The biology of the koala has received considerable attention, although comparatively little is known of its ecology. The issues of habitat protection are similar throughout the range of the koala, however, there are important differences in management perspective as dictated by local events. The widespread local extinctions that occurred in mainland Victoria resulted in a research concentration to develop specific management plans that relate to the re-establishment of the mainland populations. This situation has not occurred in either New South Wales or Queensland. Studies of koala social organisation were undertaken on French Island (Mitchell 1989). However, an island situation provides limited scope to investigate adequately the factors determining koala movements. It appears that opportunities to disperse were severely restricted as translocation is required to reduce over-browsing on French Island.
This thesis addresses broad management issues by considering:
• factors that affect population processes, e.g. population structure and disease (Chapter 3); social structure (Chapter 4) and;
• habitat use (Chapter 5);
in a fragmented rural landscape. The final chapter proposes a series of requirements for a long-term management plan. This accepts that management cannot be successful if considered only on a farm by farm basis, but requires a coordinated system of donor and acceptor habitats within a region.
The study was undertaken at Mutdapilly, 25km south-west of Ipswich in the Bremer River catchment, an area dominated by agricultural activity. The study site was within a predominantly grazing area (beef and dairy cattle) and extensive clearing caused widespread loss of forest and wood land. Remnant vegetation occurred as patches that varied in size from less that one hectare to blocks of 50 to 1 00 hectares. Eucalyptus tereticornis and E. crebra were the dominant species and both were used by koalas for food.
Regular radio-tracking was undertaken using a simple searching procedure rather than triangulation as this provided information on clinical signs of disease, movements, social organisation and tree utilisation. During capture periods, koalas were swabbed (ocular and urogenital) and subsequently tested for the presence of Chlamydia psittaci. Infection was presented as either a urogenital or ocular condition with males showing an increased level of concurrent urogenital infection compared to females. Females were more likely to have a urogenital infection with lower prevalence of ocular infection compared to males. Fertility, expressed as the proportion of independent females with young, was 40-50% (White and Timms 1994) and was lower than comparable studies. Fifty percent of non-parous females had clinical signs of disease.
Home range size was determined using a modified harmonic mean measure (Dixon and Chapman 1980) and was significantly greater than has been reported elsewhere for both males and females (Hindell and Lee 1988, Mitchell 1989). Average home range size using the 70% isopleth was 8 ha and 3.5 ha for males and females respectively. Allometric constraints explain the difference in home range due to body weight between the sexes in koalas. This does not explain why home range sizes are larger in Queensland compared to Victoria. Small home ranges recorded in Victoria may be due to demographic constraints such as high population density, as the densities reported on French Island were found elsewhere on mainland Victoria (Martin 1985a,b,c). The density at the Mutdapilly site, although much lower than those reported in Victoria, was not unusual compared to other sites in south-east Queensland (White and Kunst 1990) and elsewhere in the Bremer River catchment (White and Kunst 1988). Social organisation could therefore be more important than latitudinal or allometric relationships in determining home range size.
Further development of the model in which females, rather than males, were the primary influence on social organisation and population density (Hindell 1984) is presented. As females do not compete among themselves for mates it would be advantageous to maintain access to high quality food trees to ensure the survivorship of offspring (Emlen and Oring 1977). The mechanisms exist for koalas (male and females) to use single trees exclusively and this has also been shown to some extent by Hindell and Lee (1987). Sharpe's (1980) contention that ·female social behaviour was routine compared to males underlies the important differences in resource priorities between the sexes (males for mates and females for food) . A minimalist approach to resource defence, relying on indirect methods, would be an appropriate solution for an animal that has a low capacity for sustained activity.
Scent marking becomes especially important in males due to the large home range size and degree of overlap. This is expressed in the development of a sternal scent gland. Social organisation of males was shown to be a polygynous breeding system (Chapter 4, Mitchell 1989). Male-male aggression, although not resulting in strict territory defence, produces a lower density of males that compete for access to females. The only way in which males can make a contribution to the gene pool is by surviving to reproductive age and then successfully gaining access to at least one female. Mating with females in habitats of high quality would be important to increase survivorship of any offspring sired. Frequent home range shifts were observed for young males (Chapter 4) indicating that it may be difficult to find a breeding site.
There was a disproportionate selection of large trees since foliage abundance increases with tree size. A secondary effect of the loss of optimal habitat would be an increase in disease as C. psittaci has been in contact with koalas for a long time (Weigler et al. 1986b, Chapter 3). When exposure level is high (i.e. a high prevalence of infection within the population) adequate nutrition provides a reproductive benefit by reducing the incidence of urogenital tract and ocular disease (Lanyon and Sanson 1986b).
Koalas were not reliant on corridor systems as many regularly moved between habitat remnants across grazing paddocks. A similar finding was made by Prevatt (1981) in an urban-rural area in Victoria. Observations undertaken by White and Kunst (1988) and in this study found that koalas made frequent use of isolated trees in paddocks and used multiple patches. Radio-tracking has also revealed that long range movements (> 2 km) were frequent and that corridors were unimportant, but were treated as available habitat.
Management must therefore be considered regionally and a four level system of protection has been proposed for agricultural areas (mainly grazing), Table i. The establishment of bio-diversity reserves while providing for other species, may be unnecessary for long-term koala conservation. The next level of remnant size (source) are large within-farm blocks of high quality habitat. These areas probably have only a minimum affect on the protection of the bio-diversity of the district. The establishment of a source affects the viability of small remnants associated with it (satellites). Source remnants require identification, mapping and a high degree of protection throughout the conservation area (e .g. Bremer River catchment) . It is important that these areas be made as extinction-proof as possible and would need to be at the minimum size range of 100-200 ha.
Table i. Proposed division and purpose of remnant types for the conservation of koalas in rural areas.
Type Size Purpose Secondary Purpose
Bio-diversity As large as Maintenance of Extinction-proof source
possible ecosystem bio- diversity of dispersing koalas
Source 100-200 ha Source population Potential habitat for
of dispersing other species, farm
individuals management, apiary sites
Satellites 2-10 ha Overflow area for Prevention of salinisation,
dispersing individuals erosion, shelter for cattle,
Increase overall apiary sites.
Isolated Link between sites Shelter for cattle Trees
Small remnants provide an overflow for source areas and act as stepping-stones to distant patches. Their main function is to provide an increased abundance and would ideally be a minimum of 5 to 1 0 ha. These may be present as discrete clumps or groups of close patches and be used to provide shelter for cattle, reduction in salinisation and prevention of erosion. Isolated trees should be maintained in areas where the availability of small remnants is low or impractical, e.g. cultivation. These provide either an extension of home range or serve to reduce the inter-patch distance between isolated clumps.
Anatomical and behavioural characteristics of the koala suggest that its habitat requirements are unspecialised compared to many other arboreal folivores. These characteristics include a highly developed dental and digestive system (Lanyon and Sanson 1986a and b) that can cope with high levels of allelochemicals and tannins (Cork and Sanson 1990, Osawa 1991), low nitrogen (Harrop and Degabriele 1976) and high fibre content (Cork and Warner 1983). This is coupled with a low field metabolic rate and because of the above anatomical features, an extensive potential food supply. Other behavioural modifications that assist koalas include, leaf selection of an otherwise unpromising diet and indirect, low energy methods of maintaining space (e.g. scent marking and vocalisations, Chapter 4). The koala places no other physical demands on its environment such as the need for den sites, nest sites, display areas etc. It does not utilise the understorey and its mobility does not appear to be compromised by lack of intervening vegetation (although here, at some stage there will be a limit). It is therefore suggested that in comparison with other arboreal marsupials, it should be relatively easy to provide habitat for koalas within rural areas. However, the presence of koalas would not be a useful indicator of the suitability of the vegetation for other arboreal folivores and has no bearing on its suitability as habitat for ground dwelling species.