Roughly 10 to 11 percent of Earth’s total land area qualifies as arable, meaning it can be plowed and used to grow crops. Since land itself covers only about 29 percent of the planet’s surface, arable land amounts to just around 3 percent of Earth’s total surface area, or approximately 1.4 billion hectares. That slim fraction feeds more than eight billion people, and the pressures on it are shifting in ways that make the raw percentage only the beginning of the story.
What Makes Land “Arable” in the First Place
Arable land is not simply any patch of dirt where a seed could sprout. The classification depends on a mix of climate, soil, and terrain factors that determine whether a piece of ground can reliably produce crops at reasonable cost. Temperature and rainfall set the broadest limits. Soil properties like depth, texture, stoniness, and chemical composition narrow the picture further. Then there are site-level constraints: slope, flood risk, and whether the soil is prone to waterlogging or drought. All of these interact, so land that scores well on rainfall might still fail if the soil erodes too easily or the slope is too steep for machinery.
National land classification systems formalize these judgments. In Wales, for instance, the Agricultural Land Classification system grades parcels from best to worst based on those combined biophysical limits, which together determine the range of crops that can be grown, the consistency of yields, and the cost of farming the land over time.1Science of The Total Environment. Changes in land capability for agriculture under climate change in Wales Global-scale assessments use similar logic but apply it across the entire planet, evaluating suitability for major food and energy crops based on climatic, soil, and topographic conditions at fine resolution.2PubMed Central. Global agricultural land resources–a high resolution suitability evaluation and its perspectives until 2100 under climate change conditions The upshot is that “arable” is not a binary label stamped on a map once and forgotten. It is a graded assessment that shifts as climate, soil health, and technology change.
Where Cropland Is Expanding and How Fast
Despite a common impression that the world is steadily running out of farmland, the total area of cropland has actually been growing. Between roughly 2003 and the early 2020s, global cropland area expanded by about 102 million hectares, an increase of around 9 percent over the 2003 baseline. The rate of that expansion nearly doubled during the period, jumping from about 5 million hectares per year to 9 million hectares per year.3Nature Food. Global maps of cropland extent and change show accelerated cropland expansion in the twenty-first century
The growth has not been spread evenly. Africa accounted for the largest absolute gain, adding roughly 53 million hectares, a 34 percent jump. South America had an even higher relative increase, expanding by about 37 million hectares, which represented a 49 percent gain on its earlier cropland base.3Nature Food. Global maps of cropland extent and change show accelerated cropland expansion in the twenty-first century Much of this expansion comes from converting forests, grasslands, and savannas, which raises its own set of environmental trade-offs. The point for understanding the headline percentage is that the numerator is a moving target: arable land is being created even as it is being lost elsewhere.
What Is Eating Away at Existing Arable Land
The forces that pull land out of productive agriculture are several, and they often overlap in the same places.
Soil Erosion
Erosion by wind and water is the single most widespread form of land degradation on arable ground. Globally, erosion alone affects roughly 20 percent of arable systems. When you add in areas where erosion overlaps with other degradation processes, an additional 7 percent of the world’s arable land faces compounded damage.4Environmental Research. Arable lands under the pressure of multiple land degradation processes. A global perspective That means more than a quarter of all farmed land is dealing with some form of erosion-related stress. Erosion strips away topsoil, the biologically rich layer where most root activity and nutrient cycling happens, and rebuilding it takes decades to centuries.
Urbanization
Cities tend to be built where the farming is good, which is an unfortunate coincidence. River valleys, coastal plains, and fertile lowlands attracted both farmers and city-builders, so as urban areas sprawl outward they frequently pave over some of the most productive soil around. Researchers have termed this “urbanization-induced arable land displacement”: as cities expand, farming gets pushed to less suitable areas that may be drier, steeper, or further from infrastructure.5PubMed Central. Global Arable Land Is Shifting Toward the Tropics and Drylands Under Urbanization The net effect is that the overall count of farmed hectares might hold steady or even grow, but the quality and location of those hectares are shifting in ways that matter for productivity.
Salinization
Salt accumulation in the soil is a quieter threat but a serious one, particularly in arid and semi-arid regions that depend on irrigation. In China’s Altay region, research tracking farmland salinization over two decades found that the key drivers shifted over time. In 2000, natural factors like soil moisture were the dominant influence on salinity, but by 2022, human activities, especially irrigation practices, had become a much more prominent driver. Irrigation water use and proximity to rivers and lakes emerged as the leading factors shaping salt buildup in the soil.6Ecological Indicators. Monitoring the salinization of agricultural land and assessing its drivers in the Altay region This pattern plays out in irrigation-dependent regions worldwide: the very water that makes dry land farmable can, over time, make it unfarmable.
Climate Change Is Redrawing the Map
Warming temperatures do not simply shrink the global pool of arable land. They rearrange it geographically, opening some areas while degrading others. Northern latitudes, particularly the boreal regions of Canada, Scandinavia, and Russia, have historically been too cold for most crop production. But these areas are warming faster than the global average, and agricultural activity is projected to shift northward as a result. At the same time, regions closer to the equator face increasing drought stress, which will push yields down and force farmers to abandon fields or switch to less productive crops.7Nature Climate Change. Climate change will exacerbate land conflict between agriculture and timber production
The redistribution is not a clean swap. Boreal soils are often thin, acidic, and low in nutrients compared to the deep, fertile soils of the mid-latitudes. Permafrost thaw can make land boggy and unstable. And the same warming that opens northern fields to farming also increases wildfire risk in the surrounding forests, which creates land-use conflicts between agriculture and timber production.7Nature Climate Change. Climate change will exacerbate land conflict between agriculture and timber production So while the raw number of potentially cultivable hectares at high latitudes may increase, the actual gain in reliable, high-quality arable land will be smaller than the maps suggest.
The Per-Person Share Keeps Shrinking
Even when total arable area holds steady or grows slightly, population growth means each person’s share of that land keeps falling. In the mid-twentieth century, there were roughly 0.4 hectares of cropland per person on the planet. Today that figure has dropped below 0.2 hectares, and projections suggest it will continue to decline. One modeling study estimated that over a thirty-year horizon, the world was likely to lose 30 to 60 million hectares of cropland while bringing 100 to 200 million hectares of reserve land into use, with about half of that reserve land remaining in production after three decades.8Global Environmental Change. Population growth and loss of arable land
The practical result is that the world must produce more food per hectare every decade just to maintain the same level of per-person food availability. That pressure is what drives both the expansion of farmland into forests and grasslands and the intensification of farming on existing land, each of which carries its own environmental costs. The percentage of Earth’s surface that is arable is a useful number, but the per-capita trajectory may be more meaningful for understanding food security over the coming generation.
Can Technology Expand or Replace Arable Land
Several technological approaches aim to either expand the pool of arable land or reduce the amount of it we need.
Desert Reclamation
Efforts to turn desert into farmland have been underway for decades, particularly in the Middle East, North Africa, and parts of China. Techniques range from large-scale irrigation projects to soil amendments that improve water retention in sandy ground. Research on “regreening” deserts has shown that science and technology can convert previously barren landscapes into productive agricultural zones, though the energy, water, and financial costs remain substantial.9Journal of Advances in Biology & Biotechnology. Regreening the Desert: A Step towards Global Food Security Desert reclamation adds to the global arable total, but it tends to produce lower-quality farmland that requires constant inputs to remain productive.
Crop Breeding and Yield Intensification
The Green Revolution’s core premise was that higher yields per hectare could reduce the need to plow new land. That logic has partly held up. Analyses of the research investments made by international agricultural research centers between 1965 and 2000 suggest that improved crop varieties produced a net land saving of roughly 20 to 30 million hectares in developing countries.10Global Food Security. Does intensification slow crop land expansion or encourage deforestation? That is a real and significant effect, but it is far smaller than the most optimistic earlier claims, which predicted savings of over a billion hectares. The gap reflects how complicated the feedback loops are: when yields rise, food prices fall, which can stimulate demand and even encourage more land conversion rather than less.
Vertical Farming
Growing crops in stacked indoor facilities can cut land demand dramatically. One analysis of vertical farming’s potential in the United Kingdom found it could reduce land requirements by 93 percent compared to conventional field production for the same crops.11PubMed Central. Vertical farming as a land sparing strategy: GHG implications for UK agricultural landscapes The trade-off is energy: vertical farms consume far more electricity per kilogram of produce than outdoor fields, which means their greenhouse gas footprint can be higher unless the energy comes from renewables. The same study found that when spared land was used for solar energy generation, the overall system’s emissions dropped below the field-farming baseline. Vertical farming is currently viable only for a narrow range of high-value leafy greens and herbs, not staple grains, so its impact on the global arable percentage is marginal for now.
Food, Feed, and Fuel All Want the Same Fields
Not all arable land grows food that ends up on someone’s plate. A significant share of the world’s cropland produces animal feed: corn, soybeans, and other grains that pass through livestock before reaching human consumers. On top of that, biofuel mandates in the United States, European Union, Brazil, and elsewhere have directed a growing fraction of cropland toward energy production. Modeling of European agriculture has shown that even advanced cellulosic biofuels, which were designed to avoid competing with food crops, still compete with food and feed production through complex interactions between the crops and land used in animal breeding systems.12Land Use Policy. Competition between food, feed, and (bio)fuel: A supply-side model based assessment at the European scale
The competition matters because it means the effective amount of arable land available for growing human food is smaller than the raw acreage figure implies. How much smaller depends heavily on dietary choices and energy policy. A world that ate less meat and burned less biofuel would, in practical terms, have a larger food-producing land base without a single new hectare being cleared. That distinction between gross arable area and the share dedicated to directly feeding people is one of the most important nuances behind the headline percentage.
When Arable Land Becomes a Geopolitical Asset
As the per-person share of farmland declines and climate change reshuffles which regions can grow what, arable land has become a target for international investment and, sometimes, acquisition. Over the past two decades, foreign investors have acquired approximately 90 million hectares of land for agriculture, a total roughly the size of Nigeria’s entire land mass.13PubMed Central. Impact of transnational land acquisitions on local food security and dietary diversity These deals, often called large-scale land acquisitions, are concentrated in sub-Saharan Africa, Southeast Asia, and parts of South America, precisely the regions where cropland has been expanding fastest.
The stated rationale for these investments is often food security for the investing country or commercial profit from export crops. But the effects on local populations are contested. Preferential targeting of prime agricultural land and a shift toward export-bound crops can reduce local access to nutritious foods, even as the total amount of farmed land in the area increases.13PubMed Central. Impact of transnational land acquisitions on local food security and dietary diversity The deals also frequently spark conflicts over land rights and governance, raising questions about who benefits when arable land changes hands across borders.14PubMed Central. The parable of arable land: Characterizing large scale land acquisitions through network analysis
This geopolitical dimension adds another layer to the question of how much arable land the world has. Ninety million hectares acquired by foreign investors have not disappeared from the map, and they still count toward the global total. But for the communities that previously farmed those hectares, the land might as well have vanished. The percentage of Earth’s surface that is arable tells you something about the planet’s physical capacity. It tells you almost nothing about who has access to that capacity or what gets grown on it.
How Satellite Monitoring Has Changed What We Know
The precision of today’s arable-land estimates is dramatically better than what was available even two decades ago. Medium- and high-resolution satellite imagery, combined with machine-learning classification tools, allows researchers to track cropland at spatial resolutions fine enough to spot individual fields. The global cropland expansion figures discussed earlier rely on exactly this kind of satellite-based mapping, which can distinguish active cropland from fallow ground, pasture, and forest at a level of detail that older land-use surveys could not approach.
One practical consequence is that arable-land statistics from the 1990s and earlier, which relied heavily on national self-reporting and agricultural census data, almost certainly undercounted cropland in some regions and overcounted it in others. Countries had different definitions of “arable,” different survey methods, and different incentives when reporting to international bodies. Satellite-derived estimates have partly resolved these inconsistencies, which is one reason that recent studies sometimes show higher global cropland totals than older datasets. The percentage has not necessarily grown as much as the numbers suggest; some of the apparent increase reflects better measurement rather than more plowing.
Why the Number Is Surprisingly Hard to Pin Down
Ask five different organizations for the percentage of Earth’s surface that is arable and you may get five slightly different answers. Part of the variation is definitional. The Food and Agriculture Organization defines arable land as land under temporary crops, temporary meadows for mowing or pasture, land under market and kitchen gardens, and land temporarily fallow. It excludes permanent crops like orchards and vineyards, which some other classification systems include. It also excludes land that could theoretically be farmed but currently is not, which is what suitability models try to capture.
Another source of variation is temporal. Arable land is not a fixed quantity carved into the planet. A field abandoned after years of erosion stops being arable. A patch of forest cleared and planted with soybeans becomes arable. A rice paddy swallowed by a new suburb ceases to exist as farmland. These changes happen continuously, and the global datasets that track them are updated on different schedules using different methodologies. The commonly cited range of 10 to 11 percent of land area, or about 3 percent of Earth’s total surface, is a reasonable snapshot, but every decimal point in that figure conceals a contested boundary between what counts and what does not.