LIVING WITH LIMESTONE: SOME IMPRESSIONS OF KARST MANAGEMENT IN CHINA

Kevin Kiernan, Forestry Commission, Tasmania & Department of Geography, University of Tasmania

Few places in the world offer greater scope for learning of the interactions between humans and karst than China, which has long experience of having to come to terms with the practical implications of heavy population pressures in karst. Karst occurs over ~33% of China and occupies a similar proportion of Guangxi, the province in which the most celebrated of the karst is located. At least one major water conservation project in the Guangxi karst dates back 2200 years but many mountain villages still lack adequate water. Rural poverty remains endemic in many of the karst regions of Guangxi where the principal problems concern water supplies, difficult terrain, social isolation, deforestation, land degradation and a low level of social development. Agriculture is focussed on the karst plains and the floors of closed depressions. There is little forestry activity on the carbonate rocks. In various parts of the country the development of hydro-electric energy resources, communications, the construction of buildings and the mining of minerals have all been complicated by karst. Land surface collapse is a major problem in some areas. These difficulties have stimulated some novel planning and engineering responses. Nature conservation has very much taken a back seat to coping with the economic pressures but some prehistoric and historic sites have fared rather better. The karst landscapes are increasingly the focus of a booming tourist industry.

INTRODUCTION

Although ACKMA incorporates karst management in its embrace, the effective focus of this organisation has been principally on cave tourism and management and nature conservation in karst, and less so on broader issues of human use of karst environments and surface karst management. Partly the reasons for this are historical, partly they stem from the fact that, with a few notable exceptions, only a limited number of people in Australia and New Zealand spend their everyday lives in intimate contact with carbonate landscapes. Land-use on karst in this part of the world to date has generally been extensive rather than intensive, the impacts both on the karst and on the people being out of sight and out of mind for most of the population. Karst occurs at the surface over only about 4% of the most karstic Australian state, Tasmania. This is only a tiny fraction of the total extent of the world's karst which, at ~54,000,000 km2, totals ~34% of the terrestrial surface of the Earth, occurring mostly in the northern hemisphere mid latitudes. Although many of the problems that can arise for people who live with limestone have emerged here, their extent and broad economic impact has been very minor compared with that experienced in many other parts of the world. Perhaps nowhere are there more lessons to learn than from China, where exposed karst extends over about 910,000km2 and if covered karst is included the total extent is ~3,400,000 km2, or roughly 33% of the area of the country. China is the world's most populous nation, and about one fifth of the earth's human population are Chinese peasants, comprising some 55 nationalities. Millions of these people live in the karsts. Yuan (1988a) has observed that China faces two big geological difficulties, namely karst terrains and the less widespread loess country.

There are inevitably differences between the perspectives that developed and comparatively rich western countries are able to adopt towards the management of karst areas and those that have evolved in a developing country faced with the need to provide adequate living standards for millions of its citizens who live in karst terrain. Some of the options that remain open in New Zealand or Australia, where there are areas of relatively pristine karst, have long ago been closed in much of the Chinese karst due to environmental degradation that dates back hundreds of years. At least in southern China, limestone correlates to a large extent with human poverty. There are difficulties of water supply and alternating disasters of drought and flood, related to rapidly evolving hydrological systems, poor soils, land degradation due to deforestation, and accelerated soil erosion. The rugged terrain hampers agriculture and access. Isolated communities or families are often scattered in karst depressions that are separated from one another by steep divides. Coupled with this low level of social development basic welfare is also very low, and transport facilities are very poor with many towns lacking road access. Even the exchange of information is often restricted due to an absence of telephone, electricity, radio or television services. Some 50% of the population over 12 years of age is at best semi-literate (Li, 1988; Zhao, 1988).

Karst terrain is very prominent in at least five Chinese provinces. It comprises over 30% of Guangxi province where mountainous limestone terrain is commonly referred to as the 'Rocky Mountains'. This paper records some observations on how the Chinese live with limestone, notably in Guangxi province. The most enduring impressions were gained during October 1988, when with a small group of Chinese and western karst specialists, I travelled about 1000km on back roads and tracks through peasant communities in central Guangxi examining karst management problems. In some of these areas merely gathering drinking water is often a major effort for the people and even the passage of our vehicle filled both the air and their homes with dust. Efforts to raise living standards in this region are confounded by the practical difficulties imposed by the karst. The people are of necessity acutely aware of their landscape and, despite the difficulties it may cause, the beauty of the karst has long been recognised by Chinese artists and poets such that it is deeply interwoven into the culture and psyche of at least some of the Chinese population. Frescoes 2000 years old sensitively depict the karst mountains and rock formations. Increasingly there is also a recognition of the potential of the karst scenery to earn foreign exchange through tourism. Over 1000 people may now cruise the Lijiang River downstream from Guilin City every day, and the caves and karst landscapes around Guilin are vigorously promoted as tourist attractions. The internment of human remains in burial urns stashed in caves is still common. Caves still also form a focus for some local festivals, especially for minority peoples. Our visit to Yangshiu coincided with a Meo festival of homage for the aged in a nearby cave high in a limestone tower - and the elderly people who managed the ascent certainly deserved homage!

THE KARST

Karst occurs over ~100,000km2 of Guangxi province and this paper documents impressions of karst management after spending less than one month passing through ~40,000km2 of that area. The karst visited lies at 22°46' to 25°40'N latitude, and 107°30' to 109°40'E longitude and therefore straddles the Tropic of Cancer. The areas visited stretched from Xingan, 50km northeast of Guilin city, down the Lijiang River as far as Yangshuo and neighboring terrain to the east and west; the Liuzhou - Yongshiu - Yishan area in the Liujiang River Basin; Xincheng, Duan and Mashan in the Hongshui River basin and thence through Wuming to Nanning on the Yongjiang River 350km from the Vietnam border. A few cursory observations were also made during a 28 hour train trip back across Guangxi province to Guilin, thence into southern Hunan province and south through Guangdong province to Guangzhou (Canton). The overall drainage of central Guangxi is southeastwards following a general surface slope of 4-8%. The climate is characterised by long summers and warm winters, the mean annual temperature being ~20°C. Rainfall ranges from 1400-2000mm pa, 60-80% of that total falling between April and September. Evaporation totals 600-1000mm pa. The area is underlain by limestone and dolomite sequences of Devonian to Triassic age that are 5-10km thick.

Two broad categories of steep tropical karst are recognised in China and both are widespread in Guangxi. Fenglin (peak forest) consists of isolated towers that are scattered on a plain of some sort. Fencong (peak cluster) consists of hills that occur in conjoined groups. Fenglin is what is regarded by western researchers as classical tower karst. Subdivisions of these broad karst types are also recognised in China, hence terms such as fencong depression (equivalent to cockpit karst) and fencong valley (for which there is no English equivalent). Each of these is in turn subdivided, hence fencong depression has divide, marginal, intervalley and insular subtypes. Fenglin is classified into basin, margin, valley and polje types (Zhu, 1988). One train of thought suggests that the difference in form reflects the stage reached in landscape evolution, with fenglin evolving from fencong. However, evidence exists to suggest the generation of fenglin without the fencong stage. The fenglin is best developed in those areas least subject to late Tertiary and Quaternary uplift, with fencong best developed where uplift of this age has been most active (Sweeting, 1990). Geological factors play a significant role in shaping the karst, although variations are not always obvious in individual sections through the rocks - for instance, the Devonian/Carboniferous boundary at Yishan is defined on the basis of conodont assemblages and is not evident in the general appearance or character of the limestone. However, distinct differences are to be found in the degree of karstification and the landforms that have evolved on different lithological units (Jie, 1988). Weng Jintao (1987) indicates that 65% of the towers in the Guilin fenglin are formed from sparry allochemical limestone, only 19% from microsparite limestone, 12% from micrite allochemical limestone and 3% from dolomite. The sparry limestone has a lower corrosion rate and more restricted distribution than the micrite limestone but is thicker bedded, has deeper vertical jointing and is mechanically stronger. The steepest towers have formed where the dip of the limestone is most gentle in the centre of the Guilin Syncline (Sweeting, 1990). The behaviour of water in the epikarst is important in the evolution of the fencong because the depressions are the location of most of the near-surface water, remaining wet in the dry season and forming the locus of solution (Williams, 1988).

Perhaps the most spectacular karst is near the town of Duan within the catchment of the Disu Underground River. This karst belt extends to the Guizhou Plateau. It is characterised by steep and rocky terrain, narrow sinuous depressions along major structural lineaments and, higher in the mountains, a magnificent cockpit karst. A spectacular landscape with cones up to 800m high and cockpits as deep as 500m, its verticality and U-shaped polje cross sections recall a deeply eroded glacial landscape - the impression is of a sort of karstic Yosemite. The limestone is strongly folded and the structure is clearly revealed on the rocky hilislopes. The residual hills attain their greatest symmetry where the dip of the limestone is close to vertical, and here vertical rock ribs are prominent. Elsewhere cuesta-type hills have developed on more gently sloping beds.

One of the largest underground streams is the Disu Underground River which has been studied using a variety of techniques including remote sensing, cave mapping, dye tracing, floating geobombs, extrapolations from geological structure, spontaneous potential and resistivity studies (Chen, 1988). The estimated total linear length of the principal arteries in this the system is 200km with the main artery alone approaching 50km long. The Disu River resurges upstream from the town of Duan, its clear and cool waters contrasting with the silt-laden waters of the Hongshui River into which it discharges. The mean discharge of the Disu River is 38 cumecs and the water temperature at the resurgence is 23°C. That dramatic fluctuations in stream level occur was evident from a very former fish hanging hooked on a deadline ~20m above the Hongshui over the lip of the cave entrance. The discharge has been known to reach a formidable 545 cumecs. To put this into perspective, the maximum discharge recorded for any karst spring in the recent karst text by Ford and Williams (1989) is 515 cumecs from the Frio resurgence in Mexico. The largest karst spring complex in the world is at the source of the Manaugat River in Turkey where the mean discharge is 125-130 cumecs, with a flow of 50 cumecs from the largest single outlet at Dumanii (Jennings, 1984). The largest single spring in terms of mean discharge may be Tobio in New Guinea (85-115 cumecs?) (Ford and Williams 1989). The fact that the flood peaks of the Disu Underground River are not attenuated suggests that it flows through an open cave system.

Spectacular surface gorges are prominent features in some Chinese karsts. The Hongshui gorge downstream from Duan is overlooked by mountains that rise 400-500m above a river channel 70-100m wide. During a large flood in 1987 the discharge reached 30,000 cumecs and fresh silt was dumped on straths 50-60m above normal river level. Further north in China the Three Gorges of the Yangtze River, and another major set of gorges on one of its tributaries, are also cut in carbonate rocks (Lu, 1986). The Yangtze floods about every 10 years and during one flood in 1871 it rose 84m and left a steamboat stranded on a rock 37m above normal river level.

Smaller scale surface karst landforms are also a significant tourist drawcard such as the travertine lakes of the Huanglong Scenic Spot in Sichuan (Lu, 1986). There is considerable variety in the surface solution sculpture on the limestone and spectacular karren has also been exploited as a tourist attraction, notably the karst pinnacles of the Stone Forest in Yunnan. Smaller examples of similar phenomena are quite common. For instance, pinnacles 6-7m high are present near the town of Yishan but seem to be regarded as fairly inconsequential by the locals. Steep rock walls and planated surfaces along the banks of the Hongshui River have been deeply etched by solution along joints and scoured by floodwaters.

There are a massive number of cave entrances in some areas - between Baisha and Yanshan (to the east of the Lijiang upstream from Yeingshuo Fuli) I made some timed counts of obvious entrances as we passed by on the road, and the average of three counts was 8.5 entrances/minute. On another occasion one of my colleagues pointed out that from where we were standing (on a boat on the Lijiang) it was possible to see 34 entrances at once! Many of these caves are very old and dissected (Williams, 1987). In some cases it is possible to see right through hills pierced by massive high-level phreatic tunnels. Some tunnels appear to continue through neighboring hills, implying a very considerable age for the caverns. At least 27 Chinese caves are known to exceed 2km in length, the longest being the Tenglong system (37 km) in Licuan, Hubei. Many of the other long caves are in Guizhou province (Wang, 1989). Because most of the limestone in the more celebrated karsts around Guilin occurs as isolated steep cones and towers of limited extent the caves there are often fairly short although passage diameters can be large. However, Baimo Cave in Bama County reaches 13.7km. We visited Jingleng Cave in Mashan County, the seventh longest cave in China with 7.2km of passage. Jingleng contains a very large active streamway and one chamber of 14,000m2. Many caves have long and complex histories with evidence of multiple phases of cave enlargement. For instance, White Dragon Cave near Yishan consists of a steeply inclined lifting tube and a later tube of lesser gradient, located high above the present phreatic zone in a limestone hill. Bai Lian Cave near Liuzhou contains well exposed palaeokarstic sediments including cave breccias, while the massive Jingieng Cave is formed at least partly in a breccia that includes pieces of broken speleothem. Whereas large phreatic tunnels and collapsed caves predominate in the fenglin, the fencong is characterised more by fissures formed by vertical dissolution in a well developed vadose zone, and there is a very active epikarst (Sweeting, 1990).

In several caves there is evidence of passage enlargement by aggressive airflows. Very large scale scalloping caused by a slowly flowing erosive medium truncates both the limestone cave walls and flowstone spelothems. The scale and form of the scallops is very comparable to the 'thumb-print ice' I've seen developed by airflows in ice caves within glaciers. High atmospheric CO2 levels due to guano and bat urine may be involved (D.C. Ford, pers. comm.). I have never noticed scallops of this origin before, although aerogenic corrosion due to exhalation by visitors has, of course, been recognised to pose problems in some temperate tourist caves such as those at Waitomo in New Zealand.

Some very large speleothems have developed in this warm and wet tropical environment. Some speleothems present in the Guangxi caves I have never seen elsewhere or have seen only very incipient examples. These include 'lotus tub' rimstones that appear to be formed by dripwater falling into pools (Zhu, 1988). Shield formations are very common and reach up to ~3.5m in diameter in Ylin Cave outside Nanning. There are often multiple entrances in different parts of the hills which means that airflow through some caves is strong, speleothem dehydration is widespread and actively growing speleothems are less common than I am used to seeing in Tasmania. Radiometric dating indicates that speleothems in some caves around Guilin are more than 600ka old while palaeomagnetic evidence indicates that the clastic sediments in one cave 23m above the Lijiang are at least 900ka and possibly up to 1.6Ma old (Williams et al, 1986).

APPLIED KARST RESEARCH

China's National Program for the Development of Science and Technology has included karst research as one of its most important items (Yuan, 1981). An Institute of Karst Geology has been established in Guilin. Its staff of over 300 are poorly paid by western standards but the Institute is well equipped. In its forecourt stands a statue of Xu Xiake (1587-1641), a Chinese geographer who undertook major research into the karst and developed classification systems and genetic explanations for some of its features. In 1687 he researched extensively in the karsts of Guangxi, documenting 88 caves in the area of Guilin alone. Less comprehensive karst research has been traced back hundreds of years earlier. Attached to the Institute is a purpose-built two storey Karst Museum dedicated to 'Spread Karst Knowledge' and 'Foster Academic Exchange'. It is not uncommon to see up to six tourist buses parked outside at most times of the day. The quality of its displays is superb and reflects the wide scope of applied karst research in China. The exhibition halls extend over 11,000m2 and there are more than 1200 items on display including samples, models, dioramas, photographs, diagrams, laser pictures and videos. These illustrate the distribution of different types of karst in China, karst landforms, hydrology, engineering geology, mineral resources, prospecting and testing techniques and the history of Chinese karst study (Anon, undated). Next door to the Museum is the Science and Technology Exchange Centre. This includes a main conference hall with four-language simultaneous translation facilities, numerous smaller conference and display rooms, four accommodation blocks of 3-4 storeys for visiting researchers, a library, and a restaurant - plus trinkets for academic tourists on sale in a shop beside the compound gate.

Considerable attention is being paid by Chinese researchers to modelling karst groundwater (Chen, 1988; Kang, 1988), geophysical and remote sensing techniques (Chen and Bian, 1988; Yang et al, 1988), isotopic studies (Song et al, 1987; Shi et al, 1988), and geochemical studies (Yang, 1988; Qian et al, 1988), together with water-tracing (Mei 1988; Cai et al, 1988) and numerous other practical studies aimed at enhancing development in the karst regions. Numerous joint research projects with overseas investigators are in progress. For instance, a major Sino-French hydrological study site has been established near Yaji village on the outskirts of Guilin (Bidaux et al., 1988). This study focuses on a catchment of ~2km2 in an area of cockpit karst with a relative relief of ~500m formed on gently dipping limestone. One perennial spring and three intermittent springs are being investigated. Water tracing using fluorescein, ammonia molybdate and zinc chloride has established that some streamsink waters have multiple destinations. Precipitation, water levels in springs and boreholes, pH, temperature and conductivity of springs are recorded automatically. In addition, samples for hydrochemical and isotopic gases and tritium are taken at hourly intervals during storm events. The study is demonstrating the important regulatory function of the epikarst on the deeper karst springs, there being rapid flow, not just infiltration, with shallow epikarst springs playing a major hydrological role (Chai et al, 1988).

SURFACE KARST MANAGEMENT

Water resources

Problems of water supply loom large. Although parts of Guangxi receive rainfall of over 2000mm pa much of the water disappears underground leaving some areas virtual desert. Springs are an important resource and often form a focus for human settlement. In northern China there are at least 50 springs with a discharge exceeding 1000 l/s and as many again in excess of 50 l/s. Many springs have been tapped for irrigation. Karst aquifers account for about 25% of the total groundwater resources of China (Wang, 1988). Preliminary work has identified 2800 underground stream systems in China that it may be possible to harness for drinking water, irrigation or energy generation. Isotopic dating of the underground karst waters has permitted the development of a simple chronological classification that differentiates several vintages that range in age from less than 6 months to Tertiary, and some still older diagenetic karst groundwaters. In Guangxi about 693 underground stream systems have been identified, with a total length of 10,000km and total discharge of up to 23,000,000 m3.

Exploiting the karst water resources has long occupied the Chinese. The 2200 year old Ling Canal in the karst northeast of Guilin was constructed during the Qin dynasty as part of a program by Qinshihuang, the first Chinese emperor, to unify China. It links the Xiang River (part of the Yangtse River System) to the Lijiang (part of the Pearl River System) and was originally used for navigation and irrigation. The scheme consists of a ploughshare island, water diversion dams, spillways and a dyke. The canal is 34km long, about 5km of that having been dug by hand and the rest formed by realigning and diverting streams. This development played a major role in facilitating political, economic and social exchange between northern and southern China, and with the Great Wall and the Dujiang Dam in Sichuan province forms a major legacy of the Qin dynasty. It was reconstructed from the Ming dynasty onwards, was used for navigation until the 1930s, now waters 2660ha of farmland, and in its future economic significance tourism is likely to feature prominently.

In addition to procuring sufficient water, coping with an excess of water has often also proved a problem. Flooding along the middle and lower reaches of the Yangtze downstream of the Three Gorges has led to considerable loss of life. In 1931 the death toll was around 3.7 million, mainly from subsequent famine. Only about 300,000 people died in 1954 after Herculean efforts to reinforce dikes, one report suggesting that 10,000 peasants formed a human dike in waist-deep water until a sandbag wall could hurriedly be installed. Subsequent years saw the construction of more than 32,000 kilometres of dikes and flood-water detention reservoirs and when the river again rose nearly 30 metres in the gorges in 1981 the death toll was less than 1 000 (Clark, 1983). A majority of the 412 members of an expert committee has supported the development of a new hydro-electric project that would be the largest in the world and which would reduce the flood problem. It involves the construction of a dam 185m high and 2.8 km wide to impound 22.1 million m3 of water. It would generate 84 billion kW of electricity annually using 26 generating units with a total capacity of 17.68 million kW. The scheme would require expenditure of $11 billion (USD) over 12 years and involve a 20 year program to shift 1.1 million evacuees. However, likely adverse environmental impacts have stimulated 100 Chinese scientists, writers and academics to write and finance publication of a book strongly opposing the development (Yao, 1988; Becker, 1989). Among the practical difficulties are potential problems with sedimentation, increased seismic activity and possibly karst - a recent issue of the Beijing Review noted that the geological difficulties are considered "resoluble" (!!!). The debate remains unresolved.

Intercepting underground water

In some parts of central Guangxi karst windows provide access to underground water, but utilisation is complicated by large fluctuations in water level. At Lung Koo (Valley Pool Cave) in the Disu area, and in at least one other cave we visited, an inclined tunnel has been excavated and a small railway line installed within it to carry a pump that can be raised and lowered with changes in water level. At another site we visited pumps were installed on a concrete boat that rises and fails with the water. Aqueducts and tunnels have been constructed to convey water between the various poljes, cockpits and other karst depressions where agriculture is focussed.

The village of Gangfang lies at the bottom of a 500m deep cockpit in the catchment of the Disu Underground River. There are no springs that can be tapped, no natural access points and rainwater storage is inadequate. Underground water is present but at a depth of ~100m, hence, it is impractical to pump it. Tapping epikarstic water seems to offer the only practical solution. At another very and site we visited outside Xingping groundwater is likely to be present at -30m depth in a geosyncline but there is no obvious access point and the precise location of the conduits is uncertain. Given the geosynclinal structure at this site and the likelihood of flow through fractures, exploratory drilling along photolineaments may bear fruit. In other cases there is no incentive to utilise easily developed karstic water sources because other supplies are adequate. The Linshui spring near the town of Wuming consists of nine discharge points with a combined flow of 2 cumecs from a catchment of ~100 km2. It has not been intensively researched. A crystal clear lake of ~5000 m2 has been formed at the outlet but is used only for recreation, and as a place for washing clothes.

Dam construction

Some 28 water conservation or hydro-electric power stations of small to medium size have been built in the karst areas of China, together with about 5000 smaller schemes (Zhang et al, 1988). In Guizhou there are at least 25,000 minor surface reservoirs. The Dalongdong water supply storage in Shanglin County, Guangxi, impounds 120 million m3 by damming a streamsink. In addition, dams have been built on underground streams and canals have been constructed in caves. One underground reservoir of 20 million m3 beneath an anticline in Sichuan has involved the construction of a small dam in a tunnel (Yuan, 1981; Lu, 1986).

In some cases small to medium sized dams have been installed at springs such that the cave from which the waters emerge becomes a reservoir. Some underground water storages extend upstream into poljes to form a partial surface reservoir. In some cases the potential exists for impoundments formed by subsurface dams to be linked by tunnel to others formed behind surface dams. Damming a karst channel can back up water in a cave system to a height greatly in excess of the height of the dam - it may prove possible to build an hydro-electric power station above the level of the dam! Natural shafts from the hills above can be used for access to pump out water for irrigation (Lu, 1986).

Hydro-electric power stations on karst streams supply up to 10,000kW. The total output of 10 stations in Dushan County, Guizhou, is 3050kW. Non-carbonate rocks that form aquicludes which perch springs high above valley floors facilitate the development of schemes such as the Wujiangdu station on the Wujiang River in Guizhou. This involves a dam 165m high and will ultimately house three 210kW turbines. Cavities up to 34.6m high were found beneath river level, the deepest cavity encountered being a 10m high cavern at a depth of 220m (Li, 1981). Other hydro-electric stations have been established at the outlet to karst springs including a 650kW installation on the Tianguation River outlet from Wuyan Cave in Luota, Hunnan (Lu,1986).

The Dahua power station, the largest in Guangxi, was commenced in October 1975 and completed in 1985. The impoundment of 26km2 is surrounded by limestone peaks. River level lies at 129.0m altitude, NFSL at 155.0m and flood level at 169.3m. The power station is located inside the dam and contains four 100 MW generators. The scheme is founded on an alternating sequence of limestones and shales with well developed joints and fissures and many caves and faults. Detailed hydrological studies had to be undertaken (Lin and Zhang, 1987) and a 20m deep grout curtain 175m long established, beneath which shale interbeds impede leakage. A canal with locks to allow shipping to pass the dam had not been completed at the time I visited the project in October 1988.

Problems of karst leakage and foundation stability have arisen during the construction of multiple hydro-electric installations on rivers such as the Maotiohe River northwest of Guiyang City, the Longjiang River of northern Guangxi and several others (Cui, 1986; Tan, 1986; Yu, 1986). Initial leakage from caves that extended 40m below the level of the Maotiohe River reached 20 cumecs but this has been diminished by treatment. The largest cavity had a volume of at least 100,000m3 (Yuan, 1981; Lu, 1986; Zou, 1986 ). Leakage problems through cavernous limestone have been addressed by the construction of concrete walls in cave entrances; by blocking smaller channels with stone or concrete; by establishing coffer dams with release valves in entrances; by packing major caves with rockfill; and by establishing stone walls on aquicludes adjacent to the cave mouth. Natural surface clay layers have been used to form aquicludes in areas where the limestone is karstified only to a minimal extent, while elsewhere artificial blankets have been installed using clay or concrete on gentle and steep slopes respectively. Where negotiable caves have been explored to within close proximity of the storage but no entrance exists within the storage perimeter it has sometimes been necessary to reinforce the cave roof, either with a wall or by establishing a bridge as a foundation for an aven filling where heavy loading is anticipated. Caves have simply been filled with concrete or with clay and aggregate. Large sinkholes have also been blocked using fills over an arch structure, while in some cases leakage through very large cavities has been prevented by the construction of an enclosing chimney or well of concrete or stone that protrudes above the lake surface. Discharge gates are sometimes fitted to wells of this sort, a secondary benefit being that periodic draining inhibits siltation of the reservoir. In other cases dams or levees have been constructed around horizontal caves that have been too difficult to fill (Lu, 1986).

In some cases a risk exists of plugs being pushed out by high air pressures that may develop in the caves due to changes in water level or compression of the cave atmosphere by the hydraulic head of the lake via other leakage paths. Self-closing aeration valves may need to be installed beneath the water or floating on its surface. Where the rock mass is simply fractured rather than very cavernous grouting is undertaken. In the case of dam foundations, leakage through caverns has been reduced by hanging grout curtains; grouting to a partial or complete aquiclude; installing concrete cut-off walls to block leakage through small channels, alluvium or large caves; or employing cut-off trenches. Springs encountered during excavation of damsites or dam abutments have been diverted into filters or deflected downstream, or sometimes into the reservoir itself to avoid scour beneath the dams decreasing the buoyant support of structures. Leaks have also been drained via pipes in the dam, in the downstream riverbed or by natural channels (Lu, 1986).

Such developments are not without adverse environmental impacts (Dai, 1988). For instance, interference with the flow of scenic waterfalls in karst areas, such as the Huangguoshu Fails, may occur in future due to hydroelectric development. Similarly, considerable opposition was aroused by the proposed Three Gorges hydroelectric development on the Yangtze due to adverse environmental impacts that were anticipated, coupled with economic and social impacts.

Agriculture

The availability of soil and water dominate the distribution of agricultural activity. Cropping is focussed in karst depressions and on the plains in the areas of tower karst (Zhu et al, 1982). Peasant huts are often perched on the rocky lower slopes of sinkholes that are floored by soils that can be tilled. In addition to the soil resource, epikarstic water and sometimes springs from the deeper karst are sometimes available to support crop growth. In many cases irrigation systems have been developed that involve pumping water from underground sources. In some locations springs have been dammed, providing the possibility of raising fish. In other cases slash and burn shifting cultivation occurs on steep limestone slopes. Generally, however, there is only limited incursion onto steeper slopes except where tillable land is in short supply and terraces have been constructed. However, abandoned hillslope terraces are common, probably due to problems of water supply. Along the Hongshui Gorge steep banks of freshly deposited flood silts are tilled to near water level, while pockets of silt between limestone outcrops on the prominent straths ~50m above the river are also cultivated, the houses standing on stilts on the rockier areas.

Forestry

Perhaps the most impressive area of native forest I saw in Guangxi was around Haiyang Town in Ling Chuan county where there is a forest of more than 20,000 gingko trees. However, in many areas of karst severe deforestation problems have been caused by firewood cutting which has involved the removal of most burnable wood. Larger scale production forestry in Guangxi appears to be focussed mainly on extensive pine, eucalypt and other plantations. Large areas were aerially seeded in the late 1960s. The plantations seem to cover some thousands of square kilometres. Abundant inexpensive labour has no doubt helped facilitate their development. Eucalypts are also very common lining roadways in Guangxi. Most of the plantations have been established on non-carbonate rocks, primarily the Cretaceous red beds. In only a few places did I see plantations overlapping onto the limestone. In each case there was a rapid decline in tree size, form and vigour once the limestone boundary was crossed. I suspect this may have something to do with karst accentuated drought effects caused by transpirational withdrawal of moisture from the soil in restricted solution pockets or by the subsurface karren serving as drains.

Much of the timber is used for construction. Of all the logs I saw in forests, on log trucks and in yards through central Guangxi, and in the rail yards on the long train trip from Nanning through western Guangxi, Hunan and Guangdong provinces back to Guangzhou, I saw none greater than ~40cm in diameter. Perhaps the harvesting equipment has something to do with this - the one clear felling operation I saw in progress in Guangxi involved manual carriage of the logs to the trucks by labourers. Dry weather extraction by people in sandshoes appeared to involve appreciably less ground disturbance than I am used to seeing caused by tracked skidders or even cable logging in Tasmania! Much smaller wood is stripped from the limestone mountains for firewood and this plays a major role in the deforestation problems.

Acid rain is emerging as a major difficulty. About 90% of the trees have apparently died in the 6000ha Maocaba pine forest in Fengjie County, Sichuan, where there is reportedly no living grass within 1 km of the source of the sulphurous emmissions. Most of Sichuan and Gizhou provinces are subject to acid rain (ie pH of 5-6) while Guangxi, Zheijiang, Hunan and Shanghai are also affected. The problem arises from poor quality fuels (including impure coal interbedded with the limestones), topography, weather conditions, poor management, backward technology and poor equipment. The very limited wind inhibits dispersal (and export) of the acid rains (Li, 1987).

Communications

Rugged karst poses major impediments to the construction of roads and railways. River traffic is important in the communications system. In some areas major efforts have been devoted to establishing improved road access. However, partial collapse of a major suspension bridge across the Hongshui Gorge still had not been made good when we visited the site a year afterwards testifying to the practical and economic difficulty of even this conceptually simple task.

In areas of steep karst such as the limestone ranges of Duan County, Guangxi, the few roads sinuously contour around large enclosed depressions and ascend steep hills via extensive complexes of switchback corners. Elsewhere the construction of tunnels sometimes forms an alternative, as on the outskirts of Guilin City near Reed Flute Cave. However, tunnel construction can pose additional complications. Railway tunneling projects in the southwestern provinces in particular have encountered difficulties after breaking into caves in the limestone. These difficulties have included out-rushes of water from flooded karst cavities, the collapse of fills of cave sediment and the collapse of foundations into deeper caves. Encountering underground rivers has sometimes meant the additional difficulty of having to construct railway bridges inside caves. Tunnel construction has also required filling unstable caves; lining tunnel walls; installing artificial roof supports; drainage; and diversion of streams.

Even outside the tunnels railway engineers have encountered difficulties with the stability of constructions such as the rail-bed, bridges and buildings; the stability of excavations such as embankments; and the acquisition of a water supply. Treatment measures for embankment foundations have included filing cavities with rock fill or concrete; covering soil-filled crevices with a concrete shield and rock fill layers; the construction of arch beams over large voids; bridging unstable ground from piers on sound bedrock; and diversion and drainage of waters. Bridge foundations have been treated by filling; by bridging from foundations of solid rock; by grouting; and by construction using tube stakes (Lu, 1986).

The presence of high limestone mountains both demands and facilitates the construction of radio communications towers on prominent summits, although as one experience in Mashan taught us it can be frustrating to find a hill with an easy route to the summit only to encounter barbed wire and a locked gate that blocks access to the best scenic lookout point (although in this case not an insurmountable obstacle).

Building construction

Building on the surface

Problems of unsound foundations, foundation collapse, unpredictable foundation depth and accelerated sinkhole formation have all impacted upon the construction of buildings in the karst areas of Guangxi. As is generally the case, the greatest problems involve cavities formed in regolith by the piping of its finer components into karst voids in the underlying carbonate bedrock. There have been three forms of response. Firstly, systems of land classification have been developed to facilitate better land-use planning. Secondly, because the susceptibility of karst terrain to collapse is accentuated by fluctuations in water level, local government controls on the pumping of groundwater have been promulgated. Finally, in some cases cavities have had to be filled with stone, concrete or grout prior to the commencement of construction. Some of the most serious problems have arisen where undergound mining has necessitated the pumping groundwater, leading to a reduction in buoyant support for the regolith and greater mobility of water. Prevention of collapse and protection of karst water supplies in an unpolluted condition are accorded an increasingly high priority.

Construction in caves

Caves have been used as air-raid shelters, storage places and factory sites (Yi, 1985). Buildings up to eight storeys high have been constructed in some Chinese caves (Lu, 1986). In addition to the need to carefully assess foundation stability, additional problems that have arisen include the stability of cave walls and ceilings and also flooding by cave streams. Measures such as the use of reinforced concrete arching, solidification grouting and strengthening with steel anchors have been employed to overcome these difficulties (Lu, 1986). About 8km outside of Yangshiu town lies Laojing Cave, a stream cave 200-300m long that passes through a peak cluster formed from pure limestone of upper Devonian age. Inside the 30m high downstream entrance lies a large two story building with thick concrete walls which we were informed was a bicycle factory. A power failure precluded our seeing much inside the factory, the upper floor of which appeared to be used for storage, although I did catch a glimpse of heavy machine tools that were installed on the lower floor. A buried pipe that emerged from one wall of the cave suggested that the factory was not dependent on the cave stream for its water.

Mineral resources and mining

At least 65 different kinds of ores comprising at least 53 minerals have been identified in China's karsts including placers, carbonate and sulphate minerals in caves (Yuan, 1981; Lu, 1986). Mercury, copper, lead, zinc, iron, tin, gold, silver, bentonite, uranium and rare earths are among those present in the karst (Yao et al, 1986; Zhang, 1986). The karst context is significant to the occurrence of some mineral resources, as, for example, the presence of residual materials such as bauxite in palaeokarst cavities. Hydrothermal karst processes have formed various metallic and non-metallic mineral resources (Zhou, 1986).

Mining for coal, bauxite, pyrite, iron and non-ferrous metals takes place in some of the karst areas. The scale of mining operations varies widely. At the simplest level there are very small scale and shallow coal mines working narrow beds within the limestone formations. One which I inspected at the mouth of the Hongshui gorge downstream of Duan Town was a very simple affair where the underground workings were illuminated by bouncing light down the mine using fragments of broken mirror. However, some mining operations have penetrated more than 1000m below the level of valley bottoms and plains. This has necessitated pumping out groundwater, sometimes involving flows of up to 10 cumecs. Other problems include rapid inrushes of water undergound, invasion water often carrying with it considerable quantities of sediment. There are also serious problems of instability and collapse of the ground surface above the mines due to the draining of groundwater and this has led to the abandonment of entire villages (Yuan, 1987; Xu and Zhao, 1988). Treatment measures to resolve difficulties encountered in mining karst areas have included blocking the outlets of underground streams; injections of concrete or silicate into solution cavities; diversion of surface streams; and blocking stream sinks and sinkholes. Mine pumping and draining, grouting, and the installation of blankets on river beds and in other infiltration areas have been employed. Planning has also had to take into account the effects of mining on domestic, agricultural and industrial water supplies and drainage (Yuan, 1981; Lu, 1986).

Limestone quarrying is widespread but is fundamentally different to the usual pattern in the west. Rather than there being a few large quarries that totally change the morphology of the hills being worked, the usual pattern around Guangxi seems to be for many quarries to win only very surficial stone by manual means and limited use of explosives, which leaves widespread defacement but only limited morphological change. Quarrying often appears very hazardous with barefoot or sandal-shod workers clinging to the quarry walls as they prize loose slabs of limestone and shout a warning to other workers below them. Larger scale quarries are present adjacent to the main rail line through Guangdong province to the north of Guangzhou.

In northern China electrical energy is generated using coal quarried from within karst areas and similar development is likely to occur in the south of the country. The presence of karst is likely to complicate the draining of coal pits and mines, and the development of surface water storages. There is also considerable potential for harnessing oil and gas reserves that exist in karst reservoirs. Some of these occur in palaeokarst cavities that have been buried by Cainozoic rock units, as in the North China Plain. In other cases the karst occurs at a considerable depth beneath pre-Cainozoic rocks, some of these reserves being offshore. Thermal springs are presently subject to only low level development such as use for swimming pools, sanitoria, agriculture and minor industries, but deeper geothermal resources in karst offer potential for larger scale develeopment (Lu, 1986).

Land surface stability

Construction activity and human safety are both threatened by karst collapse. Of 30,000 recorded sinkhole collapse incidents in Guangdong, Guangxi, Hunnan, Hubei, Yunnan, Guizhou and Jiangxi provinces about two thirds are believed to have resulted from human intervention in karst processes (Jian, 1988). Collapses have damaged buildings, bridges and dams and resulted in the flow of polluted water to karst aquifers, land degradation, the flooding of mines and tunnels, impairment of opportunities to exploit water supplies, and injury or death to humans (Song, 1987). Some collapses have occurred within artificial reservoirs (Chen, 1986). Epikarstic processes rather than modern solution of bedrock, together with the structure of the regolith, are highly significant to questions of collapse. Collapse has occurred in many cities where karst aquifers are pumped for groundwater, including Kunming, Guiyang, Hangzhou, Guilin, Nanjiang, Shuicheng, Yulin and dioujang. It is most prevalent where the regolith is less than 5m thick. Where water-table fluctuations occur the major factors leading to collapse include the loss of buoyant support to the overburden; suffusion; sudden increases in water level that force gas upwards into the roof of a cavity; washing out by underground streams; and, it is argued, the development of a vacuum when the water table drops in a cavity (Jianyi and Jian, 1983; Jian, 1988). I had never considered the latter mode of formation before visiting China, but it may well be very relevant in Tasmania where many of our karsts are mantled by dolerite-rich Quaternary sediments that contain a high proportion of clay that may restrict exchange of air between subsurface cavities and the external environment. I am reminded of a car sticker I've seen about: "There's no such thing as gravity - the earth sucks".

Nature conservation

Nature conservation is accommodated in China by a spectrum of land tenure categories that range from unclaimed wild lands to Chinese gardens. In common with most western countries public recreation and biocentric perspectives have predominated in the selection of sites, but in some cases this has fortuitously given protection to natural karst features. The concept of nature conservation in China has a truly Chinese character, its roots lying in three prominent influences in Chinese gardens: Confucianism, with its stress on the simple and the ancient; Taoism and its view of humanity as being in constant search for unity with the universe; and Buddhism, the arrival of which accentuated the perception of summits as sacred places (McLaughlin, 1982). De facto protection of sorts has been afforded to some hilltops such as karst towers through the construction of temples and gardens upon them. Some caves have also been viewed as sacred places, although while this may safeguard them from deliberate destruction the modification of outcrops and caves and the carving of statuary coupled, with considerable traffic, has had a profound environmental impact.

A variety of parks and reserves exists and this system is being expanded. For example, the Guilin Miaorishan Forest Reserve in Xingan and Ziyuan Xians was approved by the Guilin District Revolutionary Committee in 1978 to protect rare conifers. It extends over 1559ha of which 720ha is strictly protected, and provides protection for karst catchments. It is managed by the District Agriculture Committee. Another protected area with implications for karst is the 7997ha Lungzhou Longgang Forest Reserve at Lungzhou-Xian which was established to protect the white-headed monkey, limestone evergreen monsoon forest, and natural ecosystems. Natural Landscape Areas have also been established in areas of natural scenery or cultural attractions that are predominantly natural in character such as the Lijiang River (McLaughlin, 1982). My general impression, however, was that life goes on as normal, and that the difficult terrain makes the karst its own best defence.

CAVE CONSERVATION AND MANAGEMENT

Cave tourism

There appears to be little organised recreational caving in China. Much of the karst is unexplored from the point of view of caves, the main exceptions being where water supply or other practical requirements have been a stimulant. On the other hand, the level of cave tourism surprised me. Zhang (1986) lists 50 caves open to the public, but advice received in October 1988 suggests that at least 200 caves have now been developed for visitors, most of them for internal tourism. Visitor numbers are prodigous. I am used to annual visitation to the tourist caves at Mole Creek at home in Tasmania totaling ~25,000, and was awe-struck when I first realised that this figure is attained just over the Easter weekend at Jenolan in NSW where annual visitor numbers exceed 250,000. However, the small caves near Liuzhou, the third city in Guangxi, attract nearly twice the annual visitation Jenolan does. This says a lot either about the patterns of recreation by the citizens of Liuzhou or the mobility of Chinese tourists. Reed Flute Cave is one of a series of caves on the outskirts of Guilin in a limestone hill known as Guangming. Although less than 1 km2 this hill contains a number of caves up to 1 km long. At the time of my visit there must have been at least 1000 people in the cave at any one lime, a continuous line entering and leaving the two entrances. Seven Star Cave, also in Guilin, consists of at least two levels each of which exceeds 1km in length. The width of the passage reaches 50m and its height ~20m. It attracts hundreds of thousands of visitors annually and has been a focus of visitor attention for hundreds of years.

Coloured lights are the norm for general cave display, mostly unshielded fluorescent tubes scattered about the cave floor. After visiting several caves lit almost exclusively by coloured lights I must concede that different colours shone from different directions does highlight textural differences to a degree we've never seen achieved using plain lighting, though I suspect their use simply reflects Chinese tastes for colour. Probably more significant in an absolute sense is the standard of the electrical installations underground - wiring is often pretty rough, conspicuous, and sometimes rather alarming. Interpretation is very much oriented around fantasy and I saw little attempt to convey any significant scientific messages to visitors - the traditional Snow White and the Seven Dwarves, Madonna and Child and similar figures that haunt Australian tourist caves are replaced by squadrons of monkey kings, fairies and dragons, sometimes with flashing lights for eyes. International visitors are well catered for - Reed Flute Cave boasts a feature known as 'Bird like Emu Gazing at Mosquito Net' (Gordon, 1990). One speleothem in the ~1100m long Ylin Cave near Nanning is known as the 'fruit tree' and the guide enthusiastically points out its remarkable assemblage of produce that ranges from rice to bananas, lotus, mushrooms and carrots! In this cave visitors sit on a clear-felled forest of concrete tree stumps to listen to piped music while a 'karst' of unmoving stalagmites sing and act out an opera on a natural stage. Overstaying the pause at this site does provide a comfortable opportunity to reflect quietly on one's surroundings, something uncommon in Australian tourist caves.

Bats continue to frequent some tourist caves, including Bai Lien Cave near Liuzhou, despite constant disturbance by tourists and cave guides shining torches at them, although in other parts of the world this sort of behaviour would be guaranteed to ensure that the bats abandoned the cave. The magnificent Jingleng Cave in Mashan County appears to have been previously lit for visitors. Its fauna includes very mean-looking snakes with an exceptional ability to climb rock walls and even cling to flat and smooth epiphreatic cave ceilings, with predictable results in adrenalin production by visitors who suddenly find themselves beneath one, although the snakes are presumably more likely to prey on bats.

Little respect seems to be shown for speleothems. In Reed Flute Cave, together with some other Chinese tourist caves I visited, the hand rails along the paths in some areas consist of a picket fence of stalagmites broken off from around the cave and cemented to the floor. At Reed Flute Cave visitors paddle through the rimstone crystal pools and clamber onto stalagmites to have their photos taken, all bathed in green, red or blue light. Litter is common along the paths. To western tastes at least the cave resource and its aesthetic appeal appears seriously degraded. Indeed, the crowded and noisy Reed Flute Cave has contributed much to my conception of what Hades must be like. Surely it is the place of eternal atonement for karst managers who have been very, very bad. Be warned! On the other hand, in at least two caves I saw 'No Smoking' signs.

My overall impression is that some of the caves must once have been quite superb, and some still are - there remain some really magnificent speleothems in caves like Ylin, Reed Flute and Jingleng. But leaving aside superficial issues such as coloured lights, the physical condition of the cave resource itself does warrant concern. Given the massive numbers of visitors, the cumulative effects of hundreds of years, economic constraints on development and the emphasis on fantasy rather than natural history, it is remarkable that many of the caves remain as fine as they do - but there is plenty of room for improved management.

Prehistoric and historic sites.

Prehistoric and historic values may have fared better than the natural values of caves. While historic sites in particular have suffered severly during the cultural revolution and probably at other times some seem to have survived relatively intact. Modern corrosion of stone inscriptions is a significant management problem in a number of areas. During my time in China I saw only one English language newspaper. One of its front page stories addressed this very issue.

Impressive prehistoric sites occur in the karst, and some of these appear to be being managed rather better now than they were when the focus of active investigation by archaeologists in the recent past. For instance, Bai Lien Cave (White Lotus Cave) ~12km southeast of Liuzhou contains cultural deposits in a southward-facing antechamber ~27m above the foot of a limestone hill. Excavation over many years has revealed ~3500 artifacts, including stone tools (large cobble choppers, flint microliths, polished tools), bone (including human remains) and ceramics. These remains span the late palaeolithic - mesolithic - early neolithic, with initial occupation dating to ~30,000 years ago. The deposits have been excavated almost totally leaving little for future study as new techniques become available. Zhenpi Cave on the outskirts of Guilin is a southward-facing single chamber about 15m in diameter with a floor area of ~200m2 and a cultural deposit ~3m deep. It contains shells, ceramics, bone and stone tools, and at least 14 intact human skeletons. Prehistoric use of Zenphi Cave occurred from 7500-10,000 years ago. Some 60% of the deposit has been excavated.

Bai Lian Cave is run as a museum that seeks to focus on "archaeology, karstology, geology, biology, mineralogy, meteorology and hydrology". Steps have been constructed to the antechamber to allow the public to view the site of the excavations through a large perspex sheet that seals its entrance, although very little remains there to be seen. Life-size human statues that depict putative prehistoric lifestyles have been installed in many parts of the cave although there is no evidence of human activity other than in the antechamber. In one case they crouch cooking food over a red coloured electric light globe. It was difficult to judge the commentary given that it was conveyed through a translator unfamiliar with the subject, but I gained the impression that considerable emphasis was placed on humanising the experience for visitors by recourse to descriptions of social systems that may or may not have been based on real evidence. In Zenphi Cave some material, including intact human skeletons, has been left partly unearthed for public display. Wooden gang-planks have been used to minimise erosion of the remaining deposits, and a corrugated iron ceiling installed to redirect potentially erosive dripwater. Visitors are not permitted past the dripline where a barrier has been erected, but are able to look right into the excavations, and to view some artifacts on display in polished wooden showcases set incongruously upon the dusty cave floor. Photography is not permitted. A comprehensive museum has been established in a building adjacent to the cave.

Many caves have important historical associations. Inscriptions on the walls of Reed Flute Cave in Guilin, and other caves in the same hill, date back to 792AD (Tang dynasty). These written records span 77 different dynasties and include 93 inscriptions from the Sung dynasty alone. They appear to be treated with great respect by visitors. The early Chinese geomorphologist Xu Xiake twice surveyed Guilin's Seven Star Cave in the summer of 1637 (Ming dynasty) and there are indications of earlier scientific investigations dating back 1400 years. Once again I gained the impression that more respect was shown for its historical values than its natural values. White Dragon Cave near Yishan is also noteworthy for inscriptions that extend back hundreds of years. Despite the fact that many of the inscriptions are outside the cave mouth in an area where visitors gather there is little evidence of any vandalism. This cave was also visited by Xu Xiake. It also contains Buddhist carvings and statues that escaped destruction during the cultural revolution. One of the most remarkable examples of respect being shown for historical sites may be Laojing Cave near Yangshiu. The cave walls bear Taoist inscribed tablets 800-1000 years old in at least two locations. Although a factory has been built in the cave it is not flush to the cave wall where inscriptions are located and I gained the impression that this may have been a deliberate attempt to preserve them.

Western Hill in Guilin City comprises a group of peaks that contain statuary that dates from the spread of Buddhism into southern China in the Tang dynasty. More than 200 statues have been carved into the limestone outcrops. They vary from 50cm to 2m high and many are set into carved niches. The statues are detailed and the faces are full of expression. Many were damaged by the excesses of the cultural revolution and some perhaps earlier, but a large number remain intact. A walking track has been established at Long Tou Feng (Dragon Head Peak) where about a third of the statues are focussed, but there does not seem to have been any defacement of statues despite this increased access. Western Hill is one of about a dozen city parks scattered around Guilin that make use of limestone towers and caves. These parks are increasingly recognised as being important to the tourist industry. The speed with which high standard walking tracks are constructed in difficult terrain is truly awesome, particularly when one takes into account that the tracks and handrails are constructed from hand-carved stone!

CONCLUSIONS

The immediacy of life is apparent everywhere in China. At one level this may assert itself to the visitor who, like I did, orders chicken for lunch and finds it being chased around the restaurant before being despatched behind a box. But at another level it is equally evident in the relationship between the rural Chinese people and the landscape in which they dwell. The imperative in China has been to survive - if it moves, eat it. Although the need for a systemic approach to karst management is recognised (Yuan, 1988b), there are not the economic resources to permit some of the niceties of karst management that we expect in the west, nor any longer much pristine karst where their adoption would offer much practical benefit. While China may contain more karst than any other country on earth it probably also contains more degraded karst than any other country on earth. China needs to improve its knowledge and management of its karst for economic reasons if for nothing else. Some remarkable things have already been achieved to improve the living standards of the Chinese people, although enormous difficulties still await resolution. On the other hand, China lacks the economic capacity to indulge unassisted in large scale nature conservation programs such as we know in the west.

These considerations emphasise that there is inevitably a major role to be played by developed countries such as ours if adequate reservation of representative and outstanding samples of the world's different kinds of karst terrain are to be safeguarded. This role includes managing our own karsts in an appropriate manner, drawing on the experience of places like China. But it should also include our provision of assistance to developing countries who may lack the economic resources to manage adequately important natural or cultural sites within their national boundaries. For instance, one of the aims of the World Heritage Treaty is to provide a mechanism for rendering assistance where a nation's economy is incapable of providing the level of management required to fully protect treasures of international significance. Obviously this treaty does not offer the only mechanism for rendering assistance, but if some of these Chinese karsts are not world heritage then I don't know what is!

ACKNOWLEDGEMENTS

For their magnificent hospitality, friendship and assistance I am grateful to Song Linhua of the Yunan Institute of Geography; Fang Jinfu of Academia Sinica in Beijing; Zhao Xun, vice secretary general of the Geological Society of Guangxi; and also Qian Xiao-e, Chen Wenjun and other members of the Hydrogeological team of Guangxi who organised and accompanied us through central Guangxi. I would also like to thank the governors and officials of Guilin, Luizhou, Yangshiu, Xingping, Yishan, Duan, and Mashan for their hospitality. Grateful thanks are also extended to Prof. Yi Guangyuan and his colleagues in Liuzhou. I also acknowledge Yuan Daoxian, Zhu Xuewen and their colleagues from the Institute of Karst Geology for their efforts in organising field activities around Guilin in conjunction with the 21st IAH Congress. I am indebted to various other companions in China for sharing their insights. The Tasmanian Forestry Commission provided financial assistance towards my attendance at the 21st IAH Congress in Guilin which provided the initial stimulus for my visit to China.

REFERENCES

ANON undated The Chinese Museum of Karst Geology - Briefing. The museum, Guilin. 16pp

BECKER, J. 1989 China greens try to halt Yangtze dam. The Guardian 12 March 1989:10

BIDAUX, P.; YUAN, D.; DAI, A. & CHAI, W. 1988. Identification of groundwater flow in a karstic aquifer from hydrochemical measures along boreholes (Experimental site of Guilin, China). Karst Hydrogeology and Karst Environment Protection. (Pap. 21 st IAH Congress, Guilin, China, 10-15 0ctober 1988):871-876. Geological Publishing House, Beijing. 2 vols

CHAI, W.; DAI, A.; YUAN, G; ZHAO, C. & BIDAUX, P. 1988 Studying the recharge boundary of a karst spring system by means of tracers. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st JAH Congress, Guilin, China, 10-15 October 1988):902-909. Geological Publishing House, Beijing. 2 vols

CHEN WENJUN. 1988 The study of Disu underground river system, Du'an county, Guangxi. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st. IAH Congress, Guilin, China, 10-15 October 1988):543-551. Geological Publishing House, Beijing. 2 vols

CHEN YUSUN & BIAN JI 1988 The media and movement of karst water. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 1 0-1 5 October 1988):555-564. Geological Publishing House, Beijing. 2 vols

CHEN ZIKANG;QIN ZHENGJIAO; LU CHENGJIE & TANG XINGYU E (1 988) A study of shallow seismics for predicting collapse in cave and soil layers. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):743-749. Geological Publishing House, Beijing. 2 vols

CLARK, C. 1983 Flood. Time-Life Books, Amsterdam

CUI QUICHAO 1986 Analysis of leakage characteristics of reservoirs and dams along the Gorge River course in Cambrian and Ordovician carbonate rocks in north China. Carsologica Sinica 5(l):40-46

DAI JINGCHUN 1988 Karst environmental effect of water-conservancy and water-electricity projects in mountain areas of Guizhou, China. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):135-143

FORD, D. & WILLIAMS, P. 1989 Karst Geomorphology and Hydrology. Unwin Hyman, London. 601pp

GORDON, H. 1990 Stretch of the imagination on the River Li. The Australian 18-19 August 1990: Review 13

JENNINGS, J.N. 1984 Karst Geomorphology. Blackwell

JIAN CHEN 1986 Percolation and collapse of reservoirs in karst regions. Carsologica Sinica 5(2):105-111

JIAN CHEN 1988 Karst collapse in cities and mining areas, China. Environmental Geology 12(l):29-35

JIANYI TAN & JIAN CHEN 1986 Preliminary study of karst collapse forecast method. Endins 13:99-104

JIE XIAN. 1988 Karst Remote Sensing Images in China. Chongquing, Chongquing Publishing House. 133pp

Ll HAIBO. 1987 Acid rain harms Southwest forests. Beijing Review 30(42):7-8

KANG FENGQUN & ZENG JIANNIAN 1988 Statistical forecasting of karst groundwater. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 0ctober 1988):628. Geological Publishing House, Beijing. 2 vols

Ll MAOQIU. 1981 The deep karsts in Wujiang Valley at Wujiangdu dam site. pp. 732-733 [in] B.F. Beck (ed.) Proceedings 8th International Speleological Congress 2. National Speleological Society, USA

Ll ZHENQIAN. 1988 The Speech on Receiving the Deputies to the IAH Twenty-First Congress. unpublished mss, 7pp

LIN JUNSHU & ZHANG ZHAOREN 1987 Primary study on environmental geomorphologic pattern for agriculture in mountain region of the northeast Guizhou Province, China. Mountain Research 5(3):129-135

LU YURU 1986 Karst in China -Landscapes. Types. Rules. Beijing. Institute of Hydrogeology and Engineering Geology, Chinese Academy of Geological Sciences, The Peoples Republic of China

McLAUGHLIN, W. 1982 Nature conservation, wildland recreation and wildlife conservation. Chap. 3, pp54-104 [in] Forestry in China. FAO Forestry Paper 35, Food & Agriculture Organisation of the United Nations, Rome

MEI ZHENGXING 1988 An outline of the underground water tracing in karst regions of China. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 0ctober 1988):894-901. Geological Publishing House, Beijing. 2 vols

QIAN XIAO'E & HE XIAOMING 1988 The hydrochemical sub-regime of karst ground water and its absolute age of ground water in Liuzhou area. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):870. Geological Publishing House, Beijing. 2 vols

SHI HUIXIN; CAI ZUHANG & XU ZHIFAN 1988 An isotopic study of the groundwater ages in region of carbonate rocks. Karst Hydrogeology and Karst Environment protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):785-792. Geological Publishing House, Beijing. 2 vols

SONG LINHUA 1987 Pumping subsidence of surface in some karst areas in China. pp.49-64 [in] J. Kunaver (ed.) Karst and Man. Proceedings of the International Symposium on Human Influence in Karst, 11-14th September 1987, Postojna, Yugoslavia. Department of Geography, Philosophical Faculty, University E. Kardelj of Ljubljana. 265pp

SONG LINHUA; FANG JINFU; ZHANG ZHONGLU & WANG CHANGFU. 1987. Stable isotope hydrology in Niukouyu area, Beijing. Carsologica Sinica 6:94-100

SWEETING, M.M. 1990 The Guilin karst. Z. Geomorph. N.F. Suppl. Bd. 77:47-65

WANG XUNYI 1989 The longest karst caves in China. Proc. 10th IUS Congress, Budapest:487

WANG ZHAOXIN 1988 Assessment and development management patterns of karst water resources in China. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):47-57. Geological Publishing House, Beijing. 2 vols

WENG JINTAO 1987 Karst and carbonaterocks in Guilin - Table 1, page 2 [in Chinese, English summary]

WILLIAMS, P.W. 1985 Subcutaneous hydrology and the development of doline and cockpit karst. Z. Geomorph. 29(4):463-482

WILLIAMS, P.W. et al Interpretation of the palaeomagnetism of cave sediments from a karst tower at Gui1in. Carsologica Sinica 5(2):119-125

WILLIAMS, P.W. 1987 Geomorphological inheritance and the development of tower karst. Earth Surface Processes and Landforms 12:453-465

WILLIAMS, P.W. 1988 Hydrological control and the development of cockpit and tower karst. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):281-290. Geological Publishing House, Beijing. 2 vols

YANG CHANGTE & CHANG SHUWU 1988 The application of remote sensing images to develop ground water and its control roles in Lizhou karst areas. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988): 774. Geological Publishing House, Beijing. 2 vols

YANG LIZHENG 1988 Some chemical characteristics of karst water in South China (abstract). Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988):846. Geological Publishing House, Beijing. 2 vols

YAO JIANGUO 1988 Experts okay Three Gorges project. Beijing Review 12-18 December 1988:5

YAO ZHENKAI; LIU YUAN & ZHU RONGBIN 1986. Types and characteristics of karst uranium deposits, Carsologia Sinica 5(2):79-85

YI HSINGYUAN 1985 Several problems on the use of natural karst caves in Peoples Air Defence and stockpile of materials. J. Quizhou Univ. 2185:138-141. (English summary only)

YU SHAOZHOU 1986 Preliminary analysis of possible leakage on the right bank at the head of Guanying reservoir. Carsologica Sinica 5(l):61-67 (English summary)

YUAN DAOXIAN 1981 A Brief Introduction to China's Research in Karst. Institute of Karst Research, Ministry of Geology, Guilin, Guangxi, China. 35pp

YUAN DAOXIAN 1987 Environmental and engineering problems of karst geology in China. pp. 1-11 [in] B. F. Beck & W.L. Wilson (eds.) Karst Hydrogeology: Engineering and Environmental Appplications. Balkema, Rotterdam. 429pp

YUAN DAOXIAN 1988a Editorial - An environment worthy of special attention. Environmental Geology 12(l):1-2

YUAN DAOXIAN 1988b. On the karst environmental system. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988): 30-46. Geological Publishing House, Beijing. 2 vols

XIANG SHIJUN & CHEN JIAN 1987 Karst collapse induced by pumping and draining groundwater, its forming conditions and factors affecting. pp. 65-76 [in] J. Kunaver (ed.) Karst and Man. Proceedings of the International Symposium on Human Influence in Karst, 11-14th September 1987, Postojna, Yugoslavia. Department of Geography, Philosophical Faculty, University E. Kardeij of Ljubljana. 265pp

XU WEIGUO & ZHAO GUIRONG 1988 Mechanism and prevention of karst collapse near mine areas in China. Environmental Geology 12(l): 37-42

ZHANG SHOUYUE 1986 The tourist caves of China. Communications 9th IUS Congr., Barcelona 1986 2:225-228

ZHANG SOUYUE; XIN KUIDE & JI CHUANMAO 1988. Research on hydrogeology in karst regions of China. Karst Hydrogeology and Karst Environment Protection. (Pap. 21st IAH Congress, Guilin, China, 10-15 October 1988): 1 -9. Geological Publishing House, Beijing. 2 vols

ZHANG XIANGXUN 1986 A preliminary study on the formation and enrichment of karst placer deposits of quartz crystal in west Guangxi. Carsolgica Sinica 5(2):87-95

ZHAO YANNIAN 1988 Exploiting the Rocky Mountains for the Prosperity of Minority Peoples. Text of address to IAH 21st Congress, Guilin. 5pp.

ZHOU ZHIQUAN 1986 Preliminary study on karstification and hydrothermal metallogenesis of the Jiepaiyu realgar deposit, Hunan province. Carsologica Sinica 5(2):71-76

ZHU JINGJIOO; LIN JUNSHU & ZHANG YAOGUANG 1982 Karst landforms and leakage analyses of Dahua Reservoir area in Guangxi. Scientia Geographica Sinica 2(4):337-347

ZHU XUF-WEN 1988 Guilin Karst. Shanghai Scientific and Technical Publishers. 188pp

ZOU CHENGJE 1986 Research on karst leakage and prevention treatment of the reservoir and dam site of the Fourth-Stage Maotiaohe River hydroelectric power station. Carsologica Sinica 5(l):1-14