Previous PageTable Of ContentsNext Page

Integrated ecological studies of pasture problems in the Tibetan Autonomous Region, P.R. China

W. Holzner,1 M. Kriechbaum,1 C. Ramble2 and H. Diemberger3

1. Centre for Environmental Studies and Nature Conservation, University of Agricultural Sciences, A-1180 Vienna, Gregor Mendel Str. 33, Austria
2. Oriental Institute, University of Oxford, England
3. Department of Social Anthropology, Free School Lane, Cambridge CB2 3RF, England


This paper presents a preliminary survey of high altitude pastureland in central and southern Tibet, comprising an outline of the methodological approach adopted and a description of the main findings. The extent, degree and age of pasture degradation are discussed, and of the many causes of recent pasture problems that are interrelated in a complex way. Particular attention is given to two factors: the cutting of peat and turf and to the alleged pika problem.

Keywords: Pasture degradation, pasture ecology, Tibet


This paper presents the results of observations and discussions carried out in 1998 during the course of feasibility studies concerning the implementation of development projects in areas of southern and central Tibet. Some problems related to Tibetan pastures such as pasture degradation, pika calamities and snow disasters were encountered. To determine which technical measures could help solve these problems, a superficial survey of the rangeland areas passed was made, applying methods we had developed during previous ecological research in areas of the Himalayas outside of China (Holzner and Kriechbaum 1998). Anthropological enquiries, particularly in the Porong area of southern Tibet, were also carried out together with the Tibetan Academy of Social Sciences and a project investigating high altitude pasture ecosystems in the Austrian Alps, which combined ecological, economic and other human science data. An important observation that may be made here is that the problems encountered in the European Alps and the Himalayas and Tibet are similar; they are general problems of human economy and society in areas marginal to agriculture. In other words, the problems of high altitude rangelands are international problems.

In agronomy and other environmental sciences, we are faced with an increasing amount of problems, in spite of the fact that research is proliferating within these fields. Scientists are often called in to solve these problems, but the results of this approach are not convincing. In this paper, we address the questions: Can the usual scientific approach solve the problems of high altitude pasturelands, and what approach has chance of success?

What is HAPIE?

High Altitude Pasture Integrative Ecology (HAPIE) tries to integrate the methods and results of the natural sciences such as biology and ecology with those of the human sciences such as anthropology, sociology and economics. Such an approach is absolutely necessary for investigations aimed at the solution of pasture problems. Pastures are areas used for agricultural production, bearing a biocoenosis of plants, wild and domestic animals and humans. Nomads, herdsmen and farmers are members of the pasture ecosystem, as are their yak and sheep, and the grasses and birds. This is a natural matter of fact, and thus it seems wise to act accordingly. It does not make any sense to deal with pasture areas from an ecological point of view alone, nor can we concern ourselves only with the economics of pasture agronomy. Instead, an integrated approach, which addresses all the pieces of the puzzle together, is the only one where the results of applied science can be put to practical use.

The goals of HAPIE are not set by science but by the concrete problems of life. HAPIE attempts to translate the knowledge of modern science and technology into practical measures with the goal of achieving optimal and sustainable use of high altitude pastures for the welfare of human society. Thus, it is necessary to view the pastoral system not only from outside as scientists do, i.e. a scholarly approach, but also from the inside, with the eyes of a herder or a yak. This approach has several advantages. For one, it can help to avoid academically generated solutions, which often hamper progress, waste time and energy, and do not particularly produce feasible results that are transferable to the people for which the study has been performed. On the other hand, a herder's eye-view can help define goals and terms more easily and more accurately. For example, 'overgrazing' simply means a mode of utilisation that prevents vegetation from recovering in the periods between grazing, with the result that the same vegetative yield cannot be achieved in the following year, provided weather conditions are comparable. Constant overgrazing causes 'degradation', a process that produces a gradual, irreversible change in pasture quality and quantity. 'Irreversible' does not mean that this process cannot be reversed, a condition that is very difficult to judge, but that a return to the former state would take at least several years or decades. This is unacceptable, or at least represents a substantial loss, from a herder's point of view because it means that the pasture should not be used at all for at least the duration of the recovery period. In the long run, degradation leads to a pastoral 'desert', a land practically devoid of palatable plants. From this point of view, 'desertification' does not mean an area free of all vegetation, but only free of usable forage.


HAPIE requires complex thinking and a general methodology that is able to interlink not only the results of different fields of science, but also to combine 'hard' scientific data with so-called 'soft' information, like 'herders' opinions or traditional knowledge. Thus, the results are products of science and technology on the one hand, and a different kind of approach to nature, which relies on direct experience derived from observation and mixed with 'inherited' or traditional knowledge.1 Both approaches are considered equal. The methods themselves, however, must be simple. They must be specially designed for the particular purposes of the project and the ways the results have been drawn must be obvious and clear. Besides that, applied research, which delivers truly applicable results, must be fast and inexpensive, with the results presented publicly in a way that everybody can understand.

For the purpose of this study, a set of simple and complex indicators was employed using specific vegetation parameters, like species numbers, relative numbers of 'pasture weeds' to palatable plants, relation between herbs and grasses, and the percentage of pioneer species (Figure 1). These indicators are easily recognised, and can be used also by non-specialists (Holzner and Kriechbaum 2000). The degree of precision that can be achieved with them is sufficient for the purpose of this study. With these indicators, the condition of a pasture can be roughly judged, in terms of whether it is in an optimal state or a state of degradation, and if the latter, whether this degradation is of ancient (centuries-old) or recent (decades-old) origin.

Adapted from Holzner and Kriechbaum (1998).

Figure 1. Development of some vegetation indicators along a gradient of grazing intensity.

Additionally, the most widespread pasture types were classified, mainly according to the dominant species or most important forage plants. These botanically-defined types match with the pasture classification of the nomads, and it was noted that results achieved from the combination of the two different sciences were particularly useful.2 For each of these pasture types, the agricultural value and the form and degree of degradation was estimated by using the indicators mentioned above (Table 1).

Table 1. Common pasture types of southern and central Tibet.

Pasture type

Dominant plants

Tibetan name1




Pygmy Kobresia turf

Kobresia pygmaea





Feather grass steppe

a) Stipa spp. and other (small) tussock grasses 
b) Stipa purpurea, Oxytropis chiliophylla


sagshi, nekarma




Jag pasture

Trikeraia hookeri





Turquois sedge pasture

Carex moorcroftii





Giant Kobresia bog

Kobresia schoenoides





Dwarf shrub pasture

Dasiphora, Juniperus, Myricaria, Hippophae





Micro-sedge pasture

Carex microglochin


RDP 1–2



Bulb Kobresia pasture

Kobresia royleana





Worm wood steppe

a) Artemisia spp.
b) Artemisia xigazeensis

yog mo








Rampa steppe

Pennisetum flaccidum





Semi-desert pasture

Astragalus spp., Stellera chamaejasme, Incarvillea, Cryptothladia, Tanacetum, Iris



ADP 4–5






High altitude pasture

Kobresia sp., 
Stipa purpurea






Androsace sp.


ADP 4–5



Edelweiss semi-desert

Leontopodium nanum





Duk tsa pasture

a) Oxytropis tatarica
b) Thermopsis lanceolata
c) Astragalus sp.

duk tsema





Stinging nettle pasture

Urtica hyperborea





OP = optimal pasture; ADP = ancient degradation, RDP = recent degradation, the numbers indicate the intensity of degradation; P = productivity; Q = quality. The numbers are the result of relative estimates between the two extremes, the Giant Kobresia pasture (5) and the Edelweiss semi-desert (1).
1. Tibetan plant names given by local people are rendered in phonetic approximation.
Source: revised from Holzner and Kriechbaum (2000).


Summarising the experience of our studies, pasture conditions prevailing in southern and central Tibet are estimated as follows: About 30% of the pastures investigated were in optimal condition, about 20% showed signs of recent but complete devastation and about 20% were heavily degraded, the causes dating back centuries. The remaining 30% fell somewhere between the optimal state and heavy overgrazing, though it was difficult to judge whether that degradation was of ancient or recent origin.

The fact that large areas3 of Tibetan pastureland has been kept in optimal condition for centuries—optimal from an economic as well as an ecological point of view—can be taken as evidence that at least some of the Tibetan pastoral systems must have been sustainable.

3. Sustainable use of the pastures did not prevail in the whole of Tibet. In quite large areas we found signs of severe ancient degradation and desertification of the pastures.

Causes of recent pasture problems

External factors—climatic

The causes of recent pasture problems are not yet completely understood, but in any case they are very complex, involving both natural and human factors, e.g. climatic and pastoral system changes. In several areas of Tibet it is quite obvious that during the last few years, wells and rivers have been drying up with increasing frequency. Whether this is an independent tendency or partially caused by the loss of vegetation cover due to local overgrazing and cutting of the forests, remains an open question. A major cause of the so-called snow disasters seems to be a shifting of summer precipitation, when the plants can make use of the moisture, to winter, when it is of less use to the plants but makes access to them difficult for livestock. It is still unclear whether this is just a transient climatic aberrance or a general trend that will continue into the future.

Internal factors—anthropogenic

From the many and interrelated anthropogenic (human-made) factors that are responsible for pasture degradation, we shall consider only two: the cutting of peat and turf, and the pika 'problem'.

The cutting of peat and turf: There are two main types of vegetation in Tibet that accumulate peat: Giant Kobresia (Kobresia schoenoides) bog and pygmy Kobresia (K. pygmaea) turf. Giant Kobresia makes for good pastures that are very conspicuous in the landscape due to their dark green colour. The intermediate states, in which, mainly because of overgrazing, the tussocks are partly decomposing and slowly and gradually becoming flatter and flatter, are also widespread, characteristic landscape elements. The decomposing tussocks are often covered by dense micro-sedges (Carex microglochin) vegetation, which also makes good pastureland but is much less productive than the Giant Kobresia bog. The thousands of tiny, spring-green hills with their soft outline set off against the desert-like mountain slopes by the rays of the low-lying sun are a motif often depicted in photographs. We call this type of pasture 'hummock pastures' (Figure 2).

Adapted from Holzner and Kriechbaum

1. Kobresia schoenoides 2. K. schoenoides remains 3. Kobresia pymaea 4. Carex microglochin 5. pioneers 6. peat.

Figure 2. Degradation cycle of a giant Kobresia bog to desert by overgrazing and peat removal.

Huge areas at high altitudes (or northern slopes in lower areas) can be covered by a turf of Kobresia pygmaea, which makes particularly good yak grazing land. The peat-like underground formed by the living subterranean parts of the plants, as well as by the partly decomposed ones, is hard but springy and extremely dense and therefore resistant against trampling by herbivores, even by the large and heavy yak.

The cutting of bricks out of these two types of turf, mainly for the purpose of making walls or wind shelters, is an old custom that is being practised increasingly nowadays. The reason is increasing demand, in part because peat bricks are even used to improve road verges and to build houses, and especially because nowadays trucks can easily transport large amounts of turfs away from the nomadic areas.

The cutting of turf or peat means the removal of the vegetation together with all or most of the soil. During subsequent years, the scarce soil remnants are removed by wind and water. An intermediate, semi-desert state is produced featuring characteristic vegetation that cannot be called 'pasture' any more because the vegetation cover and its productivity are extremely low. Recovery of the former pasture is very unlikely if Tibet's severe climatic conditions are considered. In any case, it takes such a long time that the degradation can be called irreversible from a nomads' point of view. The final result is desert, stone or gravel pavement, with very scarce vegetation nourished by the remnants of soil, which escaped erosion because it was protected under stones.

It is very difficult to assess the extent of this process in the past. However, it could be that at least some of the desert-like areas of Tibet are the result of such activity, which might have taken place in regions with a dense human population or located close to towns, where it was profitable to sell the peat.

Nowadays, turf cutting is widespread (though we were told that it has been prohibited since 1988). As the removal of turf or peat means the removal of the pastureland forever, this practice seems to be the most severe of the complex impacts destroying the pastures. Actually, it is difficult to say that the cutting induces pasture problems, because it simply removes the pastureland and therefore also the necessity to look for proper pasture management possibilities. However, far worse than having pasture problems to deal with is the notion of having no pastures at all. Thus, peat cutting is not only a problem in terms of pasture use in Tibet, it is a general ecological catastrophe for the whole country because soil and vegetation are simply sold and lost forever. Whereas the peat sponge retains moisture in the area, the water from rain and melting snow now runs off directly from the land into the rivers instead of being gradually emitted into the air. A regeneration of completely destroyed vegetation is unthinkable or only conceivable within a time span of hundreds or thousands of years.

The pika 'problem': Pikas (Ochotona curzoniae), called abra in Tibet, are relatives of hares and rabbits but are much smaller, about 20 cm long, and have shorter ears. In some pasture areas they live in astonishing numbers and are consequently considered major pests. The notion that pikas are a major cause of the poor condition of some pastures seems logical, because it is the usual reaction of humans to look for external enemies or pests as the source of our problems. From the viewpoint of HAPIE, this widespread opinion is not only questionable but also dangerous: It is easy to find a scapegoat to blame problems on, but it distracts us from looking for the true causes of those problems. In the case of pikas, huge extermination programmes have been organised instead of conducting investigations into the basic causes of pasture degradation.

Despite the extent of pika control programmes, the condition of the Tibetan grasslands apparently could not be improved. It is impossible to kill all the pikas, as others will soon fill the gaps because low population densities promote their reproduction rate. Moreover, the extermination of pikas in an area creates severe food problems for their predators, such as foxes, weasels and birds of prey. This reduces the predators' reproduction rate and ultimately leads to their starvation or exodus. Consequently, killing pikas also means reducing the numbers of their natural enemies, who assist the pastoralists in pika control. Thus, this kind of effort to control the pika population cannot have the desired results, but will exacerbate the problem still further.

According to our observations in many pika-infested areas, as well as detailed studies in certain localities, pikas take advantage of previous pasture destruction. From both an ecological and an economic point of view, the best method for addressing the pika 'problem' is to find the prior causes of pasture deterioration since it is these causes, which ultimately underlie the pika population explosion. Viewed from this angle, improving pastures will automatically reduce the number of pikas that live on them. Investigations conducted by Chinese scientists led to the same conclusion: When the amount of vegetation cover and the height of vegetation are reduced by overgrazing, pikas may be found in higher densities and consequently are in a position to do greater harm to the pasture environment (Shi 1983, and Zhong et al. 1985, both cited in Smith et al. 1990).

Additionally, efforts should be made to boost the population of pika predators. Miller et al. (1992) mention that the prior reduction or elimination of predators by hunting or poisoning appears to have been accompanied by an increase in pika populations. Laws protecting them may not be enough. Advertising campaigns promoting them as helpers in efforts to improve pastures could serve the purpose better.


Financial support for travel was provided by the Federal Ministry of Science and Transport of the Government of Austria, by the 'Kommission für Entwicklungsfragen' of the Austrian Academy of Science and by ÖKO-HIMAL, Austria. The Austrian Science Fund (P08183–BIO) financed earlier expeditions to high altitude pasture areas of the Himalayas. The study would not have been possible without the guidance and assistance of Tibetan experts, particularly the officials of the government of the Tibetan Autonomous Region of the P.R. China, and first and foremost the representatives of TARA (Tibetan Assistance to Remote Areas) and local people.


Holzner W. and Kriechbaum M. 1998. Man's impact on the vegetation and landscape in the inner Himalaya and Tibet. In: Elvin M. and Liu T. (eds), Sediments of time. Cambridge University Press, Cambridge, UK. pp. 53–106.

Holzner W. and Kriechbaum M. 2000. Pastures in south and central Tibet (China) I. Methods for a rapid assessment of pasture conditions. Die Bodenkultur 51(4):247–254.

Miller D.J., Bedunah D.J., Pletscher D.H. and Jackson R.M. 1992. From open range to fences: Changes in the range-livestock industry on the Tibetan Plateau and implications for development planning and wildlife conservation. In: Perrier G. (ed), Proceedings of the 1992 International Rangeland Development Symposium held in Spokane, Washington, USA, 11 February 1992. 105 pp.

Smith A.T., Formozov N.A., Hoffmann R.S., Changlin Z. and Erbajeva M.A. 1990. The pikas. In: Chapman J.A. and Flux J.E.C. (eds), Rabbits, hares and pikas. Status survey and conservation action plan. Information Press, Oxford, UK. 18 pp.

Previous PageTop Of PageNext Page