Farmland Restoration
There is an enormous amount of land that was farmed until it stopped answering, and then abandoned. It is neither wild nor useful. It is the most uncontested land on Earth.
The effect arrives across lifetimes. This is a gift to people you will not meet.
Origins
Agriculture leaves ruins, and unlike the ruins of cities nobody comes to look at them.
The Fertile Crescent, where farming began, is substantially desert and salt pan now. Irrigation without drainage brought salt to the surface over centuries, and the fields of Sumer went white and stopped growing barley, and the civilisation moved on. The Mediterranean was forested before Greece and Rome cut it, grazed it, and let the goats finish the seedlings; what is left is maquis and bare limestone, and most people assume that is what it always looked like.
The pattern repeats: break the land, farm it until it gives out, walk away. It has been the exhaust of agriculture for ten thousand years, and the abandoned ground it leaves behind is one of the largest land categories on Earth that nobody has any particular plans for.
The great modern example is the American Dust Bowl, and then the American recovery, and the great modern counter-example is the Loess Plateau in China — a landscape so eroded it turned an entire river yellow and gave it its name, and which, beginning in the 1990s, was deliberately restored across an area the size of a small country. Terraces, tree planting, grazing bans, and a generation of patience. It is the largest ecological restoration ever attempted by human beings and it substantially worked.
Which tells you the ruins are not permanent. They are just waiting.
What it actually is
Abandoned farmland is a genuinely unusual category, and its virtue is that almost nobody is arguing about it.
Every other land-based climate solution runs into a fight: plant trees and you may destroy grassland; grow biomass and you may displace food; protect forest and you may exclude the people living in it. Abandoned, degraded farmland has none of those problems. It is not producing food. It is not intact habitat. It is frequently still eroding. It is, in the most literal sense, spare.
Restoring it can go in several directions, and the right one depends entirely on what was there before. Where it was forest, letting forest return — usually by natural regeneration if a seed source survives, which is cheaper and better than planting. Where it was grassland, restoring the grassland, which is habitat and stores its carbon underground. Where the soil can be rebuilt, bringing it back into agriculture using the practices on the conservation agriculture and regenerative pages.
The carbon comes from soil rebuilding and, where appropriate, from returning vegetation. It is slow — soil organic matter accumulates over decades — which is why this is a generations solution. But the land is not doing anything else, and it is not going anywhere.
The one thing to be careful about: abandoned does not always mean empty. A great deal of land classed as abandoned in a database is being grazed, gleaned or gathered from by people who do not hold formal title. Restoration projects have a long history of discovering this after the fences go up.
The numbers
The scale. Estimates of abandoned and degraded agricultural land run to hundreds of millions of hectares globally. It is one of the largest land categories on Earth with no current productive use.
The carbon. Restoring degraded farmland rebuilds soil organic matter and, where appropriate, above-ground biomass. The gains accumulate over decades and are meaningful precisely because the area is so large.
The co-benefits. Erosion halted, water infiltration restored, habitat returned, and in many cases the land brought back into some productive use. Very few solutions deliver on this many axes at once.
The Loess Plateau is the proof at scale: an area of catastrophic erosion, restored through terracing, planting and grazing management, with measurable recovery of vegetation, soil and rural incomes. It is the largest deliberate landscape restoration in history.
New England is the proof that it can happen for free: farmland abandoned in the nineteenth century regrew into forest without anybody planting it, taking the region from roughly 30% forested to around 80%.
The honest caution. “Abandoned” is a category that has been used to justify land grabs. A great deal of it is being used informally by people without title. Any restoration programme that has not established who is actually on the land is a programme that will discover the answer the hard way.
Why it matters
There is a particular kind of sadness in a field that has given up.
You can see them everywhere once you know how to look. The old terraces on a Mediterranean hillside, still visible under the scrub, built by hand by somebody’s ancestors. The stone walls running through the New England woods. The gullies cutting through an African hillside where the topsoil used to be. Somebody worked that land. Somebody believed in it. And then it stopped answering, and they left, and nobody has been back.
Most of the climate conversation is about restraint — use less, emit less, take less. This is one of the few that is purely about repair. Nobody has to give anything up. There is no trade-off to negotiate and nobody to fight, because the land is already lost and everybody has already left.
And the land, given half a chance, does most of the work itself. That is the astonishing thing about the New England forest: nobody planted it. The farmers walked away and the trees came back and now there are bears in Massachusetts. Nature is not fragile. It is extraordinarily persistent, and it will take back what it can as soon as we stop actively preventing it.
Somebody broke that ground. It does not have to have been for nothing. You can put it right, and the ones who come after you will walk in the woods that grew there and never know it was ever a field, and that is the best possible outcome.
What it actually takes
Establishing who is actually on the land. First, before anything. The word “abandoned” in a land database has been the beginning of a great many injustices. Informal grazing, gathering and cultivation by people without title is extremely common on land officially classed as unused.
Matching the restoration to what was there. Forest where forest was. Grassland where grassland was. Planting trees on former grassland because trees are fundable is the same error the afforestation page describes, and it is committed constantly.
Letting it regenerate where it can. Where a seed source survives, natural regeneration is cheaper, faster and produces better outcomes than planting. It is also unfundable, because it does not look like a project. That is a problem with our funding models, not with the method.
Patience, and therefore money that does not need a result this year. Soil rebuilds over decades. This is a generations solution and it is structurally mismatched with three-year grant cycles.
A reason for it to stay restored. Land that is restored and then has no local owner or beneficiary will be cleared again. The Loess Plateau worked partly because rural incomes rose.
Where it matters most
The Loess Plateau is the great proof of concept, and the largest deliberate landscape restoration ever attempted.
The Mediterranean and Iberian interior are the oldest ruins — land farmed and grazed for millennia, now depopulating fast as young people leave. There is an enormous rewilding opportunity in the emptying of rural Spain, and a genuine social tragedy alongside it.
The Atlantic Forest of Brazil is where degraded cattle pasture is being restored to forest at the largest scale currently attempted anywhere.
The Sahel and the East African highlands are where restoration and food security are the same project, and where farmer-managed natural regeneration has quietly greened millions of hectares.
The Appalachians and New England are the accidental success: the largest reforestation in modern history, achieved by farmers walking away and nobody stopping the trees.
How to tell it’s being done well
Who is actually using this land? The first and most important question. “Abandoned” is frequently wrong and it has justified a great deal of dispossession.
What was here before? Restore that. Planting forest on former grassland is a common and damaging error.
Could it just be left to regenerate? Often yes, and it will be better and cheaper. Ask why they are planting.
Is anyone measuring soil, at depth, over years? Soil organic matter is the point, and it is slow, and surface measurements will mislead.
Who benefits in twenty years? If nobody local does, it will be cleared again.
What you can do
Anyone
- Look for the ruins. Old terraces under the scrub, stone walls in the woods, gullies where topsoil used to be. Once you can see abandoned farmland you see it everywhere.
- Support natural regeneration projects. They are cheaper and more effective than planting and they receive a fraction of the funding, because they do not photograph well.
Landowners
- If you have ground that stopped producing, it may not need much. Fence out the grazing, let the seed come in, and wait.
- Match the restoration to what was there. Forest where forest was, grassland where grassland was.
Policymakers
- Establish tenure before restoration. Abandoned land is frequently being used by people without title, and restoration has a long history of dispossessing them.
- Fund natural regeneration, not just planting. It works better and it is unfundable under most current criteria, which is a problem with the criteria.
- Fund on decadal timescales. Soil rebuilds over decades and grant cycles run three years.
Business and investors
- Degraded farmland is the least contested land on Earth for restoration, carbon and biodiversity projects. Almost nobody is fighting about it.
- Verify that it is genuinely unused. This is where the reputational risk lives.
Who is working on this
We are researching which organizations in our directory of 8,493 actively work on this solution, and we only list an organization once we have verified it. That research is ongoing. In the meantime, search the directory yourself:
Questions
How much abandoned farmland is there?
Estimates run to hundreds of millions of hectares globally. It is one of the largest land categories on Earth with no current productive use, and it is the least contested land available for restoration precisely because it is neither producing food nor functioning as intact habitat.
Why is abandoned farmland such a good target?
Because almost nobody is arguing about it. Every other land-based solution runs into a conflict: plant trees and you may destroy grassland, grow biomass and you may displace food, protect forest and you may exclude the people in it. Degraded farmland has none of those problems. It is, quite literally, spare.
Should it be replanted or left alone?
Where a seed source survives, letting it regenerate naturally is usually cheaper, faster and produces a better result than planting. The largest reforestation in modern history, the return of the New England forest, involved planting essentially nothing. The farmers left and nobody stopped the trees.
What is the biggest risk?
That the land is not actually abandoned. A great deal of ground classed as unused in a database is being grazed, gleaned or farmed informally by people without title, and the word abandoned has justified a great deal of dispossession. Establishing who is on the land is the first step, not an afterthought.
Can degraded land really be restored?
Yes, and at extraordinary scale. The Loess Plateau in China, an area so eroded it gave the Yellow River its colour, was deliberately restored through terracing, planting and grazing management. It is the largest ecological restoration ever attempted and it substantially worked.
Why is it classed as slow?
Because soil organic matter rebuilds over decades and returning vegetation takes as long as vegetation takes. It is a generations solution: the land is not going anywhere, and the people who benefit most from starting now will be the ones who come after.
Sources
- Project Drawdown - Restore Abandoned Farmland (Drawdown Explorer) Framework and classification. Cited, not reproduced.
- IPCC (2019) - Special Report on Climate Change and Land
- World Bank - Loess Plateau Watershed Rehabilitation Project
- Griscom et al. (2017), PNAS - Natural climate solutions
- Harvard Forest - the New England reforestation record
The solution taxonomy follows the framework popularised by Project Drawdown. The analysis above is our own; for their carbon modeling and rankings, visit them directly.