The Pasture Problem

Can Smart Development and Intensification Slow the Expansion of Pasture?

In this essay:


Recent decades have seen remarkable developments across the pastures of the world. Even as production of meat and dairy from ruminants (grazing animals such as cattle and sheep) increased by almost a third, the footprint of pasture has begun to decline. And this change is significant, shrinking by nearly 64 million hectares, an area larger than France, between 2000 and 2013.1 The gains have been considerable for conservation. To the benefit of endangered species from the Asiatic cheetah in Iran to the saiga antelope in Kazakhstan, pastureland is going out of production and returning to nature.2

While promising, these developments will not be enough to assure that rising demand for meat does not put new pressure on critical habitats. Global demand for ruminant meat and dairy is expected to rise by 44% between now and 2050.3 Even as pasture has shrunk in the global aggregate, it continues to expand in many parts of the world, particularly in emerging economies and the tropics, where some of the most intact and threatened areas of natural habitat remain.


Read more from our series on The Future of Food.


Pasture still occupies more space on the planet than any other land use, human or natural. More than half of the earth’s natural grasslands and savannahs and almost a fifth of cleared temperate and tropical forests are grazed.4,5 In the Amazon, as much as 70% of deforestation is attributable to ranching operations,5–8 while an estimated 73% of the world’s natural grasslands that are used for pasture have been degraded due to overgrazing.5,9

The management of pasture in the coming decades will therefore have an outsized impact upon the success or failure of global conservation efforts. Yet despite its pivotal role in the earth system, the study of pasture and grazing systems has been relatively neglected compared to other agricultural systems.10,11 Grazing systems are crucial for the global economy and food supply. Globally, they provide almost 50% of feed for ruminant production,12 while the ruminant sector as a whole provides virtually all the world’s dairy, a third of the meat, and a fifth of the animal protein. Regionally the ruminant sector has major economic significance, providing an estimated 20–40% of agricultural GDP in parts of sub-Saharan Africa.13

How we manage pasture and livestock production systems and how and where pasture expands and contracts will play a major role in what sorts of conservation outcomes will be possible. Global pastureland has declined because meat production in Western Europe and North America has improved in efficiency dramatically. There now exists a substantial yield gap between developed and emerging economies, one that will need to be closed in order to avoid further expansion of pasture.

If the improvements in production efficiencies experienced by developed countries can be repeated in emerging economies, then it may be possible to spare further expansion in critical areas such as tropical forests and to reach true peak pasture at the global scale. If not, then there exists the very real possibility that the current downtrend will be merely a temporary plateau—a short respite that heralds a second wave with potentially devastating consequences for some of Earth’s most diverse ecosystems and wildlife. In this essay, I consider where, how, and when pasture efficiency will need to be improved in order to minimize or avoid further habitat losses.


The saiga antelope is a critically endangered species that benefits as pastureland goes out of production and is returned to nature.

A Global Turnaround

What has changed in the pasture system to allow the current contraction? To get a better understanding of this phenomenon, it is necessary to take a closer look at the global picture.

First, it is worth validating that the change seen in the data from the UN’s Food and Agriculture Organization (FAO) is in fact genuine. The diversity of pasture systems makes defining and measuring pasturelands complex. After all, they span virtually every biome on the planet and can range from lush green meadows in Europe to dusty deserts in Central Asia. Indeed the reliability of the FAO data has been questioned previously for these very reasons.14

The good news is that comparison of independent land-use assessments corroborates the decline (Figure 1). These assessments utilize remote sensing data in addition to inventory data of the sort used by the FAO, and are thus a powerful confirmation that pasture is indeed contracting globally. Where they disagree is on the absolute area. This is perhaps not surprising when one considers the wide range of different geographies on which livestock are grazed. But the fact that all assessments point to a downward trend is good reason for conservationists to take heart that the decline is genuine.

Figure 1: Alternative estimates of land area under permanent pasture and meadows. Blue dots are FAO data; red dots are estimates by the HYDE model; yellow, orange, and green are estimates by the Global Landscape Initiative, SAGE model and GTAP, respectively. Note that the Global Landscape Initiative, SAGE, and GTAP are iterations of the methodology of Ramankutty & Foley (1999)15 but differ between years. Caution must therefore be applied in interpreting these trends as absolute evidence of a trend. Dotted lines are provided only to highlight estimated temporal trends from similar datasets.

Even more encouragingly, this contraction has not come at the expense of production. In fact, from 1961 to 2013, total production of meat from goat, beef, and sheep increased by 387%, 129%, and 73%, respectively, while dairy output increased by 103%. This represents a doubling of production efficiency in terms of pasture requirements such that in 2013, the area of pasture required per unit production was 52% less than it was in 1961 (Figure 2). Yields and stocking densities are also on the up. The average cattle produced 33% more meat and dairy in 2013 than in 1961, and there were 76% more animals in the production system for each unit of pasture globally. Not only is more meat and dairy being produced from each unit of pasture, but it is also being done so effectively that an absolute decoupling of land and production has been achieved.

Figure 2: Decoupling of ruminant meat production (beef, sheep, and goat) from area of pasture and population. Indices displayed are relative changes in (a) total production (tonnes, meat); (b) head of animals (total number of animals); (c) pasture per unit production (hectares per tonne of meat); and (d) pasture per capita (the amount of pasture area under production per unit global population, hectares / unit population). Data: FAOSTAT, World Bank.

Driving these trends are marked improvements in the efficiency of global production. Two important indicators of efficiency improvements are yields and stocking densities. Yield refers to how much meat or dairy is produced per animal, while stocking density refers to the number of animals on each unit of land. To maximize the amount of meat or dairy produced from each parcel of land, both need to be optimized, such that ranchers have the maximum number of animals in any one area growing to their full potential.

Livestock production efficiency can be improved in several ways. Supplemental feeding (particularly with high-energy feeds), better pasture management (fertilization, weed control, irrigation, or rotational grazing), improved animal health (veterinary care and better living conditions), and selective breeding (for beneficial traits) all help to improve these metrics. There are trade-offs associated with every production system, of course, but the goal is to ensure that animals have access to year-round, high-energy feed and are maintained in peak health to enhance their ability to grow to their maximum size.

Globally the picture looks positive. So positive, in fact, that it might be reasonable to ask if the world has already reached peak pasture, and to wonder if the main challenge now is simply to ensure that land that goes out of production is turned back to nature. There is, however, need for caution.

Regional Outlook

Where the picture starts to break down is at the regional level. It is here where the stark disparity between developed and emerging markets becomes apparent, and with it, a more cautionary tale emerges. Over the past 52 years, while developed economies have seen slower demand growth, higher meat yields, and consistent contraction of pasture, the opposite has been true for emerging economies (Figure 3).

Figure 3: Regional changes in (a) area of pasture, (b) total consumption of ruminant meat, and total (c) beef yields (kilograms per head of animal) and (d) dairy yields (kilograms per head of animal for cow milk). Pasture is defined as the area of permanent pasture and meadow, per the FAO. Consumption is the total domestic consumption of ruminant meat (beef, sheep, and goat) and dairy (cow). Consumption = Production — Exports + Imports + Changes in Inventory. Beef yields are kilos of meat produced per animal. Data: FAOSTAT.

Underpinning this divergence are stark differences in production efficiency. On average, beef yields are 65% higher in developed economies than in developing economies. It is not just the absolute yields that differ. In 2013 beef yields in all three emerging economy regions lagged behind those achieved in developed economies. The rate at which these changes have occurred is also markedly different. For example, in 1961, beef yields in Europe and Oceania were below those of Latin America. But while yields in Latin America grew by a relatively modest 15% between 1961 and 2013, in Europe and Oceania the improvement over the same period was greater than 80%.

A crucial reason for the differences in production efficiencies is the use of supplemental feeding. In North America, high-energy feeds are provided in feedlots. In the European Union, silage usually made from maize or hay is used. By contrast, much of Latin America, Africa, and Asia rely on extensive pasture systems or small-scale cut-and-carry operations where waste crop products and/or fresh grass is brought to the animals.16 These latter operations are more vulnerable to seasonal variation in the availability of forage and are thus less optimized to maintain growth and production year-round.

These regional dynamics are changing where and how pasture expands with potentially severe implications for biodiversity. As pressure mounts in Latin America, Africa, and Asia, two highly biodiverse systems have become the frontline of pasture expansion: the forests and grasslands of the tropics and subtropics.

Where and how pasture expands will have potentially severe implications for biodiversity.


Between 2000 and 2015, the area of pasture that was formerly forest contracted by 7%.17,18 However, this contraction was driven by a reduction in temperate regions hiding continued expansion of pasture in tropical forests, particularly in the Amazon and Southeast Asia.19 Tropical forests are home to an estimated half of the earth’s species.20 Many of these species are highly specialized, adapted to living in the canopy. Remove the forest and not only is habitat lost, and with it many of the species that rely on it, but the local climate is also disrupted.21 The trade-offs between production and biodiversity are arguably more acute in tropical forests than in any other ecosystem. Even relatively small-scale disturbance or fragmentation can lead to marked declines in species richness and abundance.22 In an assessment of the global impacts of agriculture, pasture, and forest use, Chaudhary and colleagues (2016) found that pasture was the primary driver of biodiversity loss in Brazil, Madagascar, and China.23 Maintaining the intactness of natural habitat, maximizing production on degraded lands, and minimizing the human footprint therefore need to be priorities in tropical forests.


Pasture is the primary driver of biodiversity loss in Brazil, China, and Madagascar (above).

Tropical grasslands, on the other hand, present a more nuanced picture. These systems have often evolved with grazing pressure for millennia. Extensive pastoral grazing is known to play an integral role in maintaining habitat and can be compatible with relatively high levels of biodiversity.24 Researchers studying the Mara ecosystem in Kenya, for example, found that where pastoral communities, livestock, and wildlife shared a landscape, the resulting mix of species was often more productive and diverse than that in nearby parks where people and livestock were excluded.25

There were trade-offs, however. While the combination of grazing, manure fertilization, and burning promoted new plant growth, which benefited native herbivores and invertebrates, it also kept grasses short, which was detrimental to large carnivores that relied on tall grass for camouflage while hunting. Ultimately this balance can only be achieved at relatively low stocking densities. As grazing intensifies, and competition for grazing and predation of livestock increases, so do conflicts with wild herbivores and carnivores. Competition with livestock has indeed been identified as a major driver of wildlife loss on African rangelands in numerous studies,26,27 and overgrazing has been associated with degradation of grasslands in Mongolia,28 Mexico,29 and China.30

The decline in the global area of pasture therefore represents something of a mixed bag. Opportunities are certainly appearing in developed economies and on more marginal lands, often on dry rangelands that have inherently low production potential. However, big challenges are presenting themselves in the tropics and subtropics. As pressure on grazing lands intensifies, the trade-offs for biodiversity begin to look increasingly stark.

Looking to the Future

The biggest determinant of where pasture is likely to expand most is demand. This is important because despite concerns that demand for beef is increasingly being met by overseas imports, there is surprisingly little evidence to support such a trend. In fact, with the exception of Oceania, whose export market has blossomed, the ratio of production to consumption has been almost constant and broadly equal to one in every global region over the past half-century.1 This means that, on balance, regional production tracks regional demand pretty well. The implication is that where demand grows most, so will the pressures on pasture.

What do demand forecasts tell us? Two major conclusions become immediately apparent from the FAO forecasts. First, at the global scale, demand for ruminant meat and dairy is set to increase by 44% between now and 2050.3 The biggest change will occur in the developing world, where demand for both is set to soar. Back in the 1990s, the developed world consumed more meat and dairy than the developing world; by 2050, that situation will have turned on its head, with almost 168% more meat and 64% more dairy consumed in developing countries (Figure 4).

Figure 4: Historic and forecast demand for (a) ruminant meat consumption and (b) dairy consumption in developed and developing countries. Solid lines are actual data. Dotted lines are FAO forecasts. All data is sourced from FAO.

Second, at the regional level, the greatest relative change in consumption of ruminant meat is likely to be in sub-Saharan Africa. Although the FAO does not break down its ruminant meat forecasts by region, the OECD provides country- and region-specific forecasts through 2026.31 These data show that consumption of beef and sheep meat is set to grow faster in sub-Saharan Africa over the coming decade than anywhere else in the world. Indeed, the region is one of the few where beef demand is still expected to be greater in 2026 than any other animal protein, and to lose less market share compared to poultry and pork.

Taking a longer-term view, it becomes apparent that these trends are only likely to increase further out to 2050. FAO forecasts suggest the growth of meat consumption in sub-Saharan Africa and South Asia is set to dwarf that of other global regions by 2050, increasing by almost 200% and 300%, respectively. In South Asia, the OECD data show that the bulk of this consumption will be in poultry, but unless diets shift dramatically, ruminants will play a significant role in Africa. The message is clear: over the next few decades, demand for ruminant meat in emerging economies, and particularly in sub-Saharan Africa, is likely to grow substantially, with significant implications for pasture.

To really appreciate how the area of land under pasture may change in the future, however, it is necessary to analyze both demand and supply dynamics. Is it possible, in other words, that improvements in production efficiencies might mitigate some of these pressures?

Wirsenius and colleagues investigated these dynamics in 2010, modelling a range of scenarios to explore how the area of pasture would respond to improvements in livestock productivity and diet between 1999 and 2030.32 They found that, while their baseline scenario led to an additional 150 million hectares of pasture brought into production, a more optimistic improvement in livestock efficiency led to a contraction of 350 million hectares (Figure 5). Where pasture did expand, it did so in sub-Saharan Africa and Latin America due to a combination of high demand, low production efficiencies, and greater land availability.

Figure 5: Predicted changes in pasture area under four scenarios of supply and demand change. Dotted lines are not indicative of a trend and are shown for purely illustrative purposes. Data presents the results of Wirsenius et al. (2010).

Of course, in order to understand how reasonable these predictions are it is necessary to take a closer look at their underlying assumptions. Both scenarios are based on FAO demand forecasts. The difference is that while the baseline scenario uses estimates for yield (meat per head of cattle), pasture productivity (yield of forage per hectare), and feed efficiency (how much feed is required for each unit of meat or dairy produced) that were as consistent as possible with FAO predictions, the more optimistic scenario for livestock productivity takes a more ambitious outlook on some of these metrics.

At the global level, the baseline scenario assumes that pasture productivity would improve by 47%, feed efficiency by 20%, and yields by 21% for beef, 7% for dairy, and 42% for mutton, compared to measured baseline data in 1998. In comparison, the improved livestock efficiency scenario assumed a 31% improvement in pasture productivity (notably lower than the baseline, but more aligned with historical improvements), a 40% and 60% improvement in feed efficiency for beef and mutton, respectively, and much more ambitious yield improvements: 61% for beef, 101% for dairy, and 72% for sheep.

A broad comparison of global and regional trends provides some perspective on just how challenging attaining these improvements is likely to be. Chang and colleagues (2015) modelled changes in grassland productivity in Europe over the past 50 years and found a 12% increase in grassland productivity over a comparable period to Wirsenius’s forecasts.33 This is below both the baseline and the improved livestock productivity scenario.

For yields, FAO data suggests that in the 33 years prior to 1998, average global yields improved by 22% for beef, 15% for dairy, and 5% for sheep. These improvements are comparable to the yield gains in the baseline, but well below those of the improved livestock productivity scenario. For global yields to come close to the more optimistic improved livestock scenario, they would need to match those of North America, which saw yields for beef improve by 47% for beef, 112% for dairy, and 32% for sheep over the 33 years prior to 1998.

Finally, with regard to feed use efficiency, Herrero and colleagues (2013) found that the developed regions of Europe, North America, and Oceania required much less feed per unit production than less developed regions due to greater availability of high-quality feed.12 Differences varied greatly by production system, but on average across all production systems, developing nations required 387% more dry matter for each unit of beef than developed nations. This suggests that with improved availability of high-quality feed, feed use efficiencies could be greatly improved in developing nations, and at levels significantly higher than both forecasts modelled by Wirsenius.

Considering the uncertainty surrounding the underlying FAO data and the complexities of modelling global livestock productivity, it would be wise to remain cautious about the absolute numbers predicted in these forecasts. However, the findings highlight two crucial points. First, the current declines in pasture are by no means guaranteed to continue. Second, if the current contraction in the global area of pasture is to be preserved, or even improved upon, the challenge lies in improving pasture productivity and yields in emerging economies in particular—one that will require a step-change in the historic rate of improvement.


Zebu, the predominant cattle in Brazil, at a recently logged ranch on the edge of the Amazon rainforest.

Reducing the Global Footprint of Pasture

What scope is there for realizing these improvements, and what will be required to enable this transition? Despite the scale of the challenge, there are reasons to be optimistic. The regions of the world where improvements in productivity are most urgently required are the same regions where the most opportunity for improvement exists. Researchers have suggested that adoption of more intensive management could potentially double or even triple yields in underdeveloped regions such as sub-Saharan Africa and Asia, and that beef production on grazing land in Brazil stands at only one-third its maximum capacity based on current pasture productivity metrics.14

Ensuring year-round access to highly nutritious feeds is perhaps the single biggest challenge for raising livestock productivity. This is particularly important in the tropics and subtropics. Unlike temperate regions, tropical and subtropical systems experience marked seasonality in rainfall that directly impacts the quality of available forage. In naturally arid grasslands, pastoral herders have adapted to this condition by ranging animals over large distances in response to sporadic availability of forage. These migrations are essential but are highly energy-intensive for the animals, and therefore dramatically reduce weight gain. While such migrations are not required on formerly forested tropical pastures, cattle can still lose a significant amount of their body weight during the dry season, when the availability and nutritional value of forage plummets.12

More intensive management could potentially double or even triple yields in regions such as sub-Saharan Africa and Asia.


To adapt to these pressures, most livestock production systems in developed economies provide supplemental feeding in some form, be it through grains in feedlots in North America, or silage in the open barns of Europe.16 These intensive systems also enable improved animal health as herds are less dispersed, facilitating diagnosis and treatment of medical conditions, and are located in more easily controllable environments, reducing exposure to heat and cold. These systems also dramatically reduce land requirements for production.34,35 However, they are costly to set up and require more advanced technical knowledge. They make most economic sense where land is scarce and costs of expansion are high. Higher operational costs also impact their competitiveness. When commercial feedlots were introduced in Kenya, for example, the beef they produced was expensive for local markets, and ultimately the operations were outcompeted by traditional extensive production.36

Unlike with commercial crops, the possibility of increasing yields on pasture has received little attention.14 Yet intensive pasture management can improve the quality and availability of forage. In seasonally dry climates, fertilization and sowing of more nutritious and productive forages such as improved grasses and legumes can improve pasture productivity markedly. For the benefits to be fully realized, however, adequate irrigation is also essential.37 Where irrigation is not available or not possible, semi-natural habitats have been shown to provide more stable pasture productivity on an annual basis, owing to improved water retention and greater drought resistance.38


A livestock market in Mali. Frequently, pastoralists manage herds to maximize their number for financial capital rather than their meat yields.

The addition of fodder crops and tree species common in the silvo-pastoral systems of Europe has also been shown to be effective in the tropics. Intercropping of shrubs and trees helps conserve water by maintaining humidity, lowering temperatures, and reducing evapotranspiration and heat stress for animals. Shrubs such as leucaena are more drought-resistant than grasses and provide high-quality fodder throughout the dry season, while simultaneously supporting a greater abundance of native vegetation. The high nutritive value of leucaena can also intensify diets such that livestock numbers can be reduced while meat production remains constant.39 These intensively managed semi-natural systems will not compete with the supplemental feed operations of Europe and North America, but they offer compromises in resource-scarce regions of the developing world and have been shown to be effective at improving livestock productivity and species diversity on smallholder farms.14,40,41

Encouragingly, a growing body of literature links intensification with sparing land for nature. Alkemade and colleagues (2013) have shown that improvements in livestock productivity in Africa, more than any other region, could lead to marked improvements in mean species richness, as more land could be brought out of production.42 In their analysis of land transitions in Brazil, Barretto and colleagues (2013) found that adoption of more intensive management strategies on consolidated agricultural land resulted in a contraction of pasture.43

However, it is not the case that intensification alone results in benefits for biodiversity. In Barretto’s study, intensification only resulted in agricultural land contraction in agriculturally consolidated areas. Where intensification occurred on frontier lands at forest margins, intensification was associated with an expansion of pasture. As a result, they concluded that ranchers only reduced acreage in areas where land was scarce. This highlights an important point: land scarcity and higher land prices are critical precursors for intensification. Indeed, the adoption of feedlots in Australia, North America, and Argentina has been attributed in part to higher land prices that incentivized more efficient use of resources.16,44 Land scarcity in the Cerrado in Brazil has also been a key driver of intensification and associated reduction in deforestation.45 What these case studies highlight is that for leakage to be prevented, improved management will need to be accompanied by the establishment and enforcement of an effective land-zoning regime. By restricting further expansion, such measures will in effect create artificial scarcity, incentivizing intensification on current land rather than expansion.

A growing body of literature links intensification with sparing land for nature.


How these transitions occur will be key. Perhaps the single biggest enabler of a move toward sustainable intensification is through engagement with the corporations that control consumption and production behaviors at the macro-level. Encouraging lenders to invest in certified zero-deforestation projects and producers to develop eco-certification standards are powerful tools for creating price incentives in the market for better environmental management.46 Zero-deforestation agreements with slaughterhouses and ranchers have been shown to be effective and integral to the reductions in deforestation achieved in the Amazon.47

Change will also be needed at the local level. Mixed crop-livestock systems typical of many smallholder operations in the tropics provide an estimated 69% of the milk and 61% of the meat from ruminants globally.12 Improving efficiencies in these production systems will be critical for meeting future demand and mitigating expansion of grazing lands. The high-energy crop residues fed to livestock in mixed systems typically achieve notably higher feed efficiencies than comparable grazed livestock. As such, these systems are well positioned to take advantage of improvements in supplemental feeding. Crop breeding programs that select for quality of crop residues as well as of the primary grains have been shown to support improved livestock productivity.14,48

There is good reason to be optimistic that change can be achieved. In their evaluation of mixed production systems in East and West Africa, Henderson and colleagues (2016) found adoption of existing best practice could improve livestock production by between 69% and 155% and milk yields by between 28% and 167%.49 A review by McDermott and colleagues (2010) suggested yields could be improved by 50% for beef, and up to 300% for small ruminants.50 Crucial for realizing these gains is improved animal health and better quality feed rations through crop improvement programs. Better access to markets is also crucial, helping herders to receive better prices and incentivizing investment in inputs.

Transitioning to more efficient production operations will require both structural and institutional reform. The role of livestock in many tropical systems extends beyond simply their productive capability.10,51 Frequently the animals themselves represent financial capital, and pastoralists often manage their herds to maximize their number rather than their meat yields. Where pastoralists have transitioned from herders (maximizing the number of animals) to producers (maximizing meat or dairy production) such as in parts of China, yields have improved, incomes have risen, and animal numbers have decreased, enabling the recovery of previously degraded grasslands.52

The Wrap-Up

Despite recent declines in the developed world, forecasts suggest pasture may be on the verge of a dramatic expansion in emerging economies, with potentially disastrous consequences for biodiversity. If this expansion is to be controlled, it will be essential that livestock production systems in these regions are intensified. The rate of change required likely exceeds historic improvements, but the potential exists if yield gaps can be closed.

How this transition is achieved will require a holistic perspective. At the macro-level, enhanced uptake of resource-efficient production technologies and knowledge as well as improved supply chain management will be essential. At the local level, transitions will need to evolve with and support the evolution of smallholder production systems. Above all, careful planning will be required to ensure that land that is removed or spared from production benefits nature. Ultimately, the intensification of livestock production systems represents a huge opportunity to enhance production, raise incomes, improve food sustainability, and save space for nature. Achieving these objectives will require trade-offs with regard to biodiversity on production landscapes. What is clear, however, is that the status quo is not an option.

References

1. FAOSTAT. FAOSTAT. (2017). Available at: http://www.fao.org/faostat/en/#home.

2. Poore, J. Back to the Wild. New Scientist (2017).

3. Alexandratos. World agriculture: Towards 2030/2050. Prospects for food, nutrition, agriculture and major commodity groups. (2012).

4. Foley, J. A. et al. Global Consequences of Land Use. Science 309, 570–574 (2005).

5. Steinfeld, H. et al. Livestock’s long shadow. (2006).

6. Fearnside, P. M. Deforestation in Brazilian Amazonia: The effect of population and land tenure. Ambio 22, 537–545 (1993).

7. Walker, R., Moran, E. & Anselin, L. Deforestation and Cattle Ranching in the Brazilian Amazon: External Capital and Household Processes. World Development 28, 683–699 (2000).

8. Barona, E., Ramankutty, N., Hyman, G. & Coomes, O. T. The role of pasture and soybean in deforestation of the Brazilian Amazon. Environ. Res. Lett. 5, 024002 (2010).

9. UNEP. One Planet, Many People: Atlas of our Changing Environment. 332 (UNEP/GRID, 2005).

10. Erb, K.-H. et al. Livestock Grazing, the Neglected Land Use. in Social Ecology (eds. Haberl, H., Fischer-Kowalski, M., Krausmann, F. & Winiwarter, V.) 295–313 (Springer International Publishing, 2016). doi:10.1007/978-3-319-33326-7_13

11. Asner, G., Elmore, A., Olander, L., Martin, R. & Harris, T. Grazing Systems, Ecosystem Responses, and Global Change. Annual Review of Environment and Resources 29, 261–299 (2004).

12. Herrero, M. et al. Biomass use, production, feed efficiencies, and greenhouse gas emissions from global livestock systems. PNAS 110, 20888–20893 (2013).

13. Haan, C. de. Prospects for Livestock-Based Livelihoods in Africa’s Drylands. (World Bank Publications, 2016).

14. Searchinger, T. World Resources Report 2013-2015: Creating a Sustainable Food Future. World Resources Institute. (2013).

15. Ramankutty, N. & Foley, J. A. Estimating historical changes in global land cover: Croplands from 1700 to 1992. Global Biogeochem. Cycles 13, 997–1027 (1999).

16. Deblitz, C. & Dhuyvetter, K. Cost of production and competitiveness of beef production in Canada, the US and the EU. (2013).

17. Goldewijk, K., Beusen, A. & Janssen, P. Long-term dynamic modeling of global population and built-up area in a spatially explicit way: HYDE 3.1. The Holocene 20, 565–573 (2010).

18. Goldewijk, K., Beusen, A., Doelman, J. & Stehfest, E. New anthropogenic land use estimates for the Holocene; HYDE 3.2. Earth System Science Data Discussions 1–40 (2016). doi:10.5194/essd-2016-58

19. Achard, F. et al. Determination of Deforestation Rates of the World’s Humid Tropical Forests. Science 297, 999–1002 (2002).

20. Wilson, E. O. Biodiversity. (National Academy of Sciences, 1988). doi:10.17226/989

21. Lawton, R. O., Nair, U. S., Pielke, R. A. & Welch, R. M. Climatic Impact of Tropical Lowland Deforestation on Nearby Montane Cloud Forests. Science 294, 584–587 (2001).

22. Gibson, L. et al. Primary forests are irreplaceable for sustaining tropical biodiversity. Nature 478, 378–381 (2011).

23. Chaudhary, A., Pfister, S. & Hellweg, S. Spatially Explicit Analysis of Biodiversity Loss Due to Global Agriculture, Pasture and Forest Land Use from a Producer and Consumer Perspective. Environ. Sci. Technol. 50, 3928–3936 (2016).

24. Eisler, M. C. et al. Agriculture: Steps to sustainable livestock. Nature News 507, 32 (2014).

25. Reid, E. Livestock and wildlife in pastoral systems of East Africa: Inevitable conflict or unexpected synergy?

26. Ogutu, J. O., Owen-Smith, N., Piepho, H.-P. & Said, M. Y. Continuing wildlife population declines and range contraction in the Mara region of Kenya during 1977–2009. Journal of Zoology 285, 99–109 (2011).

27. Okello, M. M. Land Use Changes and Human-Wildlife Conflicts in the Amboseli Area, Kenya. Human Dimensions of Wildlife 10, 19–28 (2005).

28. Gong Li, S. et al. Grassland desertification by grazing and the resulting micrometeorological changes in Inner Mongolia. Agricultural and Forest Meteorology 102, 125–137 (2000).

29. Manzano, M., Navar, J., Pando-Moreno, M. & Martinez, A. Overgrazing and Desertification in Northern Mexico: Highlights on Northeastern Region. Annals of Arid Zone 39, 285–304 (2000).

30. Zhi-qing, C. & Zhen-da, Z. Development of land desertification in Bashang area in the past 20 years. J. Geogr. Sci. 11, 433–437 (2001).

31. OECD. Meat consumption (indicator). doi: 10.1787/fa290fd0-en (Accessed on 26 September 2017). (2017). Available at: http://data.oecd.org/agroutput/meat-consumption.htm. (Accessed: 28th September 2017)

32. Wirsenius, S., Azar, C. & Berndes, G. How much land is needed for global food production under scenarios of dietary changes and livestock productivity increases in 2030? Agricultural Systems 103, 621–638 (2010).

33. Chang, J. et al. Modeled Changes in Potential Grassland Productivity and in Grass-Fed Ruminant Livestock Density in Europe over 1961–2010. PLoS One 10, (2015).

34. Nijdam, D., Rood, T. & Westhoek, H. The price of protein: Review of land use and carbon footprints from life cycle assessments of animal food products and their substitutes. Food Policy 37, 760–770 (2012).

35. Swain, M., Blomqvist, L., McNamara, J. & Ripple, W. J. Reducing the environmental impact of global diets. Science of The Total Environment 610, 1207–1209 (2017).

36. Kahi, A., Wasike, C. & Thomas, R. Beef production in the arid and semi-arid lands of Kenya: Constraints and prospects for research and development. Outlook on agriculture 35, 217–255 (2006).

37. European Commission. Profitability of permanent grassland—How to manage permanent grassland in a way that combines profitability, carbon sequestration and biodiversity. (2014).

38. Ospina, S., Rusch, G. M., Pezo, D., Casanoves, F. & Sinclair, F. L. More Stable Productivity of Semi Natural Grasslands than Sown Pastures in a Seasonally Dry Climate. PLoS One 7, e35555 (2012).

39. Thornton, P. K. & Herrero, M. Potential for reduced methane and carbon dioxide emissions from livestock and pasture management in the tropics. PNAS 107, 19667–19672 (2010).

40. Franzel, S., Carsan, S., Lukuyu, B., Sinja, J. & Wambugu, C. Fodder trees for improving livestock productivity and smallholder livelihoods in Africa. Current Opinion in Environmental Sustainability 6, 98–103 (2014).

41. Murgueitio, E., Calle, Z., Uribe, F., Calle, A. & Solorio, B. Native trees and shrubs for the productive rehabilitation of tropical cattle ranching lands. Forest Ecology and Management 261, 1654–1663 (2011).

42. Alkemade, R., Reid, R. S., Berg, M. van den, Leeuw, J. de & Jeuken, M. Assessing the impacts of livestock production on biodiversity in rangeland ecosystems. PNAS 110, 20900–20905 (2013).

43. Barretto, A. G. O. P., Berndes, G., Sparovek, G. & Wirsenius, S. Agricultural intensification in Brazil and its effects on land-use patterns: An analysis of the 1975-2006 period. Global Change Biology 19, 1804–1815 (2013).

44. AusVet Animal Health Services. A review of the structure and dynamics of the Australian beef cattle industry. 66 (Australian Department of Agriculture, Fisheries and Forestry, 2006).

45. Spera, S. Agricultural Intensification Can Preserve the Brazilian Cerrado: Applying Lessons From Mato Grosso and Goiás to Brazil’s Last Agricultural Frontier. Tropical Conservation Science 10, 1940082917720662 (2017).

46. Butler, R. A. & Laurance, W. F. New strategies for conserving tropical forests. Trends in Ecology & Evolution 23, 469–472 (2008).

47. Gibbs, H. K. et al. Did Ranchers and Slaughterhouses Respond to Zero-Deforestation Agreements in the Brazilian Amazon? Conservation Letters 9, 32–42 (2016).

48. Herrero, M. et al. Smart investments in sustainable food production: revisiting mixed crop-livestock systems. Science 327, 822–825 (2010).

49. Henderson, B. et al. Closing system-wide yield gaps to increase food production and mitigate GHGs among mixed crop–livestock smallholders in sub-Saharan Africa. Agricultural Systems 143, 106–113 (2016).

50. McDermott, J. J., Staal, S. J., Freeman, H. A., Herrero, M. & Van de Steeg, J. A. Sustaining intensification of smallholder livestock systems in the tropics. Livestock Science 130, 95–109 (2010).

51. Homewood, K. & Rodgers, W. Pastoralism, conservation and the overgrazing controversy. in Conservation in Africa: Peoples, Policies and Practice (Cambridge University Press, 1989).

52. Kemp, D. R. et al. Innovative grassland management systems for environmental and livelihood benefits. Proc Natl Acad Sci USA 110, 8369–8374 (2013).