Mission Beach, cassowaries, public survey, cassowary, personal communication, Townsville, J. Madsen, Mt Cook, forest patches, populations, Graham Range, conservation, Millaa Millaa, literature survey, Malbon Thompson Range, Nyletta Creek Mt Tyson Jarra Creek Meunga Creek Southcrn Cardwell Range, north Queensland, movements, Koombooloomba Cameron Creek, Atherton Tableland, coastal plain, management requirements, A. Moore, Tropical Forest Research Centre, field survey, southern cassowary, conservation status, rugged region, present
Aust. Wildl. Res., 1990, 17, 369-85 Cassowaries in North-eastern Queensland: Report of a Survey and a Review and Assessment of their Status and Conservation and Management Needs Francis H. J. Crome and L. A. Moore CSIRO Division of Wildlife and Ecology, tropical forest
Research Centre, P.O. Box 780, Atherton, Qld 4883, Australia. Abstract This is a report o n the distribution, status and conservation of the southern cassowary between Cooktown and Townsville, based on a field, public and literature survey. The status of birds in 77 regions is given according to a scoring system and the decline and/or loss of cassowaries in several of them is noted. Cassowary habitat and the factors controlling the distribution of the birds are discussed; plant species
diversity and history are considered the most important. We consider the species threatened in this region, and its decline, predators, disease, movements and the conservation and management requirements are discussed. Introduction The southern cassowary (Casuarius casuarius), henceforth referred to simply as cassowary, is the largest and most spectacular vertebrate in the Australian rainforests, and a major tourist attraction. It is our largest specialised frugivore and is known to disperse several hundred species of rainforest trees and vines; thus it plays an important role in rainforest regeneration and possibly also succession. Fears for the survival of the cassowary are long standing (e.g. White 1946) but only in the last 5 or 6 years have these concerns become widespread. In February 1986 cyclone Winifred stripped the forest of potential crops of fruit and forced out hungry birds. The cyclone's effects and a well publicised feeding campaign by the Queensland National Park
s and Wildlife Service (QNPWS) captured general public attention and revived fears for the survival of cassowaries. During 1988 we were contracted by QNPWS and the Australian National Parks and Wildlife Service
(ANPWS) to survey the present distribution of cassowaries between Townsville and Cooktown; review all information that was available on the species; and assess its conservation status and management and research requirements. Like most such studies the contract period was brief and the results necessarily presented at length in a report (Crome and Moore 1988) and, therefore, are not readily available despite the fact that it is an important document on the biology and conservation of this species. T o redress this issue therefore, this paper presents a summary of our findings: the results of the survey; a brief assessment of distribution in relation to selected environmental and biotic factors; an analysis of factors important in cassowary conservation and management; and suggestions as to management and further research. 03 10-7833/90/040369$03 .OO
F. H. J. Crorne and L. A. Moore Cape Tribulation J Mission Beach kilometres Townsville Fig. 1. Locations and numbers of regions described in Appendix 1 with the approximate boundaries of the rainforest.
Cassowaries in North Queensland study area
The study took place between January and September 1988 in the rainforests of the CooktownTownsville region, between Mt Cook (145°16'E,15030'S) and Mt Halifax (146°22'E,19007'S) (Fig. 1). Methods Review of Information A literature survey was performed which located 625 references, and every zoological garden and museum in Australia and most major overseas institutions were contacted for information on live birds and preserved material in their possession. The distributional survey information was gathered from four sources: extensive field surveys for signs of cassowary presence; a public survey in which information was solicited from local residents; the literature; and combination of the latter three into a scoring system. The Royal Australasian Ornithologists' Union (RAOU) Atlas scheme was consulted but a t the time of the survey, data were not readily available. At the request of a reviewer it was consulted again and about three-quarters of the records were available and purchased. They made no difference to our conclusions on distribution, being all from areas where we already knew cassowaries were regularly and easily seen. Distribution Survey Field survey Our personal field experience with cassowaries indicated that in areas where they are not attracted to orchards and/or being hand fed, they are generally shy and infrequently seen. Surveys based on sightings to estimate numbers accurately over the large study area were, if possible at all, not feasible in the short term available (9 months) over such a large area. Instead our surveys relied on signs of cassowaries to indicate their presence. I n all, 211 transects, each 5 m wide, were established between Mt Cook and Mt Halifax and surveyed twice-between January and May 1988 and May and September 1988; 176 were 1 km long and 35 were shorter, the latter in areas where longer transects could not be fitted. During a survey all droppings on the transect were mapped and collected, all footprints mapped and all birds sighted or heard recorded. Given the time and resources available, the accessibility, the nature of the terrain and the density of the vegetation realistically precluded locating transects randomly in the forest. Instead, all were on walking tracks, existing brush lines in the forest or snigging roads. None the less 1100 km were traversed on foot within the survey period. The transects may have been biased through birds preferring or avoiding such areas compared with the forest proper. T o estimate this bias we used test surveys at Mission Beach, an area known to contain cassowaries, in January 1988. Ten transects were made along tracks: at a random point along each track a brush line was cut at right angles to the track and used as a survey line. This resulted in 10 track and 10 forest transects. Comparing the number of droppings recorded on the transects by a t-test, there was no significant difference betheen transects along tracks and those within the forest proper ( t = 1.05629, d.f. = 15, P = 0 . 3 ) . Pubhc survey Information from a public survey seemed to be essential despite the problems of the excess of positive over negative records, the different but unknown effort put into observations, faulty memories and biases. In areas where the field survey failed to find signs of birds, local residents could indicate whether birds were seen there or not. Moreover, in the absence of any previous systematic field survey, only information from interested long-time residents could indicate changes in status of the birds over the last few decades. In June 1988 advertisements were put in all the newspapers from Townsville northward asking people to volunteer information, sightings, breeding records, status (rare or common, declining or increasing) and behaviour. During field surveys we contacted local residents, who may not have responded to the public survey, to canvass opinion on the status of cassowaries in their district. There were 81 responses to the public survey, many from long-time residents able to give historical accounts of cassowaries in their district. Scoring system Birds were seen only six times during the survey, confirming the difficulty of the direct-count method. The commonest sign of the presence of cassowaries was droppings, which could have been used as a quantitative index of the numbers present. However, to use dropping counts directly would require data on defecation rates and their dynamics in relation to size, sex and age of birds, and the food available. Instead, each transect was given a 'transect score', indicating increasing cassowary activity based on all signs of cassowaries.
F. H. J. Crome and L. A. Moore
Transect score 0 (absent): no droppings, sightings or any other record Transect score I (low): 1-4 droppings per km on either survey; if no droppings then a sighting and/or a bird heard and/or tracks. Transect score 2 (moderate): 5-10 droppings per km o n either survey; other signs may or may not be present. Transect score 3 (moderate to high): 11-15 droppings per km on either survey; other signs may or may not be present. Transect score 4 (high): 16 or more droppings per km on either survey; other signs may or may not be present. Transects were then grouped into 77 regions, each arbitrarily designated according t o some prominent local feature such as a mountain range, a river system, a city and its environs or a plain. Each region was given a 'regional score' representing an increasing scale of cassowary activity/ abundance, according to the transect scores, the results of the public survey, information from the literature and personal knowledge. Regional score 0 (absent): all transects score 0 and no other evidence of cassowary presence. Regional score 1 (low): at least half the transects scored zero and the rest scored 1. Other evidence birds heard and/or seen, but limited. Regional score 2 (low to moderate): at least half the transects scored 1 and/or at least one scored 2 and/or public responses indicated regular sightings, including groups of birds. Regional score 3 (moderate): at least half the transects scored 2 and public responses indicated that birds were sighted regularly. Regional score 4 (moderate to high): at least one transect scored 3 and/or public responses indicated birds were frequently sighted and moderately common. Regional score 5 (high): at least one transect scored 4 and/or public responses indicated birds were frequently sighted and common.
Measurement of habitat and other factors During the survey we collected information on four features to which cassowary presence may relate: (i) altitude-the abundance at Mission Beach and the excess of records from the lowlands suggest preference for lower altitudes; (ii) steepness of terrain-casual observations indicated preference for flat areas; (iii) degree of structural disturbance from cyclones and logging-magazine articles and the popular press report a strong negative reaction to habitat disturbance; (iv) feral pigs-popularly implicated as competing with cassowaries, eating their eggs and causing a decline. These four features were estimated for each transect and the data for each arranged into an ordinal scoring system and individually compared with the transect scores using contingency tables. (i) Altitude. The altitude at the midpoint of each transect was measured with a pocket altimeter calibrated each morning at a known altitude and converted to a score as follows: score 1, sea level120 m; score 2, 120-400 m; score 3, 400-600 m; score 4, 600-750 m; and score 5, above 750 m. The scores correspond approximately to significant changes in slope. (ii) Steepness of terrain. An assessment was made of the slope of the general area through which the transect ran. Single or several accurate measures of slope were not considered worthwhile since we wanted an assessment of the difficulty of the terrain in general for a terrestrial biped such as the cassowary. The slope was scored as: score 1 (flat), general area flat to gently undulating; score 2 (sloping), general area sloping lightly (5-lo0) or hilly with similar light gradients; score 3 (moderately steep), very hilly, slopes over 10" and up to 20'; and score 4 (steep), slopes up t o 40". (iii) Structural damage to vegetation. At three points along each transect we walked 5-10 m at right angles to the path into the forest and assessed damage to the vegetation. The three were combined into a score representing an increasing degree of disturbance on the transect as a whole: score 1, canopy mostly continuous, none to few vine and lawyer cane tangles, fallen trees and debris sparse; score 2, canopy may be broken but no obvious large gaps, few-to-moderate vine and lawyer cane tangles, fallen trees and debris sparse; score 3, obvious canopy damage, large gaps present but sparse, moderateto-heavy development of vine and lawyer cane tangles, fallen trees and debris sparse-to-moderate; score 4, heavy canopy damage and many large gaps, heavy-to-continuous lawyer cane and vine tangles, many fallen trees and much debris; and score 5, heavy damage as in score 4 but heavy grazing virtually eliminated understorey. No transects with damage less than score 4 were also grazed. (iv) Pig activity. Pig activity was scored by counting wallows, areas of rooting, sets of prints and sightings, each of which was considered a sign, along the entire transect. Scores of damage were defined
Cassowaries in North Queensland
as follows: score 1 (none), no signs; score 2 (light), 1-3 signs per transect; score 3 (moderate), 4-10 signs per transect; and score 4 (heavy), more than 10 signs per transect.
Results and Discussion Current Distribution and Status The results of the survey are summarised region by region in Appendix 1 . Fig. 1 shows the localities of the regions and Fig. 2 those of individual transects, each with the regional score. The study area was well covered but the regions between Cairns and Babinda on the Malbon Thompson Range and the Bellenden-Ker range were not adequately surveyed. Cassowaries are widely but not uniformly distributed throughout the region, with the highest indications of their activity on the coastal plain and ranges. There are low-tomoderate signs of their presence in most places with 'hot spots' of high activity and presumably density in the Mission Beach district, Coquette Point and the Graham Range, Woopen/Badgery Creeks and the Wallaman Falls district. The 'hot spots' at Mission Beach and Woopen Creek suggest that there may have been an axis of high density along the coastal plain and foothills between the Russell and Murray Rivers. Most of the habitat there has now gone and cassowaries are restricted to the foothills and the remaining forest patches. At higher altitudes we found many more signs than expected and some upland regions, notably the Black Mountain Range, the Lamb Range to Davies Creek and the slopes on the north and east shores of Lake Tinaroo, appear to be important for cassowary populations. The responses to the public survey and examination of the literature on distribution were alarming because they indicated that cassowary numbers have been reduced and are still declining in all regions where agricultural and urban development
have been intense, including the Atherton Tableland and the vicinities of Cairns, Mossman and Kuranda. As well as having been eliminated from the large areas converted to agriculture on the coastal plain they have disappeared from the vicinities of Atherton and Danbulla (although one or two may still exist) and may have been eliminated from the Lower Goldsborough Valley, the floor of the Wyanbeel Valley, the Clohesy River region and the Cassowary Range (see Appendix I). No public responses indicated they were once in these areas but they are close to and connected by forest to nearby areas with birds, and the remaining altered habitats indicate that at least some of the original forest habitats were suitable for them. It is difficult to believe that they were not in these regions before. Isolated forest patches appear to be particularly prone to loss of cassowaries. They have gone from most of the forest patches on the Atherton Tableland, including Lake Barrine National Park and the Malanda Scrub, and only one or two are left in Lake Eacham National Park (see Appendix 1). Clearing and development have also isolated populations in the Mission Beach region, Coquette Point and the Moresby Range, Flying Fish Point, the Graham Range and around Ella Bay, Edge Hill/Mt Whitfield (almost in the centre of Cairns) and the Dagmar Range. These are all in areas of intense or potential urban and/or tourist development and are all well known to the public. Some regions in which we believed cassowaries were abundant produced fewer signs than expected and inexplicably had lower measurements of activity than their locations and the forest type would suggest. These included the Daintree region, Downey Creek and the Palmerston region. Cassowaries are absent from the large tract of forest on the Windsor Tableland but we believe this to be natural, the Daintree Gorge being a formidable barrier to cassowaries. We have no indication whether they are at Mt Halifax. The reasons for the variations in cassowary activity in different regions need further investigation but we believe that plant species diversity and history are the major determinants of distribution. The cassowary needs a year-round supply of nutritious fruit (Crome 1976) and it is likely that the abundance and diversity of fruiting plants in a district determine its value to cassowaries. History may be the only way to explain adequately the absence or low numbers of cassowaries in many apparently suitable areas such as the Palmerston region, the Windsor Tableland, the Goldsborough Valley and Downey Creek. Natural barriers may have
F. H. J. Crome and L. A. Moore
Cassowaries in North Queensland
prevented them from moving to places like the Windsor Tableland, but it is possible they were exterminated from the other areas. We know that clearing has eliminated them from around Atherton and that they were shot out of the Malanda Scrub and nearby forest patches in the 1960s (see below). Small numbers in other areas may be indicative of past and present hunting pressure by both Aborigines and Europeans.
Cassowary Numbers Benntrupperbaumer (unpublished data) has calculated, by observation of individual birds, a density in the Mission Beach 'hot spot' of 1 bird per km2. This would provide an upper limit for the number in the forests of the entire region of about 10000. Considering the survey results this must be an overestimate since density outside the hot spots must be lower, and the absence from some areas must also be taken into account. With an extremely rough, but, from the conservation point of view, generous assumption that regional scores 2 and 3 represent about half this density, then the number of birds is more likely to be, as a first estimate, between 1500 and 4000. We consider this an extremely low figure.
Habitat Factors No relationships could be found between individual transect scores and any of the four variables, indicating that cassowaries are not found mainly in the lowlands, do not mostly frequent flat areas, can survive in structurally damaged habitats and have not been significantly reduced in all areas with pigs. The vegetation in the Cooktown/Townsville region is complex with over 20 types of rainforest and several other habitat types
(Tracey 1982), their features and distribution being related to soil, altitude, climate (Williams and Tracey 19841, cyclones (Webb 1958; Unwin et a[. 1988), history and fires (Stocker and Unwin 1986; Unwin et al. 1985). We were unable to determine if cassowaries select particular types of rainforest because of the distribution of our transects over forest types, the confounding of altitude, soil and forest type and the time required to assess accurately the forest types (often more than one) on each transect. However, we do not see any a priori reason for believing forest structural type to be particularly important as a criterion for habitat selection by cassowaries. C. casuarius is not confined to rainforest and has been seen in mangroves (personal observation), Melaleuca woodland and various eucalypt woodlands and savannas (personal observation; G. Stocker, J. Kikkawa and A. Graham, Personal communication
; Mayr and Rand 1937; Stronoch 1981). At Mission Beach, birds forage on the small rainforest shrubs and trees that form an understorey in eucalypt and Melaleuca woodland, and on Pandanus sp. aLong Beach
es (personal observation). Thus, cassowaries can use a range of habitats but it is unlikely that populations or even individual birds can be maintained in non-rainforest habitats permanently, since the lower plant species diversity would be unlikely to supply the fruit that cassowaries need year round. It is likely, however, that cassowaries use a range of habitats in a rainforest region, searching areas, regardless of the structure of the vegetation present, for fruiting plants or water. At times of food stress such as after cyclones in the rainforest, fruit resources in other habitats may be essential. With our indices of structural damage to vegetation we cannot separate the effects of logging from those of cyclones, as both have occurred in many areas. Cassowaries can occur commonly in areas of vegetation that have received considerable structural damage. The Mission Beach region is the best example. It is undisputably the best area for cassowaries in the north, yet has been repeatedly hit by cyclones and heavily logged and we consider the vegetation highly degraded. Similarly, the two other areas with high-tomoderate regional scores, WoopedBadgery Creeks and Wallaman Falls/Mt Fox, have been heavily logged. The association of high regional scores with severe damage may be related to the damage in those areas producing abundant food resources and high plant species diversity but this would have to be tested. This apparent tolerance by cassowaries of structural disturbance to the habitat can be taken only so far. In the two areas with the highest damage scores where cattle grazing had
F. H. J . Crome and L. A. Moore
removed much of the understorey there were no cassowaries. Thus, there appears to be an intermediate level
of disturbance that favours cassowaries, but beyond that, severe damage, with loss of understorey and much of the canopy, results in their reduction or elimination. Logging has not eliminated cassowaries as has been suggested in the popular literature
; however, as with cattle grazing, if logging were pushed to a point where severe canopy and understorey reduction occurred it would then be detrimental.
The Decline of the Cassowary As indicated above cassowaries have declined or been eliminated from several areas. Moreover, for many others we have no historical record
s from residents, so cannot say how widespread the decline has been. This, combined with the fact that sign is generally low to moderate in most areas makes us pessimistic about the status of this species. We consider the cassowary to be the most threatened species in our northern rainforests. Sightings of cassowaries in an area does not mean they are secure since cassowaries live long, attaining over 40 years in captivity (Crome and Moore 1988), and individuals may stay in an area of many years, giving the impression of a healthy population. We believe the major factors causing their decline are loss of habitat, predation, traffic accidents
, disease and possibly competition and nest predation by pigs. Habitat loss The major period of habitat clearance has passed, but this has perhaps resulted in elimination of most of the best country for the species, i.e. the coastal plain and foothills between the Russell and Murray Rivers. The clearance would probably have been much greater had not large areas of forest been left as State Forests and Timber Reserves. Habitat loss, however, is still a problem on the coastal lowlands, where development is progressing rapidly. Of particular concern are the proposals for tourist development in those few and important areas where cassowaries have large populations such as at Mission Beach and Coquette Point. coastal development
for tourism, residences, roads, etc., is a major problem in the lowlands, and the possible flooding of areas of forest resulting from sea-level rises due to the greenhouse effect may become another. Management is difficult since the problem is one of attrition with no single area identifiable as being a 'key' to cassowary conservation and therefore no individual development could be regarded as 'the' problem for cassowaries; instead the totality of development is 'the' problem. We offer no panacea for this but have noted the lack of coastal zone planning in the north. With planning, the placement and engineering of roads and signs, the location of developments and the amount, shape and position of forest retained with them, could take into account the requirements of cassowary conservation. Predarion by humans Cassowaries are the largest native game in the Australian and New Guinean rainforests and have been an important part of Aboriginal life and diet for centuries. Gilliard and Lecroy (1967) describe extensive cassowary trading around New Guinea and between islands and their importance as a cultural commodity beyond simple food value is well known (Reid 1981/82, 1982). In some places they are semi-domesticated (Reid 1978/79, 1987). . Thompson (1935) says of the cassowary on Cape York that '. . elaborate rituals and taboos are associated with the eating of its flesh'. Like New Guineans, Aboriginal Australians sometimes kept tame cassowaries (Frizelle in White 1913) and they still hunt them, catch and fatten the young for later slaughter and collect the eggs for food, but we do not know how frequently (unpublished observations). Predation in the past could have been heavy but we have no data on this. They are also shot by other Australians. We have reports that they were shot out of Malanda scrub (A. Macauliffe, personal communication) and were shot around Millaa Millaa in the 1960s (R. Hill, personal communication). In 1987 we found a bird that had been shot and, in 1988, a freshly butchered carcass.
Cassowaries in North Queensland Predation by dogs We received one report (J. Madsen, personal communication) of dingoes attacking a quarter-grown bird in the Millaa Millaa district, and numerous reports of domestic dogs harassing the birds. We observed this ourselves on five occasions. While one or two dogs are not a real problem for adult cassowaries, packs may harass them until exhausted, but we d o not known whether or not this occurs frequently. Chicks and subadults, on the other hand, fall prey to dogs and we have received first-hand accounts of this. In settled or built-up areas such as Mission Beach and the Daintree, dogs could also have subtle effects o n feeding, movements and behaviour. Pig dogs are reputed to be the worst offenders but we have no evidence that this is so. Traffic accidents Traffic is a major killer of cassowaries in the Mission Beach district, and possibly in other areas. In the south Mission Beach area, J. Benntrupperbaumer (personal communication) recorded 17 adults and chicks killed by cars out of 26 known to have died from February 1986 to August 1988. A significant problem is the attraction of birds to roadsides where they are fed by people; thus they frequent roads and are then in danger of being run over. Disease Of particular concern is the presence of diseases. We found two dead birds in the Mission Beach area. One had died of a tuberculosis-related disease and the other of a fungal disease like aspergillosis. We found a third bird in April 1989 which showed similar symptoms to the tuberculosis-infected bird but specimens were not distributed to veterinarians. The possibility of diseases in some populations is serious in view of the disease-related decline of the Hawaiian bird fauna (van Riper et al. 1986). Pigs Pigs are known to destroy nests and we received first-hand accounts of this (W. Dunn, personal communication, J. Madsen, personal communication). Being ground-feeding omnivores they are also important potential competitors of cassowaries. Several respondents in our public survey were convinced that pigs were responsible for the decline of cassowaries in the Millaa Millaa district (R. Hill, personal communication, J. Madsen, personal communication), the Daintree (P. Mason, personal communication), the Graham Range (W. Dunn, personal communication), and a 'problem' for the birds in the Walter Hill Ranges (D. Hogan, personal communication). If they are so highly detrimental there is a real problem since pig control in rainforests may be a difficult task. We were unable to determine from our survey any simple negative relationship between cassowary and pig presence but the matter needs further investigation. Movements The importance to management of understanding cassowary movements cannot be stressed enough. Cassowaries have regular routes to and from feeding sites and are presumably contained within a home range but they may also undertake regular movements (Frizelle in White 1913; personal observation). Crome (1976) and Stocker and Irvine (1983) recorded a scarcity of droppings early in the year suggesting that at such times birds move into different areas, perhaps to other parts of the forest or other habitats in the same district that were not sampled. We have no evidence that the species moves altitudinally, but it is important to determine this. Cassowaries can move long distances but it is not known how frequently. The only data are records of birds west of Kennedy. Atkinson (1984) reported sightings of a bird at Glen Dhu station '60 miles' from the rainforest after the 1918 cyclone and, in our public survey, W . Sneath (personal communication) saw a bird on the same station on 26 June 1988 'about an hour's drive west of Kennedy'. There are three records from this same area in the RAOU Atlas database.
F. H. J. Crome and L. A. Moore
Open country is not an absolute barrier to cassowary movements. Reports after cyclone Winifred indicated that birds were leaving the rainforest and wandering open areas on the coastal plain in search of food and we have several accounts of birds moving through grazing country on the Atherton Tableland and cane fields on the coast (personal observation; G. Anderlini, L. Christoph, personal communication) and savannas on Cape York. Open country is probably a partial barrier, however, if only because of greater exposure to shooters, cars and dogs. Ordinary cattle fences d o not appear to be a complete barrier. A pad of a cassowary at Mission Beach goes over a fence, and a bird near Lake Eacham was seen running full speed across a paddock and slipping through a tightly strained fence (D. Kitchin, personal communication). Ramsay (1876) records a captive bird jumping a '6 foot high' fence. Cassowaries have frequently been recorded bathing and swimming (C. Holloway, J. Madsen, personal communication; MacGillivray 1917; Thompson 1935; Anon. 1965) and Rand and Gilliard (1967) saw a Casuaris casuaris sclateri crossing the half-mile wide Fly River in New Guinea. Water therefore would not act as a complete barrier to movement but may reduce it. It certainly increases the risk of being devoured by crocodiles!
Conservation and Management The conservation prognosis for cassowaries is not good in lowland, tableland and other settled regions, but the areas of montane forest maintain populations. However, if, as we postulate, the cassowary is an important component of forest dynamics, being a major disperser of seeds of rainforest plants and possibly the only long-distance disperser of some Tree species
with large fleshy fruits, then areas from which cassowaries are lost may undergo dynamic shifts in plant species composition
. The conservation of cassowaries, therefore, may not only be a matter of the preservation of a single, albeit spectacular, species but also an essential part of maintaining forest dynamics. This makes conservation more difficult and not adequately soluble by means of a reserve system. We have made 30 recommendations for management and future research to QNPWS and ANPWS (Crome and Moore 1988). Many of these are of local significance only but the major ones are outlined as follows. 1 Conservation should be considered in an overall context for the region and aim to retain cassowaries in all areas of forest where they now occur. There has been an undue emphasis on the Mission Beach region. 2 Conservation should not be solely concerned with rainforests but rather the complex of habitats that cassowaries use. 3 If cassowaries are lost from tracts of forest then they should be reintroduced. 4 The movement of cassowaries between the already fragmented patches of forest should be facilitated. Despite the fact that they can cross open country the extent of the cleared and developed areas makes it hazardous. A rural Nature conservation
program involving the replanting of suitable rainforest species would make movement safer and provide extra food resources for cassowaries. 5 A coastal zone development plan should be devised for north Queensland and the retention of cassowaries in all parts of the coastal lowlands in which they still occur should be part of a planning and development strategy
. 6 The Mission Beach area, Coquette Point and the Moresby range, Woopen/Badgery Creeks, and the Wallaman Falls regions should receive special consideration and protection, as should the isolated populations in the Edge Hill/Mt Whitfield area, the Graham Range and Ella Bay, and the Dagmar Range. An assessment should be made of Development proposals
in Cairns for the Edge Hill/Mt Whitfield area to determine how much of the area will be saved from urban development. If only limited habitat can be retained the cassowary population may have t o be relocated in the ranges to the west. Development on either side of Innisfail in the Graham Range and Ella Bay area, and Coquette Point and the Moresby Range should be carefully regulated. 7 The cassowary has a high public profile, many hand feed the birds and several tourist ventures want their own 'wild birds'. This is a probIem (QNPWS, personal communication)
Cassowaries in North Queensland
and will become more severe. Hand feeding should be discouraged and where an individual or tourist venture has and wants to fence land used by cassowaries adjoining State Forest or National Park, the fence should be one that cassowaries can negotiate, e.g. an ordinary cattle fence. 8 Habitat where cattle activity has been intense is unsuitable for cassowaries; cattle should as far as possible be excluded from rainforest. 9 Translocation of cassowaries is an important management tool, both for the reasons outlined above and for removal of problem birds from residential area
s. When reintroductions are necessary in the future, a stock of known origin captive birds would be useful. However, for a release to be successful, three things must be known. Will the bird (1) fit into the cassowary population in the new area; (2) be disease free; and (3) be of the same genetic stock as the birds in the new area. Our knowledge of cassowary social organisation, behaviour and movements is insufficient yet to answer (1) and we know nothing about the genetic differentiation within north Queensland populations. Disease is potentially a major problem and could have catastrophic effects if introduced into small populations. Therefore, we recommend that there be a survey of the genetics and disease load of populations of cassowaries and the establishment of a Research Project
into their behaviour, social organisation and the effects of pigs. 10 Finally, this study was restricted to the Cooktown/Townsville region. The status, both genetic and ecological, of the populations on Cape York remains unknown. In view of the development proposals for the Cape the populations in this region should be surveyed.
Acknowledgments This was a cooperative project between CSIRO, QNPWS and ANPWS. We thank Ross Hynes of the QNPWS for his encouragement, support and efforts for cassowary management. Thanks also to the members of the Tully and the Innisfail Wildlife Preservation Societies for their encouragement and moral support, and to Ben and Joan Benntrupperbaumer for everything. Although we cannot name them, we thank the 81 people who responded to our public survey. The staff of QNPWS in north Queensland took a particular interest in this project and generously shared their experiences, views and opinions with us. In particular, we thank Buzz Simmonds, Darren Storch, Brent Vincent, Karl Seiner, Dave Bender and the staff of the Cardwell Information Centre. Finally, we thank Graham Harrington, Jiro Kikkawa, Denis Saunders, Graeme Caughley and Andrew Graham who constructively criticised drafts of this manuscript.
References Anon. (1965). Cassowary swimming in Queensland. Bird Observer 65, 4. Anon. (1976). Fauna of Eastern Australian Rainforests: preliminary report o n sites in Mid-Eastern and North-Eastern Queensland. (Queensland Museum: Brisbane.) Atkinson, R. L. (1984). Bush Tales and Memoirs. (Pinevale Publications: Townsville.) Barnard, H . G . (1926). Birds of the Cardwell district, Queensland. Emu 26, 1-13. Bourke, P . A. (1947). The Atherton Tablelands and its avifauna. Emu 47, 87-116. Crome, F. H . J. (1976). Some observations on the biology of the Cassowary in northern Queensland. Emu 76, 49-58. Crome, F. H . J., and Moore, L. A. (1988). The Southern Cassowary in North Queensland-A pilot study. Vols I-IV. A report prepared for the Queensland National Parks and Wildlife Service and the Australian National Parks and Wildlife Service. Gilliard, E. T., and Lecroy, M. (1967). Annotated list of birds of the Adelbert Mountains, New Guinea. Results of the 1959 Gilliard Expedition. Bull. Am. Mus. Nut. Hist. 138, 51-82. MacGillivray, W. (1917). Ornithologists in North Queensland. Emu 17, 63-87. Mayr, E., and Rand, A. L. (1937). Results of the Archbold Expeditions 14. Birds of the 1933-34 Papuan Expedition. BUN. Am. Mus. Nut. Hist. 73, 1-248. Ramsay, E. P . (1876). The cassowary. Proc. Zool. Soc. Lond. 1876, 119-23. Rand, A. L., and Gilliard, E. T. (1967). Handbook of New Guinea Birds. (Natural History
Press: New York.) Reid, B. (1978/79). History of domestication of the Cassowary in Mendi Valley, southern highlands, Papua New Guinea. Ethnomed. 3/4, 1407-32.
F. H. J. Crome and L. A. Moore
Reid, B. (1981/82). The Cassowary and the Highlanders. Present day contribution and value to village life of a traditionally important wildlife resource in Papua New Guinea. Ethnomed. 7, 149-240. Reid, B. (1982). Cassowaries as currency. N.Z. J. Ecol. 5, 152-3. Reid, B. (1987). food intake
and growth rate
of Cassowary chicks Casuarius spp. reared at Mendi, Southern Highland Papua New Guinea. Int. Zoo. Yb. 26, 189-98. van Riper, C. 111, van Riper, S. G . , Goff, M . L., and Laird, M. (1986). The epizootiology and ecological significance of malaria in Hawaiian land birds. Ecol. Monogr. 56, 327-44. Stocker, G. C., and Irvine, A. K. (1983). Seed dispersal by cassowaries (Casuarius casuarius) in North Queensland Rainforests. Biotropica 15, 170-6. Stocker, G. C., and Unwin, G. L. (1986). Fire and the functioning of tropical plant communities
. In 'Tropical Plant Communities'. (Eds H. T. Clifford and R. L. Specht.) pp. 91-103. (University of Queensland
: Brisbane.) Storr, G. M. (1953). Birds of the Cooktown and Laura districts. Emu 53, 225-48. Stronoch, N. (1981). Notes on some birds of the Bensbach region, Western Province. Papua New Guinea Bird Society Newsletter 1981, 178-9. Thompson, D. F. (1935). Birds of Cape York Peninsula. (Government Printer: Melbourne.) Thorsborne, M., and Thorsborne, A. (1988). Hinchinbrook Island. The Land Time Forgot. (Weldons: Sydney.) Tracey, J . G. (1982). The vegetation of the Humid Tropical Region of North Queensland. (CSIRO: Melbourne.) Unwin, G . L., Stocker, G. C., and Sanderson, K . D. (1985). Fire and the forest ecotone in the Herberton highland, north Queensland. Proc. Ecol. Soc. Aust. 13, 215-24. Unwin, G. L., Applegate, G. B., Stocker, G. C., and Nicholson, D. I. (1988). Initial effects of cyclone Winifred on forests in north Queensland. Proc. Ecol. Soc. Aust. 15, 283-96. Webb, L. J. (1958). Cyclones as an ecological factor in tropical rainforest, north Queensland. Aust. J. Bot. 6, 220-8. White, H. L. (1913). Notes on the Cassowary (Casuarius a u s t r a h , Wall). Emu 12, 172-8. White, S. R. (1946). Notes on the bird life of Australia's heaviest rainfall region. Emu 46, 81-122. Williams, W. T., and Tracey, J. G. (1984). Network analysis of Northern Queensland tropical rainforests. Aust. J. Bot. 32, 109-16.
Appendix 1 . Summary of dislributional information on southern cassowary in north Queensland
Regions where we had no transects but from which we received information from the public survey or for which there are literature records do not
receivc a regional score
1 Mt Cook 2 Anna River Flats 3 Big Tableland/Mt Amos/ Mt Hedley 4 Cedar Bay and Mt Finlay 5 Wyalla Plain near Mt Muir 6 Shipton's Flat 7 Mt Misery 8 Bloomfield Valley 9 Daintree R. to Emmagen Creek. Foothills and lowlands
10 Belinda Creek 11 Whyanbeel Valley 12 Dagmar Range 13 Windsor Tableland 14 Mossman Gorge 15 Rex Range 16 Cassowary Range
No. signs. Steep rocky country. Absent
No signs. Absent
5 Records from Big Tableland and east of Rossville (Storr 1953)
Reported at Mt Finlay (Anon 1976). Local residents see birds but they are
No evidence they are present but they may be in ranges t o the west towards
None on the flat proper but occur in surrounding ranges (Roberts, personal
communication). Much cattle damage
Two footprints on six transects. Cattle grazing
Status indefinite. One record from public survey of bird drinking, other
respondent said birds never seen
The transects indicated a regional score of 1. The public response was good
but it was of individual birds at Cow Bay, Mt Alexander Range, Noah
Creek, Emmagen Creek and the ferry. Several respondents said birds had
declined due to pigs. Only two respondents reported several in an area
(B. Galbraith at Shipwreck Bay and J. St John Wood at Cape Kimberley).
There were no responses indicating birds were common
Report from P . Halsebosch of droppings
Locals d o no1 see birds, but they may be present in surrounding ranges
Birds still occur a1 Rocky Point (S. Kingston, personal communication) but
it is unlikely that there are many there. Early records show birds pqesent
at Mossman (White 1946)
Six years' intensive work by us shows there are none here
None in Gorge and none seen by QNPWS ranger
There were good signs in this region
This region is highly modified and we have no indications birds are present.
L. Kleinschmidt (personal communication) saw no sign o n a recent walk 2
in this region to the Mowbray Valley
NO." 17 Mt Lewis 18 Jullaten 19 Black Mountain Range to Kuranda 20 Foothills and lowlands north of Cairns 21 Cairns district 22 Clohesy River (Shotee Creek) 23 Copperlode area to West Cairns 24 Yarrabah/Pine Creek 25 Malbon Thompson Range" 26 Goldsborough Valley/Mulgrave Valley 27 Behana Gorge
Appendix 1 (continued)
~ a t a ~ Regional scoreC
86 02 20 22 24 00 81 20
Sign right to the top of the range
Birds seen at the junction of the Jullaten and Black Mountain roads
(C. McCracken, personal communication). Jullaten area mostly cleared
and sightings likely to be of birds from surrounding Black Mountain,
Mt Lewis and Rex Range
Cassowaries occur all along this range to Kuranda and the top of the high-
way to Cairns. They are declining or retreating from the vicinity of
Kuranda and are now rarely seen within 5 krn of Kuranda
The lowlands north of Cairns to Mossman have no birds according to ourpersonal observations
spanning 15 years and discussions with locals. We
have one positive public response of a bird behind the Hartley's Creek
school. There must be birds in the foothills of the Black Mountain range
as would be expected
Birds occur on the outskirts of Cairns itself and in the ranges to the west.
There is a group of birds in the Freshwater/Edgehill/Brinsmead Rd
area well known to and fed by locals. They are almost certainly isolated
from birds in the Redlynch/Kamernnga district by major roads and
Although only two transects are in this region we have walked it exten-
sively. There are no signs of birds, much clearing and heavy cattle activity
in the rainforest
We received few responses about this region relative to Cairns' population
We did not have time to survey the southern part of this important region
nor the Aboriginal reserve. Public responses indicated reasonable numbers.
Regional score may he higher. We know cassowaries are still shot here
We have been unable to survey this potentially important region as yet.
It is likely to have a moderate-to-high regional score
They are absent except in the upper parts of the valley on the slopes of
Bellenden Ker. The country is very dry on the valley floor and western
The valley itself is very dry with mixed rainforest/eucalypt associations, but
they may occur on the higher slopes of the valley
elle end en-~er~ Babinda/Boulders Graham Range Josephine Falls/Mirriwini Waugh's Pocket Woopen Creek/Badgery Creek Lamb Range/Mt Haig/Davies Creek/Robson Creek Danbulla State Forest Severiu Creek-Cathedral Fig to top of Gillies Highway AthertodKairi district Lake Eacham Lake Barrine Gadgarra Wongabel Mt Baldy Crater/Longland's Gap/Seamark Road Lamin's HiWBartle Frere/Topaz Malanda district
We were unable to survey this extensive region
Birds are present but we need more surveys here
Public responses indicate that there are reasonable numbers of birds here 4
but they are declining
Birds may have been released into National Park 1351. The situation on the
south slopes of the Bartle Frere massif needs to be clarified
Birds present but declined since 1956 cyclone (G. Anderlini, personal
One of the best regions surveyed and known for many years t o be a good 0
area for cassowaries
The ranges north of the Atherton Tableland and the hills around Tinaroo F
have surprisingly many signs of cassowaries
This region is Danbulla proper south-west of Lake Euramoo and based o n
Python Hill. We last saw birds there in 1983. Present indications are
that they are absent or transitory
Although fragmented in parts, birds with chicks are regularly seen There used to be birds in the 'Kairi Scrub' (Bourke 1947) but they n o longer exist here. There is virtually no habitat left We failed to find birds here but know that at least one still exists. They were common in the early 1970s (personal communication) and bred here in 1983 We are fairly sure they have become extinct here. We saw no signs, and P . Sweedman has not seen any, although they were common in the 1950s (S. and K. Turner, personal communication) Surprisingly many signs in this highly disturbed region Known to be absent There appear t o be few birds here and they may be declining A much logged and disturbed region. The National Park needs more surveying A bird has been seen as high as Broken Nose on Bartle Frere, but they have retreated from the farming regions Cassowaries have been almost exterminated from the district and are absent from most of the remaining forest patches (personal observation). They were shot out of the Malanda Scrub in the 1960s (McAuliffe, personal communication). They are still present but rare near Tarzali. Used to be 'moderately common' in the 1920s (G. Crowther, personal communication)
NO.* Millaa Millaa and south-west Atherton Tableland North Johnstone/Beatrice River/ Elinjaa Creek region Mt Fisher and the Maalan Lower Palmerston/Henrietta Creek/Johnstone Gorge Mt Father Clancy Suttees Cap Lower Tully Corgc Coolamon Creek/Tully Falls Major's Mountain/Ravenshoe district Koombooloomba Cameron Creek (north-west of Kirrima) Kirrama Range Seymour Range/Flying Fish Point/outskirts of Innisfail Coquette Point/Moresby Range South Johnstone District/Pingin Hill/Gregory Palls/Meena Creek Downey Creek
Appendix 1 (continued)
p p Comments
Indications are that they were common until the 1960s and have declined drastically since then. Several of our respondents considered pigs to be responsible for the decline. They are still present in the ranges around
Birds present Most sightings are on the Maalan at 'Lager Bend' No sign was found o n any transect and public responses indicated decline since the 1950s No sign on transects but present A large region of continuous forest, more surveys needed Data not very reliable Odd birds present throughout the Coolamon Creek area. More surveys needed in western parts
Present in district Birds are thinly diqperscd through this region
This region may be similar to Koombooloomba but more surveys are needed Low but regular sign from throughout range Birds are still here and possibly in reasonable numbers but are declining. Thcy were common at nearby Oaradgee in 1957 (D. Soley, personal communication) An important region with birds around Coquette Point well known to residents. Thcy are declining and under threat This important region has not been adequately surveyed. Our respondents indicate good numbers of birds in higher country but there appears to have been a drastic decline of birds associated with habitat clearing. They once were seen up to the Innisfail aerodrome (D. Soley, personal communication) Despite the fact that this valley appears to be excellent country there were few signs of birds
Macnamee/Liverpool Creek Cowley Beach Mission Beach Nyletta Creek Mt Tyson Jarra Creek Meunga Creek Southcrn Cardwell Range/ Mt Graham to Mt Leach Hinchinbrook Island Ingham District
North Wallaman, near Black
Adder Flat and National
Blue Water Creek, south-west
of National Park 547
Arnott Creek/Gardiner Mountain
Although showing the effects of cyclone Winifred this region has good sign O ! Appear to use several habitats in addition to rainforest (A. Graham, personal
The best region in the north Heavily affected by cyclones No recent sightings but seen between 1979 and 1983 ( 0 . Zeirner, personal communication)
''E-. z S
Fauna surveys in 1985 by CSlRO showed birds present with low-to-moderate 5
More data arc likely to indicate a higher scorc
This rugged region may be important but we were unable to visit it during
Present (Barnard 1926; Thorsborne and Thorsborne 1988)
Wc drove extensively throug-h this region and failed to find suitable habitat
to establish transects between Ingham and the coast. Birds probably
occur in the ranges to the wcst
Data poor, this region south of the Murray River undoubtedly has a higher
Good region. A male with chicks was seen during the survey
Birds easily found and regularly sccn
Present, public survey indicatcd breeding
Regular sighiings of what might be a few wide-ranging birds
Absent but more transects needcd. Much of region hard to access and birds
mav. be .nresent
A Refer to Fig. 1.
First column is the number of transects and the second is the number of responses to the public survey.
No regional scorc given if no transects done.
These regions are major mountain regions and have been included for completeness.