyclones occurring as a result of Greenhouse gas-induced climate change. It would be useful to understand the effects of cyclones and severe storms on goshawk populations to help predict the impact of any increases.","sortOrder":22},{"sectionNumber":"Small population size","sectionType":"division","heading":"Small population size","content":"### Small population size\n\nThe risk of extinction varies inversely with population size (Lacy 1987) due primarily to demographic and environmental variability the effects of which can become marked in small populations. Demographic variability is the individual variation in reproductive success, which is masked when population size is large. Very small populations (< 30 individuals) can easily become extinct as a result of random demographic variation between individuals (Caughley & Sinclair 1994).\n\nThe reproductive success of a population is also likely to vary with fluctuations in the environment. The degree of this fluctuation is determined by how much seasonal and annual variation in habitat quality there is. The influence of this on the risk of extinction increases markedly with decreasing population size. For example, a cyclone, an extreme example of environmental variation, causing the death of half the population of 10000 animals has an insignificant affect on the probability of extinction of that population. However, if half a population of 200 animals dies, the risk of extinction of that population increases significantly (Caughley & Sinclair 1994).","sortOrder":23},{"sectionNumber":"Inbreeding Depression","sectionType":"division","heading":"Inbreeding Depression","content":"### Inbreeding Depression\n\nThe risks of inbreeding depression increases with decreasing population size due to the increase in frequency of homozygous alleles. Populations at risk of inbreeding depression are probably a couple of dozen individuals or smaller in size which have been at that size for several generations, as a population contraction must last for several generations to lead to a significant loss of heterozygosity (Caughley & Sinclair 1994). Any action on the island which might permanently, rather than temporarily, reduce the number of mature individual Christmas Island Goshawks (such as permanent forest clearance rather than temporary storm damage) would raise the possibility of inbreeding depression as a concern in their conservation.","sortOrder":24},{"sectionNumber":"Other Potential Threats","sectionType":"division","heading":"Other Potential Threats","content":"### Other Potential Threats\n\nKilling of Christmas Island Goshawks, especially by poultry owners, has been considered to be a significant threat to goshawks in the past (Stokes 1988). Given the small population size, any killing of either juveniles or adult Christmas Island Goshawks would be considered a significant threat to this taxon. Another possible threat is posed by cats, especially near settled areas.\n\nRoad killed Christmas Island Goshawks have been reported. A substantial increase in vehicular traffic will be associated with the proposed satellite launching station and the new immigration, reception and processing centre (IRPC). This will likely increase the number of roadkills in high traffic areas. Thus it is possible that collisions with cars will become an issue for the conservation of populations in high traffic areas.\n\nWeeds, especially newly introduced invasive species, could potentially impact on Christmas Island Goshawk nest sites, for example by forming vine towers over nesting trees.","sortOrder":25},{"sectionNumber":"Recovery Objectives and Timelines","sectionType":"part","heading":"Recovery Objectives and Timelines","content":"## Executive Summary\n\nThe Christmas Island Goshawk is currently listed as Endangered under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). It is currently considered a subspecies of Accipiter fasciatus, but its taxonomy needs resolution. It is considered to be the rarest endemic bird on Christmas Island, where it occurs in all habitats from primary and marginal rainforests to suitable areas of secondary regrowth vegetation. The total population size is thought to be very small, perhaps as few as 100 adults, and is probably limited by the availability of suitable rainforest habitat. Crazy Ants pose an unknown but potentially critical threat to the survival of this bird. The recovery plan objective is to downgrade the Christmas Island Goshawk from Endangered to Conservation Dependent, primarily through successful implementation of the Invasive Ants on Christmas Island Action Plan and protection of habitat critical to the survival of the species from clearance. An assessment of goshawk population dynamics is the most essential requirement of this recovery plan, and community awareness and participation in the conservation of this endemic raptor are also important actions.\n\n##   \nBackground Information\n\n### Conservation Status\n\nThe Christmas Island Goshawk (Accipiter fasciatus natalis) is currently listed as Endangered (EPBC Act), however, according to Garnett & Crowley (2000) it now fulfils the criteria for Critically Endangered.\n\n### Taxonomic Status\n\nSince the Christmas Island Goshawk was described by Lister in 1888 as a full species Accipiter natalis, debate has continued over whether it is more closely allied to the Brown Goshawk Accipiter fasciatus, or to the Grey Goshawk A. novaehollandiae. Lister (1888) thought the Christmas Island Goshawk was allied to the Grey Goshawk from the Maluku A. n. griseogularis. Chasen (1933) disagreed, thinking it a member of the A. fasciatus group. The Christmas Island Goshawk may have been isolated from populations of either of these species for a long period of time, and Carter (1994) raised the possibility of resurrecting the Christmas Island subspecies natalis to a full species.\n\n### International Obligations\n\nThe Christmas Island Goshawk is not listed under international agreements.\n\n### Distribution and location\n\nThe Christmas Island Goshawk is restricted to Christmas Island, a volcanic island in the Indian Ocean (10o0'S; 105o40' E), approximately 1400km northwest of Australia. The island sits on the northernmost edge of the Australasian continental plate, and immediately north the ocean floor drops into the Java Trench and depths of up to 6000m. The nearest land is Java in the Republic of Indonesia, 360km to the north. Christmas Island is truly oceanic and all its native biota has colonised by sea or by air (Gray 1981). The island is 135km2 and 75% is covered with original vegetation (Environment Australia 2002).\n\nGibson-Hill (1947) reported that Christmas Island Goshawks seemed to prefer areas of ‘slightly thinner growth on the edge of thick jungle or the borders of clearings’. During 1994 and 1995 goshawks were observed in all major habitats on the island (Hill unpubl. data,). Goshawks were regularly seen hunting in regrowth vegetation along roadsides. This may be merely because they were much more visible in this habitat compared to intact forests. Parks Australia North (PAN) data from a systematic survey over all habitat of the island (over 1000 survey points) suggest that more goshawks are located within rainforest (95% of sightings) than in cleared areas. The Christmas Island Goshawk is the least frequently encountered diurnal forest bird and might best be described as ‘widespread and uncommon’.\n\nThere are no data on adult or juvenile movements. Adult Christmas Island Goshawks are probably resident on their territories year-round. Grey Goshawks defend a territory around the nest site which appears to be significantly smaller than the home range in which they feed (Burton 1991).\n\nThere are no detailed population estimates for Christmas Island Goshawks. New Caledonian Brown Goshawks, a similar-sized closely related bird, occurred at a density of 0.89 - 0.96 adults/km2 in rainforest (Thiollay 1993). There is approximately 101 km2 of primary forest on Christmas Island which suggests that it is unlikely that there are many more than 50 pairs or 100 mature Christmas Island Goshawks on Christmas Island.\n\n###   \n\n### ![](image.002.png)\n\n###   \nBiology\n\nThe Christmas Island Goshawk has been recorded eating a wide range of vertebrates and invertebrates including birds, small mammals and insects. Prey includes Christmas Island White-eye Zosterops natalis, Island Thrush Turdus poliocephalus, Emerald Dove Chalcophaps indica, Java Sparrow Lonchura oryzivora, domestic chickens Gallus gallus, (Lister 1888, Gibson-Hill 1947, Hill, unpubl. data); as well as nestling Red-footed Boobies Sula sula (Hill, unpubl. data) and an adult Golden Bosun Phaethon rubricauda westralis (J. Tranter pers. comm.). It has also been recorded catching rats Rattus rattus (Gibson-Hill 1947) and ‘lizards’ (Lister 1888). It is frequently seen catching grasshoppers (probably Valanga sp.) which are common in secondary vegetation around the island (Gibson-Hill 1947, Hill unpubl. data). Gibson-Hill (1947) also recorded it feeding on the large centipede Scolopendra morsitans, which is widespread on the island, and on beetles (Coleoptera) and mantids (Mantidae).\n\nThe Christmas Island Goshawk captures its prey off the ground or in the air by seizing it in flight with its long, strong legs. Like other goshawks, it will hunt from a perch where it may sit for some time waiting for suitable prey. Christmas Island Goshawks will chase small flying birds through the forest as well as pounce on birds foraging on the ground (Hill unpubl. data). Hill observed goshawks ‘short-stay perch-hunting’ (after Newton 1986) flying along a track, perching for a few minutes then flying 50-100m further and landing again, as well as pursuing insects on the ground and snatching insects off foliage. There is likely to be some sex differences in hunting and in diet, as is found in both Brown and Grey Goshawks (Marchant & Higgins 1993). Given the large sexual dimorphism (similar to Grey Goshawk, Hill unpubl. data) and the opportunity for niche expansion on the island in the absence of other diurnal raptors, this difference may be quite pronounced.\n\nGibson-Hill (1947) described three Christmas Island Goshawk nests all of which were in near-horizontal forks of forest trees, one 10m high and two 15-20m in tall forest trees. Hill (unpubl. data) observed four nests: two in Syzygium nervosum in Primary rainforest; and two in Terminalia catappa in Marginal rainforest. All nests were 25-35m above the ground, in the fork of two or more branches, three towards the edge of the crown, and one in the middle of the tree. One nest (probably not used) was 30-40cm in diameter, two others were 50-80cm in diameter, and the fourth was under construction. Nest size and location within the tree share more similarities with Grey than Brown Goshawks (J. Young pers. comm.; see Burton et al. 1994).\n\n### Habitat\n\nApproximately 75% of Christmas Island is still covered with natural vegetation and 84% of this (or 63% of the island) is protected within National Park (Figure 1). The remaining Primary rainforest is currently protected from clearance by a Federal Government moratorium, in conjunction with the requirements for approval under the EPBC Act and Environment Protection and Biodiversity Conservation Regulations 2000 (EPBC Regulations). Marginal rainforest and Second-growth rainforest or Secondary Vegetation outside the National Park may only be cleared if approval is given under the EPBC Act and Regulations.\n\nSchedule 12 of the EPBC Regulations use Du Puy’s classification to describe the Christmas Island forests. Primary Rainforest is restricted to the central plateau and is evergreen closed forest 30-40m high with emergent trees up to 45m. Common canopy trees include Syzygium nervosum, Planchonella nitida, and Hernandia ovigera. The canopy formed by these trees is irregular, especially on the western side of the island where it is sheltered from strong southeast trade winds which blow throughout the dry season. The understorey comprises dense thickets of Pandanus elatus up to 4m tall, and more open areas which commonly include the plants Arenga listeri, Leea angulata, Ochrosia ackeringae, Pisonia umbellifera and Aidia aff. racemosa (Environment Australia 1994).\n\nMarginal Rainforests grow on generally shallower soils on the coastal terraces and scree slopes surrounding the island. Many Marginal Rainforest trees lose their leaves in the dry season. Marginal rainforest is generally lower than Primary Rainforest, often less than 30m, and may include many vines. The canopy height and shape are variable, determined primarily by the degree of exposure to the southeast trade winds. Marginal rainforest facing south and east has a smooth, wind-pruned canopy and tends to increase in height with increasing distance from the sea cliff. The inland cliffs and scree slopes may have no vegetation or carry a closed forest, depending on the degree of the slope. Trees such as Ficus microcarpa and Dendrocnide sinuata are common there.\n\nOld stockpiles and cleared areas that have not been mined may support low second-growth forest of colonising trees such as Macaranga tanarius and Claoxylon indicum and an introduced tree Leucaena leucocephala generally less than 10m high. Previously mined areas tend to have very little remaining soil and on them grow dense herblands of a fern Nephrolepis multiflora to 2m high along with introduced scramblers and occasional low trees.\n\n### Habitat Critical to Survival\n\nIdentification of habitat critical to survival of the Christmas Island Goshawk is difficult due to the lack of information on specific habitat requirements. Although Christmas Island Goshawks may be generalists which forage in most habitats, it is highly likely that they require rainforest to breed as these habitat contain suitable trees. Applying the precautionary principle and using the criteria provided by the EPBC Act, habitat critical to survival of the Christmas Island Goshawks is defined as all Primary Rainforest, Marginal Rainforest (Du Puy 1993) and possibly second-growth forest suitable for nesting. Primary Rainforest and Marginal Rainforest has been identified and mapped (Figure 1). Dexter (2000) recommended mapping the ecological quality of second-growth forests outside the national park. As this has not occurred, and there is insufficient information on the habitat requirements of the Christmas Island Goshawks, suitable second-growth forest cannot be identified at this time.\n\n### Affected Interests\n\nCommonwealth Parks Australia North (PAN), Shire of Christmas Island, Christmas Island Phosphates, Union of Christmas Island Workers, the Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services (DOTARS), and the Department of Immigration, Multicultural and Indigenous Affairs.\n\n### Role and interests of indigenous people\n\nNot applicable. Christmas Island does not have an indigenous population.\n\n### Social and economic impacts\n\nThe actions in this plan may have positive and negative social and economic impacts. Positive social impacts will arise from community education actions that will increase Christmas Islanders knowledge and interest in their own environment. The rainforest rehabilitation program provides on-island jobs, as will the goshawk monitoring program. Christmas Island endemic birds attract specialist bird watching groups each year which is high value, low-impact tourism.\n\nNegative social and economic impacts arising from implementation of the plan could include greater restrictions due to review of the quarantine barrier, and potential economic effects of preventing illegal destruction of Christmas Island goshawks by residents. The EPBC Act already provides a regulatory framework for the protection of rainforest on Christmas Island, and one element of this is assessing potential impacts of proposed developments on listed threatened species. These provisions have the potential to impact on economic activity, for example by adding additional obligations for industry and other development on the island in order to minimise impacts on listed species. This arises from the listing of the species under the EPBC Act invoking a range of protective provisions and offences where a population is to be affected. The magnitude of this potential impact is unknown, as it will vary with the location, size and extent of the activity, proposed or current. However, the plan will aid in determining what could be significant impacts on the Christmas Island Goshawk (through defining habitat critical to survival and developing a management plan outside the national park).\n\n### Biodiversity Benefits:\n\nProtection of the habitat of the Christmas Island goshawk provides protection for numerous other listed species (Table 1). Community education targeted at this species will promote awareness of all the endemic forest birds and their conservation needs.\n\nTable 1: Native species listed under the EPBC Act that will benefit from recovery actions listed in this plan (after Environment Australia 2002).\n\n| Taxon                                                     | National Status |\n| --------------------------------------------------------- | --------------- |\n| Tectaria devexa var. minor                                | E               |\n| Carmona retusa                                            | V               |\n| Christmas Island Pipistrelle (e) Pipistrellus murrayi     | E               |\n| Christmas Island Shrew 1 (e) Crocidura attenuata trichura | E               |\n| Christmas Island Blind Snake (e) Ramphotyphlops exocoeti  | V               |\n| Christmas Island Gecko Lepidodactylus listeri             | V               |\n| Christmas Island Hawk-owl (e) Ninox natalis               | VJ              |\n| Abbott’s Booby (e) Papasula abbotti                       | E M S J         |\n| Red-footed Booby Sula sula rubripes                       | M S C J         |\n| Christmas Island Frigatebird (e) Fregata andrewsi         | V M S C J       |\n| Great Frigatebird Fregata minor minor                     | M S C J         |\n\nNotes:\n\n(e) species and subspecies endemic to the island E Listed under the EPBC Act as Endangered\n\nV Listed under the EPBC Act as Vulnerable M Listed under the EPBC Act as a Migratory species\n\nMF Within a family listed under the EPBC Act as Migratory S Listed Marine species under the EPBC Act\n\nC Listed under China-Australia Migratory Bird Agreement\n\nJ Listed under Japan-Australia Migratory Bird Agreement\n\n### Relationship to other plans\n\nThe Christmas Island National Park Plan of Management is the strategic nature conservation document for the island. This recovery plan makes numerous recommendations in common with other recovery plans for Christmas Island threatened taxa. Opportunities for sharing resources and points shared in common with other recovery plans are identified in the implementation section of this plan.\n\n### Recovery Team\n\nA Recovery Team for the Christmas Island Goshawk has not yet been formed. The team should comprise representatives of PAN, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate.\n\n##   \nThreats to the species\n\n### Birds on Islands\n\nIsland birds are particularly vulnerable to extinction and a very high number of island birds relative to continental species have become extinct in the past century (Diamond 1985, King 1985). This is because (i) island birds tend to have smaller population sizes than continental species and small populations are more vulnerable to extinction (Lacy 1987); (ii) island birds have often evolved in the absence of many predators, diseases and competitors and they can be catastrophically affected by the introduction of one of these (King 1985); (iii) there are often no opportunities for dispersal and thus recolonisation on islands, and this factor also predisposes a population to extinction (Lacy 1987); and (iv) there is often less opportunity to make reserves on islands because land area is limited (Diamond 1985). Predators in general are often naturally rare because their prey, other animals, are less abundant than other food resources. Thus island raptors are a particularly vulnerable group of island birds.\n\n### Crazy Ants\n\nThe exotic invasive yellow crazy ant (Anoplolepis gracilipes) arrived on Christmas Island more than 70 years ago, and is now widespread throughout rainforest (Orchard et al. 2002). The ants can form multi-queened 'super-colonies', in which they occur at very high densities. This has apparently been a relatively recent phenomenon; with the first supercolony discovered in 1989, with further dramatic increases probably beginning around the mid-1990s.\n\nAt supercolony densities, the ant is having a devastating impact on the island's ecosystems. Red crabs, robber crabs, blue crabs and most other ground dwelling animals such as reptiles, have been and are continuing to be severely impacted, sometimes to the point of local extinction in heavily infested areas. By eliminating local populations of red crabs, the ants are also having a marked effect on forest composition and structure, and litter dynamics in infested areas. The feeding activities of the ants and their mutualistic scale insects can stress large trees to the point of death, and appear to be causing canopy dieback in areas of supercolony formation (Orchard et al. 2002). In addition groundings of birds in crazy ant supercolonies would most probably result in the death of the bird. Although research has shown that crazy ants have little impact on canopy insects, it is possible that the ants may reduce insect populations. In the recent Risk Watch List compiled for the Christmas Island National Park and Conservancy, the crazy ant invasion was rated as an Extreme Risk to biodiversity and conservation values, with catastrophic consequences of failure to implement effective control measures.\n\nPAN field workers and assistants conducted an island-wide ant survey during May-August, 2001. The survey was designed by CAMBI (Centre for the Analysis and Management of Biological Invasions, Monash University). Of 972 sites surveyed; 741 surveyed points fell in natural forest. Crazy ants were recorded at 46.7% of sites in natural forest. Super-colony densities were recorded at 22.7% of sites in natural forest. Using these survey points as a representative sample of the forest, it was estimated that 2,379 ha of the estimated 10,492 ha of forest on the island was infested with crazy ants at supercolony densities (Orchard et al. 2002). Crazy ants are not evenly distributed throughout natural forest but are more commonly found on terrace forests and less commonly found in deep-soil tall-closed forest, which make up most of the natural forests on the island plateau. Deep-soil tall-closed forest made up 31% of census sites in natural forest but only 6.5% of supercolony records (Orchard et al. 2002).\n\nIn September 2002 an aerial baiting program was undertaken, and all known supercolonies treated with insecticide. Results so far indicate the program was successful in controlling supercolonies over 2500ha of Christmas Island. Crazy ants are still present in low densities, however, further high densities of crazy ants may establish in the terrace forests without warning. PAN staff will continue to monitor any new supercolony formation and treat by hand baiting over the next few years.\n\n### Disease\n\nA serious threat to all island birds is the introduction of new diseases. Island birds have often evolved in the absence of diseases common in continental bird faunas and the introduction of such diseases to island birds can be disastrous. An example of this is the introduction of avian malaria to Hawaii, which caused the extinction of almost the entire endemic bird fauna from below 600 m altitude, and was probably the main cause of the total extinction of several bird species (Hay 1986). The range of many surviving species was severely reduced and fragmented which in turn markedly increased their chances of extinction. Avian malaria arrived with the accidental introduction of a new species of mosquito (Hay 1986). Christmas Island has been very vulnerable to the accidental introduction of new disease. In 1994 a quarantine barrier was established between the island and Indonesia and Australia and this has reduced that risk.\n\n### Habitat Loss\n\nApproximately 25% of the island’s original forests have been cleared and replaced by shrublands of ferns on minefields, Secondary Vegetation on stockpiles, and roads and housing (Environment Australia 1994). This has meant a loss of available forested habitat, although secondary vegetation growing along roadsides and on stockpiles contributes to the feeding habitat for Christmas Island Goshawks at least at some times of the year. Nonetheless, based on the decrease in available habitat, the total Christmas Island Goshawk population has probably decreased by at least 25% since settlement (Stokes 1988).\n\nPrimary forests on Christmas Island have been fragmented in places by clearing and most Primary Rainforest is dissected by roads and old mining 'grid lines' The grid line system, a series of parallel lines 120 m apart and a bulldozer blade in width was installed in the 1970's to explore the island's phosphate resources. After 25 years these are mostly overgrown with dense stands of Secondary Vegetation.\n\nThere are some major development proposals and associated infrastructure in particular the Immigration Centre, Satellite launching station, airport upgrade, and expansion of phosphate mining that have the potential to reduce the area of goshawk habitat. One development condition for the satellite launching facility is to develop and implement an environment management plan, which is currently underway. The EMP will address threatened fauna and their habitat on the sites. The immigration centre is exempt from assessment under the EPBC Act, however, the DEH is being consulted on environmental aspects of the development, including monitoring of habitat for potential impacts. The airport upgrade and phosphate mining expansion proposals are undergoing assessment for environmental impact under the EPBC Act. Survey and monitoring of the Christmas Island Goshawk will aid in determining potential impacts of these and future developments on goshawk habitat.\n\nThe Christmas Island Rainforest Rehabilitation Program (CIRRP) is a long term program to revegetate old minefields. The CIRRP is funded by a conservation levy that is collected from Phosphate Resources Limited at an indexed rate per tonne. Rehabilitation priorities are primarily determined to protect Abbott’s Booby sites, although other threatened species are also considered in determining the priority of sites. A significant additional result of CIRRP will be the reestablishment of habitat that supports other species such as the goshawk.\n\n###   \nNatural Catastrophes\n\nAlthough goshawks, as generalist raptors, are likely to cope well with habitat modifying natural catastrophes, the effect of events such as cyclones are more severe on small populations. There is a possibility of an increase in severe storms and cyclones occurring as a result of Greenhouse gas-induced climate change. It would be useful to understand the effects of cyclones and severe storms on goshawk populations to help predict the impact of any increases.\n\n### Small population size\n\nThe risk of extinction varies inversely with population size (Lacy 1987) due primarily to demographic and environmental variability the effects of which can become marked in small populations. Demographic variability is the individual variation in reproductive success, which is masked when population size is large. Very small populations (< 30 individuals) can easily become extinct as a result of random demographic variation between individuals (Caughley & Sinclair 1994).\n\nThe reproductive success of a population is also likely to vary with fluctuations in the environment. The degree of this fluctuation is determined by how much seasonal and annual variation in habitat quality there is. The influence of this on the risk of extinction increases markedly with decreasing population size. For example, a cyclone, an extreme example of environmental variation, causing the death of half the population of 10000 animals has an insignificant affect on the probability of extinction of that population. However, if half a population of 200 animals dies, the risk of extinction of that population increases significantly (Caughley & Sinclair 1994).\n\n### Inbreeding Depression\n\nThe risks of inbreeding depression increases with decreasing population size due to the increase in frequency of homozygous alleles. Populations at risk of inbreeding depression are probably a couple of dozen individuals or smaller in size which have been at that size for several generations, as a population contraction must last for several generations to lead to a significant loss of heterozygosity (Caughley & Sinclair 1994). Any action on the island which might permanently, rather than temporarily, reduce the number of mature individual Christmas Island Goshawks (such as permanent forest clearance rather than temporary storm damage) would raise the possibility of inbreeding depression as a concern in their conservation.\n\n### Other Potential Threats\n\nKilling of Christmas Island Goshawks, especially by poultry owners, has been considered to be a significant threat to goshawks in the past (Stokes 1988). Given the small population size, any killing of either juveniles or adult Christmas Island Goshawks would be considered a significant threat to this taxon. Another possible threat is posed by cats, especially near settled areas.\n\nRoad killed Christmas Island Goshawks have been reported. A substantial increase in vehicular traffic will be associated with the proposed satellite launching station and the new immigration, reception and processing centre (IRPC). This will likely increase the number of roadkills in high traffic areas. Thus it is possible that collisions with cars will become an issue for the conservation of populations in high traffic areas.\n\nWeeds, especially newly introduced invasive species, could potentially impact on Christmas Island Goshawk nest sites, for example by forming vine towers over nesting trees.\n\n##   \nRecovery Objectives and Timelines\n\n### Overall objective:\n\nTo shift the status of the Christmas Island Goshawk from Endangered to Conservation Dependent within 10 years.\n\n### Specific objectives:\n\nTo reduce the risk of extinction and improve the conservation status of the taxon by:\n\n1.  Determine taxonomic status\n2.  Determine and maximise total population size and area of occupancy\n3.  Implement threat abatement strategies\n4.  Increase community involvement and awareness\n5.  Implement the Recovery Plan through a Recovery Team\n\n### Performance Criteria\n\n-   Taxonomic position of Christmas Island Goshawk clarified.\n\n-   Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring.\n-   Density of Christmas Island Goshawks increased as demonstrated by population monitoring\n\n-   Crazy Ants having an insignificant impact on Christmas Island Goshawks.\n-   Protection of all habitat critical outside the national park on Christmas Island.\n-   A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n-   Maintenance of effective quarantine against the introduction of all avian diseases\n-   Demonstrated increase in community awareness and support for habitat protection\n-   Demonstrated successful operation of the Recovery Team over five years.\n\n## Actions\n\nSpecific objective 1. Determine taxonomic status\n\nPerformance Criteria Taxonomic position of Christmas Island Goshawk clarified.\n\nAction 1\\. Investigate the taxonomic status of the Christmas Island Goshawk.\n\nThis project is designed to describe the relationships of the Christmas Island Goshawk to the Brown and Grey Goshawks by comparing DNA sequences of selected mitochondrial genes. Mitochondrial DNA may be collected from feathers allowing specimens already in collections to be used for analysis. The Museum of Victoria has material from Christmas Island Goshawks.\n\n  \n\nSpecific objective 2\\. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring\n\nDensity of Christmas Island Goshawks increased as demonstrated by population monitoring\n\nAction 2\\. Conduct detailed population survey\n\nThe recovery team will obtain estimates of population size, distribution, age structure, fledging success, and survival rates. This is required to assess the status of this taxon. Data on age structure and fledging success will be used to monitor the population, and will, over time, indicate whether or not any significant change in population size is occurring.\n\nTrap and colour band the Christmas Island Goshawk population across the island with the aim of individually marking the majority of the population. A majority of the population will remain individually marked by periodic supplementary trapping after this initial program. This action will provide data on the age-structure of the population, distribution on Christmas Island (including some information on habitat), ranging behaviour and spacing of individuals from resightings during the 18 month study, and thus some data on population size.\n\nThe members of the Australasian Raptor Association (ARA), part of Birds Australia, include many of the Australian professional and amateur experts on birds of prey. They may be able to provide unpaid assistance in trapping and marking birds, and in training of local staff in trapping techniques. The cost of their involvement would be airfares from Australia and accommodation and transport on Christmas Island. With the involvement of the ARA this project lends itself to being done by a post-graduate student. If ARA involvement was not possible dedicated researchers would re required, and the cost would be substantially higher.\n\nAction 3\\. On-going monitoring of the population\n\nIt is essential to formally and regularly monitor the size of the population, and breeding success. The recovery team will monitor the total population size at regular intervals (initially annually) such that, if a decline takes place, it provides sufficient warning to managers to allow remedial management and research to take place. The monitoring program shall be based on observations of the individually marked population of Christmas Island Goshawks. PAN Christmas Island staff, tour operators, and the community in general will be encouraged to record all observations of goshawks, and to report their locations accurately. This will provide an ongoing data set on the proportion of resightings of marked birds, and on the habitats in which birds are seen. Once every five years, volunteers from the ARA will visit the island and trap and mark additional birds to maintain a majority of marked birds in the population.\n\nIn addition a sample of goshawk nests should be monitored annually. Monitoring of ten nests over a number of years will provide a considerable amount of information on breeding success, diet, and habitat requirements, and would be relatively inexpensive. Breeding success data will allow evaluation of the species’ status and population viability. Knowledge of diet is essential to evaluate the effects of habitat change, and predict the likely impact of any proposed changes in land management on the island. Information on diet and breeding requirements, combined with data from the population survey on distribution, ranging and spacing, will enable refinement of the definition of habitat critical to the survival of Christmas Island Goshawks, and determination of at what stage/s they utilise rehabilitated areas.\n\nTen nests will be monitored for signs of breeding, and once breeding has commenced monitored regularly (fortnightly) while the breeding attempt continues. In addition roost sites of parents should be located and pellets collected for dietary analysis. The pellets should be relatively straightforward to analyse due to the limited number of potential vertebrate prey. Insects should be identified as accurately as possible. Monitoring of nesting and collection and analysis of pellets could be part of the general duties of the resource management staff of PANCI.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Protection of all habitat critical outside the national park.\n\nAction 4\\. Develop and implement wildlife management plan outside the national park\n\nAction 5\\. Ensure protection of habitat critical outside the national park.\n\nMost forested land outside the National Park is publicly owned, by the Commonwealth. PAN staff are routinely involved with the other stakeholders in negotiations over development proposals. This is ongoing. These proposals can occur unexpectedly and as there is no overall wildlife management plan outside the park, are assessed on an ad hoc basic. While consultation with other agencies and proponents on projects will be essential and ongoing (under the EPBC Act referrals process), achievement of long-term protection of threatened species outside the park requires a more comprehensive and considered approach through the development of a wildlife management program. The EPBC Regulations include that actions under an approved management plan are not considered offences in relation to protected species, and that a management plan must not, or not be likely to, adversely affect the conservation status of a protected species or a population of a protected species. The Goshawk meets the definition of a protected species.\n\nThe plan should allow for the input of relevant landholders and decision makers, and provide more strategic and comprehensive protection of flora and fauna, while facilitating greater certainty for environmentally sensitive developments. The plan should also allow for adaptive management as better information becomes available on the Christmas Island Hawk-Owl and other threatened species, in particular greater definition of the areas of habitat which are critical to survival.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n\nAction 6\\. Continue effective and long-term rainforest rehabilitation program managed by DOTARS & supported by PAN and the mining company.\n\nThe Christmas Island Rainforest Rehabilitation Program has been occurring over the last three decades. Currently, the main aim is to revegetate mined areas adjacent to Abbott's Booby nesting habitat to reduce wind turbulence caused by the clearings and implicated in reduced nesting success of the booby (Reville et al. 1990). There are approx. 70 clearings covering 3200ha or 24% of the island area (Carew-Reid 1987). It is not known how Christmas Island Goshawks use these areas and at what stage of rehabilitation this habitat become important for foraging and for nesting – this will be investigated as part of the monitoring program.\n\nIn 2002, the Government decided to redirect the payment of the conservation levy to the portfolio agency with responsibility for the mine lease (DOTARS). A MOU between PAN and DOTARS was signed in February 2004 to allow PAN to continue the CIRRP within Christmas Island National Park. Rehabilitation is critical in terms of increasing rainforest habitat to further protect a range of species including the goshawk. The CIRRP is funded through a conservation levy that forms part of Christmas Island Phosphate’s lease. Effective management of the lease by DOTARS will ensure optimal outcomes for rainforest rehabilitation.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Crazy Ants have a negligible impact on Christmas Island Goshawk\n\nAction 7\\. Implement the Invasive Ants on Christmas Island Action Plan.\n\nPAN is currently implementing an Invasive Ants Action Plan, which aims to control and minimise the impact of Crazy Ants across the island. This program will require ongoing monitoring of Crazy Ant numbers and control of populations as detected. The recovery team must ensure that the control program does not have any negative impacts on Christmas Island Goshawks.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Maintenance of effective quarantine against the introduction of all avian diseases\n\nAction 8. Maintenance \\[and regular review\\] of a quarantine barrier between Christmas Island and all other lands which minimises the risks of new avian diseases establishing on Christmas Island.\n\nAs part of the regular review process of the Christmas Island Quarantine Service, the effectiveness of the quarantine barrier preventing the man-made introduction of avian diseases onto Christmas Island should be assessed. Review of the Christmas Island Quarantine Service is undertaken by the Australian Quarantine Inspection Service, with input from PAN and in consultation with DOTARS. There is no additional cost for this action.\n\nSpecific objective 4. Increase community involvement and awareness\n\nPerformance criteria Demonstrated increase in community awareness and support for habitat protection\n\nAction 9. A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks.\n\nAs part of the Christmas Island Education and Interpretation Plan, which has a key objective of increasing knowledge of and local support for habitat and species protection, develop a multi-lingual pamphlet on Christmas Island Goshawk stressing their uniqueness and how they are part of Christmas Island ecosystems. This pamphlet would encourage the community to report sightings of colour-banded goshawks. It should also discuss managing problem goshawks around poultry, emphasising that goshawks are protected, and offering solutions to dealing with problem birds.\n\nThis should be further supported by a conservation related curriculum unit for both primary and secondary school students which raises awareness of the Christmas Island habitats and species generally.\n\n  \n\nSpecific objective 5\\. Implement the Recovery Plan through a Recovery Team\n\nPerformance criteria Demonstrated successful operation of the Recovery Team over five years\n\nActions 10. Establish a recovery team which meets regularly\n\n11\\. Carry out a major review of the recovery plan\n\nA recovery team shall be formed to implement this recovery plan. The team should comprise representatives of PANCI, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate eg. other affected Christmas Island interests, and other specialists where necessary. Progress will be monitored and evaluated yearly by members of the recovery team through an annual review. The monitoring process will include: compiling information from resighting of banded birds, success of monitored nests, and assessing progress of all actions against the criteria and objectives of the recovery plan (data comparisons performed by participating biologist). A major review of this performance must be conducted at the end of the first five years of implementation, by an independent reviewer.","sortOrder":26},{"sectionNumber":"Overall objective:","sectionType":"division","heading":"Overall objective:","content":"### Overall objective:\n\nTo shift the status of the Christmas Island Goshawk from Endangered to Conservation Dependent within 10 years.","sortOrder":27},{"sectionNumber":"Specific objectives:","sectionType":"division","heading":"Specific objectives:","content":"### Specific objectives:\n\nTo reduce the risk of extinction and improve the conservation status of the taxon by:\n\n1.  Determine taxonomic status\n2.  Determine and maximise total population size and area of occupancy\n3.  Implement threat abatement strategies\n4.  Increase community involvement and awareness\n5.  Implement the Recovery Plan through a Recovery Team","sortOrder":28},{"sectionNumber":"Performance Criteria","sectionType":"division","heading":"Performance Criteria","content":"### Performance Criteria\n\n-   Taxonomic position of Christmas Island Goshawk clarified.\n\n-   Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring.\n-   Density of Christmas Island Goshawks increased as demonstrated by population monitoring\n\n-   Crazy Ants having an insignificant impact on Christmas Island Goshawks.\n-   Protection of all habitat critical outside the national park on Christmas Island.\n-   A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n-   Maintenance of effective quarantine against the introduction of all avian diseases\n-   Demonstrated increase in community awareness and support for habitat protection\n-   Demonstrated successful operation of the Recovery Team over five years.","sortOrder":29},{"sectionNumber":"Actions","sectionType":"part","heading":"Actions","content":"## Actions\n\nSpecific objective 1. Determine taxonomic status\n\nPerformance Criteria Taxonomic position of Christmas Island Goshawk clarified.\n\nAction 1\\. Investigate the taxonomic status of the Christmas Island Goshawk.\n\nThis project is designed to describe the relationships of the Christmas Island Goshawk to the Brown and Grey Goshawks by comparing DNA sequences of selected mitochondrial genes. Mitochondrial DNA may be collected from feathers allowing specimens already in collections to be used for analysis. The Museum of Victoria has material from Christmas Island Goshawks.\n\n  \n\nSpecific objective 2\\. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Distribution of Christmas Island Goshawks widespread on the island in all suitable habitats as demonstrated by population monitoring\n\nDensity of Christmas Island Goshawks increased as demonstrated by population monitoring\n\nAction 2\\. Conduct detailed population survey\n\nThe recovery team will obtain estimates of population size, distribution, age structure, fledging success, and survival rates. This is required to assess the status of this taxon. Data on age structure and fledging success will be used to monitor the population, and will, over time, indicate whether or not any significant change in population size is occurring.\n\nTrap and colour band the Christmas Island Goshawk population across the island with the aim of individually marking the majority of the population. A majority of the population will remain individually marked by periodic supplementary trapping after this initial program. This action will provide data on the age-structure of the population, distribution on Christmas Island (including some information on habitat), ranging behaviour and spacing of individuals from resightings during the 18 month study, and thus some data on population size.\n\nThe members of the Australasian Raptor Association (ARA), part of Birds Australia, include many of the Australian professional and amateur experts on birds of prey. They may be able to provide unpaid assistance in trapping and marking birds, and in training of local staff in trapping techniques. The cost of their involvement would be airfares from Australia and accommodation and transport on Christmas Island. With the involvement of the ARA this project lends itself to being done by a post-graduate student. If ARA involvement was not possible dedicated researchers would re required, and the cost would be substantially higher.\n\nAction 3\\. On-going monitoring of the population\n\nIt is essential to formally and regularly monitor the size of the population, and breeding success. The recovery team will monitor the total population size at regular intervals (initially annually) such that, if a decline takes place, it provides sufficient warning to managers to allow remedial management and research to take place. The monitoring program shall be based on observations of the individually marked population of Christmas Island Goshawks. PAN Christmas Island staff, tour operators, and the community in general will be encouraged to record all observations of goshawks, and to report their locations accurately. This will provide an ongoing data set on the proportion of resightings of marked birds, and on the habitats in which birds are seen. Once every five years, volunteers from the ARA will visit the island and trap and mark additional birds to maintain a majority of marked birds in the population.\n\nIn addition a sample of goshawk nests should be monitored annually. Monitoring of ten nests over a number of years will provide a considerable amount of information on breeding success, diet, and habitat requirements, and would be relatively inexpensive. Breeding success data will allow evaluation of the species’ status and population viability. Knowledge of diet is essential to evaluate the effects of habitat change, and predict the likely impact of any proposed changes in land management on the island. Information on diet and breeding requirements, combined with data from the population survey on distribution, ranging and spacing, will enable refinement of the definition of habitat critical to the survival of Christmas Island Goshawks, and determination of at what stage/s they utilise rehabilitated areas.\n\nTen nests will be monitored for signs of breeding, and once breeding has commenced monitored regularly (fortnightly) while the breeding attempt continues. In addition roost sites of parents should be located and pellets collected for dietary analysis. The pellets should be relatively straightforward to analyse due to the limited number of potential vertebrate prey. Insects should be identified as accurately as possible. Monitoring of nesting and collection and analysis of pellets could be part of the general duties of the resource management staff of PANCI.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria Protection of all habitat critical outside the national park.\n\nAction 4\\. Develop and implement wildlife management plan outside the national park\n\nAction 5\\. Ensure protection of habitat critical outside the national park.\n\nMost forested land outside the National Park is publicly owned, by the Commonwealth. PAN staff are routinely involved with the other stakeholders in negotiations over development proposals. This is ongoing. These proposals can occur unexpectedly and as there is no overall wildlife management plan outside the park, are assessed on an ad hoc basic. While consultation with other agencies and proponents on projects will be essential and ongoing (under the EPBC Act referrals process), achievement of long-term protection of threatened species outside the park requires a more comprehensive and considered approach through the development of a wildlife management program. The EPBC Regulations include that actions under an approved management plan are not considered offences in relation to protected species, and that a management plan must not, or not be likely to, adversely affect the conservation status of a protected species or a population of a protected species. The Goshawk meets the definition of a protected species.\n\nThe plan should allow for the input of relevant landholders and decision makers, and provide more strategic and comprehensive protection of flora and fauna, while facilitating greater certainty for environmentally sensitive developments. The plan should also allow for adaptive management as better information becomes available on the Christmas Island Hawk-Owl and other threatened species, in particular greater definition of the areas of habitat which are critical to survival.\n\nSpecific objective 2. Determine and maximise total population size and area of occupancy\n\nPerformance Criteria A continuing increase in suitable habitat through implementation of the Christmas Island Rainforest Rehabilitation program.\n\nAction 6\\. Continue effective and long-term rainforest rehabilitation program managed by DOTARS & supported by PAN and the mining company.\n\nThe Christmas Island Rainforest Rehabilitation Program has been occurring over the last three decades. Currently, the main aim is to revegetate mined areas adjacent to Abbott's Booby nesting habitat to reduce wind turbulence caused by the clearings and implicated in reduced nesting success of the booby (Reville et al. 1990). There are approx. 70 clearings covering 3200ha or 24% of the island area (Carew-Reid 1987). It is not known how Christmas Island Goshawks use these areas and at what stage of rehabilitation this habitat become important for foraging and for nesting – this will be investigated as part of the monitoring program.\n\nIn 2002, the Government decided to redirect the payment of the conservation levy to the portfolio agency with responsibility for the mine lease (DOTARS). A MOU between PAN and DOTARS was signed in February 2004 to allow PAN to continue the CIRRP within Christmas Island National Park. Rehabilitation is critical in terms of increasing rainforest habitat to further protect a range of species including the goshawk. The CIRRP is funded through a conservation levy that forms part of Christmas Island Phosphate’s lease. Effective management of the lease by DOTARS will ensure optimal outcomes for rainforest rehabilitation.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Crazy Ants have a negligible impact on Christmas Island Goshawk\n\nAction 7\\. Implement the Invasive Ants on Christmas Island Action Plan.\n\nPAN is currently implementing an Invasive Ants Action Plan, which aims to control and minimise the impact of Crazy Ants across the island. This program will require ongoing monitoring of Crazy Ant numbers and control of populations as detected. The recovery team must ensure that the control program does not have any negative impacts on Christmas Island Goshawks.\n\nSpecific objective 3. Implement threat abatement strategies\n\nPerformance Criteria Maintenance of effective quarantine against the introduction of all avian diseases\n\nAction 8. Maintenance \\[and regular review\\] of a quarantine barrier between Christmas Island and all other lands which minimises the risks of new avian diseases establishing on Christmas Island.\n\nAs part of the regular review process of the Christmas Island Quarantine Service, the effectiveness of the quarantine barrier preventing the man-made introduction of avian diseases onto Christmas Island should be assessed. Review of the Christmas Island Quarantine Service is undertaken by the Australian Quarantine Inspection Service, with input from PAN and in consultation with DOTARS. There is no additional cost for this action.\n\nSpecific objective 4. Increase community involvement and awareness\n\nPerformance criteria Demonstrated increase in community awareness and support for habitat protection\n\nAction 9. A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks.\n\nAs part of the Christmas Island Education and Interpretation Plan, which has a key objective of increasing knowledge of and local support for habitat and species protection, develop a multi-lingual pamphlet on Christmas Island Goshawk stressing their uniqueness and how they are part of Christmas Island ecosystems. This pamphlet would encourage the community to report sightings of colour-banded goshawks. It should also discuss managing problem goshawks around poultry, emphasising that goshawks are protected, and offering solutions to dealing with problem birds.\n\nThis should be further supported by a conservation related curriculum unit for both primary and secondary school students which raises awareness of the Christmas Island habitats and species generally.\n\n  \n\nSpecific objective 5\\. Implement the Recovery Plan through a Recovery Team\n\nPerformance criteria Demonstrated successful operation of the Recovery Team over five years\n\nActions 10. Establish a recovery team which meets regularly\n\n11\\. Carry out a major review of the recovery plan\n\nA recovery team shall be formed to implement this recovery plan. The team should comprise representatives of PANCI, the Shire of Christmas Island, the Department of Transport and Regional Services, an independent environmental consultant, and other members as thought appropriate eg. other affected Christmas Island interests, and other specialists where necessary. Progress will be monitored and evaluated yearly by members of the recovery team through an annual review. The monitoring process will include: compiling information from resighting of banded birds, success of monitored nests, and assessing progress of all actions against the criteria and objectives of the recovery plan (data comparisons performed by participating biologist). A major review of this performance must be conducted at the end of the first five years of implementation, by an independent reviewer.","sortOrder":30},{"sectionNumber":"Guide for Decision-makers","sectionType":"part","heading":"Guide for Decision-makers","content":"## Guide for Decision-makers\n\nThe following management actions are required to aid the Christmas Island Goshawk viability and recovery:\n\n1.  Adequate resourcing of the identified management activities for the species, in particular, the crazy ant action plan;\n2.  Prevention of illegal destruction of goshawks by Christmas Island residents;\n3.  Quarantine prevention of the introduction of new avian pathogens;\n4.  Quarantine prevention of the introduction of new invasive weeds; and\n5.  Quarantine prevention of the introduction of new predators.\n\nThis taxon is dependent on active conservation management.\n\nAny action which would remove nesting sites, potential nesting sites or feeding habitat could result in a significant impact on the species and requires referral to the Commonwealth Environment Minister under the EPBC Act.","sortOrder":31},{"sectionNumber":"Community Participation","sectionType":"part","heading":"Community Participation","content":"## Community Participation\n\nThis recovery plan aims to involve the community in a significant way in monitoring Christmas Island Goshawks. They will be encouraged to report all sightings or suspected sightings of goshawks to PANCI. In addition, goshawks will be colour-banded with an easily legible code or colour-combination that the community will be encourage to report. Validated reports will provide important population monitoring data.","sortOrder":32},{"sectionNumber":"Tools to Assist Implementation","sectionType":"part","heading":"Tools to Assist Implementation","content":"## Guide for Decision-makers\n\nThe following management actions are required to aid the Christmas Island Goshawk viability and recovery:\n\n1.  Adequate resourcing of the identified management activities for the species, in particular, the crazy ant action plan;\n2.  Prevention of illegal destruction of goshawks by Christmas Island residents;\n3.  Quarantine prevention of the introduction of new avian pathogens;\n4.  Quarantine prevention of the introduction of new invasive weeds; and\n5.  Quarantine prevention of the introduction of new predators.\n\nThis taxon is dependent on active conservation management.\n\nAny action which would remove nesting sites, potential nesting sites or feeding habitat could result in a significant impact on the species and requires referral to the Commonwealth Environment Minister under the EPBC Act.\n\n## Community Participation\n\nThis recovery plan aims to involve the community in a significant way in monitoring Christmas Island Goshawks. They will be encouraged to report all sightings or suspected sightings of goshawks to PANCI. In addition, goshawks will be colour-banded with an easily legible code or colour-combination that the community will be encourage to report. Validated reports will provide important population monitoring data.\n\n## Tools to Assist Implementation\n\n### Costs\n\n| Action                                                                                                                                                         | Likely expenses                                                             | Cost estimate                          | Responsibility | Timeframe                   |\n| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------- | -------------------------------------- | -------------- | --------------------------- |\n| Investigate taxonomic status                                                                                                                                   | Museum Victoria                                                             | $6000                                  | PANCI          | Yr 1`1-2                    |\n| Detailed population survey                                                                                                                                     | PANCI staff, training consultant, and local staff/contractor/ARA volunteers | $73,800#                               | PANCI          | Yr 1-2                      |\n| On-going population monitoring                                                                                                                                 | PAN staff timeARA volunteers                                                | $11,000 pa, + $7,000 /5 year^          | PANCI          | Ongoing                     |\n| Develop and implement wildlife management plan for habitat outside national park                                                                               | PANCI staff time                                                            | $10,000 then $5000pa                   | PAN            | Ongoing                     |\n| Ensure protection of habitat critical outside the national park*                                                                                               | PANCI staff time                                                            | $2,000 pa                              | PANCI          | Ongoing                     |\n| Continue effective and long-term rainforest rehabilitation program managed by DOTAR and supported by other government departments, PAN and the mining company. | DOTARS staff time, PANCI supervision and monitoring, contractors.           | $750,000** pa                          | DOTARS         | Until mine ceases operation |\n| Implement the Invasive Ants on Christmas Island Action Plan. *                                                                                                 | PANCI staff, ant bait, contractors, research                                | $475,000 in 2002/3) approx $100,000 pa | PANCI          | Ongoing                     |\n| Maintenance [and regular review] of a quarantine barrier between Christmas Island and all other lands to minimise risks of new avian disease establishing*     | AQIS staff time, PANCI assistance to AQIS                                   | $2000 pa (PANCI)                       | AQIS           | Ongoing                     |\n| A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks. *                                              | PANCI                                                                       | $2000 pa                               | PANCI          | Ongoing                     |\n| Establish a recovery team which meets regularly *                                                                                                              | PANCI                                                                       | $2000 pa                               | PANCI          | Ongoing                     |\n| Carry out a major review of the recovery plan                                                                                                                  | Recovery Team                                                               | $2500                                  | PANCI          | Yr 5                        |\n\nNotes: PAN: Parks Australia North; PANCI: PAN Christmas Island staff.\n\n\\* Actions that are wholly or partly shared with other Christmas Island recovery plans.\n\n\\# MSc student on island for 18 months with stipend paid- savings may be made if offer to support a student with a scholarship; Assumes involvement of ARA volunteers rather than dedicated researchers in trapping, marking birds and training local staff\n\n^ Local trained staff est. 30 nights work @$250/night + $1500 mileage: $9000/year, Staff time to analyse data: $2000/year; 2 ARA volunteers visit every 5 years (airfares + accom. and transport Total $7000 /5 yr);\n\n\\*\\* Funds provided by conservation levy, with total dependant on phosphate production levels\n\n### Summary table of costs\n\n| Year        | 1          | 2        | 3        | 4        | 5        |\n| ----------- | ---------- | -------- | -------- | -------- | -------- |\n| Total Costs | $1,015,800 | $924,000 | $924,000 | $924,000 | $926,500 |\n\n## Landholder incentives\n\nOpportunities exist for the Natural Heritage Trust to be used to protect remnant vegetation on private land. Management agreements can be put in place with landholders’ permission to protect and enhance native vegetation on the island.\n\n## Role and interests of indigenous people\n\nNot relevant.\n\n## Interests that will be affected by the plan’s implementation or adoption\n\nParks Australia North, Shire of Christmas Island, Union of Christmas Island Workers, Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services, the Department of Immigration and Indigenous Affairs, Christmas Island Phosphates.\n\n## Consultation with, and advice considered from relevant State and Territory Governments and the general public\n\nThere are no relevant State or Territory governments to consult with. Whilst Christmas Island is a Commonwealth non-self governing Territory, Western Australian laws are applied to the island as Commonwealth applied laws. A number of Western Australian agencies have roles on Christmas Island through agreements with the Department of Transport and Regional Services (DOTARS). DOTARS has contributed to the development of this plan.\n\nThis plan was originally written in 1996. The current version was completed by Birds Australia in consultation with PAN, staff of Christmas Island National Park, and the Department of Environment and Heritage, Canberra. Both Christmas Island Phosphates and Shire of Christmas Island were consulted during the preparation of this current plan.\n\nAs part of the requirements for the EPBC Act, a three month public consultation phase is undertaken whereby the public are able to contribute comments on draft plans, and where relevant these comments are then incorporated.\n\n## Community Participation\n\nRegular press articles in Christmas Island newspapers. Participation by community and visitors in reporting sightings of colour-banded goshawks will be an important part of the proposed goshawk monitoring program. Also see action 9 - includes production of a multi-lingual pamphlet on Christmas Island seabirds and a potential curriculum unit on endemic seabirds for both primary and secondary school students.\n\n##   \nReferences\n\nBurton, A.M. 1991. Resource partitioning between two sympatric goshawks in the Australian wet tropics. PhD thesis, James Cook University of North Queensland.\n\nBurton A.M., Alford, R.A. and J. Young. 1994 Reproductive parameters of the Grey Goshawk (Accipiter novaehollandiae) and Brown Goshawk (Accipiter fasciatus) at Albergowrie, northern Queensland, Australia. J. Zool. London 232:347-363.\n\nCarew-Reid, J. 1987. Rehabilitation of mined clearings on Christmas Island, Indian Ocean. Unpublished report to the Director of the Australian National Parks and Wildlife Service, Canberra.\n\nCarter, M. 1994. Christmas Island Birds. Wingspan: 1994, Summer.\n\nCaughley, G. and A.R.E. Sinclair. 1994. Wildlife Ecology and Management. Blackwell Science, Cambridge.\n\nChasen, F.N. 1933. Notes on the birds of Christmas Island. Bull. Raff. Mus.: 55-87.\n\nDexter, N. 2000. A Draft Conservation Management Plan for Terrestrial Wildlife on Christmas Island outside the National Park. Environment Australia, Canberra.\n\nDiamond, J.M. 1985. Population processes in island birds: immigration, extinction and fluctuations. In Conservation of Island Birds. ICBP Tech. Pub. 3.\n\nDu Puy. 1993. Flora of Australia, Volume 50. Oceanic Islands 2. Australian Biological Resource Study, Canberra.\n\nEnvironment Australia. 1994. Christmas Island National Park Management Plan. Commonwealth of Australia, Canberra.\n\nEnvironment Australia. 2002. Christmas Island National Park Management Plan. Commonwealth of Australia, Canberra.\n\nGarnett, S. and G. Crowley. 2002. Threatened and extinct birds of Australia. Royal Australasian Ornithologists Union. Australian National Parks and Wildlife Service, RAOU Report 82. Melbourne.\n\nGibson-Hill, C.A. 1947. Notes on the Birds of Christmas Island. Bull. Raffles Mus. XVIII 1947: 87-169\n\nGray, H.S. 1981. Christmas Island Naturally. Geraldton, H. Gray.\n\nHay, R. 1986. Bird Conservation in the Pacific Islands. ICBP Study Rep. 7. Cambridge, England.\n\nKing, W.B. 1985. Island Birds: will the future repeat the past. In Conservation of Island Birds. ICBP Tech. Pub. 3.\n\nLacy, R.C. 1987 Loss of Genetic Diversity from Managed Populations: Interacting effects of drift, mutation, immigration, selection, and population subdivision. Conserv. Biol. 1(2): 143-58.\n\nLister, J.J. 1888. On the natural history of Christmas Island. Proc. Zool. Soc. Lond. 1888: 512-31.\n\nMarchant, S. and P.J. Higgins. 1993. Handbook of Australian, New Zealand and Antarctic Birds, Vol II Raptors to Lapwings. Oxford University Press, Melbourne.\n\nNewton, I. 1986. The Sparrowhawk. Poyser, Staffordshire, UK.\n\nOrchard, M., Comport, S. and P. Green. 2002. Control of the Invasive Yellow Crazy Ant (Anoplolepis gracilipes) on Christmas Island, Indian Ocean; progress, problems, and future scenarios. Unpublished Discussion Paper 1 February 2002\n\nReville, B.J., J.D. Tranter and H.D. Yorkston. 1990. Conservation of the Endangered Seabird Abbott's Booby on Christmas Island. 1983-1989. ANPWS Occ. Paper No. 20.\n\nStokes, T. 1988. A review of the birds of Christmas Island, Indian Ocean. Australian National Parks and Wildlife Service Occ. Pap. No. 16\n\nThiollay J.M. 1993. Habitat segregation and the insular syndrome in two congeneric raptors in New Caledonia, the White-bellied Goshawk Accipiter haplochrous and the Brown Goshawk Accipiter fasciatus. Ibis 135 : 237-246.","sortOrder":33},{"sectionNumber":"Costs","sectionType":"division","heading":"Costs","content":"### Costs\n\n| Action                                                                                                                                                         | Likely expenses                                                             | Cost estimate                          | Responsibility | Timeframe                   |\n| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------- | -------------------------------------- | -------------- | --------------------------- |\n| Investigate taxonomic status                                                                                                                                   | Museum Victoria                                                             | $6000                                  | PANCI          | Yr 1`1-2                    |\n| Detailed population survey                                                                                                                                     | PANCI staff, training consultant, and local staff/contractor/ARA volunteers | $73,800#                               | PANCI          | Yr 1-2                      |\n| On-going population monitoring                                                                                                                                 | PAN staff timeARA volunteers                                                | $11,000 pa, + $7,000 /5 year^          | PANCI          | Ongoing                     |\n| Develop and implement wildlife management plan for habitat outside national park                                                                               | PANCI staff time                                                            | $10,000 then $5000pa                   | PAN            | Ongoing                     |\n| Ensure protection of habitat critical outside the national park*                                                                                               | PANCI staff time                                                            | $2,000 pa                              | PANCI          | Ongoing                     |\n| Continue effective and long-term rainforest rehabilitation program managed by DOTAR and supported by other government departments, PAN and the mining company. | DOTARS staff time, PANCI supervision and monitoring, contractors.           | $750,000** pa                          | DOTARS         | Until mine ceases operation |\n| Implement the Invasive Ants on Christmas Island Action Plan. *                                                                                                 | PANCI staff, ant bait, contractors, research                                | $475,000 in 2002/3) approx $100,000 pa | PANCI          | Ongoing                     |\n| Maintenance [and regular review] of a quarantine barrier between Christmas Island and all other lands to minimise risks of new avian disease establishing*     | AQIS staff time, PANCI assistance to AQIS                                   | $2000 pa (PANCI)                       | AQIS           | Ongoing                     |\n| A community education program to raise awareness and interest in the conservation of Christmas Island Goshawks. *                                              | PANCI                                                                       | $2000 pa                               | PANCI          | Ongoing                     |\n| Establish a recovery team which meets regularly *                                                                                                              | PANCI                                                                       | $2000 pa                               | PANCI          | Ongoing                     |\n| Carry out a major review of the recovery plan                                                                                                                  | Recovery Team                                                               | $2500                                  | PANCI          | Yr 5                        |\n\nNotes: PAN: Parks Australia North; PANCI: PAN Christmas Island staff.\n\n\\* Actions that are wholly or partly shared with other Christmas Island recovery plans.\n\n\\# MSc student on island for 18 months with stipend paid- savings may be made if offer to support a student with a scholarship; Assumes involvement of ARA volunteers rather than dedicated researchers in trapping, marking birds and training local staff\n\n^ Local trained staff est. 30 nights work @$250/night + $1500 mileage: $9000/year, Staff time to analyse data: $2000/year; 2 ARA volunteers visit every 5 years (airfares + accom. and transport Total $7000 /5 yr);\n\n\\*\\* Funds provided by conservation levy, with total dependant on phosphate production levels","sortOrder":34},{"sectionNumber":"Summary table of costs","sectionType":"division","heading":"Summary table of costs","content":"### Summary table of costs\n\n| Year        | 1          | 2        | 3        | 4        | 5        |\n| ----------- | ---------- | -------- | -------- | -------- | -------- |\n| Total Costs | $1,015,800 | $924,000 | $924,000 | $924,000 | $926,500 |","sortOrder":35},{"sectionNumber":"Landholder incentives","sectionType":"part","heading":"Landholder incentives","content":"## Landholder incentives\n\nOpportunities exist for the Natural Heritage Trust to be used to protect remnant vegetation on private land. Management agreements can be put in place with landholders’ permission to protect and enhance native vegetation on the island.","sortOrder":36},{"sectionNumber":"Interests that will be affected by the p","sectionType":"part","heading":"Interests that will be affected by the plan’s implementation or adoption","content":"## Interests that will be affected by the plan’s implementation or adoption\n\nParks Australia North, Shire of Christmas Island, Union of Christmas Island Workers, Asia Pacific Space Centre Pty. Ltd., Department of Transport and Regional Services, the Department of Immigration and Indigenous Affairs, Christmas Island Phosphates.","sortOrder":38},{"sectionNumber":"Consultation with, and advice considered","sectionType":"part","heading":"Consultation with, and advice considered from relevant State and Territory Governments and the general public","content":"## Consultation with, and advice considered from relevant State and Territory Governments and the general public\n\nThere are no relevant State or Territory governments to consult with. Whilst Christmas Island is a Commonwealth non-self governing Territory, Western Australian laws are applied to the island as Commonwealth applied laws. A number of Western Australian agencies have roles on Christmas Island through agreements with the Department of Transport and Regional Services (DOTARS). DOTARS has contributed to the development of this plan.\n\nThis plan was originally written in 1996. The current version was completed by Birds Australia in consultation with PAN, staff of Christmas Island National Park, and the Department of Environment and Heritage, Canberra. Both Christmas Island Phosphates and Shire of Christmas Island were consulted during the preparation of this current plan.\n\nAs part of the requirements for the EPBC Act, a three month public consultation phase is undertaken whereby the public are able to contribute comments on draft plans, and where relevant these comments are then incorporated.","sortOrder":39},{"sectionNumber":"References","sectionType":"part","heading":"References","content":"##   \nReferences\n\nBurton, A.M. 1991. Resource partitioning between two sympatric goshawks in the Australian wet tropics. PhD thesis, James Cook University of North Queensland.\n\nBurton A.M., Alford, R.A. and J. Young. 1994 Reproductive parameters of the Grey Goshawk (Accipiter novaehollandiae) and Brown Goshawk (Accipiter fasciatus) at Albergowrie, northern Queensland, Australia. J. Zool. London 232:347-363.\n\nCarew-Reid, J. 1987. Rehabilitation of mined clearings on Christmas Island, Indian Ocean. Unpublished report to the Director of the Australian National Parks and Wildlife Service, Canberra.\n\nCarter, M. 1994. Christmas Island Birds. Wingspan: 1994, Summer.\n\nCaughley, G. and A.R.E. Sinclair. 1994. Wildlife Ecology and Management. Blackwell Science, Cambridge.\n\nChasen, F.N. 1933. Notes on the birds of Christmas Island. Bull. Raff. Mus.: 55-87.\n\nDexter, N. 2000. A Draft Conservation Management Plan for Terrestrial Wildlife on Christmas Island outside the National Park. Environment Australia, Canberra.\n\nDiamond, J.M. 1985. Population processes in island birds: immigration, extinction and fluctuations. In Conservation of Island Birds. ICBP Tech. Pub. 3.\n\nDu Puy. 1993. Flora of Australia, Volume 50. Oceanic Islands 2. Australian Biological Resource Study, Canberra.\n\nEnvironment Australia. 1994. Christmas Island National Park Management Plan. Commonwealth of Australia, Canberra.\n\nEnvironment Australia. 2002. Christmas Island National Park Management Plan. Commonwealth of Australia, Canberra.\n\nGarnett, S. and G. Crowley. 2002. Threatened and extinct birds of Australia. Royal Australasian Ornithologists Union. Australian National Parks and Wildlife Service, RAOU Report 82. Melbourne.\n\nGibson-Hill, C.A. 1947. Notes on the Birds of Christmas Island. Bull. Raffles Mus. XVIII 1947: 87-169\n\nGray, H.S. 1981. Christmas Island Naturally. Geraldton, H. Gray.\n\nHay, R. 1986. Bird Conservation in the Pacific Islands. ICBP Study Rep. 7. Cambridge, England.\n\nKing, W.B. 1985. Island Birds: will the future repeat the past. In Conservation of Island Birds. ICBP Tech. Pub. 3.\n\nLacy, R.C. 1987 Loss of Genetic Diversity from Managed Populations: Interacting effects of drift, mutation, immigration, selection, and population subdivision. Conserv. Biol. 1(2): 143-58.\n\nLister, J.J. 1888. On the natural history of Christmas Island. Proc. Zool. Soc. Lond. 1888: 512-31.\n\nMarchant, S. and P.J. Higgins. 1993. Handbook of Australian, New Zealand and Antarctic Birds, Vol II Raptors to Lapwings. Oxford University Press, Melbourne.\n\nNewton, I. 1986. The Sparrowhawk. Poyser, Staffordshire, UK.\n\nOrchard, M., Comport, S. and P. Green. 2002. Control of the Invasive Yellow Crazy Ant (Anoplolepis gracilipes) on Christmas Island, Indian Ocean; progress, problems, and future scenarios. Unpublished Discussion Paper 1 February 2002\n\nReville, B.J., J.D. Tranter and H.D. Yorkston. 1990. Conservation of the Endangered Seabird Abbott's Booby on Christmas Island. 1983-1989. ANPWS Occ. Paper No. 20.\n\nStokes, T. 1988. A review of the birds of Christmas Island, Indian Ocean. Australian National Parks and Wildlife Service Occ. Pap. No. 16\n\nThiollay J.M. 1993. Habitat segregation and the insular syndrome in two congeneric raptors in New Caledonia, the White-bellied Goshawk Accipiter haplochrous and the Brown Goshawk Accipiter fasciatus. Ibis 135 : 237-246.","sortOrder":41}],"analysis":{"flash_summary_failed":{"failed":true,"reason":"Unauthenticated. Configure AI_GATEWAY_API_KEY or use a provider module. Learn more: https://ai-sdk.dev/unauthenticated-ai-gateway","source":"analysis-cron"},"kimi_summary":{"_metrics":{"completionTokens":575},"content_quality":"ok","complexity_score":4,"scope_assessment":{"changed":false,"description":"The plan remains focused on its original purpose: recovery of the Christmas Island Goshawk. While it addresses related issues like crazy ant control and rainforest rehabilitation, these are directly linked to the goshawk's survival rather than scope creep. The plan appropriately nests within broader island management frameworks without expanding beyond its core species mandate."},"complexity_factors":["Multiple stakeholder agencies with overlapping responsibilities (PAN, DOTARS, AQIS, DEH)","Cross-references to other plans and legislation (EPBC Act, EPBC Regulations, Invasive Ants Action Plan, National Park Management Plan)","Conditional logic around habitat protection (different rules inside vs outside National Park)","Scientific uncertainty requiring adaptive management (taxonomy unresolved, exact habitat requirements unknown)","Implementation dependencies on external programs (rainforest rehabilitation funded by mining levy)","Performance criteria with subjective thresholds (e.g. 'negligible impact', 'demonstrated increase')"],"plain_english_summary":"The recovery plan sets a 10-year strategy to move the Christmas Island Goshawk from Endangered to Conservation Dependent by guiding targeted management, monitoring and community engagement to secure its tiny (~100 adults) population. \n\nIt focuses on controlling invasive yellow crazy ants, protecting remaining rainforest from development and mining impacts, reducing disease and mortality risks, and coordinating government, residents and industry action—vital because the goshawk is a critically range-restricted endemic whose fate shapes the island’s wider endangered fauna."},"flash_summary":{"complexity_score":6,"scope_assessment":{"changed":true,"description":"This version enlarges and operationalises an earlier plan (originally drafted 1996) by adding detailed costings, formal assignment of responsibilities, and explicit linkage to island‑level programs and funding mechanisms. Notable scope changes include: incorporation of the Invasive Ants Action Plan as a central threat‑abatement mechanism (Action 7); formal reliance on the conservation levy and DOTARS’ lease management to fund and prioritise the Christmas Island Rainforest Rehabilitation Program (Action 6; Costs notes); clearer requirements for EPBC referral where habitat is affected (Guide for Decision‑makers); and a structured Recovery Team and five‑year independent review requirement (Actions 10–11). These additions move the document beyond a descriptive recovery plan into a resourced, multi‑agency implementation program with specific budgetary and administrative linkages."},"complexity_factors":["Lengthy document with repeated sections and duplicated text increasing reading load","11 discrete recovery Actions (Actions 1–11) with sub‑actions and linked Performance Criteria","Multiple performance criteria and indicators (taxonomic clarification, distribution, density, ant impacts, habitat protection, quarantine, community awareness, recovery team operation)","Costs table with 11 budget line items and a 5‑year summary (Year 1 ≈ $1,015,800 then ≈ $924,000 pa)","Dependence on external, separately authored plans (Invasive Ants on Christmas Island Action Plan; Christmas Island Rainforest Rehabilitation Program; National Park Plan of Management)","Cross‑references to EPBC Act and EPBC Regulations including Schedule 12 forest classifications (legal thresholds for referrals)","Conditional and variable funding sources (conservation levy tied to phosphate production), creating contingent obligations","Operational dependence on volunteers and external specialists (ARA) and on multiple agencies (PANCI, DOTARS, AQIS), increasing coordination complexity","Data gaps and contingent rules (habitat critical to survival partly undefined for second‑growth; many actions conditional on monitoring outcomes)","Implementation timing and review structure: annual monitoring + a required major independent review at 5 years"],"plain_english_summary":"**What this instrument is and what it does (mechanics)**\n\n- This is a recovery plan for the Christmas Island Goshawk (Accipiter fasciatus natalis) prepared under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). It sets an overall management program — objectives, measurable performance criteria, and specific actions — to halt decline and support recovery of the taxon (Overall objective; Actions sections).\n\n- The plan’s formal objective is to move the goshawk from “Endangered” to “Conservation Dependent” within 10 years by: (a) resolving taxonomic status, (b) estimating and increasing population size and area of occupancy, (c) abating main threats (notably invasive yellow “crazy” ants), (d) increasing community involvement, and (e) implementing the plan through a Recovery Team (Overall objective; Specific objectives; Actions 1–11).\n\n- It identifies a set of 11 principal actions (investigate taxonomy; population survey and on‑going monitoring; habitat protection and wildlife management planning outside the National Park; rainforest rehabilitation; implement the Invasive Ants Action Plan; maintain quarantine; community education; form a Recovery Team; and review the plan) and links each to responsible agencies and estimated costs (Actions; Tools to Assist Implementation; Costs table).\n\n- The plan defines “habitat critical to survival” conservatively as all Primary Rainforest and Marginal Rainforest (and possibly suitable second‑growth) on Christmas Island, and says any action that would remove nesting or feeding habitat could be a significant impact requiring referral to the Commonwealth Environment Minister under the EPBC Act (Habitat Critical to Survival; Guide for Decision‑makers).\n\n**Who it affects and who pays / decides**\n\n- Affected parties named in the plan: Parks Australia North (PAN/PANCI), Shire of Christmas Island, Christmas Island Phosphates (mining company), Union of Christmas Island Workers, Department of Transport and Regional Services (DOTARS), Australian Quarantine and Inspection Service (AQIS), the Department of Immigration, and local residents and businesses (Affected Interests; Costs table).\n\n- Primary planners and implementers: PAN (on‑island management and monitoring), DOTARS (management of the mine lease and rainforest rehabilitation funding since 2002), AQIS (quarantine), and a Recovery Team to be established that will include those stakeholders (Recovery Team; Actions 4–6; Action 8).\n\n- Who pays: the plan lists specific cost items. A large component of ongoing rehabilitation funding is supplied by a conservation levy paid by the phosphate mining company under the mine lease; other costs are budgeted to PAN/PANCI, DOTARS, AQIS and to occasional volunteer contributions (Costs table; Notes under Costs). The plan’s Year 1 total estimated cost is about $1,015,800 with subsequent years around $924,000–$926,500 (Summary table of costs).\n\n**Why it matters (official rationale and a practical test of mechanisms)**\n\n- Official rationale: the goshawk is a small, island‑restricted raptor (perhaps ~100 adults) vulnerable to habitat loss, small‑population dynamics, introduced predators/diseases, and especially the invasive yellow crazy ant. The plan identifies ant control and habitat protection as the two strategic levers to reduce extinction risk and restore population viability (Executive Summary; Threats: Crazy Ants; Habitat Loss).\n\n- How the plan tries to achieve that: (i) reduce or neutralise the ant threat by implementing the island’s Invasive Ants Action Plan and ongoing monitoring (Action 7); (ii) protect and increase rainforest area through the Christmas Island Rainforest Rehabilitation Program (CIRRP) funded by the conservation levy (Action 6); (iii) stop direct loss of nesting/feeding sites by requiring EPBC referrals and developing a wildlife management plan for land outside the national park (Actions 4–5; Guide for Decision‑makers); and (iv) build a monitoring, research and community‑reporting system (Actions 2–3, 9).\n\n**Immediate practical implications and likely behaviour changes**\n\n- Developers and proponents operating on Christmas Island face more definite environmental obligations: developments that remove or affect mapped Primary or Marginal Rainforest or potential goshawk habitat may trigger a referral to the Commonwealth Environment Minister under the EPBC Act (Guide for Decision‑makers; Habitat Critical to Survival). That raises the administrative and compliance cost of some projects.\n\n- The mining company funds rehabilitation through a conservation levy; DOTARS now administers the lease and rehabilitation program, so mine operations and rehabilitation priorities are formally linked to biodiversity outcomes (Action 6; Costs notes).\n\n- Operational behaviour shifts: PANCI staff and volunteers will run trapping, banding and nest monitoring programs; AQIS will maintain quarantine checks; PAN/PANCI will coordinate ant control and community education; community members will be encouraged to report banded birds (Actions 2–3, 7–9; Community Participation).\n\n**Costs, incentives, trade‑offs and implementation risks**\n\n- Costs: the plan quantifies both one‑off and recurring expenses (taxonomic study $6,000; population survey ~$73,800; ant control and CIRRP are the big line items — ant control ~$475,000 in 2002/3 then ~$100,000 pa; CIRRP ~ $750,000 pa but dependent on phosphate production levels) (Costs table; Summary table of costs).\n\n- Funding incentives and trade‑offs: rehabilitation relies on a mining‑linked conservation levy. That channels a concentrated funding source to conservation but creates dependency on phosphate production levels and on DOTARS’ lease management decisions (Costs notes; Action 6). This arrangement creates negotiating and priority‑setting points between conservation managers and the mining lease administrator.\n\n- Implementation risks and opportunity costs: the plan repeatedly notes data gaps (no reliable current population estimate; unresolved taxonomic status; unclear use of rehabilitated areas), so many actions depend on research outcomes. The recovery outcome is therefore conditional on successful ant control, sustainable funding, effective quarantine, and reliable monitoring (Background Information; Actions 1–3; Threats). Reliance on volunteer specialists (ARA) reduces immediate cash cost but increases operational risk if volunteers are unavailable (Action 2; Costs notes).\n\n- Administrative and compliance burden: increased monitoring, management planning for land outside the national park, EPBC referrals and quarantine reviews increase workload for PANCI, DOTARS and AQIS and introduce potential delays and additional conditions for developers and industry (Actions 4–5; Action 8; Social and economic impacts).\n\n- Concentrated benefits vs diffuse costs: benefits are concentrated among conservation agencies, endemic species and eco‑tourism interests; potential costs (compliance, constraints on developments) are dispersed among developers, landholders and possibly local businesses. The plan itself recognises both positive social benefits (education, eco‑tourism, jobs from rehabilitation and monitoring) and possible negative economic effects where restrictions apply (Social and economic impacts).\n\n**Key implementation uncertainties to watch**\n\n- Whether ant control can be sustained at landscape scale without unacceptable non‑target impacts (Action 7; Threats: Crazy Ants).\n- Whether the conservation levy continues at levels sufficient to fund CIRRP and long‑term rehabilitation (Costs notes).\n- Whether monitoring (trap‑mark‑resight, nest monitoring) yields robust population trend data fast enough to detect declines and trigger remedial actions (Actions 2–3; Performance Criteria).\n- Legal/administrative clarity around what counts as “habitat critical to survival” for EPBC referral purposes where second‑growth habitat mapping is incomplete (Habitat Critical to Survival).\n\n**Where responsibility and discretion lie**\n\n- PAN/PANCI: deliver monitoring, community education, ant control operational input and lead/co‑ordinate recovery actions on the ground (Actions 2–3, 7, 9; Costs table).\n- DOTARS: manage mine lease and conservation levy funding; oversee CIRRP funding and rehabilitation work (Action 6; Costs notes).\n- AQIS: maintain quarantine barrier and perform periodic reviews (Action 8; Costs table).\n- Commonwealth Environment Minister: decides EPBC Act referrals and approvals when proposed actions may significantly affect the listed species or critical habitat (Guide for Decision‑makers).\n\n"}},"importantCases":[],"_links":{"self":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis","history":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis/history","analysis":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis/analysis","conflicts":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis/conflicts","importantCases":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis/important-cases","documents":"/api/acts/national-recovery-plan-for-the-christmas-island-goshawk-accipiter-fasciatus-natalis/documents"}}