3: What Does an Acre Produce? Farming, ethanol and the economics of living landscapes

Thorsten Arnold is an environmental scientist and farmer in Grey County, Ontario, working with the National Farmers Union on its ethanol campaign.

Stopping agricultural expansion is necessary. It is not sufficient. Agriculture also needs enough ecological function distributed through the land that stays in production — and a country that has already learned this once, the expensive way, should not have to learn it again.

Three ledgers, not two

This series has argued that land carries two different way to look at opportunity costs. Michael Grunwald reminds us to count the carbon that a forest or wetland would have stored [1] – that is the social GHG ledger. The farmer counts the mortgage payment a hedgerow does not earn – that is the farm business ledger. The first misses from the farm ledger; the second from the climate models and the land sparing debate. My previous essay showed how three farm-economic treadmills — technology, land debt and machinery — keep intensifying the second while the first stays invisible.

There is a third accounting problem. Land produces more than either carbon or commodities. Land moves and stores water. It holds soil. It cycles nutrients. It shelters crops from wind. It traps snow that otherwise blow away, endangering winter drivers. It carries pollinators and the predators of crop pests. It provides habitat and the corridors between habitats. Neither the climate ledger nor the farm ledger records any of these broader ecosystem functions and services. And an agricultural system that measures only one output of land will systematically underproduce all the others.

Where Grunwald’s metric fails

Grunwald is at his harshest on what he calls “the soil fantasy”: regenerative and organic farming sold as climate solutions [1]. On one point he is right, and I strongly believe that farmers and ranchers should concede it: Soil carbon makes a weak climate offset. It saturates, it can be lost again with one tillage pass or one drought, and it is hard to measure. Even if on our own farm we are astonished how much we can build soil, many industry-scale claims are exaggerated.  Ecological farming should not be defended mainly on carbon sequestration claims as it does so much more.

But Grunwald’s negative verdict on agroecology rests on a single metric and within his land sparing viewpoint. If an ecological practice lowers yield, he argues, the lost production must be replaced somewhere else by clearing land, so the land’s carbon opportunity cost can exceed any benefit. His logic assumes the marginal bushel is for food that someone needs. In North American commodity agriculture, the marginal bushel is more likely to end up in a fuel tank. A yield penalty on an Ontario field shrinks a surplus that policy is currently paying to burn. Remove the ethanol mandate first, he argues, and the land it releases dwarfs any regenerative yield penalty.

His metric is not wrong. What I criticize is how he applies it within a boundary that is far too narrow. So narrow that it can pervert and undermine the larger system.

Grunwald’s camp will answer that lower yields in Ontario from regenerative farming would  simply move production to Brazil, with negligable benefits for Ontario. His leakage argument holds only if the lost bushel was feeding someone. About a third of Ontario’s grain corn already goes to ethanol [12]. While that is true, a smaller Ontario crop combined with a mandate moratorium would first shrink what we burn, not what anyone eats. Keep fuel demand constant and his leakage argument has merit.

This is exactly why ecological function has to be paid for as part of a package that also manages supply, the subject of the final essay.

Land sparing is not enough

Grunwald’s framework tends toward a spatial division: farm intensively on less land, so that more land can remain or return to nature elsewhere. Protecting large intact ecosystems has real merit, and expansion into forests, wetlands and native grasslands must stop. But agriculture cannot outsource all ecological function to wilderness somewhere else. This is simply not how landscapes work.

As a watershed scientist, I see this first in water. Water does not recognize the boundary between “agriculture” and “nature.” Rain falling on a cornfield becomes evapotranspiration or infiltration or recharge or soil storage or runoff, depending on the soil and vegetation it meets on the way. A wetland or pond in a depression is not only an acre failing to grow corn, it is also storing water on the landscape and releasing it slowly. A riparian corridor is not just lost productive width. It also holds flowing soil from being lost to erosion. Trees on an exposed field are not just obstacles to machinery.They prevent wind erosion and snow drifts and provide essential corridors for wildlife.

Grunwald exposes a boundary error in bioenergy accounting, when he points out that indirect land use change creates real emissions that are not counted. Yet, Grunwald commits a similar boundary error when accounting for farmland’s ecosystem services: he only counts carbon storage and productive yield but omits the landscape’s vital  functions for water and soil health, biodiversity and climate regulation. In his logic, land that produces no commodity is counted as producing nothing.

The features that he misses are the infrastructure that keeps the productive landscape functioning:

Only some of these benefits return to the farm itself: sheltered crops yield more than exposed ones [2][3], and field edges, wet spots and poor soils often lose money when cropped [4]. The rest of costs and benefits appear downstream and impact neighbours and towns and lakes – society at large.

At the global scale, neglecting these functions has two names: soil degradation and desertification. About a third of the world’s soils are moderately to highly degraded by erosion, compaction, salinization, acidification and pollution [16]; the UN’s Global Land Outlook puts degraded land at up to 40% of the planet’s land area [17]. Degraded fields yield less and cost more, until some are abandoned — and the demand they served continues, incentivizing farmers to convert new wilderness. Between 2003 and 2019, total global cropland grew by about 102 million hectares, half of it on former natural vegetation and forest. At the same time, some 115 million hectares of existing cropland were abandoned or converted to other uses [18] – an area that adds to agriculture’s wilderness demand on top of the net expansion. Degradation is not the only reason land is abandoned, but the pattern is clear: agriculture advances like a roller, flattening wilderness at its front edge and leaving exhausted land behind. Intensification that neglects the landscape’s ecosystem function accelerates this dynamic. Yield bought by mining soil and water and habitat is borrowed from the next decade. Soil-degrading high-yield agriculture becomes another treadmill — one that runs on land.

So Grunwald’s question — how can humanity grow enough food on less land so that nature survives elsewhere? — needs a second one beside it: how much ecological function must remain throughout farmland so that farms, watersheds and communities keep functioning? The two questions are complementary. Protect large intact ecosystems. Stop agricultural expansion. Raise productivity where it genuinely reduces land demand. And keep wetlands, trees, riparian corridors, perennial cover and windbreaks inside the agricultural matrix so we don’t loose this land.

Ontario has done this before

Ontario learned this lesson the expensive way, an entire generation before the prairies experienced its Dust Bowl. Clearing and cultivating drought-prone sandy soils in the late 1800s and early 1900s produced extensive water and wind erosion and sand blowouts — notably in Norfolk County [5], on the Oak Ridges Moraine in Durham and Northumberland [6], on the sand plains of Simcoe County [7], and on the Sauble and Paisley sands of Bruce County [8]. Simcoe farmers were already noticing soil loss and drying streams in the 1870s [7].

The province bought degraded farmland and in 1908 opened its first forestry station at St. Williams, to grow seedlings and demonstrate reclamation through reforestation [5]. Edmund Zavitz’s report on the waste lands of southern Ontario launched the public reforestation programs that followed [9]. Later floods and droughts confirmed that deforestation, soil loss and water could not be managed field by field. Ontario’s farmers not only planted trees but also raised landscape health into the political arena; in 1919 they elected one of their own, Ernest Drury of the United Farmers of Ontario, as premier, and his government passed the Reforestation Act of 1921. After Hurricane Hazel exposed Ontario’s flooding risks further, the province institutionalized watershed management in the Conservation Authorities Act, 1946.

History shows what happens when ecological infrastructure is stripped from the agricultural matrix:

• soil erodes and wind strips topsoil;
• water leaves too quickly, flood peaks rise, and summer baseflows decline;
• nutrients leave fields and organic matter declines;
• crop resilience deteriorates.

Without an explicit strategy to protect Ontario commons, individual farmers are left to compensate with more drainage and fertilizer and irrigation and pesticides and machinery — each a new fixed cost, each accellerating the three treadmills, each externalizing costs by taking more from Ontario’s commons.

This biophysical feedback is not really a treadmill, yet it interacts with the three economic ones:

landscape simplification → erosion and hydrological degradation → declining resilience and productivity → more inputs and capital → more pressure for production and acreage → further simplification.

The treadmills strip the landscape of its ecosystem functions; the stripped landscape turns the treadmills faster. Degradation is itself a driver of agricultural expansion.

The farmers who replanted

I believe that we pay a disservice to our farmer ancestors if we don’t celebrate their role in regenerating Ontario. They deserve celebration. The recovery of Ontario’s desertifying farmland was farmers’ work. Let’s remember that and never forget.

Norfolk landowners took free seedlings from the St. Williams station and planted windbreaks and woodlots on blow sand — more than seven million seedlings across the county in the 1920s alone. Their windbreaks still divide fields into long, narrow strips. Norfolk today has about 27% forest cover, the highest in southwestern Ontario, and was named Forest Capital of Canada in 2008 [10]. Municipalities joined in: Simcoe County signed Ontario’s first Agreement Forest in 1920 and planted the Hendrie Tract in 1922; by 1982 the program covered 107,000 hectares under 59 agreements [7]. On the Prairies, after the droughts of the 1930s, farmers working with the Prairie Farm Rehabilitation Administration planted more than 2,000 km of shelterbelts; the Indian Head nursery distributed more than 570 million tree and shrub seedlings between 1901 and 2001 [11].

Farmers planted trees because they watched their soil leave, and because public institutions put trees, knowledge and a shared purpose within reach. They were not paid for carbon. They were just given the means to keep their land working – and they understood because they experienced degradation’s cost first hand.

There is an irony here. Some of the ecological infrastructure these farmers rebuilt now carries an opportunity cost again — and the land-cover record shows it being spent. Deforestation in southern Ontario rose from 1,382 hectares in 2008 to 3,674 hectares in 2018, driven largely by agriculture [13]. More than a quarter of the pasture and hay land in the Mixedwood Plains of Ontario and Quebec went to annual crops, mostly corn and soy, between 2011 and 2022 [14]. Southern Ontario lost about 18,000 hectares of wetland between 2000 and 2020, mostly small wetlands under half a hectare [15]: exactly the wet corners the treadmills make expensive to keep. How much of that loss was once-restored land has not been measured, and should be.

Necessary, not sufficient

Stopping agricultural expansion is necessary. This point of Grunwald is not sufficient though – we have to ensure that current farmland stays viable. A landscape spared at its edges but stripped at its centre will degrade from the inside. It will erode, flood, dry out and demand more land to make up for what it lost. Agriculture needs enough ecological function distributed through the land that stays in production.

Ontario’s farmers rebuilt that function once, when public institutions made it possible. The treadmills are now eroding it again, and ethanol is turning them faster. The question for the final essay is what institutions would let farmers rebuild it — and keep it.

Next in the series: “Instead of Burning Corn, Pay Farmers for Ecological Function.”

References

1. Grunwald, M. (2025). We Are Eating the Earth: The Race to Fix Our Food System and Save Our Climate. Simon & Schuster.
2. Kort, J. (1988). Benefits of windbreaks to field and forage crops. Agriculture, Ecosystems & Environment 22–23: 165–190.
3. Brandle, J. R., Hodges, L. & Zhou, X. H. (2004). Windbreaks in North American agricultural systems. Agroforestry Systems 61: 65–78.
4. Brandes, E. et al. (2016). Subfield profitability analysis reveals an economic case for cropland diversification. Environmental Research Letters 11: 014009.
5. Government of Ontario. St. Williams Conservation Reserve Management Plan.
6. Parker, W. C. et al. (2008). Restoring southern Ontario forests by managing succession in conifer plantations. The Forestry Chronicle. (Journal details to be checked.)
7. County of Simcoe. Forest History.
8. Saugeen Field Naturalists. Tree Planting in Southern Ontario: A Brief History.
9. Zavitz, E. J. (1909). Report on the Reforestation of Waste Lands in Southern Ontario. Ontario Department of Agriculture. (Title and year to be checked against the original.)
10. Sustainable Heritage Case Studies (2019). Heritage of Forestry in Norfolk County.
11. Encyclopedia of Saskatchewan. Prairie Farm Rehabilitation Administration (PFRA) Shelterbelt Centre.
12. Grain Farmers of Ontario (2022). Ontario corn ethanol. Industry source.
13. State of Ontario’s Biodiversity. Afforestation and Deforestation. Citing State of Ontario’s Natural Resources – Forests 2021.
14. Fyson, V. et al. (2024). The changing agricultural landscape in Canada’s Mixedwood Plains Ecozone 2011–2022 and the implications for biodiversity. FACETS.
15. State of Ontario’s Biodiversity. Extent of Wetland Cover and Wetland Loss. 6,152 ha (2000–2011), 7,303 ha (2011–2015), 4,562 ha (2015–2020).