AUSTRALIAN 310 WATCHER

AUSTRALIAN BIRD WATCHER 1994, 15 , 310-313

Notes on the Lord Howe Strepera graculina crissalis

by DAVID C. McFARLAND, 15 Currong Street, Kenmore, Queensland 4069

Summary Observations on the Lord Howe Currawong Strepera graculina crissalis were made during several visits to over a number of years. This paper provides information on Currawong abundance, especially in the southern mountains, and breeding activity. The numbers of and Woodhens Tricholimnas sylvestris in the areas visited are compared.

Introduction The Lord Howe Currawong Strepera graculina crissalis is considered a threatened of a secure endemic (Garnett 1992) and is confined to Lord Howe Island. The bird's status arises primarily from its small population size. However, only Knight (1987) has made any attempt to quantitatively assess Currawong numbers. Previous reports are vague, ranging from 'common' and 'fairly numerous' to 'more than 10 but less than 100' (Sharland 1929, Hindwood 1940, Fullagar eta!. 1974). The most recent estimate is 73 with a suggested maximum of 80 breeding individuals (40 pairs) (Knight 1987). In the past the Currawong was shot because of its reputation for preying on both wild and domestic birds and attacking humans in the vicinity of nests (Hutton 1991). Currawongs are also suspected of taking the and chicks of the rare Lord Howe Island Woodhen Tricholimnas sylvestris (Miller & Mullette 1985, J. Hardy pers. cornm.). Information on breeding by the Lord Howe Currawong is limited to data on clutch size (3 eggs) and season (July-October) (Basset Hull1909, Hindwood 1940). Nests have been found high in both the Banyan and Greybark Drypetes australasica (Sharland 1929). In this paper I provide additional information on the abundance of Currawongs on the island, particularly in the southern mountains, and examine a possible relationship between Currawong and Woodhen numbers. Some data are also presented on Currawong breeding activity.

Study area Lord Howe Island (31 °33 'S, 159 "05 'E) lies approximately 700 km north-east of and covers an area of I 455 ha. A remnant of a large shield volcano, the crescent-shaped island is 11 km long and 2.8 km wide at its widest point (Figure 1) . The southern part of the island is dominated by Mount Gower (875 m) and Mount Lidgbird (777 m). A chain of smaller basaltic peaks occurs along the northern edge of the island. The island experiences ;; maritime climate with warm summers, wet winters and high humidity year round. Vegetation on Lord Howe is diverse, ranging from rainforest and montane mossy forest to palm forest, shrublands and grasslands (Pickard 1983). Ornithologically, the island is most famous for the successful rehabilitation of the endangered Lord Howe Island Woodhen (Miller & Mullette 1985). Methods Since 1985 I have been involved in Woodhen censuses undertaken by the National Parks & Wildlife Service of New South Wales. In November 1987 and 1989, transects located in the southern mountains (Mounts Lidgbird and Gower, Intermediate Hill) were walked and a tape of a calling Woodhen used to elicit responses by birds in the area. I noticed the tape also attracted other species, particularly the Lord Howe Currawong and Lord Howe Whistler Pachycephala pectoralis contempta. When walking a transect a record was kept of the total number of Currawongs and Woodhens found. Birds that followed the recording along the transect, i.e. kept constant contact or flew from the same direction as a previous encounter, were counted only once. Whenever nests were found their location, the tree species used, and the number of young present (if seen) were recorded. In November 1991 the transects were not VOL. 15 (7) SEPTEMBER 1994 Lord Howe Currawong 311

TNt

( •lot~d~;;,IIII250ml ~'(OUR

Qh1t Lldgblra(765ml FAR FLATS-) '\{) 1 ~ ERSKINE VALLEY

LITTLE SLOPE JBIG SLOPE ·~++--:,:-Mt Gower Plateau (866ml 0 2

Figure 1. Map of Lord Howe Island with localities mentioned in text. used, but most areas visited in previous years were again searched for Woodhens. In each area Currawongs and Woodhens were counted. A few notes on Currawong numbers and breeding were also made at locations away from the southern mountains. Results No attempt was made to systematically assess the number of Lord Howe Currawongs in the lowlands or northern mountains. Based on all observations, the highest number of birds recorded was 43 in 1991 with 37 of these in the southern mountains (Table 1). Numbers varied little in areas revisited over the three years (Table 1) and, given a minimum of five birds on Big Slope, the Currawong population in the southern mountains was at least 42. These results also assume that individuals were site-faithful AUSTRALIAN 312 McFARLAND BIRD WATCHER

Table 1 Numbers of Lord Howe Currawongs recorded during visits to the southern mountains in 1987, 1989 and 1991. The areas referred to in the table are A = Far Flats/Intermediate Hill/Boat Harbour; B = eastern slopes of Mt Lidgbird; C = Erskine Valley; D = Little Slope; E = Big Slope; F = Mt Gower summit. Dash indicates area not visited or counts not made. For locations of areas, see Figure 1. Year Southern mountains area A B c D E F Total 1987 18 2 3 5 28 1989 14 2 5 5 9 35 1991 16 4 4 4 9 37

during the breeding season and so there was no recounting of birds related to movements between areas. Assuming that Currawongs were attracted to the tape call from 100 m either side of the transect (about the same distance found for Woodhens) then the average density was 0.18 ± 0.02 birds/ha (mean ± S.E., range = 0-0.5 birds/ha, n = 28 transects). The number of Currawongs in a given locality was positively and significantly correlated with Woodhen abundance in each year (using transect and non-transect results: 1987 - r [correlation coefficient] = 0.54, df = 15, P( 0.05; 1989 - r = 0.'17, df = 9, P( 0.01; 1991 - r = 0.79, df = 12, P( 0.01) . Although there was only one instance of Woodhens being found and no Currawongs, there were 10 cases of 1-4 Currawongs occurring in an area without Woodhens. On five occasions I saw Lord Howe Currawongs diving at adult and immature Woodhens and twice they were seen coming to ground to search for hidden chicks. When Woodhens were being banded, Currawongs often perched in the branches above or in nearby trees. (However, Currawongs appear to be attracted to other human activity in the absence ofWoodhens: R. Harden in litt.) Of the 13 nests found with adult birds in attendance, 10 were in the southern mountains. Where the nest contents could be determined (usually only when well­ feathered chicks were present) the number of young seen was one (1 nest), two (5 nests) , and three (4 nests). The tree species used included Scalybark Cleistocalyx fullagari (7 nests), Blackbutt Cryptocarya triplinervis (3 nests), Greybark (2 nests) and Thatch Palm Howeaforsteriana (1 nest). All were over 8 m from the ground in trees located on the lower slopes of gullies in undisturbed forest. All of these records were from November and December. During other visits in February and March there was no indication of breeding. Despite five November ascents since 1985 and the high numbers of Currawongs on the Mount Gower summit (Table 1), I have no evidence of breeding in that locality. However, Currawongs were often seen flying up and down between the summit and the eastern slopes of Mount Lidgbird and Erskine Valley.

Discussion In 1972 Fullagar et al. (1974) estimated the Lord Howe Currawong population at 30-50 birds and considered this a decline from an earlier assessment of 70-100 birds (McKean in Fullagar et al. 1974). How these estimates were arrived at was not described. The most recent estimate was of 73 birds in 1985 (Knight 1987). These values are coarse because of the difficulties in counting a small population of highly mobile that cannot be individually identified. An accurate measure of the population size can only be obtained through intensive colour banding, territory mapping and transects/total counts. The Currawong population was and always will VOL. 15 (7) SEPTEMBER 1994 Lord Howe Currawong 313 be small (Fullagar et al. 1974) given the bird's size, trophic position (high in food chain, i.e. predator role) and limited habitat area, and consequently it will always be at some risk (Fullagar & Disney 1975). Detailed information is needed, along with follow-up monitoring, to determine whether the subspecies is under any current threat. On the Australian mainland, densities in eucalypt forest range from 0.02 to 0.46 birds/ha (Blakers et al. 1984). On Lord Howe, in the tall undisturbed forest that the subspecies prefers (Knight 1987, Hutton 1991), the average density is within that range at 0.18 birds/ha. Although human settlement has resulted in some loss of preferred habitat (Pickard 1983), the birds may have benefited from the increased diversity and abundance of foods, e.g. introduced fruits and rodents (Hutton 1991). The positive correlation between Currawong and Woodhen numbers can be explained by both species independently selecting the same habitat, whose productivity (i .e. invertebrate numbers in canopy and litter used by Currawong and Woodhen respectively) determines bird numbers. As seen in the results, Currawongs do occur in areas without Woodhens. However, it is likely that Woodhens, particularly chicks, are taken opportunistically by Currawongs. The facts that Lord Howe Currawongs have a broad diet, occur in areas without Woodhens and have suffered no apparent historical decline paralleling that of Woodhens (Hutton 1991) argue against any dependence on Woodhens as a food source. The Currawong is an occasional predator but, unlike the introduced owls, rats and cats, it is a natural one that Woodhens and other island wildlife have survived with since before human settlement. Lord Howe Currawongs nest in a wider range of trees than previously known and have a breeding season somewhat longer than the July-September mentioned by Basset Hull (1909). The nest site fits well with Knight's (1987) general description of Currawong territories including 'a section of stream or gully with tall timber.' Pairs appear to raise two or more chicks but fledging success is not known.

Acknowledgements My thanks to Bob Harden for providing comments on this paper and for use of the map, and the National Parks & Wildlife Service of New South Wales for inviting me to participate in the Woodhen project. References Basset Hull, A.F. (1909), 'The birds of Lord Howe and Norfolk Islands', Proc. Linn. Soc. N.S.W 34, 636-693. Blakers, M., Davies, S.J.J.F. & Reilly, P.N. (1984), The Alias ofAustralian Birds, RAOU & Melbourne University Press, Melbourne. Fullagar, P.J. & Disney, H.J. deS. (1975), 'The birds of Lord Howe Island : a report on the rare and endangered species,' XII Bull. International Council Bird Preservation 12, 187-202. --, McKean, J.L. & Van Tets, G.F. (1974), 'Report on birds', 55-72 in Recher, H.F. & Clark, S.S. . (Eds.), Environmental Survey of Lord Howe Island, Government Printer, Sydney. Garnett, S. (Ed.) (1992), Threatened and Extinct Birds of , RAOU Report 82. Hindwood, K.A. (1940), 'The birds of Lord Howe Island', Emu 40, 1-86. Hutton, I. (1991), Birds of Lord Howe Island Past and Present, I. Hutton, Coffs Harbour. Knight, B.J. (1987), 'A population survey ofthe Lord Howe Island Currawong', Aust. Birds 21, 28-29. Miller, B. & Mullette, K.J. (1985), 'Rehabilitation of an endangered Australian bird: the Lord Howe Island Woodhen Tricholimnas sylvestris (Sclater),' Bioi. Conserv. 34, 55-95. Pickard, J. (1983) , 'Vegetation of Lord Howe Island', Cunninghamia l , 133-265. Sharland, M.S.R. (1929), 'Land birds of Lord Howe Island', Emu 29, 5-11. Received 2 November 1993 •