Conservation of cave fauna in Australia

Stefan Eberhard 1, Elery Hamilton-Smith 2

1Cave biologist, 745 Leslie Rd, Leslie Vale, Tas 7054, Australia, 2Rethink Consulting, P O Box 36, Carlton South, Vic 3053, Australia

Introduction

Despite a high level of interest in caves during the 19th century, and some promising early beginnings in the study of cave biology in Australia, research on cave fauna has been extremely limited until the last 30 years. Until very recently, Australian cave communities have been perceived as being depauperate compared with the well-studied communities of Europe and North America, especially with regard to the representation of taxa that display a high degree of troglomorphism. A higher proportion of troglobitic species, perhaps comparable with those in some Northern Hemisphere countries, was only recognised in Tasmania, with its more humid climatic history and episodes of glaciation. However, during the last decade in particular this traditional perception has been overturned by dramatic discoveries of highly diverse communities containing very diverse and ancient troglomorphic taxa. It must be appreciated that most Australian karst areas have not been subject to any thorough or systematic biological examination, and that much of the known fauna remains undescribed. This in itself is a major impediment to conservation planning and management. Further, some cave communities have been lost entirely, and a great many more have been seriously degraded or are threatened by human activities. This paper describes the major conservation issues and the nature of impacts threatening Australian cave fauna, and discusses the conservation strategies which have been applied to them.

Background

In the prehistoric past Australia was part of the southern super-continent Gondwana, and was attached to Antarctica. These lands shared a Gondwanan flora and fauna, and when the final separation between them occurred (45 million years ago), both lands were well watered and supported cool temperate and subtropical forests (White 1994). As Australia separated from Antarctica and drifted northwards, the continent began to dry out, so that its forests contracted and were replaced with arid-adapted vegetation. The formation of the Antarctic ice cap 15 million years ago saw the beginning of a series of marked climatic fluctuations that have greatly stressed the Australian (and other Gondwanan) flora and fauna. Warm and wet interglacial periods alternated with very dry, cool and windy glacial stages, but only a small area of the Eastern Highlands and Tasmania were subject to extensive ice cover. The cyclic fluctuation of the climate in the late Cainozoic, superimposed upon a generally increasing and spreading aridity over the last 2.4 million years, provided conditions under which subterranean refugia played an important role (Flannery 1994).

Many Australian karst landscapes are quite ancient, and may have been available for colonisation by animals over a very long period of time. Palaeokarst features of Devonian age (350 to 400 million years ago) are known in New South Wales for instance, although subsequent episodes of burial, marine transgressions, flooding, sedimentation and major climate changes may well have eliminated faunas from that time (Osborne & Branagan 1988). The occurrence of highly modified stygobiontic taxa belonging to morphologically conservative groups, such as the Anaspidacea, imply that subterranean freshwater environments have been continuously available for a very long period, possibly since the end of the Palaeozoic era. The distribution of remipede Crustacea and their associated fauna can be explained by cave colonisation along the shores of the Tethys Seaway during the Triassic and Jurassic (225 - 160 million years ago), whereas, during the early Cretaceous, the plate that is now Western Australia formed the eastern shore of Greater Tethys (Yager & Humphreys, in press). Bats have been present in Australia since the Miocene, and they are found in the cave fossil record of some two million years ago. Available methods of dating cave sediments are generally restricted to the last few millions of years, but the occurrence of isolated caves of probable phreatic origin situated high above valley floors of the present-day landscape suggest great ages for such sites.

The composition of Australian cave communities at the higher systematic levels is broadly similar to that of other continents, but frequently involves taxa that are uniquely southern in origin, such as the phreatoicidean isopods and anaspidacean syncarids. The communities naturally contain a high proportion of taxa with Gondwanan affinities, although others, such as the Remipedia, have Pangean or Tethyan affinities. Like cave faunas elsewhere, many taxa are phylogenetic or distributional relicts. There is a very high degree of endemism, with many taxa displaying vicariant distribution patterns. Other obligate cave taxa have extant relatives dwelling in nearby surface habitats. Many of the cave-dwelling taxa are derived from the terrestrial or aquatic habitats found within mesic forest environments. Some of the obligate terrestrial cave taxa from temperate regions are derived from only weakly hygrophilous groups, and many are only slightly troglomorphic. The majority of stygobionts have freshwater origins, although some have a marine origin. Troglophiles and trogloxenes make up a substantial portion of the diversity within many cave communities. Bats are widespread in continental Australian caves, and their guano supports a rich assemblage of guanophilic invertebrates.

Although the proportionate area of carbonate rock is less in Australia (Figure 1) than the world average (Gillieson & Spate, in press) there are about 500 discrete localities documented. In addition, there are cave and groundwater communities established within non-carbonate rocks such as basalt, sandstone, quartzite, acid igneous rocks, unconsolidated sediments and others. These include lava tunnels, sea caves, boulder and talus caves, aeolian caves, piping caves beneath silcrete crusts, and interstitial environments.

Karst, pseudokarst and groundwater localities are distributed across the Australian continent, which encompasses climate types ranging from tropical through to temperate. We have used regional climates, which reflect the surface vegetation and animal communities, as a basis for delineating four major cave-fauna provinces within Australia. In most cases, these regions also each have a dominant pattern of karst landforms. As defined by Bridgewater (1987) and depicted in Figure 1, these provinces are: a tropical-climate zone (II); a subtropical dry-climate zone (III); a transitional zone with winter rain (IV); and a warm-temperate zone extending to warm-temperate/tropical transitional climates zone (V, II-V).

Threats to Australian Cave Fauna

Ever since European settlement of Australia a little over 200 years ago the natural environment has endured a legacy of neglect and abuse. This has been the result of a colonial ideology that has fostered wholesale exploitation of the country's natural resources. Many karst and cave communities have suffered the brunt of Australian cultural tradition, but the last decade or so has witnessed a dramatic improvement in awareness as environmental issues have figured prominently within the social and political arena. Indeed, there is cause for optimism in the future outlook as a number of major conservation battles have been recently fought, and won, with karst issues being a key ingredient; but there is little room for complacency, as some highly significant sites have already been lost whilst others remain under imminent threat.

Owing to a paucity of earlier baseline data, and until recently a general lack of impact-related studies undertaken in Australia, it has proved difficult to quantify and predict precisely the effects of various human activities on cave fauna. Nevertheless, there are numerous examples where human activities have caused the local extinction of cave-dwelling bat and invertebrate populations, or where they have been seriously degraded or compromised. A great many sites remain highly vulnerable to a range of threatening processes that include quarrying and flooding, land clearance (deforestation) and agricultural activities (crop-raising and stock-raising) and forestry. The nexus between land clearance and agricultural activities is the most widespread and insidious threat, affecting cave microclimates, nutrient inputs, hydrological regimes, and water-quality parameters such as sedimentation, nutrient enrichment and pollutants. A further threat to cave communities is direct disturbance caused by cave visitors, guano miners, cave management and research activities. The Australian experience of these threats is described below.


 
 

Figure 1. Distribution of karst areas and climate provinces in Australia. Tropical-climate zone (II); subtropical dry-climate zone (III); transitional zone with winter rain (IV); warm-temperate zone extending to warm-temperate/tropical transitional climates zone (V, II-V). Map complied by Ken Grimes from data in Bridgewater (1987); Gillieson & Spate (in press).

QUARRYING AND FLOODING

A number of caves, along with their resident fauna, have been obliterated through quarrying activities, whilst others have been inundated following the construction of water impoundments. Some of these sites were the focus of high-profile conservation campaigns that were finally settled in the law courts, but decisions have fallen both ways in recent times. A determined conservation campaign was waged over some 25 years at Mount Etna in Queensland, where quarrying of limestone for cement has removed a number of significant caves (Sprent 1970; Hamilton-Smith & Champion 1975; Bonyhardy 1993). Eventually, during the course of lengthy legal proceedings, speleologists resorted to occupying caves in order to prevent further blasting. This phase of the battle ended for Speaking Tube Cave, which was an important roost site of the endangered Ghost Bat, Macroderma gigas, when the quarry company illegally blasted the entrance and in 1989 finally destroyed it. The high cost of legal proceedings prevented their final determination, but eventually in 1991, largely as a result of a change in government, the site was included in a new National Park.

The Mount Etna experience does not stand alone; a number of sites were lost before the importance of karst and cave communities was recognised beyond the speleological fraternity. Texas Caves in Queensland for example, was the type locality of an endemic troglobitic silverfish (Smith & Shipp 1978) and other undescribed invertebrates, but they were inundated by waters of the Glenlyon Dam (Archer 1978). In New South Wales, construction of the Burrinjuck Dam during the 1930s flooded the type locality of the horseshoe bat, Rhinolophus megaphyllus (Ryan 1965). In 1993, a South Australian quarrying company, with the apparent support of the State Mines Department, used explosives to destroy the entrance of a highly significant cave system at Sellicks Hill. This event resulted in extensive legal and political action, but the future of the site remains unresolved (Buswell 1994). Other karsts have been inundated at Dartmouth in Victoria, Canberra in the Australian Capital Territory and Lorinna and Moina in Tasmania. Quarrying of karst has occurred, and is still occurring, at Flowery Gully in Tasmania, Lilydale in Victoria, Marulan in New South Wales, Cape Range in Western Australia, and elsewhere.

However, there have been a number of successful conservation campaigns against quarrying. During the 1970s, the Colong karst in New South Wales and the Precipitous Bluff karst in Tasmania were both spared from quarrying proposals due to concerted efforts by speleologists and other interested groups. In the late 1980s cave archaeological sites on Tasmania's Franklin River provided a key piece of evidence that helped to save the river valley from being dammed. In 1992 a limestone quarry at Yessabah in New South Wales ceased operations following legal challenge by a speleologist who successfully demonstrated that he held a vested interest in the caves. His case was enhanced by the fact that one of the caves threatened was a major bat-roosting site. Cave faunal values were a significant criterion in the decision to include Exit Cave within the boundaries of the World Heritage area in Southwest Tasmania. Then in 1992, evidence of the deleterious impacts of quarrying upon this fauna was a major consideration in the decision to close the quarry concerned.

Many other limestone quarries are not known to have impacted upon caves (Gillieson 1989, pp.48-49), and there are large resources of non-cavernous limestones scattered throughout the country. Nevertheless, a number of highly significant karsts continue to be placed under considerable pressure for mining development, owing to their geographical proximity to population centres or the limestone quality, as well as other economic and political considerations. At the time of writing, a proposal for quarrying over part of the highly significant Cape Range karst has been approved by the Western Australian government. Regrettably, further protection of this region, or any other cave faunal site, is hindered by the universal problem of inadequate recognition being accorded to invertebrate communities by political and administrative decision-makers (Hill & Michaelis 1988).

LAND CLEARANCE AND AGRICULTURAL ACTIVITIES

Land clearance (deforestation), and agricultural activities (crop-raising and stock-raising) and forestry are the most common and widespread threat facing karst biota in Australia. The impacts resulting from these activities are often subtle and difficult to quantify because much of the degradation occurred long before any biota had been recorded. The impacts which potentially affect the biota include alteration to cave microclimates (including moisture, carbon dioxide and other parameters) and increased nutrient inputs from both grazing animals and artificial fertilisers, as well as changes to hydrological regimes and water quality, particularly sedimentation and pollutants.

It is very likely that a substantial proportion of the cave communities within New South Wales, Victoria, southwestern Western Australia and southeastern South Australia have undergone a marked alteration of composition or structure as a result of widespread land-clearance activities with their resultant wide range of environmental impacts. Of the approximately 120 karst areas in New South Wales and Victoria, some 60% have been subject to removal or major modification of their native vegetation cover (Eberhard & Spate 1995).

This has clearly caused a loss of biodiversity associated with removal of the native vegetation, and has reduced the input into caves of food, such as leaves, twigs and accidental fauna. Deforestation has increased the desiccation of cave environments, particularly around entrances, resulting in depletion of the twilight and entrance-zone sub-communities. It has altered watertable levels and caused changes in water quality and flow regime which impact upon the aquatic communities.

The extent of these impacts can best be demonstrated from extensive studies undertaken in South Australia on the effect of exotic pine (Pinus radiata) plantations on groundwater levels and atmospheric moisture regimes. As obvious examples, the water level in Sheathers Cave dropped by approximately one metre over a five-year period following the establishment of a pine plantation above the cave, whereas conversely, the level in the nearby Mount Burr Cave rose significantly, probably to approximately the same extent, when the pine plantation above it was destroyed by wildfire (Grimes et al, 1995, p.16). A significant increase in humidity, which resulted in the regeneration of previously dormant speleothems, occurred at Naracoorte Caves following the removal of pines. Similarly, caves beneath pine plantations at Yarrangobilly in New South Wales are less humid and contain less biota than those beneath indigenous forests (New South Wales 1983).

Observations have indicated a general decline in the numbers of cave-dwelling bats in central and southeastern New South Wales (Hall & Dunsmore 1974) and, anecdotally, elsewhere. Throughout New South Wales there are many caves that previously had extensive use by large numbers of bats, but which now are used only infrequently by smaller numbers, or not at all. The evidence includes the absence of bats from large numbers of caves that contain piles of bat guano and roof-staining from bat urine. The decline may be attributed, at least in part, to widespread deforestation which has severely restricted the area of suitable foraging habitat available to the insectivorous bats. Use of inappropriate insecticides may lead to the accumulation of residues in bats and consequent mortality (Dunsmore et al, 1974), although direct Australian evidence of this leading to mortality of cave-dwelling species is lacking.

Local extinction of guano-dependent invertebrates has occurred at the sites that are no longer used by bats. One of the few well-documented examples involves the disappearance, between 1866 and 1895, of a large maternity colony of the large bent-wing bat (Miniopterus schreibersii) from Mount Widderin Cave in Victoria (Simpson & Hamilton-Smith 1964). The bats probably abandoned the cave as a result of the clearing of native scrubland, which also caused more rapid percolation of water into the cave. In turn, the loss of the bats impacted upon the invertebrate fauna of the cave, which then came to support less than half the number of taxa recorded from other Miniopterus maternity sites in the region (Hamilton-Smith 1968).

The effects of agricultural and forestry activities, and industrial and metropolitan development, upon stream ecosystems has been the subject of a number of studies within Australia, and clearly these activities can result in major physico-chemical and ecological disturbances. Of these, agriculture and forestry impinge most widely on Australian karst areas, and the impacts derived from quarrying may extend some distance downstream from the initial site of disturbance. The price of agricultural prosperity has been a substantial blight upon the natural integrity of karst environments in continental Australia, and many Tasmanian karst areas are located within forestry concessions subject to clearfell logging operations.

A variety of hydrological impacts consequent upon deforestation, agricultural or other land-use activities threaten aquatic cave and groundwater communities. The most apparent and widespread threat involves the lowering of water tables through groundwater abstraction for irrigation purposes, watering of stock, or human consumption. Other hydrological impacts include changes to flow regimes and water quality, particularly sedimentation, nutrient enrichment and pollutants.

Water-table lowering in caves is acutely evident in Western Australia in the Augusta-Margaret River region, and at Yanchep where a uniquely rich community associated with underwater mats formed from tree roots is at risk (Jasinska & Knott 1991). The Cape Range stygofauna is also threatened by water abstraction.

Changes in flow regime and water quality become evident following alteration or removal of the native vegetation cover. The changes include an increase in water yield, thus exacerbating flooding in caves (Kiernan 1988). More rapid run-off after a precipitation event increases the rapidity in rise and peak discharge of streams flowing into caves. Previously perennial stream courses may become ephemeral or intermittent. A limestone quarrying operation caused significant adverse impacts to aquatic cave-dwelling fauna at Ida Bay in Tasmania. Changes in flow regime and water quality (including sedimentation, eutrophication, and toxins) are responsible for causing the extinction of fauna in Bradley Chesterman Cave (Eberhard 1995).

A number of recent Australian studies of surface waters (e.g. Doeg & Milledge 1991; Growns & Davis 1994; O'Connor & Lake 1994) implicate sedimentation as a primary cause of stream degradation and faunal decline, and this appears to be the case in underground streams also. In Exit Cave at Ida Bay in Tasmania for example, sedimentation originating from the limestone quarry altered stream habitats and restricted the distribution of hydrobiid snails. During February 1992, when the quarry was operational, a study showed that snail abundance was significantly lower in sediment-affected streams when compared with control streams. After closure of the quarry operation in October 1992, a rehabilitation program has sought to minimise further environmental degradation of the cave system by restoring natural inflow regimes and limiting the further influx of sediment. Continued monitoring of the snail populations now indicates similar abundance between the sediment-affected and control sites (Eberhard 1995). Since European settlement, many other stream cave systems in eastern Australia have been subject to increased rates of sediment deposition as a result of land-use activities within their catchments (Gillieson 1989).

The protection of water resources in general, and karst groundwater resources in particular, is an acute problem within rural communities throughout Australia. Ignorance and disregard of the nature of karst hydrological systems have severely compromised water resources and subterranean fauna in some localities. These impacts have been most often caused by improper rubbish disposal in sinkholes, and poor management of pollutants. Municipal disposal sites at Mole Creek in Tasmania and Buchan in Victoria both drain directly into karst aquifers. On one occasion, some 300,000 litres of waste milk was dumped in a cave at Allensford in Victoria (White 1976).

An extreme example is the dumping of thousands of sheep carcasses in Earls Cave, in the Mt Gambier region of South Australia, and other dumping at this site over a long period (Aslin 1972a). One study in the region identified 26 other caves, 21 industrial drainage bores and a variety of other natural depressions as potential points of entry to the aquifer for pollutants (Aslin 1972b). The extent of the problem is indicated by the nitrate pollution of this aquifer, which was measured as ten times in excess of safe levels (Waterhouse 1973). However, there is evidence of impacts to the biota occurring much earlier than this. The first European settlement in the region was at a site in Mount Gambier now known as the Town Cave, selected because of its ready access to good water. Woods (1862, pp. 359-360) described this cave, saying that the water in the bottom of the cave was "so deep as to give it, clear as it is, a deep sea-blue tint". He then added "the water is full of a cypris and cyclops, the shells of which seem to strew the bottom. There is also much conferva, a shrimp-like brachiopod and a minute paludina". The site later became a drain for the waste of the growing town, and today the water level is lower, with accumulations of silt and rubbish; the water is dirty and foul-smelling, and the aquatic community described by Woods has disappeared.

Nutrient enrichment may displace stygobiont communities and facilitate colonisation of underground waters by epigean taxa. This has occurred at Bradley Chesterman Cave in Tasmania, although petroleum hydrocarbons have also been implicated in elimination of the original community at this site (Eberhard 1995). Pollution from sewage has been identified at both Mole Creek in Tasmania and Buchan in Victoria, although action has been taken to eliminate the problem at the latter site. Subsidence in a limestone area also ruptured a sewage main in Tasmania (Kiernan 1988, pp.9-10). Eutrophication is the major threat to stromatolite colonies in karst lakes and cenotes in Western and South Australia (McNamara 1992; Hallam & Thurgate 1992; Thurgate 1995). Run-off laden with agricultural fertilisers has increased the concentration of phosphates in Lake Clifton ten-fold. The resultant algal blooms have stifled continuing growth of the stromatolite colonies by reducing available light levels, as well as directly smothering them (McNamara 1992, pp.23-24).

HUMAN VISITORS

Direct impacts to cave communities have been caused by human visitors, including guano miners, cave managers and tourists, recreational cavers, and scientists. Mining of bat guano has long ceased in Australia, but during the last century large quantities were removed from caves throughout eastern Australia. Doubtless these operations caused major disturbance to resident bat populations at the time, as well as to the associated guano communities of invertebrates. At Naracoorte in South Australia for example, guano miners cut an access hole through the roof of the Bat Cave that is the maternity site for the regional population of the bent-winged bat, Miniopterus schreibersii. The resultant ventilation and lowering of temperature would have made the cave unsuitable as a maternity site. Caves that provide the right micro-climatic conditions for breeding in this species are few and far between, and there are only a handful of maternity sites known throughout the entire range of the species in Australia (Dwyer & Hamilton-Smith 1965). Although it is not known what happened to the bat population at the time, the hole was sealed on the cessation of mining and the bats have occupied the cave ever since. The cave, now located in a World Heritage area, has more recently been accorded a very high level of protection.

Traditionally, cave managers concerned with developing caves for tourism purposes have seen their conservation responsibility as being confined to the preservation and display of the most beautiful speleothems, and until recently, scant regard was given to the habitat requirements of the fauna. Cave microclimates have been significantly altered by the excavation of new entrances and connecting tunnels, or the enlargement of existing entrances. Additionally, entrances have been gated or sealed-off entirely, thus inhibiting access for bats and other trogloxenes, and altering the input of nutrients such as leaf litter and accidentals. At a number of sites, cave entrance gates have subsequently been modified to allow the free passage of bats, or to restore the original micro-climatic conditions. The Alexandra Cave at Naracoorte, for example, was discovered by digging open the entrance, thus allowing entry to rhaphidophorid crickets which established a flourishing colony and created considerable interest amongst visitors. Later it was realised that the free airflow permitted through the wooden gate placed on the new entrance had desiccated many speleothems, so a new air-tight door was installed to further their natural regeneration. This succeeded, but did of course result in loss of the cricket colony. In this example, the crickets are found widely throughout the region and the loss of one cave is not a matter of concern. However, in the Northern Territory, a badly designed gate resulted in the destruction of the largest recorded population of the rare golden horseshoe bat, Rhinonycteris aurantius.

The metabolism of large numbers of human visitors affects cave microclimates by increasing carbon dioxide levels, moisture and temperature. Visitors also introduce lint, skin flakes, soil particles, food scraps, microflora, and fungi. These provide a rich food source which benefit some cavernicolous species but may also permit colonisation by adventitious species. This phenomenon is evident at Jenolan Caves in New South Wales, where considerable populations of springtails dwell alongside the tourist paths. The growth of algae, mosses and other vegetation near artificial lighting (Lampenflora) may also encourage colonisation of the dark zone by invertebrates that are normally restricted to the twilight zone.

Other foreign materials that tend to accumulate in tourist caves include soot from carbide and other early combustible lighting sources, pitch from old electrical wiring seals, verdigris from scraps of electrical wiring or from coins placed in wishing wells, broken light globes, scrap metal and timber (Bonwick & Ellis 1985). Old and disused timber left dumped in caves attracts cavernicolous species, and clean-up programs, although well-intentioned, can cause mortality to many individuals, and perturbation of the ecological balance established in the faunal community when the timber is removed.

Recreational caving activities have caused widespread and profound disruption to bat populations, as colonies that are repeatedly disturbed may abandon their cave sites and not return to them; this has occurred at Cotter Cave in the Australian Capital Territory and elsewhere. Even minor disturbances of bats at their over-wintering sites and maternity roosts may result in significant fatalities (Hamilton-Smith 1970). Caving activities have increased in popularity over the last 25 years especially, and may be part of the cause for the decline in bat populations throughout New South Wales and Victoria during this period, although hard data are lacking. Most cave visitors are simply unaware of the disturbance they may cause, but occasionally bats are deliberately killed.

Floor-dwelling invertebrates in many caves are also widely under threat from recreational cavers who inadvertently trample them underfoot. A side effect of intensive trampling is the compaction of soft floor sediment that may render it less suitable as invertebrate habitat. In extensively trampled caves the area of substrate inhabited by invertebrates has been greatly reduced, and the animals are confined to untrampled substrates alongside the walls, or in accessible crevices, and may even become extinct. This phenomenon is evident in many popular caving areas. Regrettably, the residual invertebrate population in Mount Widderin Cave at Skipton in Victoria has been almost totally exterminated by opening of the cave to large visitor parties who have been allowed to trample and compact the total floor area (Hamilton-Smith 1968; Spate & Hamilton-Smith 1993). Similarly, conduct of guided parties through a cave at Mount Etna in Queensland resulted in destruction of the root mat that had provided the habitat for planthoppers (Cixiidae and Meenoplidae). In this case, the management agency argued that the limited root destruction would not impact upon the surface vegetation, and so permitted the tours to occur; the fauna was not even considered (Vavryn, D & J 1995, pers. comm.). Other micro-habitats that are vulnerable to damage from trampling are pools and small watercourses, which often contain a distinctive and specialised fauna.

Speleological activities such as digging can alter cave microclimates and the distribution of the fauna. There is one documented instance of a cave which when first dug into, contained a high relative humidity and many invertebrates. Shortly after opening of the entrance, the cave dried out and the fauna disappeared (Hamilton-Smith 1970).

The final direct human threat to cave fauna concerns the activities of scientific researchers. At least part of the decline in numbers of cave-dwelling bats during the 1960s and 70s may be attributed to the pressure of specimen-collecting and banding programs. The former activity is known to have decimated local populations on at least two occasions with minimal scientific return (Hamilton-Smith 1970). At the same time, although participants in the banding scheme took particular care to minimise the disturbance they caused, and this research yielded much useful information on the ecology of species (Simpson & Hamilton-Smith 1965), there were occasional negative impacts. More recently, concern has been expressed about the possibility that invertebrate specimens may be over-collected. Slaney and Weinstein (1995) found that continuous trapping over a one-month period in Rope Ladder Cave in Queensland caused a significant decline in numbers of invertebrate taxa. These authors recommended that the general ecology of organisms in a cave be investigated before any intensive sampling is carried out, and that, where possible, sampling with replacement should be used.

Strategies in Conservation

Present understanding of the nature of karst landscapes emphasises the importance of the interrelationships between environmental conditions prevailing on the surface and those underground. We argue that this is fundamental to the development of effective strategies for the conservation of subterranean faunal communities. This approach may involve, for instance, the management of water- catchment areas which extend well beyond the geological boundaries of the karst outcrop. Aside from the highly publicised conservation campaigns which have been waged in attempts to conserve imminently threatened sites, the conservation strategies that have been applied in Australia to date include: legislative protection of threatened species and development of recovery plans for them; the protection of areas of sensitive habitat and at individual cave sites; legislative recognition of endangered populations and communities, and key threatening processes; habitat restoration; the location of karstlands within National Parks and other protected or recognised areas; and community involvement and public awareness campaigns. Each of these strategies is discussed in turn below.

SPECIES PROTECTION

A number of cave dwelling species are protected under legislation in Australia. A person may not knowingly take, damage, or kill a protected species without a permit. The legislation has restrained the shooting of bats for example, but was not implemented until long after the widespread loss of bat foraging habitat through land-clearance activities had already occurred.

In recent years there has been widespread recognition of the importance of maintaining biodiversity, and substantial funding has been allocated to the protection of species whose continued survival may be threatened. Threatened species are generally classified as either vulnerable, endangered, or extinct, although a number of variations on this scheme are recognised (e.g. IUCN Red Data Book). In New South Wales, for example, two of the four species of habitual cave dwelling bats are listed as vulnerable under the Threatened Species Conservation Act 1995. This means that any land development proposals, forestry operations, or farming activities which do not conform to accepted agricultural practices, must first consider the potential impact upon listed species or the habitat of listed species. The assessment is usually made by an environmental survey conducted prior to the commencement of works, although in the case of private landholders especially, the onus is on the state fauna agency to ensure that landholders are informed of protected species and their habitat requirements. Consequently, both developers and state fauna agencies have been obliged to invest considerable resources towards meeting the requirements of the TSC Act. The legislation potentially offers considerable security to listed species, whilst both developers and state fauna agencies have been obliged to invest considerable resources towards meeting the requirements of the Act.

Legislative protection of cave-dwelling invertebrates has generally lagged far behind that for vertebrates although in 1971 a number of Tasmanian cave species were wholly protected under that states Parks and Wildlife Act 1970. Although a remarkable step forward at the time, a number of problems have become evident. When the legislation was enacted the dozen or so species listed comprised, virtually, the known extent of Tasmania's cave fauna; thus despite the discovery of many additional taxa in the meantime, it is still widely and erroneously believed by park managers and academic biologists alike that all species of cave fauna are protected. A permit is required to collect any of the listed species, including those located outside national parks or state reserves. The Threatened Species Protection Act 1995 (Tasmania) incorporates a number of the recently described taxa, but the majority remain undescribed and therefore cannot be listed. In Western Australia, 13 out of the 26 invertebrate species listed as threatened are troglobites (W Humphreys, pers. comm.).

Once a species becomes listed as threatened, the relevant state fauna agency has an obligation to ensure its protection. This usually involves the implementation of a research plan or action plan, which is followed in turn by the development of a recovery plan. Recovery plans may involve habitat restoration and protection, public education programs, and enhanced breeding. Currently an action plan for Australian bats is being developed, and will include attention to cave-dwelling species (Richards & Hall 1996). The Parks and Wildlife Service Tasmania has produced listing statements for the state's listed cave invertebrates and is pursuing the development of recovery plans for the species most under threat. A few plans have been prepared for non-cavernicolous invertebrates, but there are formidable problems involved with invertebrates generally, including lack of sufficient baseline information to identify species demanding such action, lack of research upon which to base plans, and the complexity of implementation (Yen & New 1995). Limited funding resources or shifts in the corporate priorities of state fauna agencies may well inhibit the implementation of recovery plans for listed species.

PROTECTION OF POPULATIONS AND COMMUNITIES, KEY THREATENING PROCESSES

A substantial portion of Australia's invertebrate cave taxa remain undescribed, and are likely to remain so in the foreseeable future. The lack of adequate taxonomic knowledge will continue to be a major stumbling block in gaining protection for individual species. One solution is to seek protection of endangered populations or ecological communities, which is provided for under the Australian Endangered Species Protection Act 1992 (Eberhard & Spate 1995a). The degree of adoption of this type of legislative protection into the respective acts for each State varies. In New South Wales there are a number of bat maternity, over-wintering, and important roosting caves which have been proposed for listing as critical habitat for bat populations which are considered endangered within their local distributional range (Eberhard & Spate, in prep.). Cave communities, the distributional limits of which can usually be readily defined, are likely candidates for listing under this legislation. However, the legislation only provides for the recognition of endangered, but not vulnerable, communities or populations. Under these strict criteria only one endangered cave community has been proposed in New South Wales, despite the fact that many others would evidently qualify for listing as vulnerable if this category were available. Endangered communities or populations are not, at present, written into the respective Acts in Tasmania and Western Australia, although a number of endangered cave communities in WA have been recognised as such and recovery plans implemented (V English, pers. comm.).

The New South Wales TSC Act 1995 also provides for the listing of key threatening processes which may impact upon threatened species, populations, or communities. It is perhaps significant that out of more than a dozen populations or communities which have been nominated for listing, human visitors to caves have been identified as a key threat in each case, either solely or in combination with other threats. Accordingly, cave visitors have been nominated as a key threatening process impacting upon cave fauna in New South Wales (Eberhard & Spate, in prep.).

HABITAT PROTECTION AND HABITAT RESTORATION

The protection of areas of sensitive habitat within individual caves, coupled with an education program, has proved to be a pragmatic and successful conservation strategy which works much along the lines of a community recovery plan. Thus, in Mullamullang and Nurina Caves on the Nullarbor Plain, certain key habitat areas have been delineated with protective strings and signs that explain the reason for trying to exclude visitors from these areas (Poulter 1991, 1994). Similarly in Kubla Khan and Little Trimmer Caves in Tasmania, so-called substrate protection zones have been maintained by marking out pathways and no-go areas. This approach relies upon voluntary compliance by the caves visitors, and its relative efficacy will depend on a number of factors such as the accessibility and popularity of the site, the presence of a dedicated management authority, and the type of people who visit the cave (e.g. speleologists vs the general public). There are many caves where the impacts of trampling are ongoing, and these sites would greatly benefit from some form of habitat protection.

Similarly, bat conservation has been fostered by some land management agencies and caving groups. For example, a number of caves at Bungonia are closed annually by the New South Wales National Parks and Wildlife Service during critical periods of the bat breeding season and winter torpidity. Another example is at Anticline Cave in eastern Victoria, where a major maternity site used by the horseshoe bat Rhinolophus megaphyllus had been abandoned, and the species was no longer seen in the area for a period of almost 20 years. Cavers took the initiative, with the support of the landowner, in restricting visits to the cave and as a result the population has been re-established (Hamilton-Smith 1991). A similar return by bent-winged bats to a cave at Jenolan has been recorded after the responsible manager had closed the cave to visitors, but its significance has not yet been fully assessed.

The importance of restoring previously damaged and degraded habitats is only now being realised. Both speleologists and tourism managers have been involved in aesthetic restoration (e.g., Bonwick & Ellis 1985), and this clearly has some benefits in terms of habitat restoration. The experiences at Horse Cave in the United States and Waitomo in New Zealand clearly demonstrate the potential importance of restoration. Restoration of climatic patterns or breaking-up of compacted floors are two examples where action is being taken at the time of writing, or might be taken, in Australia. Rehabilitation of the former quarry at Ida Bay in Tasmania has been undertaken with a full recognition of the impacts upon fauna, and, as noted above, the hydrobiid snail population is already returning to its former condition as a result of siltation control. Furthermore, the aquatic community in Bradley Chesterman Cave has recently re-colonised after suffering extinction at this site when the quarry was operational Artificial fertilisers, which would have accelerated the surface revegetation program, have been excluded because of their negative impacts upon cave biota (Gillieson 1995).

PROTECTED OR RECOGNISED AREAS

Land reserves for the specific protection of caves, as opposed to cave communities per se, were first gazetted at Wombeyan in 1865, and then elsewhere later. Although some of these reservations were adequate, many were not, often only encompassing the cave entrance rather than the whole cave or karst system, and very rarely protecting the total catchment. Hence, substantial portions of some reserved cave systems extend beneath private land, state forest, or other land tenure, thus compounding the task of management and conservation. In some instances this problem has been solved by further land acquisition, but it has often involved protracted negotiations to settle conflict of interest.

Quite a number of important karst areas are located within National Parks or other conservation reserves, although, more often than not, their inclusion has been by fortuitous circumstance rather than deliberate intention. Consequently, some parts of the karst system, or its catchment, may not be included within the reserve boundaries. Such land protection is no panacea, and a range of other problems may arise, particularly in relation to fauna conservation. Lack of training for park managers in karst issues, combined with limited funding availability for karst-related conservation projects, is one of the more important. Many caves in National Parks are under great pressure from large numbers of recreational visitors and in some parks this problem has been addressed by limiting access through institution of a permit system. Some caves have been closed entirely and enforcement of the closure sometimes requires the installation and maintenance of entrance gates which in turn may interfere with access for bats.

Park authorities walk a difficult line between rigorous enforcement of conservation ideals on one hand versus their need to maintain public and political legitimation on the other, and this may discourage strong action on conservation issues. There have been a number of incidents where park rangers have apprehended persons for contravening regulations, but the institution of legal proceedings has been halted at the ministerial level. Further, depending on the political climate at the time, conservation agendas may be discarded in favour of economic initiatives. Certainly financial problems have led to a shortage in management resources, particularly in personnel on the ground. Nevertheless, there has been a continuing improvement in the overall quality of cave management, aided by scientific research and monitoring over the last 20 years.

A great many significant karst and cave communities are located on privately-owned land, or they are located in areas of state forest subject to logging operations, especially in Tasmania. Only in recent years, the Forestry Department of the Tasmanian government has played a leading role in conserving karst resources within production forests by developing a forest practices code which embraces proper recognition of karst. In addition, logging operations have been excluded from within some sensitive karst catchments, and operations have been deferred in other areas pending further investigations of their significance and vulnerability. For karst-lands and cave communities under private ownership there is no legislative mechanism in place which specifically ensures their conservation, although recent developments pertaining to land clearance and catchment protection within New South Wales may prove to be relevant. Unfortunately, many karst communities on private land are already seriously degraded. The future of these communities depends to a large extent upon the attitude of the landholder, which varies between individuals. Some caves and sinkholes continue to be utilised as rubbish dumps, whereas other landowners provide a high level of conservation management by excluding or limiting access to visitors. More often than not however, this action is based on sheer convenience, or unsubstantiated fears of legal liability in case of a caving accident, rather than a genuine concern for conservation. The key to any conservation program is essentially one of enhanced understanding through education and interpretation at all levels, and insufficient attention has yet been given to the involvement of private landowners as partners in conservation, particularly in respect to karst areas.

The Register of the National Estate provides a mechanism for the recognition of areas containing important natural or cultural values. National Estate listing can be applied to sites anywhere regardless of the land tenure. The Cape Range karst in Western Australia has recently been listed, primarily on the basis of its subterranean faunal values. At the very least, such recognition can be used as a basis for further conservation action.

International treaties are playing an increasing role in resource protection. The stromatolites of Shark Bay, together with caves in southwestern Tasmania, at Riversleigh in Western Queensland and Naracoorte in South Australia fall within World Heritage areas. Most recently, a number of karst sites have also been recognised under the Ramsar Convention on Wetlands. These include Cape Range and several sites in the Kimberley of Western Australia, part of the Katherine karst in the Northern Territory, caves at Chillagoe and Undara in Queensland, and a number of sites in the Naracoorte - Mount Gambier region of South Australia (Australian Nature Conservation Agency 1996). Cape Range has been listed (1997) by the Karst Waters Institute as one of the world's top ten most endangered karst aquifers.

PUBLIC EDUCATION AND COMMUNITY INVOLVEMENT

It would appear that the future management and conservation of cave communities will rest both on a legislative footing and on better public recognition and understanding of the complexities of karst processes and karst environments (Eberhard & Spate 1995a). The privatisation of guiding services may, and often does, result in less emphasis on conservation and public education, but at Undara in Northern Queensland, privatisation has led to perhaps the best guiding practice in Australia. The development of a Minimal Impact Caving Code by the Australian Speleological Federation has been adopted by some operators within the adventure tourism industry (Webb 1993). Another example is a pamphlet published by speleologists which has facilitated raising awareness of the plight of cave-dwelling bats in Victoria, and co-operation between the speleologists, a private landholder, and the state conservation department has enabled recovery of the Buchan population (Hamilton-Smith 1991; Friends of Buchan Caves 1993).

More generally, collaboration between cavers, academic researchers, and cave managers has proved to be a fertile starting point, from which there is a positive flow-on effect to the general public. The fostering of community involvement in the management of privately owned or leasehold karstlands is seen as a crucial ingredient in the process. In New Zealand, protection of the catchment of the Glowworm Cave at Waitomo relied on co-operation between a number of diverse parties including cavers, tourist operators, farmers, traditional Maori landowners, and the national conservation agency. The New Zealand experience has set an exemplary standard in community involvement, an approach to conservation which urgently needs to be forged in rural karstlands in Australia. There may be scope for community awareness and involvement with karst management issues to be fostered through programs such as Landcare for example. Experience and expertise continue to evolve, and this is transmitted through the auspices of organisations such as the Australian Speleological Federation and the Australasian Cave and Karst Management Association.

References

Archer, M  1978, Modern fauna and flora of the Texas Caves area, Southeastern Queensland. Memoirs of the Queensland Museum, 19: 111-119.

Aslin, F  1972a, Caves of Australia, No. 6 : Earls Cave, Newsletter of the Australian Speleological Federation, 57 : 15-16.

Aslin, F  1972b, Preliminary report on potential pollution of underground water in and around Mt. Gambier, Newsletter of the Australian Speleological Federation, 57 : 17-18.

Australian Nature Conservation Agency  1996, A Directory of Important Wetlands in Australia, The Agency, Canberra, 964 pp. 2nd ed.

Bonwick J & Ellis R  1985, New caves for old: cleaning, restoration and redevelopment of show caves,  Cave Management in Australasia, 6 : 134-153.

Bonyhardy, T  1993, Places Worth Keeping: Conservationists, Politics and the Law, Allen and Unwin, Sydney, 192 pp.

Bridgewater, P B  1987, The present Australian environment: terrestrial and freshwater, in Fauna of Australia, eds G R Dyne & D W Walton, Australian Government Publishing Service, Canberra, Vol. 1A, pp. 69-100.

Buswell,  C 1994, Sellicks Hill Quarry Cave, Australian Caver, 136 : 18-23.

Doeg, T J & Milledge G A  1991, Effect of experimentally increasing concentrations of suspended sediment on macroinvertebrate drift, Australian Journal of Marine and Freshwater Research, 42: 519-526.

Dunsmore, J D , Hall, L S & Kottek, K H  1974, DDT in the bent-winged bat in Australia, Search, 5(3) : 110-111.

Dwyer, P D & Hamilton-Smith, E  1965,  Breeding caves and maternity colonies of the bent-winged bat in southern Australia, Helictite, 4 : 3-21.

Eberhard, S M  1995, Impact of a limestone quarry on aquatic cave fauna at Ida Bay in Tasmania. Paper presented to 11th Conference, Australasian Cave and Karst Management Association, Tasmania, May 1995.

 Eberhard, S M & Spate, A P  1995a, Cave Communities. Submission to the Endangered Species Unit, Australian Nature Conservation Agency, Canberra, 15 pp.

Eberhard, S M & Spate, A P  1995b, Cave Invertebrate Survey: Toward an Atlas of NSW Cave Fauna. Report to the National Parks and Wildlife Service of New South Wales, 112 pp.

Eberhard, S M & Spate, A P  (in prep.) Cave Fauna of New South Wales - Nomination of Key Threatening Processes, Endangered Ecological Communities, Endangered Populations, & Threatened Species for Inclusion on the Threatened Species Conservation Act 1995. Report to the Scientific Committee, National Parks and Wildlife Service of New South Wales.

Flannery, T  1994, The Future Eaters, Melbourne: Reed Books. 

Friends of Buchan Caves  1993, Cave Dwelling Bats of East Gippsland. Brochure, 6 pp. with poster on verso.

Gillieson, D S  1989, Effects of land use on karst areas in Australia, in Resource Management in Limestone Landscapes : International Perspectives, eds D S Gillieson & D I Smith 1989. Department of Geography and Oceanography, University College, Australian Defence Force Academy, Canberra, pp.43-60.

Gillieson, D S  1995, Rehabilitation of the Lune River Quarry, Tasmanian Wilderness World Heritage Area, Australia, Journal of the Australasian Cave and Karst Management Association, 19 : 4-10.

Gillieson, D S & Spate, A P  (in press). Caves and karst in Australia and New Guinea, in Global Karst, ed Yuan Diaxion, UNESCO and China Science Press.

Grimes, K G, Hamilton-Smith, E & Spate, A P  1995, South East Karst Province of South Australia. Report on behalf of the Australasian Cave and Karst Management Association for the National Parks and Wildlife Service of South Australia, 110 pp.

Growns, I O & Davis, J A  1994,  Effects of forestry activities (clearfelling) on stream macroinvertebrate fauna in South-western Australia, Australian Journal of Marine and Freshwater Research 45: 963-975.

Hall, L S & Dunsmore, J D  1974, A survey of cave-dwelling bats, Australian Speleological Federation Newsletter, 64 : 9-10.

Hallam, N D & Thurgate M  1992, The Biology of Ewens and Piccaninnie Ponds and Woolwash, Blacks and Gouldens Holes, South Australia. Department of Ecology and Evolutionary Biology, Monash University, Melbourne, 79 pp.

Hamilton-Smith, E  1968, The insect fauna of Mt. Widderin Cave, Skipton, Victoria, Victorian Naturalist, 85 : 294-6.

Hamilton-Smith, E  1970, Biological Aspects of Cave Conservation, Journal of the Sydney Speleological Society, 14(7): 157-164.

Hamilton-Smith, E  1972, The bat population of the Naracoorte Caves area, Proceedings of the Eighth Biennial Conference, Australian Speleological Federation, pp. 66-75.

Hamilton-Smith, E  1991, The return of the bats, Nargun, 23 (10) : 81.

Hamilton-Smith, E & Champion, R  1975, Mount Etna and the Caves: A Plan for Action, Queensland Conservation Council, Brisbane, 147 pp.

Hill, L & Michaelis, F B  1988, Conservation of Insects and Related Wildlife, Australian National Parks and Wildlife Service, Occasional Paper No. 13, Canberra, 40 pp.

Humphreys, W F (ed.)  1993, The Biogeography of Cape Range, Western Australia. Records of the Western Australian Museum, Supplement 45: 1-248

James, J M  1996, Evidence for a sulfureta in Weebubbie Cave, Nullarbor Plain, Western Australia. Abstracts, Karst Studies Seminar, Naracoorte 9-11 February, pp. 11-12.

Jasinska, E J & Knott, B  1991, The importance of caves at Yanchep, Proceedings of the Eighteenth Biennial Conference, Australian Speleological Federation, pp.55-57.

Jennings, J N  1968, Syngenetic karst in Australia, in Contributions to the Study of Karst, J N Jennings & P W Williams, Department of Geography Publication G/5, Australian National University, pp. 41- 110.

Jennings, J N  1969, Karst of the seasonally humid tropics in Australia, in Problems of the Karst Denudation, ed. O Stelcl, Institute of Geography, Brno, Czechoslovakia, pp. 149-158.

Kiernan, K  1988, The Management of Soluble Rock Landscapes, Speleological Research Council, Sydney, 61 pp.

McNamara, K  1992, Stromatolites, Western Australian Museum, Perth, 27 pp.

New South Wales  1983, Harvesting and Rehabilitation of Jounama Pine Plantation, Kosciusko National Park: Environmental Impact Statement, National Parks and Wildlife Service and Forestry Commission, 198 pp.

O'Connor, N A & Lake, P S  1994,  Long-term and seasonal large-scale disturbances of a small lowland stream, Australian Journal of Marine and Freshwater Research, 45: 243-255.

Osborne, R A L & Branagan, D F  1988, Karst landscapes of New South Wales. Earth Science Reviews 25: 467-480.

Poulter, N  1991, Cave rights for troglobites, Proceedings of the Eighteenth Biennial Conference, Australian Speleological Federation, pp. 15-24.

Poulter, N  1994, Protecting caves from people II, Australian Caver, 137: 17-26.

Rainbow, W J  1904, Studies in Australian Araneidae III, Records of the Australian Museum, 5: 326-336.

Richards, G C & Hall, L S  1996, Draft Action Plan for Australian Bats. Report to the Australian Nature Conservation Agency for project 115. 162 pp.

Ryan, R M,  1965. The type locality of the Australian horseshoe bat, Rhinolophus megaphyllus, Australian Journal of Science, 27: 259.

Simpson, K G & Hamilton-Smith, E  1965, Third and fourth annual reports on bat banding in Australia, CSIRO Australia: Division of Wildlife Research Technical Paper 9: 1-24.

Slaney, D P & Weinstein, P  1995,  Conservation of cave fauna: more than just bats, paper presented at Conference on Invertebrate Biodiversity and Conservation. Melbourne, November 1995.

Smith, G B & Shipp, E  1978, A new species of cave-dwelling Nicoletid silverfish (Thysanura: Insecta) from the Texas Caves, Queensland, Memoirs of the Queensland Museum, 19: 121-123.

Spate, A & Hamilton-Smith, E  1991, Cavers impacts - some theoretical and applied considerations, Australasian Cave and Karst Management Association; Proceedings of the ninth conference, pp. 20-30.

Sprent, J K (ed.)  1970, Mount Etna Caves, University of Queensland Speleological Society, Brisbane, 116 pp.

Thurgate, M  1995, Sinkholes, caves and spring lakes, South Australian Underwater Speleological Society, Occasional Paper no. 1.

Waterhouse, J D  1973, Contamination of groundwater in the Mount Gambier area. Mineral Resources Review, 139: 27-40.

Webb, R  1993, Minimal impact caving code - Fourth Draft, Australian Caver, 135: 24- 25.

White, M E  1994, After the Greening: The Browning of Australia, Kangaroo Press.

White, N J  1976, Dumping of milk, Allansford, Western Victoria. Nargun, 8(7): 11.

Woods, J E (Tenison-)  1862, Geological Observations in South Australia. Longman, Green, Longman, Roberts and Green, London, 404 pp.

Yager, J & Humphreys, W F (in press), Lasionectes exleyi, sp. nov., the first remipede crustacean recorded from Australia and the Indian Ocean, with a key to the world species. Invertebrate Taxonomy 10.

Yen, A L & New, T R  1995, Some problems affecting implementation of recovery programmes for threatened Australian non-marine invertebrate species, in People and Nature Conservation: Perspectives on Private Land Use and Endangered Species Recovery, eds A Bennett, G Backhouse & T Clark, Royal Zoological Society of New South Wales, Sydney, pp. 9-14.