Regional Environmental Technical Assistance 5771
Poverty Reduction & Environmental Management in Remote Greater
Mekong Subregion Watersheds Project (Phase I)

BIODIVERSITY & PROTECTED AREAS

Regional Report

By

J E Clarke, PhD

CONTENTS

STRATEGIC SUBREGIONAL ANALYSIS

1. Introduction

Methods

Regional and country reports were compiled during a six-month study. Most of the period was spent in Vientiane, Lao PDR, where desk research and interviews were used to gather local and regional information, and where the reports were written. Visits were also made to the capital cities of the other GMS countries: in this context the Yunnan Province of China is treated as a country. Two visits were made to protected areas: Xishan Nature Reserve near Kunming in Yunnan, and Hlawga Wildlife Park on the outskirts of Yangon, Myanmar

Methods of information gathering were highly dependent upon the organisations contracted to provide support in each country, and the manner in which support was given. The organisations were varied: government institutions (Myanmar and Yunnan), IUCN (Thailand and Viet Nam) and private consultancies (Cambodia and Lao PDR). In some cases, locally employed counterparts acquired information and transmitted it to the Specialist; in others, the Specialist acquired information directly from reports and interviews—access to which was facilitated by the organisation employed to provide support. Local counterparts were assigned in Cambodia, Lao PDR and Yunnan. Information was therefore variable in depth and focus.

The World Conservation Monitoring Centre (WCMC) in Cambridge, UK, provided additional data on threatened species status.

Two other related studies were carried out under this project. The first was a biodiversity and ecotourism assessment in selected areas of Yunnan, by two Junior Biodiversity/Ecotourism Specialists. Their Mission Report is annexed to the country report for Yunnan and some information contained in it has been incorporated in the country report. The second was a study of ecotourism in the GMS, carried out by an Ecotourism Specialist as a desk research exercise in Helsinki. Some material from the latter has been added to country and regional reports.

Language differences imposed limitations on quality and quantity of information and on speed of acquisition. Where reports were unavailable in English, the Specialist was dependent upon counterparts for translations. This could be time-consuming and produced versions that may or may not have accurately represented their originals. This applied in particular to statutory laws of countries whose definitive versions are in a local language and to place and protected area names which different translators spelt differently.

Close questioning sometimes revealed discrepancies in information given by sources within a country. For example, in two countries different spokesmen gave conflicting information on international conventions to which their country had signified accession. There may have been other instances where incorrect or out-dated information was acquired—more likely in countries where, due largely to linguistic differences, the Specialist had direct access to only a relatively small number of persons and less scope for cross-checking information.

Another factor more likely to arise when the Specialist dealt directly with small numbers of persons was that information based on subjective value judgements was influenced by the focus of the individual from whom they were acquired. Each person has his or her own perceptions and fields of interest that bore upon assessments of, for example, which protected areas are the most valuable and orders of priority for threats to conservation. Given longer in each country, and access to field operations, the Specialist might have produced different sets of priorities based upon a broader range of locally expressed ideas, his own observations and past experiences and his own (equally subjective) set of values.

2. Geography & Demography

The Greater Mekong Subregion (GMS) spans 2,336,000 km² of diverse terrain between latitudes 5⁰35' to 29⁰10' north and longitudes 92⁰10' to 109⁰25' east. The distance from its extreme northerly point in northwestern Yunnan Province to the southernmost tip of Thailand is approximately 2,500 km; at its broadest the subregion measures some 2,000 km from east to west. Individual countries range in size from 181,000 km² (Cambodia) to 677,000 km² (Myanmar): see Table 1. Elevations extend from sea level along the coasts of Cambodia, Myanmar, Thailand and Viet Nam to 5,881 metres in northern Myanmar and 6,740 metres in north-western Yunnan. Lao PDR and Yunnan are landlocked.

Marked differences in climate exist across this broad sweep of diverse landscapes—from coral reefs and sun-drenched beaches in equatorial climes, through tropical, subtropical and temperate forests to alpine habitats and glaciers. Summer temperatures in southern coastal areas reach around 38⁰ C, while, in northern areas bordering on the Himalayas, frosts occur for more than 100 days a year. Annual rainfall varies from 500 or 600 mm in south-east Yunnan and along the south-east coast of Viet Nam, to 6,000 mm on the Cardamom Mountains of Cambodia and in southern and northern Myanmar. Most of the GMS is dominated by monsoon winds that blow from the south-west during the warmer, wet season (about May to September), and from the north-east during the cooler, drier October to April period.

The estimated 1999 human population is 252 million (average density 108/km²). Annual growth rates mostly exceed 2.0 per cent. There are marked differences between countries (see Table 1).

TABLE 1: Greater Mekong Subregion country sizes,
1999 estimates for demographic data and forest cover

Sources: Various

Originally, most of the subregion was forested but, in all countries, forest cover has undergone marked reductions over the past 50 years—a trend that continues, including in Thailand, where logging has been outlawed since 1989. Extant forest cover is an estimated 877,000 km²—about 38 per cent of total land area. Estimates of forest cover by country are shown in Table 1.

Throughout the subregion, rural men and women depend upon access to forests and their plant and animal products for fuelwood, food, medicines, building materials, feed for livestock and cash-generating products. Biodiversity and its maintenance are of paramount importance to them.

3. Biodiversity

Biodiversity Indicators

Two indicators of biodiversity are used in this report: the biodiversity index and endemic bird area (EBA). They are explained below.

A measure of biodiversity, used both here and in the individual country reports, is MacKinnon’s (1997) biodiversity index. MacKinnon calculated biodiversity indices for biounits and countries in the Indo-Malayan Realm. They are cited in this report under descriptions of the biounits and in the country reports. The methods used to score are described in MacKinnon (1997) but, in brief, the process took account of the total numbers of species in selected taxa, numbers of endemics and narrow range species and size of biounit or country. The resultant scores were relative and adjusted around a mean of 10 for the Realm. Biodiversity indices are unavailable for Yunnan Province or for biounits that lie wholly within Yunnan.

Biodiversity indices for biounits were based upon vertebrates only although MacKinnon (1997) claimed a close correlation between vertebrate and plant diversities. For countries, they drew upon data for mammals, birds and higher plants.

The second indicator of biodiversity is the endemic bird area (EBA). Slattersfield et al, (1998) identified 250 EBAs globally, based upon studies of restricted-range species. An EBA contains the breeding ranges of two or more restricted-range species, the latter being land birds whose breeding ranges are smaller than 50,000 km². All or parts of six EBAs occur in the GMS (see Table 2 and map).

Bird diversity or endemism might correlate positively with diversity of other taxa—plant as well as animal—and thus be usable as a general indicator of biodiversity. This may intuitively appear plausible but the hypothesis has yet to be tested rigorously.

The occurrence of EBAs is mentioned under the brief descriptions of the biounits in the GMS that follow, and they are described in greater detail in the relevant country sections. Although they are not EBAs there are also three important sites for migratory birds in the GMS: the Mekong and Red River Deltas (Viet Nam) and Tonlé Sap Lake (Cambodia).

TABLE 2. Distribution of endemic bird areas
in the countries of the Greater Mekong Subregion.

Biounits of the Greater Mekong Subregion

The Indo-Malayan Realm extends from Pakistan to Papua New Guinea. It reaches into the Indo-Chinese and Sundaic Subregions and the Southwest China Unit although the Sundaic accounts for only a minute proportion of its total land area. The GMS occupies a roughly central position in this Realm, with Yunnan protruding to the north.

TABLE 3. Distribution of biounits (Udvardy 1975) and sub-units
(MacKinnon & MacKinnon 1986, MacKinnon et al 1996)
in the countries of the Greater Mekong Subregion.

Udvardy (1975) divided the Realm into 27 biogeographic units (biounits) while MacKinnon & MacKinnon (1986) produced a finer classification of 70 sub-units. MacKinnon (1997), ‘on the basis of respective levels of similarity and distinctiveness in species communities’, reclassified the first two subregions into 24 biounits and 90 sub-units. Based upon a separate report on biodiversity in China (MacKinnon et al, 1996), a further two biounits (four sub-units) complete the biogeographic classification for the GMS, which comprises 10 units and 20 sub-units. These are described below and their occurrence in the GMS countries tabulated in Table 3 and illustrated on the accompanying map. The numbers in parentheses are those assigned to the biounits and sub-units described by MacKinnon et al (1996) and MacKinnon (1997).

Forested southerly slopes of the Himalayas. Within the GMS this biounit occurs only in the northern corner of Myanmar, from which it extends westwards across northern regions of the subcontinent. MacKinnon (1997) recognised four sub-units, one of which is referred to in the country report for Myanmar.

It was originally densely forested but much has been destroyed or modified by human activity.

The biodiversity index is 12.1. Levels of endemism are moderate. A small portion of one EBA (Eastern Himalayas) occurs, which is described in the country report for Myanmar.

Coastal rainforests plus deltas and estuaries of the Irrawaddy and Salween Rivers. Within the GMS this biounit occurs only in Myanmar although it extends beyond into Bangladesh.

The original vegetation comprised lowland evergreen forest and freshwater swamp, plus some semi-evergreen forest, montane evergreen forest, mangrove forest and deciduous dipterocarp forest near the mouth of the Salween. Most of the rainforest and freshwater swamp have been cleared for agriculture, and much of the upland forest replaced by creeping bamboo.

The biodiversity index is 9.6. Endemism is low. A small portion of one EBA (Irrawaddy Plains) occurs, which is described in the country report for Myanmar.

The Irrawaddy catchment, extending from the dry central zones of Myanmar to the surrounding mountains lying north, east and west. Within the GMS this unit occurs only in Myanmar and a small western area of Yunnan although it extends beyond into Bangladesh and India. MacKinnon (1997) identified three sub-units, which are referred to in the country sections for Myanmar and Yunnan.

The original vegetation comprised a wide range of types. Mixed deciduous forests with patches of dry dipterocarp forest covered most of the Irrawaddy Basin, with thorn scrub in the driest areas. Semi-evergreen forests surround the basin, grading into moister montane forest on the uplands or, to the north, with sub-tropical moist lowland forests. At higher elevations subtropical montane forests appear, while the highest peaks support sub-alpine vegetation.

Natural vegetation covers 52 per cent of the unit (MacKinnon, 1997). Much is being logged or is under cultivation. Shifting agriculture is converting moister hill forest to grassland.

The biodiversity index is 10.5. Endemism is fairly low. There is one EBA (Irrawaddy Plains), which is described in the country report for Myanmar.

Coastal Indochina biounit (05)

Coastal areas extending from the Red River Delta in northern Viet Nam to the coast of southern Myanmar. This unit occurs in Viet Nam, Cambodia, Thailand, Myanmar and marginal areas of southern Lao PDR. MacKinnon (1997) identified four sub-units, referred to in the country sections.

Most of the unit is semi-evergreen forest. True ever-wet rainforest occurs only in restricted parts of Cambodia and Viet Nam. Other vegetation types include peat forest (Mekong Delta), mangrove and freshwater swamps (Mekong and Chao Phraya Deltas), mixed deciduous forest and very small areas of forest on limestone and dry dipterocarp forest.

Human population densities are high. Most lowland forest has disappeared but some forests survive in upland areas.

The biodiversity index is 10.6. Endemism is moderately high. There are two EBAs (South Vietnamese Highlands and Da Lat Plateau) described in the country report for Viet Nam.

Annamese Mountains biounit (-M)

Two mountain blocks in Viet Nam on the Da Lat Plateau and around Ngoc Linh Mountain. The unit extends marginally into Lao PDR. MacKinnon (1997) identified two sub-units, which are referred to in the country section for Viet Nam.

Montane evergreen forest predominates; some tropical wet evergreen and semi-evergreen forest occurs at lower elevations. Patches of coniferous forest and moss forest are found on the highest peaks.

About 40 per cent of forest cover remains but disturbed by fuelwood collection.

The biodiversity index is 12.0. Endemism is moderately high. There is one EBA (Da Lat Plateau), which is described in the country report for Viet Nam.

All land east of the Red River Delta in Viet Nam and on into China but not Yunnan Province. MacKinnon et al (1996) identified two sub-units referred to in the country section for Viet Nam.

Originally the unit was clothed mostly with tropical semi-evergreen rainforest, subtropical lowland moist forest and sub-montane dry evergreen forest. Smaller areas of forest occurred on limestone, mangrove forest, freshwater swamp, subtropical broadleaf forest, subtropical pine forest and montane deciduous forest.

Most lowland forest has been cleared for agriculture. That which remains is disturbed. Secondary bamboo has replaced much of the original cover on upland areas.

The unit extends from tropical regions near sea level to temperate sub-alpine regions, resulting in a broad range of vegetation types, including dry dipterocarp forest and semi-evergreen lowland and tropical montane evergreen forests. Areas of true wet rainforest occur in the Salween Valley, and there are also pine forest, forest on limestone, dry evergreen forest, montane deciduous forest, subtropical moist lowland forest and subtropical montane forest.

The original vegetation has been cleared for agriculture in areas where human population densities are high, and in areas of lower densities where ethnic minorities practice patterns of land use that are destructive of forests. Elsewhere where densities are lower (especially Cambodia) extensive tracts of forest remain. MacKinnon (1997) estimated that about 34 per cent of the unit was still under natural vegetation.

The biodiversity index is 8.3. Endemism is moderate. The unit contains portions of two EBAs (Annamese Lowlands and Yunnan Mountains) described in the country reports for Lao PDR, Viet Nam and Yunnan.

This is the major biounit in Yunnan, covering most of the province north of unit 10. Within Yunnan, MacKinnon et al (1996) identified three sub-units, referred to in the country report for Yunnan.

Terrain varies from high mountain and mid mountain to flat plateau. Pinus yunnanensis and Parmandii forest are typical of mid mountain areas, plus areas of Castanopsis delavayi, Corthacantha and Quercus franchettii. Platycarya strobilacea and Pyunnanensis forest occur in karst areas, and Cyclobalanopsis glancoides and Pyunnanensis in plateau areas in Kunming and Dali. Other species include Lithocarpus echinotholus, Schima noronhae and Tsuga dumosa on high mountains and in deep gorges. A more detailed account is in the Yunnan country report

No biodiversity index has been calculated. Within the context of China, species richness for Yunnan as a whole is very high—a reflection of the large size and geographical diversity of this biounit. The same is probably true of endemism, which is also high. Most of one EBA (Yunnan Mountains) occurs, which is described in the country report for Yunnan.

Part of a small sub-unit of this biounit occupies the north-eastern corner of Yunnan. It comprises forests of Lithocarpos cleistocarpus and Castanopsis platyacartha growing on mid mountain terrain.

No biodiversity index has been calculated. Species richness and endemism for Yunnan as a whole are high but this biounit is small and geographically uniform. Part of one EBA (Yunnan Mountains) extends into the unit, and is described in the country report for Yunnan.

The peninsula and offshore islands. Within the GMS, this biounit occurs only in the extreme south of Myanmar and Thailand. MacKinnon (1997) identified two sub-units, which are referred to in the two country reports.

The vegetation is predominantly rainforest rich in species, including areas of freshwater swamp, montane forest, forest on limestone, heath forest, peat swamp and mangroves. Intensive rice cultivation and extensive plantations have greatly modified the original plant cover, of which only 25 per cent remains.

The biodiversity index is 11.4. Endemism is moderately high.

Flora and fauna

The flora of the Indo-China region (excluding Yunnan) comprises at least 20,000 species, which is the estimated minimum number for Thailand alone (MacKinnon, 1997). At least 5,000 are endemic but endemism at generic level is lower. In a separate report (MacKinnon et al, 1996) the flora of Yunnan Province is estimated to contain 14,000 species. There is probably considerable overlap between the two but the total number for the GMS seems unlikely to be less than 25,000, and is possibly much higher.

TABLE 4. Numbers of known species of vertebrates and higher plants
in the countries of the Greater Mekong Subregion.

Sources: MacKinnon (1997) for all except Yunnan; MacKinnon et al (1996) for Yunnan. The latter offers the statement that vertebrates total 1,638 but numbers are not cited for amphibians or fishes. Fish data for some countries are ambiguous as to whether marine species are included, and have therefore been omitted.

The fauna of the GMS comprises four elements: taxa shared with northern India, Himalayan Palearctic species, Chinese Palearctic species and Sundaic species. The latter have entered southern coastal areas from the Malay Peninsula. Judging from data gathered at national levels (Table 4), there are probably in excess of 1,800 terrestrial or semi-terrestrial vertebrates: 300 mammals, 1,000 birds, 400 reptiles and 100 amphibians. Numbers recorded depend both upon the numbers that occur, the nature of the country and efforts taken to survey and collect data. The more intensively an area is studied, the greater the numbers of plants and animals likely to be discovered.

Country levels of biodiversity

Biodiversity indices for the GMS countries except Yunnan are presented in Box 1, together with indices for all 18 countries of the Indo-Malayan Realm.

Within the context of the Indo-Malayan Realm, biodiversity indices for five of the GMS countries are relatively modest: all are below the mean of 10.0. MacKinnon et al (1996) state that Yunnan Province ‘has an extremely high biological richness’ although this is within the context of China where Yunnan has the richest faunal diversity of all provinces, and scores highest in endemism.

Threatened species

Information on status of threatened species was taken from data sheets provided by WCMC during November and December 1998. The information is summarised in Tables 5 and 6, and elaborated in country reports. Categories of threat for fauna are 1994 IUCN Red List categories.

TABLE 5. Numbers of threatened species of wild plants and animals in the countries
of the Greater Mekong Subregion. Information for Yunnan is incomplete.

Source: Country lists of threatened species supplied by World Conservation Monitoring Centre, November & December 1998. The two species extinct in the wild are Schomburgk’s deer (Cervus schomburgki) and a cyprinid fish (Epalzeorhynchos bicolor).

TABLE 6. Occurrence of critically endangered wild vertebrates
in the countries of the Greater Mekong Subregion. Information for Yunnan is incomplete.

Source: Extracted from country records provided by WCMC. Some revisions made after checking with specialist in-country sources.

Levels of threat represent global status, which is not necessarily the same as status in individual countries. For example, although Asian tapir's (Tapirus indicus) global level of threat is only 'vulnerable', it may in Thailand may be almost or possibly extinct (MacKinnon, 1997). Checks made within countries with specialists for selected taxa suggest that some WCMC records are out of date. For example, the Vietnamese warty pig, listed by WCMC as an extinct former endemic to Viet Nam is now known to occur in Lao PDR and possibly survives still in Viet Nam; and it seems improbable that both Javan and Sumatran rhinoceroses survive in five countries as suggested by Table 6.

Global threat status has yet to be assessed for four species of ungulates due to their being only recently described. They are:

• Pseudoryx nghetinhensis (saola): Endemic to the Annamite Range along the borders of Lao PDR and Viet Nam.
• Muntiacus truongsonensis (small dark muntjac): Endemic to the Annamite Range along the borders of Lao PDR and Viet Nam.
• Muntiacus putaoensis (little leaf deer): Discovered in 1998. Believed to be endemic to a small locality in northern Myanmar
• Muntiacus (Megamuntiacus) vuquangensis (giant muntjac): Endemic to the Annamite Range along the borders of Lao PDR and Viet Nam.

4. Policy and Law

Policies

There are no specific policies for protected areas in any GMS country although intentions and guidelines are expressed in policy initiatives that have broader scope, and in legislation.

In Cambodia and Lao PDR, national Environmental Action Plans provide general guidance; and, in Lao PDR, biodiversity conservation is addressed in the Tropical Forest Action Plan. Policy objectives for protected areas in Lao PDR are also defined in forest legislation. In Myanmar, the National Forest Policy of 1995 includes provision for protected areas. In Viet Nam, the National Conservation Strategy and Biodiversity Action Plan fulfil similar functions. Provisions for biodiversity and protected areas are included in Thailand’s National Forest Policy and in the far-reaching Policy and Prospective Plan for Enhancement and Conservation of National Environmental Quality 1997-2016. For Yunnan, the national Biodiversity Action Plan 1995 (prepared in response to Agenda 21) prescribes policy.

All policies pay at least lip service to the values of biodiversity and the role that protected areas play in conserving it, and some express this very forcefully; but, in general, government actions tend not to live up to their policy pronouncements—a theme that is elaborated below.

Laws

All GMS countries have laws governing biodiversity conservation and management of protected areas. Most are of recent origin or are older laws that have been up-dated. In spite of the existence of statutory laws, a widely expressed opinion in all countries was a lack of effective enforcement.

Cambodia’s legislation appears to be least effective. A Law on the Environment and a ministerial decree declaring a substantial protected area system are in place but are said to be without force unless further sub-decrees are issued.

In Lao PDR, biodiversity conservation is provided for in the Forest Law of 1996, and a 1993 decree that designated the first national biodiversity conservation areas (NBCAs).

Myanmar recently up-dated its legislation, producing the Protection of Wildlife and Wild Plants and Conservation of Natural Areas Law of 1994.

In Thailand, there are the National Park Act 1961, the Wildlife Protection and Preservation Act 1992 and the National Forest Act 1964. There is talk of overhauling legislation with the object of producing a unified law for protected areas.

In Viet Nam, biodiversity conservation is addressed in the Law on Environmental Protection 1993, while the Law on Protection of Forests 1972 provides for protected areas. Other relevant laws, developed decades ago, need revision and consolidation: Ban on Elephant Hunting 1960; Regulations on Hunting of Forest Wildlife 1963; and Law on the Protection of Forests 1970.

In Yunnan, the Wildlife Conservation Law 1988 focuses on biodiversity conservation and provides for nature reserves, which form the greater part of the protected area system. The Provincial Forest Law of 1993 was revised in 1998.

International conventions

All GMS countries (China in the case of Yunnan) have acceded to two or more relevant international conventions: see Table 7.

TABLE 7: Adherence to international conventions relevant to biodiversity conservation
by the Greater Mekong Subregion countries in February 1999.

5. Protected Area Systems

Inventories

All GMS countries have protected areas—in total 550, of which 380 are judged to have biodiversity conservation as a major function. They are listed in their respective country reports. Their distribution between countries is illustrated in Table 8. Fourteen classes of protected areas exist although many are, perforce, translations into English from other languages and may vary from one translator to another. Some classes appear not to fulfil major roles in conserving biodiversity: for example, the forest parks of Thailand and Yunnan, Thailand’s non-hunting areas and the protected landscapes of Cambodia and Viet Nam. Table 8 therefore distinguishes between protected areas judged to have biodiversity as a major function (in bold face) and those that do not. The former broadly fit IUCN management categories I to IV.

TABLE 8. Areas (km²) of land in legally designated protected areas of the
Greater Mekong Subregion. No proposed areas or areas of dubious validity included.
Areas that perform a major biodiversity conservation function are in bold typeface.

Conservation coverage

All biounits and sub-units that occur in the GMS are represented in the combined protected areas of the six countries: see Table 9. Biodiversity representation within individual countries is almost complete: the only ‘lapses’ are the absence of sub-unit 5a from the protected areas of Cambodia, and 10c and 5d from Myanmar. Both sub-units are amply represented elsewhere.

Cambodia, Lao PDR and Thailand have relatively large systems, well in excess of the 10 per cent ‘target’.

Thailand’s system is long established although several areas have been added in recent years. Those in Cambodia and Lao PDR are of recent origin, and designed specifically with biodiversity conservation and representation in mind. In these larger systems, cover of biotic communities and their species is very good although there is a paucity of lowland dry evergreen forest. This forest type, where it occurs inland has been extensively destroyed in Southeast Asia.

Apart from limited coverage of lowland forest and inadequate provision for pheasants, Viet Nam's system offers good representation but individual areas are relatively small (see Table 10). It is difficult to see how this protected area system can be increased in size in the face of high human population density and annual increment rates leading to rising demands for land.

Myanmar and Yunnan’s protected area systems are also small—their average sizes least for the subregion—but Myanmar has scope for enlargement. The National Forest Policy 1995 prescribed an ultimate target of 10 per cent cover by protected areas, and Brunner et al (1998) believe this to be realistic because large, well-forested and sparsely populated tracts of country exist in the north.

TABLE 9. Representation of biounits (Udvardy 1975) and sub-units (MacKinnon & MacKinnon 1986, MacKinnon et al 1996) in terrestrial protected area of the six countries
of the Greater Mekong Subregion.

Number = Approximate number of protected areas or parts thereof in which sub-unit is represented
- x - = Sub-unit present but not represented in a protected area