This blog is part of the Agriculture and Ecosystems Blog’s month-long series on Restoring Landscapes.
Pastoral land covers 400 million hectares of China, 41.7% of the entire country. Much of it is semi-arid and high plateau land where China’s ethnic minorities live, traditionally using the land to produce meat, fur, wool and cashmere. Over the past century, increasing pressure from human and livestock populations has resulted in 90% of the pastoral land being classified as degraded to some degree, especially in the more arid regions where ecosystems are the most fragile.

This degradation is of increasing concern for many, not only for local livelihoods and the conservation of endemic species, but also because the headwaters of several major Asian rivers are found in these areas. What happens in these rangelands will have important implications for millions of people downstream, and difficult decisions have to be made to safeguard this vital resource for China, and Asia’s future generations.
Unfortunately, using the land’s ecosystem services sustainably is a challenge for most pastoralists. Degraded lands lack good quality forage, reducing the overall productivity and quality of livestock. Lack of good forage also increases sheep mortality in the winter, which inevitably reduced pastoralists’ incomes. To compensate, pastoralists increase their livestock numbers, which only makes matters worse, trapping them in a vicious cycle of further degradation.
Potential solutions
The Chinese government has tried to implement restoration activities and grazing management guidelines, such as no-graze areas and rotational grazing. But most of these activities depend on government support, fail to guarantee pastoralists’ livelihoods, are complicated by tenure issues, and do not address market forces that cause intensified pastoral land use in the first place. It is no wonder that pastoral participation is hard to come by.
To try to address the situation, the Chinese Ministry of Agriculture embarked on a five-year project in 2010, led by the Inner Mongolia Agricultural University. Bioversity International has been a key partner in evaluating the resilience and sustainability of the pastoral ecosystem, providing concrete suggestions for pastoralists and policymakers to sustainably manage these natural resources. After all, how can we manage what we cannot measure?
Working with optimization models, Bioversity has been working with pastoralists for three years on balancing their livestock numbers and improving their feeding methods in accordance with the ecosystem’s capacity. The work is already contributing to an increase in the pastoralists’ income, and reducing grazing pressure on the land. Our study covers five core pastoral areas in China: the Northeast region, the Inner Mongolia plateau, the Loess plateau, Xinjiang and the Qinghai-Tibet Plateau.
Case study: Maqu county, Gansu province
Maqu County is a typical pastoral county in the southern part of Gansu, high up on the Tibetan Plateau. With over 89% of the total area being pastoral land, livestock production on average contributes to 97% of the county’s net per-capita income. However, 90% of its pastoral land is considered to be in deterioration, with 25.30% in severe degradation.
When the project began, data on the county’s livestock production system was relatively sparse and insufficient. Therefore, farm surveys and field experiments were carried out to gain a better understanding of the ecosystem’s characteristics, climatic conditions, feeding resources, as well as the economic status of pastoralists and the quality and quantity of livestock they keep. From this data, several optimization models were used to investigate tactics and strategies for farm improvement.
Feed-livestock balance

The traditional livestock in Maqu County are Oula sheep, a breed that survives well under the typically harsh conditions of the plateau. While its wool is neither particularly prized nor abundant, Oula meat is favoured by locals.
Yet their resilience to degraded pastoral land is unable to match their popularity. We found that most sheep were underfed, particularly in winter months (Fig. 2). Our research also suggests that the number of livestock exceeds the carrying capacity of the pastoral land, leading to overgrazing, landscape degradation, and insufficient pasture for livestock to feed on.
But simply reducing the amount of livestock raised is not a realistic solution for most pastoralists, so we investigated several solutions that could work together to optimize pastoral resources.
Solution One: Seasonal grazing
To reduce overgrazing, one option could be to graze different areas in different seasons, or feed the sheep in a warm pen during winter months to minimize weight loss from the cold. We found that having a two-field rotational grazing system (from May through September on one area and from October through April on another) was most strategic for pastoralists.
This type of rotational grazing was also compared with pen-feeding the sheep in the winter months, but the high costs from pen-feeding outweighed any benefit from minimizing weight loss by keeping sheep out of the cold.
Solution Two: Changing lambing time and sale dates
Traditionally, lambs are born in January (mid-winter) when temperatures are low and the ewes lose weight. The lambs are therefore undernourished until June, when grass growth is adequate for grazing. Through a series of trials with several lambing times, we found that by lambing in spring or summer, the lambs could have more access to high quality forage and less need for supplementary feed, thus reducing costs and increasing net income.
Working together
While our models cannot include and analyze all parameters of a pastoral ecosystem, the project supports the idea that good management and local participation can enhance the services that we get from ecosystems, and help reduce increasing population and market pressures. Using both solutions outlined, lambing in the spring (April specifically) and using a two-field rotational grazing system, farmers could maintain their current stocking of 0.6 sheep units/ha while reducing grazing pressure on the land.
By providing pastoralists with accurate information and developing realistic management strategies together, we were able to achieve some positive results within a short period of time. Hopefully this can be a model for further restoring this vital ecosystem, and we encourage more local pastoralist participation in the future.
We would like to thank and acknowledge the following partners: Gansu Agricultural University, Lanzhou, China; Inner Mongolia Agricultural University, Hohhot, China; Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, China; Australian Centre for International Agricultural Research, Orange, Australia