Key Issues in Grassland Studies
The Grassland Review Panel appointed by the National Academy of Sciences to review this study concluded that the best way to understand the current state of grassland science in China is to let Chinese scientists speak for themselves—through their published research, their work in situ, and their descriptions of both. These views have been presented in Chapters 2 through 13. At the same time, the members of the panel would like to advance the dialogue with their Chinese colleagues by commenting on some of the key issues raised in the preceding chapters, including how these issues have been treated in China and abroad, and what this means for the challenge that grassland science and scientists inside and outside China face in the years ahead. The panel's comments on these issues make up this concluding chapter.
THE PASTORAL FRONTIER
For more than two millennia, Chinese have been building fortifications (often misleadingly referred to as "The Great Wall") along a line that runs across Asia, marking the frontier between pastoral societies and practices in the north and settled agriculture in the south. The line has been drawn by nature, for it distinguishes the lands that have sufficient moisture to support cultivation from those that do not. It has been reinforced by man, for it marks the divide, economically, between intensive agriculture and extensive animal husbandry; politically, between large centrally controlled states and dispersed tribal units; militarily, between the mounted horsemen and the wall builders; and culturally, between different ethnic and linguistic groups. This frontier has been one of the most persistent and important zones of contact between com-
peting forms of human adaptation, and nowhere has the contrast been sharper or its significance greater than in China.
From the eighteenth century, when the French philosophes first took an interest in this subject, until the early 1900s, Western historians thought of this frontier as hard and absolute, dividing two self-contained, alien worlds. Foreign scholars, new to East Asia, were influenced by the well-articulated and pervasive Chinese view, that drew a sharp line between (Chinese) civilization and (nomadic) barbarism. The cultural distinctness and geographical isolation of the nomads reinforced this interpretation. In the literature, nomads were presented as survivors of mankind's primitive past, following a way of life that offered nothing to and demanded nothing from outsiders.
Although accepting the essential differences between pastoral and agrarian peoples, Lattimore (1940) was the first to draw attention to the importance of interactions between these ways of life. Later work, by Barth (1961), Khazanov (1984), and other anthropologists, went further to stress the interdependence of nomadic and settled societies. More recently, Barfield (1989), Jagchid and Symons (1989), and Waldron (1990) have portrayed China's inner Asian frontier historically as the site of contact, influence, and change in both directions. Steppe pastoralists and Han farmers engaged in significant, mutually beneficial trade of grain, metals, medicines, and luxury goods for livestock, furs, and other animal products. Demographic equilibrium among the pastoral peoples depended on slow, steady settlement, which drew off surplus population and kept a favorable balance between people and resources on the steppe. The ebb and flow of political power, rather than dividing enemies, gave first one side then the other an opportunity to embrace, influence, and be influenced by its opposite. The Mongols of the Yuan Dynasty (1279–1368) conquered China on horseback but ruled it by adopting Chinese techniques. When the Chinese of the Ming (1368–1644) retook their country, they erected a state apparatus along lines inherited from the Mongols and enrolled thousands of Mongols in their armies and civil service. Meanwhile, Yuan loyalists on the steppe attracted Chinese followers from among believers in proscribed religions, such as the White Lotus sect of Buddhism, and erected Chinese-style cities, such as Koko Khota, site of Hohhot, the present capital of Inner Mongolia. Out of this dialogue emerged the late Imperial Chinese state—a blend of institutions, particularly military institutions, borrowed from the steppe, with a culture that was essentially Chinese.
The balance of power and influence along the frontier has shifted with changes in technology. The supremacy of the nomads was based on the mounted archer, who combined rapid mobility with formidable firepower. The introduction of firearms gradually rendered mounted warfare obsolete, whereas the railroad, iron plow, irrigation pump, and other modern agricultural devices paved the way for the expansion of land-hungry Chinese peasants north. Today, the farmer reigns supreme in northern China; the spread of
agriculture is limited only by nature or, where nature is momentarily bowed, by sporadic natural disaster; and the pastoralists have been reduced to areas where nothing else works.
In 1991 the focus of attention along this frontier returned to its source, the environment. As indicated, the pastoral frontier has been defined by a difference in climate, principally rainfall, and its history by the different and changing human adaptations to this reality. Now, for the first time, the dominant issue has become the human impact on the environment and the degree to which that impact can be limited, directed, or controlled. The members of this panel are encouraged by the fact that our Chinese colleagues are addressing these problems and seeking solutions to them.
Scientists have long recognized that climate influences the vegetation species composition of landscape, but only in recent years have they begun to explore the feedback of landscape to local and regional weather and climate. Our understanding of climate change, due to either natural or anthropogenic causes, can profit from further study of the response of landscape composition to perturbations in climate and the ways landscape affects changes in climate.
The distribution of photosynthetically active plants over the grasslands of northern China by using a Normalized Difference Vegetation Index (NDVI) is expected to show considerable spatial structure such as observed over the grasslands of the United States (Pielke et al., 1991). During winter, of course, this area would be dormant. Such imagery demonstrates, first, that there is a large seasonal response of the grasslands to the changing weather; and second, that there is considerable spatial structure in this landscape.
Pielke and Avissar (1990) and Pielke et al. (1990a,b) review observational and modeling evidence to show that atmosphere boundary layer structure and the generation of local wind circulations as strong as sea breezes can occur over grasslands in the United States, when located between irrigated land and adjacent prairie. Influences of regions of different species composition have also been observed elsewhere (e.g., Andre et al., 1990, for southwest France).
Similar responses should be expected for the grasslands of northern China. Such an evaluation is critical not only to understanding local and regional climate and weather, but also as input into global circulation models. Analysis of this type could help answer a number of interesting questions: Do intense grazing and wildfire cause changes in grassland climate due to major alterations in albedo and the portioning of latent and sensible heat fluxes, compared to what might be expected in the absence of such activities? When material- or man-caused global climate changes occur, what is the expected influence on the seasonal and spatial landscape of the northern China grasslands?
SOCIAL DIMENSIONS OF GRASSLAND STUDIES
For thousands of years, humans have played a key role in the grassland ecosystems of China. The social dimension of grassland ecosystems should be an integral part of any analysis. Yet in China, as elsewhere, the integration of social and natural scientific research has been impeded by both organizational divisions between academic disciplines and the intellectual assumption that views human beings as separate from their natural environment. The result has been for scholars to neglect such issues as the effects of pastoral systems on grassland ecology, the dynamics of herd growth and risk taking in pastoral economies, and the impact of mass migration of Han settlers into grassland areas previously dominated by indigenous minority groups. The reforms of the 1980s, which dissolved the collectives and turned livestock and in many cases land over to individual households, have given renewed importance to indigenous systems of production. Future grassland studies must take greater account of social and economic structures and their effects on the exploitation of natural resources.
Traditional pastoral nomadism in China depended on the exploitation of extensive, seasonal pastures. The herds normally consisted of sheep, goats, horses, cattle, and camels, with the addition of the yak in high-altitude areas. Of these, sheep and horses were most important, but most herders preferred to keep a variety of animals, in an effort to achieve security against sudden, catastrophic losses of any one species, and self-sufficiency by offering the widest range of animal products. The proportion of each species within a herd normally reflected the constraints imposed by local ecological conditions: more cattle in wetter regions, more goats than sheep in marginal pastures, more camels along desert margins and yaks in the highlands. Pastoralists also cultivated relations with settled agrarian communities, exchanging wool, meat, milk products, and hides for grain, cloth, tea, and manufactured goods.
Extensive pastoralism required regular movements to take advantage of seasonal pastures. Most herders lived at least part of the year in tents or yurts and migrated as complete families from one seasonal camp to the next. The cycle took two basic forms: horizontal movements across the steppe, and vertical movements up and down mountain slopes. The migrations were limited to a defined territory, and the use of pastures was controlled by extended kinship groups. The number of migrations depended on the quality of the pasture and the availability of water: groups with dependable pastures and water supplies returned each year to a few fixed campsites, whereas those with access to more marginal resources moved more often.
Although extensive pastoralism is associated with the minority peoples of China, not all minorities in the grassland areas are pastoralists. In the northern grasslands, only three major groups traditionally engaged in pastoral nomadism: the Mongols in the east, the Kazakhs in northern Xinjiang, and
the Tibetans in Gansu and Qinghai. Other large minorities, such as the Hui in Gansu, Qinghai, and Ningxia, and the Uighurs in Xinjiang, are farmers or traders who practice animal husbandry as an adjunct to village-based agriculture.
Perhaps the greatest impact on China's grasslands, both ecological and social, during the past century has been the massive influx of Han settlers. During the Qing Dynasty (1644–1911), Han farmers were forbidden to settle in the northern grasslands, a region the ruling Manchus sought to preserve for other minority peoples who were their natural allies. By the end of the nineteenth century, however, these restrictions had lapsed, and Han farmers began to move into Manchuria and southern Mongolia, putting grasslands under cultivation and pressing indigenous pastoralists and their livestock into smaller, less-productive areas. The grasslands that survive in China today are steppe and semidesert areas that have not been colonized by crop agriculturalists.
The expansion of the Han and of the agricultural, industrial, and commercial practices they brought with them have had important, continuing effects on the economics, politics, and ecology of the grasslands. This is a politically sensitive topic in China, and Chinese reports sometimes allude to problems that they fail to describe in detail. However, the issue is so important for grassland ecology and other fields of interest, that it merits more direct and frank discussion.
Finally, analysis of the sociology of pastoralism must rest on a careful treatment of statistics. For example, some of our Chinese colleagues point to figures that show an exponential growth in number of livestock in China as an explanation for degradation or even desertification of the grasslands. Without checking the figures or measuring the effect of heavier stocking rates, it is impossible to confirm or refute such claims, but the panel believes that these statistics should be reexamined in their historical context. Pre-1949 statistics on grasslands and livestock in China are fragmentary and, when available, of uncertain value. It is difficult to say what "traditional" stocking rates might have been. Given the upheavals in China during the first half of the twentieth century, the number of livestock in 1949—the first year of the People's Republic—must have been below, perhaps far below, levels achieved earlier or obtainable under normal conditions. Therefore, it is difficult to say whether increases after 1949 represent absolute gains or simply a return to some historically sustainable norm. There is ample evidence that statistics gathered after 1949, particularly during political campaigns such as the Great Leap Forward (1958–1960), are highly unreliable. As recent research in western Tibet has shown, the assertion that pastures are overstocked may be due to new accounting procedures rather than actual increases in herd size (Goldstein and Beall, 1989).
It is especially important to consider the problem of livestock numbers in the context of herding systems and strategies. Pastoralists have generally attempted to maximize the size of their herds, because livestock represent wealth
on the hoof. Although it is difficult to increase livestock numbers by deliberate planning, the herder may obtain more animals because of good weather, good pasture, or just plain luck. Yet without such innovations as fodder supplies, winter shelters, or veterinary care, large herds are difficult to sustain. Because pastoralism is a risky business in which animals quickly gained can be just as quickly lost to bad weather, disease, or theft, large numbers of animals act as insurance against disaster. Yet if done properly, the introduction of modern ranching techniques to produce animals for sale to national markets could lead to stocking rates far above the level that can be sustained by traditional means. In either case, researchers must analyze the organization and dynamics of animal husbandry before drawing conclusions about their impact on the grasslands ecosystem.
DESERTIFICATION AND DEGRADATION
Throughout much of this century, scientists and other observers have been concerned about long-term changes in the species composition, productivity, stability, and utility of and and semiarid terrestrial ecosystems. In North America, attention has focused on the Great Plains, particularly during the droughts of the 1930s. Europeans have expressed similar concerns about droughts in Africa in the 1930s and during the past quarter century. More recently, Chinese and foreign scholars have begun to take note of the problem of degradation and desertification in China.
The debate over changes in arid and semiarid lands revolves around questions of the causes of these changes, particularly the relative importance of natural versus human factors; their duration; whether they are temporary or irreversible; and what, if anything, can or should be done about the problem. Because nearly all and and semiarid environments are affected by year-to-year variations in weather, it is difficult to distinguish between short-and long-term trends, temporary and permanent changes, and human-and climate-driven ecosystem dynamics. These uncertainties make it difficult to define and distinguish desertification, degradation, and other types of changes in vegetation.
Hellden (1988) cites several definitions of "desertification" and finds that all include the notion of decreasing productivity leading to long-lasting, possibly irreversible desertlike conditions. Most definitions fix the responsibility for desertification on humans or on a combination of human and natural factors. Gorse and Steeds (1987) present the commonly held view that desertification is the sustained, irreversible decline in biological productivity of arid and semiarid land, resulting from both human and abiotic pressures.
Degradation also refers to decreases in productivity or to unfavorable changes in species composition, but generally indicates that these changes are less severe or long-lasting. Stebbing (1935) was among the first to define degrada-
ion as a process of change in ecosystems toward more arid states. Binns (1990) associates desertification with irreversible changes but says that degradation is reversible given favorable weather and adequate time. Still, inconsistency in terminology reflects an incomplete understanding of and lack of agreement on the causes, duration, and results of changes in ecosystem status.
The history of studies of desertification in Africa demonstrates how difficult it has been to reach agreement on whether, how much, and for what reasons desert areas have expanded. In the most famous case, Stebbing (1935) declared that in recent times the Sahara had moved 300 km to the south and that this encroachment was caused by humans. Rodd (1938) disputed both claims, insisting that Stebbing did not understand variable weather conditions and ecosystem response in the Sahara region. Based on field studies, Lamprey (1988) found that from 1958 to 1975, the southern boundary of the Sahara had advanced 100 km. However, analyzing the same area by using remotely sensed data for the period 1962–1979, Hellden (1988) concluded that the desert boundary had not moved at all. Hellden conceded that crop yields and probably rangeland productivity were severely reduced during a 10-year drought, but maintained that the end of the drought was followed by rapid recovery of both rangeland and cropland productivity. Dregne and Tucker (1988) evaluated the work of Lamprey and Hellden, and concluded that Lamprey had failed to prove the advancing desert thesis. Recently, Tucker et al. (1991) reported that analysis of satellite imagery suggests that the Sahara experiences massive between-year changes and may actually have shrunk since 1984. Binns (1990) also reckons that the margins of the Sahara have been degraded (reversible change) rather than desertified (irreversible change).
Opinions about the causes of vegetation change in and lands also vary. Wade (1974), Sinclair and Fryxell (1985) and other writers during the past 20 years attribute desertification in Africa to human activities. On the other hand, Mace (1991) warns, "Sometimes we are so sure of something that we don't need to see the evidence. That Africa's rangelands are being reduced to desert through overgrazing by domestic livestock is received wisdom. But...such a view may be seriously flawed."
The rangeland degradation/desertification problem is not so much a question of human and livestock impacts as a lack of understanding of exactly how abiotic factors regulate these systems. There is little doubt that domestic animals change the ecological character of the relatively small proportion of rangelands in which they are highly concentrated, but the reaction of rangeland vegetation to the abiotic environment is not well understood. Long-term research on extensively used rangelands is required to explain the interactive effects of grazing, weather, fire, and fire suppression. We need to know which types of rangelands are controlled by the abiotic environment and which are controlled by a combination of abiotic and biotic factors. This is a challenge that awaits grassland scientists in China and throughout the world.
MANAGEMENT OF COMMON POOL RESOURCES
The problem of managing common pool (or common property) resources in an ecologically sustainable manner is posed in acute form in extensive dry grasslands of the sort that cover large parts of China. The debate that has occurred around this issue in other parts of the world and with respect to other resources, such as forests and fisheries, can inform policy choices and management strategies in the grasslands of China.
This debate has both theoretical and practical implications. Until recently, theory was dominated by the notion of the "tragedy of the commons," which explains grassland degradation as the inevitable consequence of rational strategies pursued by individual owners who stock privately owned animals on publicly or collectively owned land. This theory holds that under such circumstances, the benefit of each animal accrues to the owner, whereas external costs (essentially a reduction in the amount of grazing available by one animal ration) must be shared by all. Because under these circumstances private benefit exceeds private cost, each owner is encouraged to continue adding animals to his herd, leading inevitably to overuse and degradation. Theorists who pursue this line of analysis generally arrive at one of two policy preferences: to increase privatization of common resources (land, water, forage, etc.) so that both costs and benefits accrue to the same owner/decision maker; alternatively, to expand the regulatory powers of natural resource bureaucracies, enabling them to enforce scientifically determined stocking levels.
The first challenge to this view came from empirical studies of grassland users, especially traditional herding societies, and other customary users of common forests and fisheries. These studies showed that far from being the object of abuse by private owners, common pool resources such as pastures are often subject to well-defined access and management rules enforced by effective customary institutions. Such rules specify who has access to the resources and under what conditions, regulate access and levels of use, and provide for the resolution of conflicts and enforcement of sanctions. Although true open-access commons, such as the atmosphere or ocean floors, exist, they are the exception. Controlled-access commons, where natural resources have customary users such as herders, farmers, or fishermen, are more often the norm.
Recent theory on common pool resources (National Research Council, 1986; Ostrom, 1990) has caught up with these findings, by examining the conditions under which user groups, whether organized in a traditional (e.g., kinship) or modern (e.g., cooperative) unit, can be assured that individual owners will respect collective rules about resource use. Where members of such groups share similar production objectives and methods, where there are no large differences in wealth or social status, where group membership has important benefits in addition to those connected with production, and especially where rules governing resource use are effectively enforced by the group or by some
superior authority, it is likely that common resources will be managed in a sustained manner.
Conversely, where these conditions do not apply, it is unlikely that common resources such as pasture can be sustained. This is the case, for example, where members of the group do not follow similar production strategies, where there are large wealth or status differentials, where group membership has few benefits, and especially where rules about resource use and management are unenforced or unenforceable. The last condition prevails where governments attempt to manage common resource use through ill-equipped centralized bureaucracies and inappropriate regulations and, in so doing, undermine the rules and procedures adopted by the resource user groups themselves. In this situation, individual producers may find it rational to ignore the rules, which result in a tragedy of the commons.
This debate has important implications for grassland management and research in China, especially in the search for land tenure policies and institutions to promote sustainable resource use. Common pool resource management theory and experience under conditions similar to the Chinese grasslands can illuminate several issues important to Chinese policymakers. These include the role of pastoral land tenure structures; the effect of these structures on various types of resources, especially valuable, small-scale resource patches (such as wetlands) within larger, low-value resource areas; the role of leases, contracts, and the conditions that attach to them; and the relationship of local decision making and enforcement to wider state structures.
Existing knowledge about common pool resource management could be of immediate use to Chinese policymakers. Similarly, research on these questions as they relate to the Chinese grasslands can add to the general theory and to our understanding of its practical application. Ideally, such research should be carried out through collaboration between natural and social scientists, using methods that give a prominent place to the perceptions and capabilities of pastureland users themselves.
RATIONAL RANGELAND MANAGEMENT
There is no doubt among Chinese scientists and government officials that some rangelands of northern China are being overused and degraded. For example, official figures for Inner Mongolia suggest that more than half of the grassland area of the Inner Mongolia Autonomous Region (IMAR) is degraded and that almost one-fourth of this degraded land is "unusable." Similarly, 21.2% of all grasslands in northern China are judged unusable (see Table 1-1). Observations made by the two CSCPRC delegations confirm that degradation does occur at specific sites throughout northern China. Serious cases of desertification appear on sandy soils where cultivation and large concentrations of people or livestock in agricultural villages have resulted in soil denudation and
activation of sand dunes. However, some Chinese authorities assert that more general degradation is occurring throughout the rangelands of northern China and that this degradation has been caused by overstocking, overgrazing, and other irrational rangeland management practices adopted by minority pastoralists such as Mongols, Tibetans, and Kazakhs. It is important to differentiate among damage patterns caused by concentrations of people in villages, faulty agricultural practices, and improper or irrational rangeland management. The extent and intensity of damage and the policies needed to reverse these trends are different in each case.
The concepts of irrational pastoral land use practices and resulting land degradation originated with Herskovitz's (1926) hypothesis that pastoralists accumulate vast numbers of livestock for reasons of social power and prestige. Hardin's (1968) notion, the tragedy of the commons, has also been invoked to illustrate the irrational and destructive nature of pastoral management. Brown (1971) argued that pastoralists were irrational because they conducted dairy operations in environmental settings suited for beef production. The belief that pastoral livestock management is irrational and inherently destructive has a long history and has been widely accepted by scholars and officials in the international development community (Sandford, 1983).
On the other hand, a number of scientists who have worked with pastoralists and their ecosystems disagree with this premise. They find many pastoral strategies perfectly rational, given the circumstances facing the herders in question (Helland, 1980; Sandford, 1983; Swift and Maliki, 1984; Ellis and Swift, 1988; Goldstein et al., 1990; Mace, 1991). These studies cast doubt on the premise that pastoralism leads inevitably to the destruction of rangelands. The scientific literature contains many cases around the world (including Tibet) where pastoral practices are not irrational and are not degrading the environment. Instead, these studies demonstrate that ill-founded assumptions, including herder irrationality, have contributed to the adoption of poor livestock development strategies (de Haan, 1990) and in some instances have actually caused environmental damage and reduction in the economic welfare of the very people the strategies were designed to help. For example, it is often argued that pastoralists should be settled and become agriculturalists for their own good as well as that of the environment. Yet the greatest environmental damage is generally found around wells or agricultural villages where pastoralists have been settled (Lusigi, 1981; Sinclair and Fryxell, 1985).
There is no doubt that improper rangeland management has contributed to environmental damage and economic loss—in China as elsewhere. However, the record also shows that traditional pastoral people may be no more likely to cause these problems than are scientists and development personnel who make easy but incorrect assumptions about unfamiliar ecosystems or modern ranchers who find themselves in an economic squeeze. Inappropriate practices and destructive strategies are often found in both traditional and modern grazing
systems. Yet analysis of these practices usually reveals that they are quite rational. Often, long-term sustainability is sacrificed for short-term survival. Rather than dismiss such behavior as irrational, rangeland scientists should concentrate on understanding the causes that drive people to sacrifice the sustainability of their environment for short-term gain and on finding solutions that will help them satisfy both short-and long-term needs.
CONSERVATION AND WILDLIFE
The study of grasslands in China as elsewhere must take account of wild as well as domesticated animals and of the importance of preserving the grasslands not only as an economic resource, but also as a natural reserve. The fate of wildlife on the grasslands of China gives cause for concern. There has been in recent years a considerable reduction in the number, variety, and range of wild animals, especially large ungulates, in the area covered by this study. In 1932, the central Asian explorer Roy Chapman Andrews described the huge herds of Mongolian gazelles on the eastern steppe. ''The entire horizon appeared to be a moving line of yellow bodies and curving necks,'' wrote Andrews. "Thousands passed in front of us." Sixty years later, the range of this gazelle has decreased by more than two-thirds. The saiga antelope and Przewalski's horse are extinct in China, while the wild Bactrian camel has been reduced to perhaps 500 individuals, all in the most remote desert tracts. The goitered gazelle, wild ass, wild yak, and Tibetan antelope have declined to a fraction of their former numbers.
The reduction in the number of ungulates in China, as well as such predators as wolf and snow leopard, has been the result of several factors. Unrestricted hunting, to eliminate a threat to livestock, reduce competition for forage, or provide meat, hides and other products for subsistence or commercial use, has taken a heavy toll. The decline of wild animals was particularly sharp during the Great Leap Forward (1958–1960), when agricultural production dropped and many animals were slaughtered for food, and during the Cultural Revolution (1966–1976), when conservation directives were generally ignored.
Despite these problems, China's rangelands continue to support a variety of wild ungulates: the goitered gazelle (Gazella subgutturosa ), Przewalski's gazelle (Procapra przewalski), Tibetan gazelle (Procapra picticaudata), Mongolian gazelle (Procapra gutturosa), Tibetan antelope (Pantholops hodgsoni), wild ass (Equus hemionus), wild yak (Bos grunniens ), argali sheep (Ovis ammon), blue sheep (Pseudois nayaur), Asiatic ibex (Capra ibex), white-lipped deer (Cervus albirostris), red deer (Cervus elaphus), and wild Bactrian camel (Camelus bactrianus). In northwestern Tibet and southwestern Qinghai, both remote and almost uninhabited areas, the Tibetan antelope, wild ass, and other unique upland fauna survive in moderate abundance. More than 200,000 Mongolian gazelles migrate between the eastern steppes of Mongolia and Inner Mongolia.
The preservation of these wild ungulates, which is essential for conserving biodiversity, may also make better economic sense than the recent practice of displacing wildlife with domestic livestock. As this study points out, vast areas of China's grasslands are unusable, because they are too dry, too remote, too degraded, or suffer from some other limitation. Such marginal rangelands could be devoted to protecting and in some instances managing wild ungulates on a sustained-yield basis. Practices in other parts of the world suggest that the economic return per hectare from some rangelands can be increased by using a mix of livestock and wildlife rather than livestock alone. Mongolia, for example, has attempted to manage gazelles by protecting and harvesting a certain number each year for the European luxury meat market—a program that suffers in part because the gazelles migrate seasonally into China where they are slaughtered indiscriminately.
Conservation, for whatever purpose, must begin with research. We know little or nothing about the current status, distribution, or habits of most wild ungulates in China; about how their presence affects domestic livestock; or about the ways domestic and wild animals might coexist for the benefit of all. The one area of research on wild mammals covered in this study is on grassland rodents. In China, as elsewhere, the assumptions that herbivorous rodents and pikas (Ochotonidae) damage the grasslands and should be eradicated or at least controlled are apparently shared by scholars and policymakers alike. Pikas—the ecological equivalent of the American prairie dog—have been greatly reduced in many areas of China by the application of zinc sulfide poison. This program is reminiscent of the mass poisoning of prairie dogs in the United States and may be similarly unwarranted. It is sometimes overlooked that pikas eat various forbs unpalatable to livestock and that the digging activities and underground defecation of pikas and other rodents help to recycle nutrients. For these and other reasons, the elimination of small mammals could have a long-term negative impact on rangelands. More research is needed, however, to provide an adequate understanding of these questions and a sounder basis for rangeland management practices.
Given the rate of environmental destruction in China and other parts of the world, rangeland research in all countries should incorporate a conservation component, whether it involves concern for the protection of rare species, such as the wild Bactrian camel, or of unique ecosystems, such as the eastern Mongolian steppe. Rangeland surveys should look beyond livestock production to consider the protection of key habitats. Studies of rangeland economy should explore the ways in which domesticated and wild animals can coexist, while providing the greatest long-term benefit to the balance between man and nature. To protect the diversity and richness of China's rangelands and to guarantee that future generations of Chinese enjoy the benefits of this great resource are important tasks that deserve the attention of scholars in China and abroad.
THE ORGANIZATION AND CONDUCT OF SCIENCE
One of the perennial challenges to science in general and ecosystem science in particular is finding the right balance between disciplinary and interdisciplinary approaches. Working within an established discipline contributes a focus, standards, and consistency that can make the results of research comprehensible and useful to other scholars. In recent years, however, scientists who study grasslands and other ecosystems have widened the boundaries of inquiry and developed new techniques that help integrate work from various disciplines into a more holistic view of nature. In fact, reorganization of this science has been in part a response to the intellectual transformation that considers an adequate understanding of natural ecosystems to be dependent on contributions from a wide range of disciplines.
This study demonstrates that scientists in China are also taking part in that transformation. One feature of Chinese science that many foreign visitors have noted is its segmentation into islands of disciplinary expertise. This phenomenon has been attributed to various factors: an organizational structure that places resources in the hands of research institutes concerned only with their distinct, separate missions; a personnel system that allows for little mobility of students and scholars from one institution or part of the country to another; a centrally planned economy that offers few incentives for communication between researchers and producers; or cultural traits that mitigate against the free and open sharing of information. Whatever the reasons, fragmentation of scholarship along disciplinary and institutional lines remains a feature of grassland science in China and is evident in the findings of this study.
Two fields that have not been adequately integrated into the agenda of Chinese grassland research are animal husbandry and human behavior. This reflects, in part, the division of labor among China's research institutions: the Chinese Academy of Sciences is responsible for basic research in the natural sciences; the Academy of Agricultural Sciences, for applied research related to agriculture and animal husbandry; and the Academy of Social Sciences, for work on human behavior. None of China's major centers of grassland science has paid significant attention to the role of domesticated animals in the context of the grazing system. Social studies have been even further removed from the grassland agenda. Despite widespread awareness of and acute concern for the human impact on grasslands, research in China on grassland ecosystems has made little use of economics and none whatsoever of political science, sociology, anthropology, demography, history, or other disciplines that might help explain the relationship between man and nature. Similarly, work within established scientific disciplines has been largely self-contained; there has been too little effort to draw together scholars in the fields of botany, zoology, meteorology, soil science, and so forth to develop interdisciplinary and collaborative research.
Happily, there are signs that things are changing. It is probably impossible to find a Chinese scientist engaged in the study of the grasslands who is not aware of the fragmentation of research in China and does not condemn it. These men and women recognize that ecological problems involve complex relationships among diverse natural and human factors. They know that ecosystem science in the West has developed conceptual frameworks, methodologies, and techniques that take account of this complexity, and they want to learn and apply these approaches to the study of natural resources in China. Some institutions have already taken steps in this direction. The Gansu Grassland Ecological Research Institute in Lanzhou has introduced a variety of new disciplines, technologies, and collaborative projects that are weaving together a more comprehensive view of China's grasslands. The Inner Mongolia Grassland Ecosystem Research Station at Xilingele is demonstrating how a common research site, open to scholars from throughout China and the world, can produce greater synergy and better results than the all too often insulated research institute, that has been the standard form for the conduct of science in China. Panel members were impressed by the commitment to such changes expressed by scholars in other Chinese universities and research institutes as well.
One of the keys to this transition is new technologies, such as computer modeling and remote sensing, and the skill to use them. Whole ecosystems are too complex to describe and analyze with ordinary language or simple mathematical formulas. Simulation models integrate various parts of the system, such as precipitation, plant productivity, and animal nutrition, either for analytical purposes or to predict future outcomes. These models perform a synthetic function, enabling scientists from different disciplines to address common problems, and for this reason, they have been central to the evolution of contemporary ecosystem science.
Simulation modeling is still poorly developed in China. In part, the problem is inadequate hardware and software. But a more serious problem is access to training. Chinese with knowledge of this area are few in number and limited in experience, which generally requires study abroad. The development of mathematical models depends on the understanding and cooperation of field scientists who provide the data, and scientists in China, as elsewhere, have sometimes been slow to adopt these unfamiliar techniques. In sum, even though Chinese scientists sense the importance of modeling for developing an integrated view of man and nature, progress in this area remains difficult.
The Chinese are more advanced in the use of remote sensing, a technique that has helped to develop regional and global approaches to ecosystem science. They have considerable experience with satellite technology, maintain a Landsat receiving station that supplies Multispectral Scanner (MSS) and Thematic Mapper (TM) data, and have been successful in using remote images for mapping grasslands and other biomes. A wide array of maps and accompany-
ing texts, scheduled for publication in the next two to three years, should make an enormous contribution to knowledge of the state of China's grasslands and other natural resources. However, Chinese efforts to process and use digital data, to coordinate this date with ground-based measurements, and to construct and manage complex geographical information systems (GIS) are just beginning and are subject to some of the same human and material limitations that have slowed the development of simulation modeling.
The problems of handling large bodies of data and integrating information from various disciplines to create new forms of knowledge is the central challenge of ecosystem science at the end of the twentieth century. This is a challenge that unites scientists inside and outside China in a common quest to understand nature and man's relationship to it. Members of this panel look forward to joining our Chinese colleagues in this adventure.
Andre, J.-C., P. Bougeault, and J.-P. Goutorbe. 1990. Regional estimates of heat and evaporation fluxes over non-homogeneous terrain. Examples from the Hapex-Mobilhy programme. Boundary Layer Meteorology 50:77–108.
Andrews, Roy Chapman. 1932. The New Conquest of Central Asia. New York: American Museum of Natural History.
Barfield, Thomas. 1989. The Perilous Frontier: Nomadic Empires and China. Oxford: Basil Blackwell.
Barth, Fredrik. 1961. Nomads of South Persia: The Basseri of the Khamseh. London: George Allen & Unwin.
Binns, T. 1990. Is desertification a myth? Geography 75:106–13.
Brown, L.H. 1971. The biology of pastoral man as a factor in conservation. Biol. Conser. 3(2):93–100.
de Haan, C. 1990. Changing trends in the World Bank's lending program for rangeland development. Pp. 43–54 in Low Input Sustainable Yield System: Implications for the World's Rangelands. R. Cincotta, G. Perrier, C. Gay, and J. Tiedeman, eds. Range Science Department, Utah State University, Logan.
Dregne, H.E., and C.J. Tucker. 1988. Desert encroachment. Desertification Control Bulletin 16:16–19.
Ellis, J.E., and D.M. Swift. 1988. Stability of African pastoral ecosystems: Alternate paradigms and implications for development. Journal of Range Management 41(6):450–459.
Forse, B. 1989. The myth of the marching desert. New Scientist 4:31–32.
Goldstein, M.C., and C.M. Beall. 1989. The impact of China's reform policy on the nomads of western Tibet. Asian Survey 29:619–641.
Goldstein, M.C., C.M. Beall, and R.P. Cincotta. 1990. Traditional conservation on Tibet's northern plateau. National Geographic Research 6(2):139–156.
Gorse, J.E., and D.R. Steeds. 1987. Desertification of the Sahelian and Sudanian Zones of West Africa. World Bank Technical Paper Number 61. World Bank, Washington, D.C.
Hardin, G. 1968. The tragedy of the commons. Science 162:1243–48.
Helland, J. 1980. Five Essays on the Study of Pastoralists and Development of Pastoralism. Occasional Paper No. 20, University of Bergen, Norway.
Hellden, U. 1988. Desertification monitoring: Is the desert encroaching? Desertification Control Bulletin 17:8–12.
Herskovitz, M.J. 1926. The cattle complex in East Africa. American Anthropol. 28:230–272.
Jagchid, Sechin, and Van Jay Symons. 1989. Peace, War, and Trade Along the Great Wall: Nomadic-Chinese Interaction Through Two Millennia. Bloomington: Indiana University Press.
Khazanov, Anatoly M. 1984. Nomads and the Outside World. Translated by Julia Crookenden. Cambridge: Cambridge University Press.
Lamprey, H. 1988. Report on the desert encroachment reconnaissance in northern Sudan: 21 October to 10 November 1975. Desertification Control Bulletin 17:1–7.
Lattimore, Owen. 1940. Inner Asian Frontiers of China. New York: American Geographic Society.
Lusigi, W.J. 1981. Combatting Desertification and Rehabilitating Degraded Population Systems in Northern Kenya. Technical Report A-4, UNESCO-UNEP Integrated Project in Arid Lands, Nairobi.
Mace, R. 1991. Conservation biology: Overgrazing overstated. Nature 349:280–281.
National Research Council. 1986. Proceedings of the Conference on Common Property Resource Management. Washington, D.C.: National Academy Press.
Ostrom, E. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.
Pielke, R.A., and R. Avissar. 1990. Influence of landscape structure on local and regional climate. Landscape Ecology 4:133–155.
Pielke, R.A., G. Dalu, J. Weaver, J. Lee, and J. Purdom. 1990a. Influence of landuse on mesoscale atmospheric circulation. Fourth Conference on Mesoscale Processes, Boulder, Colo., June 25–29, 1990.
Pielke, R.A., T.J. Lee, J.F. Weaver, and T.G.F. Kittel. 1990b. Influence of vegetation on the water and heat distribution over mesoscale sized areas. Eighth Conference on Hydrometeorology, Kananaskis Provincial Park, Alberta, Canada, October 22–26, 1990.
Pielke, R.A., G. Dalu, J.S. Snook, T.J. Lee, and T.G.F. Kittel. 1991. Nonlinear influence of mesoscale landuse on weather and climate. Journal of Climate (forthcoming).
Rodd, F. 1938. The Sahara. Geographical Journal 91:354–355.
Sandford, S. 1983. Organization and management of water supplies in tropical Africa. ILCA Research Report (Addis Ababa, Ethiopia) December 1983, 20–31.
Sinclair, A.R.E., and J.M. Fryxell. 1985. The Sahel of Africa: Ecology of a disaster. Canadian Journal of Zoology 63:987–994.
Stebbing, E.P. 1935. The encroaching Sahara: The threat to the West Africa Colonies. Geographical Journal 85:506–524.
Swift, J., and A. Maliki. 1984. A Cooperative Development Experiment Among Nomadic Herders in Niger. Overseas Development Institute, Paper 18c.
Tucker, C.J., H.E. Dregne, and W.W. Newcombe. 1991. Expansion and contraction of the Sahara desert from 1980 to 1990. Science 253:299–301.
Wade, N. 1974. Sahelian drought: No victory for Western aid. Science 185:234–237.
Waldron, Arthur. 1990. The Great Wall of China: From History to Myth . Cambridge: Cambridge University Press.