An open proposal · Draft v0.9 · October 24, 2026

Proof of Soil

A proposal to measure what an economy is worth by the health of the ground it stands on, and to pay the people who make that ground healthier.

OOrganic layer · leaf litter, humus
ATopsoil · where most life lives
ELeached layer
BSubsoil · minerals collect
CParent material

The paper in one minute

  • GDP counts money changing hands. It cannot tell whether the land under an economy is getting richer or poorer. Soil is the clearest signal of that.
  • Soil is the one form of natural wealth people can actively build. Clean water, stored carbon, food and forests all rest on it.
  • Proof of Soil measures soil health from five independent sources at once: satellites, drones, ground sensors with lab samples, analysis that cross-checks them, and the people who work the land. Faking all five costs more than doing the work.
  • When verified soil health improves, new units of value (this paper calls them soil credits) are issued to the steward who did it. Value is created from regeneration instead of from debt.
  • Soil is a commons, so its health belongs on an open public ledger. People who tend it are rewarded. Land whose soil measurably declines, measured across the whole parcel, pays for the damage, and that payment goes straight to the stewards, automatically, on a ledger anyone can watch.
  • Land in the system can be used and enjoyed but not destroyed. The old trio of ownership rights (use, fruits, destroy) becomes use, fruits and care.
  • No single group decides how soil is scored. Scientists, practitioners and stakeholders share that job, and when they disagree, the land's own measurements break the tie.
  • AI makes goods and governance cheaper, and it runs on land. Proof of Soil gives it one rule: build what you need, as long as the land ends up healthier.
  • The first piece already works: SoilScope shows what land does for the people around it, following the Stanford method (InVEST), and puts a yearly dollar figure on each kind of work from public prices. It is most detailed in the United States and rough elsewhere.

An open proposal

This paper describes a concept, not a product. SoilDAO is not building the full Proof of Soil system at this time. The idea is offered openly so that anyone, anywhere, can use it, adapt it and build on it, with credit. The one piece in operation today is SoilScope (Part X), which shows what living land does for people and puts a public price on that work.

Abstract

Every year the world loses about 24 billion tonnes of fertile soil,1 and roughly a third of the world's soils are already degraded.2 The tools built to stop this, from conservation grants to subsidies to carbon markets, mostly ask people to act against their own economic interest for the common good. That has never worked at scale.

This paper proposes a different approach. It ties the creation of value directly to measured improvements in soil health, verified on the ground from several independent sources. People who regenerate land are paid because regeneration creates value, without depending on charity or subsidy. The paper sets out the argument, the measuring system, the way value flows, the property principle that protects the land, and what has already been built. It is offered as an open proposal: a concept anyone may use and build on, with credit.

Part I

Three things we treat as one

Three ideas are bundled so tightly that we treat them as laws of nature. Each is really one way of doing something, and the thing itself is bigger. Pull them apart and better designs become possible.

Money ≠ currency

Money is a technology for coordinating human effort across time and distance. A national currency is one way to build that technology. Today most money is created when banks make loans, as the Bank of England described in 2014,3 so most money in circulation is a record of promises to repay.

That design is useful, and it has a blind spot. Money tracks human labor and debt. It has no way to count ecological labor: the water a wetland filters, the fertility soil life builds, the carbon held in the ground. That work keeps every economy alive, and none of it appears on the books.

Money can be anchored to anything a network agrees is valuable. The agreement lasts only as long as what sits underneath it. Agreements built on stories change when the story changes. An agreement built on the living ground, the infrastructure all land life depends on, rests on something that cannot be argued out of existence.

Money tracks human labor. Nothing on the ledger tracks the work the land does.

Government ≠ governance

Governance is how a group of people makes decisions together. Government is one way to organize it, with authority gathered into central institutions. Centralization made sense: coordinating at scale needed someone to keep the ledger, enforce the rules and hold the keys.

Much of what central authority has done through history concerns land: who holds it, who works it, who taxes it. The enclosure of England's common lands, from the Statute of Merton in 1235 onward,4 is one long example of law used to move shared land into private hands.

Two things are new. Shared digital ledgers let many parties keep one record that no single keeper controls. And sensors can now read the condition of land directly, from orbit and from inside the soil. Together they let decisions about land be informed by the land itself.

Institutions govern what their accounts can see: GDP, tax receipts, property values. What the accounts cannot see, such as soil fertility, watershed health, biodiversity and the vitality of the people who live there, gets filed as an "externality." These things have value. They lack a price, and what has no price is invisible to a system that only reads prices.

Proof of Soil extends the ledger so that living systems show up on it. The land cannot speak, but it sends signals all the time: soil life that thrives or collapses, water that soaks in or runs off. These signals can be measured. Measured honestly, they give the land a voice in decisions about it, and the people who steward the land should be the ones who govern it.

For the first time, the land votes.

Economy ≠ GDP

The word economy comes from the Greek oikonomia: oikos, the household, and nemein, to manage and share out fairly. In its first meaning, an economy is the running of a household so that everyone in it thrives.

GDP is one way of measuring that. Simon Kuznets built the first US national income accounts in 1934 at the request of the Senate. In the same report he warned against reading too much into them:

"The welfare of a nation can scarcely be inferred from a measurement of national income."

Simon Kuznets, report to the US Senate, 19345

After the Second World War, GDP became the world's standard scorecard anyway. It measures transactions. That produces strange results. A clean river adds nothing to GDP. Pollute it and GDP rises: cleanup crews, medical bills, lawsuits. A forest cleared for pasture raises GDP through timber and beef sales, and the soil, water and carbon lost along the way are never subtracted. A farm whose topsoil thins every year can report record output.

What gets measured gets managed. A scorecard that only sees transactions rewards turning living systems into transactions.

What happensWhat GDP recordsWhat a soil ledger records
A river is polluted and cleaned up+ cleanup, medical and legal spending− damage to the watershed and the soils it feeds
A forest is cleared for pasture+ timber and beef sales− soil carbon, water retention and soil life lost
A farm loses topsoil every season+ the harvest it sells− a falling Soil Health Score
A farm raises soil organic matter by 1%0 no transaction, nothing counted+ about 5 more tonnes of carbon stored per acre, and more water held
Figure 1 · The same events on two ledgers

People have tried to widen the scorecard for decades: Bhutan's Gross National Happiness (1972), the UN's Human Development Index (1990), the Genuine Progress Indicator, Kate Raworth's Doughnut.6 Each made an important case. What they still lack is a physical anchor, something measured on the ground parcel by parcel, and a way to reward improvement directly.

Proof of Soil is designed to add both. An economy is growing when its soil is growing.

Part II

Why now: AI

The argument in this paper began before the age of AI. AI makes it more urgent in three ways: it makes governance cheaper, it makes goods cheaper, and it runs on land.

Governance gets cheaper

Any group of more than two people needs governance: a family, a company, a community. That need is permanent. Government, the large central institution, is how we have met it at the scale of a country, because coordinating millions of people needed a center to keep the records and enforce the rules.

Much of that work is administration: keeping records, issuing permits, checking rules, moving payments. AI agents working from open data can already do a growing share of it faster, cheaper and in public view. Meanwhile, trust in government has fallen steeply in many countries. In the United States, only about one in five people say they trust the federal government to do what is right most of the time.7

The direction is clear. As administration becomes software, the important question shifts from who holds the center to how governance is verified. For land, Proof of Soil's answer is to let the land report its own condition in real time, and to pay the people who care for it when it improves.

When goods get cheap, value moves to the ground

Today's economies measure themselves by the goods and services they produce. AI and robotics are pushing the cost of many of those toward zero. For people, that is good news. For a money system built on debt, it is awkward, because debts don't shrink when prices fall. Fiat currencies that have failed have usually failed through inflation. AI raises the opposite pressure.

If almost anything can be produced cheaply, value moves to what cannot be produced: land, water, energy and living soil. Whether AI will be deflationary overall is still debated. Its own appetite for energy, land and materials is pushing those prices up, which only strengthens the point.

AI can produce goods. It cannot produce topsoil.

An economy measured by the health of its ground keeps its meaning however cheap everything else becomes.

AI runs on land

AI is physical. It lives in data centers that need land, power and water, and global data center electricity use is projected to more than double by 2030.8 Communities increasingly resist new sites over water, power and land use, and the largest technology companies have made public pledges, such as becoming water-positive by 2030.9

Proof of Soil offers one simple rule: build what you need, as long as the land ends up healthier than you found it.

Buy 100 acres. Build on 10. Regenerate the other 90.
10 acres · the data center
90 acres · land under regeneration, verified every season
Figure 2 · An illustrative split. The real ratio comes from measuring the whole site.

Large sites already buy buffer land for security and future expansion. Regenerating it costs little next to the building, and the five measurement layers verify it season by season. Waste heat can warm greenhouses. Solar arrays can share ground with grazing. Training AI models doesn't need to happen near users, so, as with Bitcoin mining before it, compute can go where energy is plentiful and where land most needs restoring.

To be clear about what this covers: regenerating 90 acres makes the land and the local watershed better off, through rain soaking in instead of running off, cooler local air and waste heat put to use. It does not cancel the power a data center draws from the grid or the water it takes for cooling. Those need clean energy and water reuse alongside it. Proof of Soil makes the land claim checkable. It doesn't stretch it.

Net-positive is measured across the whole site. The building's footprint counts as a loss, because paved land stops doing most of what living land does. The regenerating land around it has to make up that loss and more, measured the same way as everything else. The 10-and-90 split is an illustration. The real ratio for any site comes from measurement, and some sites will need more land than that.

A data center that leaves its land measurably healthier earns something no pledge can buy: proof. It is the clearest modern case of the net-positive rule in Part IX: replenish at least twice what you consume. A site that leaves its land measurably worse would pay for that damage instead, and the payment would go to the people regenerating land elsewhere (Part VI).

Keep the market. Change the incentive.

None of this requires giving up open markets. Markets are the most powerful coordination tool people have built, and people should be free to build, invest, profit and live well. The problem is where today's incentives point: toward extraction. Proof of Soil points them toward regeneration.

Picture an investor choosing between two data centers with the same technology. One drains the land it sits on. The other regenerates it, and the regeneration is measured and paid for. The technology upside is the same. Only one of them also grows the asset underneath.

Keep the market. Change the incentive.

Part III

First principles: an honest economy

Before designing a new system, it helps to define its parts plainly. Two words do most of the work: token and economy.

What a token is

A token is a unit of representation that carries meaning or value by collective agreement. A shell on a beach is just a shell. A shell a community agrees to use in trade is a token. Nothing about the shell's physics changed. The agreement changed.

So the real questions are always the same two. What is being represented? And what agreement makes the representation hold?

The map is not the territory

Any measurement has three layers. The thing being measured. The system that measures it. And the marks the system produces. Distance, a tape measure, inches. Gold, a banking system, dollars. Soil, a verification protocol, tokens.

Three layers of measurement: the marks sit on the measuring system, which sits on the territory LAYER 3 · THE MARKS Dollars, tokens, scores. What people look at. LAYER 2 · THE MEASURING SYSTEM The tape measure, the banking system, the protocol. Its honesty decides whether the marks mean anything. LAYER 1 · THE TERRITORY The real thing: the distance, the gold, the living soil. Without it, the marks measure nothing.
Figure 3 · The map is not the territory

Human attention goes to layer three, because the marks are what we handle every day. But a measurement is only as good as the thing measured and the honesty of the system doing the measuring. A token with no real asset under it and no open rules around it is a mark on a broken stick.

Stewardship and scarcity

The textbook definition most economics students learn comes from Lionel Robbins in 1932. It frames economics as the study of "scarce means which have alternative uses."10 Gregory Mankiw's widely used textbook puts it as how society manages its scarce resources.11

Compare that with the original oikonomia, the fair management of a household so it thrives. The modern definition centers on scarcity. The original centers on stewardship. The modern definition also leaves a question unasked: who manages the scarcity of money itself? If the measuring stick can be stretched or shrunk out of view, managing real resources well becomes very hard.

Tokenomics is an honest economy

"Tokenomics" joins token and economics. It forces a reckoning. You cannot design the token without designing the economy it measures: the underlying value, the measuring system and the marks, all at once. Four rules follow.

  1. Territory first. There must be a real asset or service underneath. If the territory is empty, the token measures nothing.
  2. Rules in the open. How units are created, retired and shared out is written into public rules that cannot be changed quietly.
  3. Marks that move with the territory. New units appear only when real value appears, and are retired when value is used up, so the marks keep tracking what is real.
  4. Fast feedback. A dishonest national economy can bleed slowly for a century. A badly designed digital economy can collapse in days, because anyone can see the flaw and act on it. That speed is harsh, and it is useful. It makes honesty a condition of survival.

Done right, tokenomics is simply an honest economy.

Nature as an economic actor

If you take goods from a store without paying, that is theft, because it breaks the economy. If you take from nature without replenishing it, that is called business. An honest economy puts nature on the ledger, so the soil and the systems it feeds are counted, compensated and protected like any other party to a transaction.

Part IV

Why soil

If the ledger should see ecological value, which ecological value should anchor it? Each candidate has a problem except one.

Water cycles. It evaporates, condenses and falls again; the planet holds roughly the same water it held a billion years ago. Nobody can create more of it, so its supply cannot grow with good work.

Carbon is a proxy for a process. It is invisible and dispersed, which is why carbon markets have struggled so much with verification.12

Trees and biodiversity matter enormously, and they are outputs of healthy soil. They show the result after the fact.

Soil is the base the others stand on. Healthy soil holds water, stores carbon, grows food and supports the life above it. A teaspoon of healthy soil holds more microorganisms than there are people on Earth.13 Soil holds more carbon than the atmosphere and all plant life combined.14

Nature builds an inch of topsoil over roughly a hundred to a thousand years. Tilled land often loses soil many times faster than that.15 People are also the one species that can deliberately reverse the trend: well-managed land can raise its soil organic matter measurably within a few years. A one percent rise in organic matter in the top six inches is roughly nine to ten more tonnes of organic matter per acre, or about five tonnes of carbon.16 It also helps the soil hold water, though how much is debated. The US Natural Resources Conservation Service has estimated up to about 20,000 more gallons per acre. A 2018 global analysis by Minasny and McBratney found the effect is usually much smaller.17 Proof of Soil treats water gains as something to measure on each parcel, not something to assume.

To anchor value in water, carbon or trees is to anchor it in derivatives. To anchor it in soil is to anchor it in the underlying.

Soil is the one form of natural wealth that people can actively build, that nature cannot replace at the speed we are losing it, and that decides the health of every system above it.

Part V

How Proof of Soil measures

A digital ledger knows only what is recorded on it. It cannot see whether a field got healthier. Something has to carry that fact from the ground to the record.

In ancient Greece, people consulted the Oracle at Delphi for knowledge they could not reach themselves. In digital systems, an "oracle" plays the same role: it is the bridge that brings real-world facts onto a ledger so rules can act on them. A lending system cannot react to a price drop it never hears about. A soil-based system cannot reward improvement it cannot see.

That bridge is the most important and most attackable part of any system that claims to represent something real. Get the measurement wrong and nothing else matters. Get it right and the whole system is as trustworthy as its data. Proof of Soil therefore reads the same piece of ground through five independent layers.

Five independent measurement layers converge into one verified score 1 · Satellites Vegetation, cover, erosion · every 5–12 days 2 · Drones Field-level detail, water paths · each season 3 · Ground sensors + soil samples Moisture, temperature, pH live · lab cores for carbon 4 · Analysis Combines all layers, flags what doesn't add up 5 · Stewards Field observation, local knowledge, can trigger review One verified Soil Health Score
Figure 4 · Five layers, one piece of ground

1. Satellites. Public missions such as the European Space Agency's Sentinel-2, plus commercial imagery, track vegetation health, land cover and erosion every few days.18 This layer works from day one with nothing installed.

2. Drones. Multispectral cameras map topography, plant health, erosion channels and water paths at field resolution, usually each season.

3. Ground sensors and soil samples. Small probes in the soil report moisture, temperature, acidity and compaction around the clock. Some properties, soil carbon above all, still cannot be measured reliably by sensors in real time. For those, the anchor is periodic soil cores tested in accredited labs, with a documented chain of custody from field to lab.

4. Analysis. Models combine every layer into one Soil Health Score and look for readings that don't fit together: a carbon jump with no matching plant growth, moisture with no rain to explain it, a sensor that disagrees with the satellite view. This layer collects no new data. It checks the others against each other.

5. Stewards. The people who work the land submit observations and local and traditional knowledge through a simple phone app. They can flag anomalies and trigger a review. Technology can be fooled; generations of knowledge about a place are much harder to fake.

No single source is trusted alone. It should always cost more to fake regeneration than to do it.

What goes into the score

The Soil Health Score combines a small set of properties that soil scientists already use to judge soil health:

  • Organic carbon, the backbone of soil health, measured from lab-tested cores.
  • Structure: how well the soil holds together in crumbs, and how compacted it is.
  • Water infiltration: how fast rain soaks in instead of running off.
  • Soil life: biological activity, such as how much the soil breathes and how much living microbial mass it holds.
  • Cover and erosion: how much of the year living plants cover the ground, and signs of soil loss, read from satellite and drone.

Each property is compared with the parcel's own baseline and with what similar soils in the same climate can reach, then weighted into one score. The exact indicators, weights and sampling rules are for each implementation to set and publish.

Verification has to cost less than the soil it pays for. On a large ranch, lab cores and drone flights are a small cost. On a two-hectare farm, they could eat the whole reward. So small parcels enroll together: neighbors or a cooperative share a pooled set of lab samples and regional calibration, a statistical sample stands in for every field, and satellites watch each parcel continuously. The shared regeneration fund covers baseline costs for first-time stewards.

The score and the dollars are different things. Soil credits are issued from verified gains in the score. The dollar figures SoilScope shows (Part X) translate those gains into terms people already use, such as soil kept on the hills, clean water, storm water held back and carbon stored. The dollar estimate explains the value. It does not decide how many credits are issued.

Keeping it honest

Goodhart's Law warns that when a measure becomes a target, it stops being a good measure.19 If rewards follow a metric, some people will try to move the metric without doing the work. The design answers this on several fronts.

  • Convergence. A chemical input might fool one sensor. It cannot also fool satellite vegetation readings, drone imagery, lab cores and a neighbor's observations at the same time.
  • Baselines from history. Each parcel's starting point is set from years of satellite record before it joins, so degrading land just before enrolling does not pay.
  • Weather adjustment. Results are compared with similar land nearby and with normal climate, so a wet year doesn't count as stewardship.
  • Pause on anomalies. When the layers disagree, new issuance for that parcel waits until a review resolves it.
  • Something at stake. Stewards put reputation, and in some designs a bond, behind their parcel. Detected gaming leads to penalties larger than any gain.
  • Decline costs more than improvement earns. Sustained healthy land earns ongoing maintenance rewards. The best long-term strategy is steady stewardship, never boom and bust.
  • Open scoring rules. The scoring method is public, every change is published, and changes follow the process below. The method can evolve faster than any gaming strategy.

Who decides: many voices, one judge

Someone has to set and update the rules for scoring soil health. Handing that job to a panel of experts brings a familiar problem: specialists see deeply but narrowly, and many people have learned to be wary of being told to "trust the experts." An expert in one tradition can dismiss methods from another, whether in building, medicine or farming. A builder using techniques that have worked for thousands of years may still fail a modern inspection. A permaculturist and a soil scientist may disagree about the same field. Handing it only to the people with money at stake brings the opposite problem: the people paid by the score would be writing the score.

Proof of Soil gives the job to three voices, and makes the land the judge.

Knows the methods

Science

Soil scientists, labs and measurement specialists. They keep the methods rigorous and repeatable.

Knows the land

Practice

Farmers, permaculturists, traditional and indigenous knowledge holders, the grandmother who has grown food for fifty years. Each is an expert in their own tradition.

Has skin in the game

Stake

Stewards, landowners and funders whose returns depend on the land actually improving.

  • No voice rules alone. A change to the scoring method needs support from at least two of the three voices.
  • Every change must pass the data test. Before it takes effect, a proposed rule is run against the measured history of enrolled land. A rule that would reward land the measurements show declining does not pass.
  • When experts disagree, the land breaks the tie. If a soil scientist and a permaculturist disagree, both approaches run side by side on comparable land, and a few seasons of measurement decide.
  • Standing comes from results. Within each voice, influence grows with a verified track record: land you stewarded or advised that measurably improved. A grandmother whose soil keeps getting better earns as much say as a PhD. Proof of Soil applies to the experts too.
  • Everyone's interest points at the soil. Loosening the rules would issue credits for land that isn't improving, which weakens every credit already held. Stewards and funders both lose from that. The rules answer to the land, not to outside markers like a market price.

You don't have to trust the experts. You can watch the land.

Part VI

How value flows

In today's system, most new money starts as debt. In Proof of Soil, new value starts as verified life in the ground.

  1. Land enters stewardshipHeld in a structure whose purpose is the land's health.
  2. Use rights go to stewardsThe right to work, grow, harvest and benefit from the land.
  3. Stewards work the landRegenerative practice suited to the place.
  4. The land is measuredFive layers read the same ground, continuously.
  5. Improvement is verifiedScore rises above the parcel's baseline.
  6. Credits are issuedNew units in proportion to the verified gain.
  7. Value reaches the stewardPaid to whoever did the work on that parcel.
  8. More land, more capitalResults attract stewards and funding, and the cycle repeats.

Step 8 feeds back into step 1. Healthier soil grows more food, stores more carbon and holds more water each cycle.

Figure 5 · The regeneration cycle
Credits issued = (Current score − Baseline score) × Parcel area × Rate

Soil can't improve forever. Every soil has a natural ceiling set by its climate and texture, and as land approaches it, rewards shift from growth to upkeep. Land that is already healthy earns a steady maintenance reward based on how healthy it is, measured against that ceiling, not only on this year's gain. A farmer who has regenerated for twenty years is paid for keeping it that way, not penalized for starting high. That upkeep reward is paid from system fees and the shared regeneration fund, not from new credits, so the rule stays exact: one new credit in circulation for each unit of verified net soil gain on the whole ledger. Land that declines loses more than the same improvement would have earned.

Three jobs, kept separate

Any system built on this idea has three jobs to do. Keeping them separate avoids a common failure, where speculation and decision-making get tangled together.

Circulates value

Soil credits

The unit that moves. One credit is worth the same as any other, and each carries a record of the parcel and season that produced it, like a serial number. Supply grows only when verified soil health grows.

Records each place

A record for each parcel

Its boundaries, its steward, its verification history and its score over time. Credits issued from that parcel go to its steward.

Steers the rules

The three voices

Science, practice and the people with a stake, as described in Part V, so no single group can rewrite how soil is scored.

A unit backed by land that is getting healthier does not need interest-rate manipulation to hold its value. It needs stewards, sensors and sunlight. Credits like these could start as payment for verified regeneration and, in time, grow into something closer to money, which is the point of Part I. How a unit like this stays stable in its early years, before markets for it grow, is one of the first design questions anyone who builds it will face.

Who needs soil credits

A unit of value holds its worth only if someone needs it. Soil credits are designed to be used, not just held. When a company or public body uses credits to back a claim, those units are retired for good, the way a carbon credit is retired once it is claimed.

  • Companies proving a site is net-positive. A data center, mine or developer retires credits from the land around its site to show that land ended up healthier (Part II).
  • Food and fiber brands. Retiring credits from the farms in a supply chain is how a "regenerative" label becomes checkable.
  • Water districts and flood agencies. Healthier upstream soil means less runoff and sediment downstream. Paying for verified improvement can cost less than paying for the damage.
  • Insurers and lenders. Land whose soil is measurably improving carries less drought and erosion risk, which can earn it better terms.
  • Stewards. Stewards post a bond in credits behind their parcels (Part V), which keeps units committed instead of for sale.

New credits appear only when verified soil health rises. Credits leave circulation when someone retires them to make a claim. Supply follows real regeneration in both directions.

For buying and selling, one credit is as good as another. For a claim, the record matters. A company proves its site is net-positive by retiring credits that its own land produced, backed by that parcel's record and score history.

Soil is a commons: reward the care, pay for the damage

Soil is a public good. The ground under one farm feeds the creek, the town downstream and children not yet born, whoever holds the deed. So its health should be part of the commons: recorded on an open public ledger that anyone can read and check, parcel by parcel, season by season.

A ledger like that makes a simple rule possible. People who tend the soil are rewarded. Land whose soil the ledger shows getting measurably worse pays for that damage, and the payment goes to the stewards whose land is getting better. What counts is the net change across the whole parcel, not any single square foot. Roads and buildings are part of living on land; one paved corner beside thriving soil can still add up to a gain. A data center site whose land, measured across the whole site, ends up less healthy than before pays for the net loss; one that regenerates enough of its buffer land to come out ahead, as in Part II, pays nothing and earns credits instead. The same rule holds on a farm. A farm whose soil is thinning pays, wherever it is, and a ranch that is building soil gets paid. It is the direction of change that counts, not whether land is farmed, built on or wild.

This is not punishment. Part III made the point: take goods from a store without paying and it is theft; take from nature without replenishing it and it is called business. Paying for the damage closes that gap. Destroying soil stops being free, and building it starts to pay. Design the environment, not the sermon (Part IX).

The supply follows the whole ledger. Stewards are issued credits for their verified gains, by the formula above. Land that loses soil health covers its measured loss, either by retiring soil credits to cover it, bought from the stewards who earned them, or by paying a charge that is used to do the same. Either way the payment reaches the stewards, and the credits used to cover it leave circulation. So the number of credits in circulation can grow by no more than the net gain of the whole ledger. A nation's money supply tends to grow as its economy grows. This supply grows as the living ground grows, anchored to soil instead of debt. When losses outweigh gains, the supply does not grow, and any charge that cannot be covered with credits goes into the shared regeneration fund to bring more land into care.

How the money moves. Charges and payments move by published rules on the open ledger, automatically. No treasury holds the money and no official decides where it goes: the payment for a measured loss goes straight to the stewards whose land shows a verified gain, and anyone can follow every transfer, from the parcel that paid to the parcel that earned it. The regeneration fund sits on the same ledger, under the same rules, in plain view.

You don't have to trust a government. You can watch the money.

In numbersSay that in one season the stewards' land gains enough soil health to earn 100 credits under the formula above, and other land loses the equivalent of 30. The stewards are issued 100 credits. The land that lost 30 covers its loss, at the simplest one-for-one rate, by buying 30 of those credits from the stewards and retiring them. The stewards are paid for all 100, and the supply of credits grows by 70, exactly the net gain of the ground.

What keeps it fair.

  • Measured changes, not dollar estimates. Charges and new credits follow the measured net change in each parcel's soil health score against its own baseline, across the whole parcel. Ground sealed under roofs and pavement counts inside that change, never as a separate charge. They never follow SoilScope's dollar estimate. As Part V says, the dollar estimate explains value. It does not decide issuance, and it does not decide charges. This paper sets no charge amount, per hectare or otherwise.
  • Set in the open. Within the system, the size of a charge is set the same way as the scoring rules: publicly, by the three voices, with two of the three in favor. It must also pass the data test: run against measured history, a charge rule that would bill land the measurements show holding steady or improving does not pass.
  • The same proof both ways. A loss has to clear the same bar as a gain. The five layers must agree, the baseline comes from years of history, and the weather is accounted for, so a dry year is not damage. When the layers disagree, the charge waits for review, just as issuance does.
  • Damage, not use. Farming, grazing and building that leave the land as healthy as before, measured across the whole site as in Part II, pay nothing. Only measured decline pays. That is Custodia (Part VII) in practice: use and fruits stay free, and damage carries its cost.
  • Looking forward. Charges apply only to decline measured after a rule is adopted, never to damage done before it.

Who it reaches. A ledger can only charge land that has joined it. Land joins by its owner's choice (Part XI) and posts a bond behind its parcel (Part V); a measured decline is charged against that bond automatically. Land outside the system comes in because someone needs the proof: a company that wants its site shown to be net-positive, a brand that wants a checkable label, a buyer who wants to know the soil under a purchase. A town, county, region or country could also choose to have its land follow the same rules. What changes is who is covered, not how the money moves: it still moves on the open ledger, straight to the stewards, with no treasury in between.

None of this is new in spirit. In 1972 the Council of the Organisation for Economic Co-operation and Development recommended that its member governments follow the polluter-pays principle: the polluter bears the cost of the measures public authorities decide on to keep the environment in an acceptable state. The Council reaffirmed the principle in 1974, in a recommendation that is still in force.20 And under section 404 of the United States Clean Water Act, a project that fills a wetland must first avoid and reduce the harm, and can then be required to make up for what remains, often by buying credits from a "mitigation bank" that has restored wetlands elsewhere. A 2008 federal rule requires those offsets to meet measurable ecological standards and to be monitored, in support of the long-standing national goal of "no net loss" of wetlands.21 In both, the money passes through a public agency that decides where it goes. Proof of Soil applies the same logic to soil, measured on every parcel, season after season, and moves the money by published rules where anyone can see it.

As a voluntary path. No one has to wait for a law. A company can cover its own land's measured losses by retiring soil credits, the same retirement described above, and the money it pays reaches the stewards who earned them. On an open ledger, anyone can check that the losses were covered, parcel by parcel.

Destroying soil should not be free. Building it should pay.

Extractive money in, regenerative money out

The system does not need investors who care about soil. It needs a design that turns ordinary capital into regeneration.

  1. BootstrapCapital looking for returns funds the start. It doesn't have to care about soil. It is the fuel.
  2. BuildThat capital pays for measurement, steward onboarding and the ledger itself.
  3. FlipVerified regeneration starts earning: companies and brands retire credits to prove their claims, insurers price lower risk, data has users. Stewards are paid from that revenue instead of investor money.
  4. FlywheelMore verified land, more demand, more stewards. The regenerative economy grows larger than the capital that started it.

No one needs to be altruistic. The design makes regeneration the obvious choice.

Landowners keep a lasting stake in land that can't be lost to speculation. Stewards get access to land through proven skill instead of capital. Funders get an asset whose foundation strengthens with time. Communities keep the value created on their land, because the land itself cannot be sold away. Fees and a shared regeneration fund are set so that a family garden and a thousand-acre operation can both take part.

Part VII

Stewardship instead of domination

What does it actually mean to own something? Not how you got it, but what you are allowed to do with it.

The civil-law tradition that grew out of Roman law22 describes full ownership as three rights: usus, the right to use; fructus, the right to its fruits; and abusus, the right to consume or destroy. You can use someone else's property and enjoy its fruits with their permission. Only an owner can destroy. That is why abusus marks full ownership.

For things people make, such as a cup, a tool or a machine, this works. Land is not something people made. It breathes, metabolizes, reproduces and evolves. It is a living system, and a right to destroy a living system is a different kind of right.

Ownership as we know it

Usus
the right to use
Fructus
the right to the fruits
Abusus
the right to consume or destroy

Ownership as control

Proof of Soil

Usus
the right to use
Fructus
the right to the fruits
Custodia
the duty to care for it

Ownership as responsibility

Figure 6 · Replacing the right to destroy with the duty to care

Dominium est ius utendi, fruendi, custodiendi re sua.
Ownership is the right to use, to enjoy the fruits of, and to care for what is yours.

A parent says "my child" knowing full well the child is its own being. The word my signals responsibility, not possession. Custodia applies the same idea to land.

Custodia is not "don't touch it." Locking land away with no one caring for it is neglect with paperwork. It is not "extract every possible yield" either, which is the right to destroy dressed in green. The steward builds, grows and innovates: lays a road where the land needs access, raises shelter where people need it, digs a well, plants a greenhouse. The question is always whether that progress serves the land or uses it up. A steward who plants a forest that will take centuries to mature, for the awe of people not yet born, is practicing Custodia.

A society grows great when old men plant trees whose shade they know they shall never sit in.23

Making it last

Most structures that protect land have an off switch. A company can be sold, and the buyer gets the right to destroy. A business can go bankrupt and its land goes to creditors. A charity's status can be revoked. A government program is one election from reversal.

Custodia needs permanence. In Proof of Soil, land is held in a mission-locked structure whose stated purpose is the land's own health. Depending on the country, that might be an irrevocable trust, a community land trust, a foundation or a conservation covenant. Whatever the vehicle, it must answer three questions: who has authority over it, how it survives a sale, a bankruptcy or a change of government, and how stewards can operate day to day. The right vehicle depends on local law.

This idea has relatives. Conservation easements limit what can be done to land forever. Community land trusts take land out of speculation so communities can stay. New Zealand recognized the Whanganui River as a legal person in 2017, and Ecuador's 2008 constitution recognizes rights of nature.24 Custodia joins that family and adds measurement and pay.

Part VIII

The Membrane

Human infrastructure has usually forced itself on top of nature. Movements that try to reverse this often make the opposite mistake and wall themselves off from the world. Both approaches ignore how living things actually work.

A tree is sovereign in its own biology: its own DNA, structure and life cycle. Yet it cannot survive without the fungal networks in the soil, the water moving through the watershed, the sunlight and the air. Its independence is defined by how well it interacts with what surrounds it. Every cell has a membrane that protects what is inside while letting nutrients, water and signals pass.

Proof of Soil works the same way. Land being regenerated still needs pathways for water, lines for power and communication, and immediate access for emergency responders. When infrastructure has to cross regenerating land, it is designed to serve the soil as well: a utility corridor that doubles as a water-harvesting swale or a pollinator strip. Emergency access to protect people or a forest is immediate and frictionless, like a cell's stress response.

A local values layer. Soil health is the shared core that every participating place measures the same way. On top of it, each community or country can add indicators of what it values most, such as happiness, culture, health or biodiversity. The core makes places comparable. The layer above keeps each one its own.

We don't isolate the land from the world. We set the terms on which the world meets the land.

Part IX

Abundance by design

Life builds more life than it uses. One seed becomes a tree that drops ten thousand seeds. Scarcity is often a product of how systems are designed.

Proof of Soil follows two principles drawn from that observation:

  1. No project should begin unless its margin of success allows full restoration of the resources it consumes.
  2. No natural resource should be consumed without the intention to replenish it by a factor of at least two.

Soil credits follow the same rule. They are issued in proportion to verified regeneration and cannot be created from debt, speculation or accounting tricks.

Design the environment, not the sermon

Almost every failed sustainability effort shares one flaw: it asks people to choose long-term collective benefit over short-term personal gain. That asks people to override human nature, and at scale it fails. The better approach is to design an environment where ordinary self-interest, profit-seeking and competition produce the outcome we want.

Paying for damage is part of that environment, not a sermon. Land whose soil the ledger shows declining pays for the loss, and the payment goes to the people rebuilding soil (Part VI). No one is lectured or punished. Destroying soil simply stops being free, and caring for it starts to pay.

Proof of Soil doesn't ask anyone to care about soil. It makes caring for soil the most rational economic choice available.

Part X

What already exists

Before land can be paid for its work, people have to be able to see that work. That is what SoilScope does.

SoilScope (soilscope.land) is a free map of what land does for the people around it. It looks at nine kinds of work: soil held in place, storm water held back, carbon held in trees, cleaner air, cooler ground, clean water, wildlife habitat, pollination and nutrient filtering. It draws on public data including national land-cover maps, global soil maps, Sentinel-2 imagery, Landsat summer ground temperatures and rain maps. The most detailed figures are for the United States. Outside the United States it borrows California's values for rain, storms and some prices, so its figures there are a rough first look, not a local measurement.

The method. SoilScope follows InVEST, a set of models from the Natural Capital Alliance (formerly the Natural Capital Project at Stanford).25 InVEST works out how much the land does, in tons of soil, cubic metres of water, kilograms of nitrogen or degrees of cooling. It has no built-in prices: the user supplies them. SoilScope's soil, flood and water math follow InVEST's models, and its cooling dollars use InVEST's energy-savings formula. Each amount then gets a public price: what the work would cost to replace, what damage it avoids, or what the same thing sells for. Where no local price exists, SoilScope uses the best general public value instead of counting the work as zero. For wildlife habitat, that is the standard value for each kind of land cover in the Ecosystem Services Valuation Database, a global collection of published studies.

A worked example: Marin's protected farmland

The first full portfolio SoilScope studied is the Marin Agricultural Land Trust in California. It mapped 90 easement records from the state's conservation easement database, covering 58,768 of the trust's 58,917 acres, in 57,840 squares of land 64 metres on a side.26 (The trust counts 98 easements, because one state record covers several properties. The newest easement, Tunnel Hill Ranch, added in 2026, is not yet included.)

It starts with what the land does. Each amount is labelled. Measured means a satellite or survey reading. Estimated from measurements means worked out from those readings with published methods.

What the land doesAmountHow we know
Land under easement58,768 acresMeasured: the easements' deeded acreage
What covers it11,465 hectares of shrub and scrub, 6,564 of grassland and pasture, 4,900 of forest, 356 of wetland, 352 developed, and 54 of open water, bare ground and cropsMeasured: the national land-cover map, read from satellite images
Summer ground temperature57.8% of the land is cooler than comparable built-up land nearby on summer mornings. On average, forest is 7.22 °C cooler and dry summer grassland 1.41 °C warmer.Measured: Landsat satellites, summers of 2022 to 2025
Rain1,021 millimetres in an average year; 804 in 2025Measured: Oregon State University's rain maps
Soil kept on the hills1.81 to 3.00 million short tons a yearEstimated from measurements: the Department of Agriculture's soil-loss equation, compared with heavier grazing on the steepest hills, or with farming all of it
Storm water held back5.00 to 5.17 million cubic metres more than the comparison land would hold, in the standard 2-year storm (92 millimetres in a day)Estimated from measurements: the federal runoff method, compared with row crops everywhere, or with hard-grazed pasture on the steepest hills
Clean water flowing off the land73,552 acre-feet in a typical year; 45,919 on 2025's rain. It feeds the creeks, wells, Tomales Bay and the Nicasio, Soulajule and Stafford reservoirsEstimated from measurements: rain minus what plants and soil give back to the air
Nitrogen kept out of the creeks28,100 to 54,600 kilograms a yearEstimated from measurements: how much nitrogen each kind of land and soil sheds a year, from a California water board's published rates, compared with heavier grazing on the steepest hills, or with farming all of it
Carbon dioxide taken up by forest12,702 tonnes a yearEstimated from measurements: forest only, because field measurements on grazed grassland show no net yearly gain without added compost
Figure 7 · What the land does, before any dollar sign

Only then does it put a price on that work:

$58.9M–$90.5M
a year, for all nine kinds of work
$1,002–$1,540
per acre per year
9
kinds of work, each priced from a public source

It is a range, not one number, because two choices are still open, and SoilScope names them instead of quietly picking one:

  • What a ton of soil kept on the land is worth. Either $9.18 a short ton, the Department of Agriculture's figure for California, Oregon and Washington together, which the report's authors call credible, or $15.95, the same report's figure for Marin County alone, which they call credible only when added up across several states.
  • What the steepest hills, over 30% slope, would become without protection. Heavier grazing, or farmland.

The low end pairs the regional soil price with heavier grazing on the steepest hills. The high end pairs the Marin County price with farmland on every slope. Two kinds of work carry most of the value. Soil kept on the hills instead of washing into creeks and bays, where it clouds the water people fish, swim and boat in, clogs ditches and adds to the cost of treating drinking water, is 28% to 53% of the total, depending on the choices. Clean water flowing off the land is 27% to 42%. Storm water held back before it reaches roads and homes is 10% to 15%, and wildlife habitat 8% to 13%. The other five kinds of work add about 2% together.

What the land doesHow it is pricedA year
Soil kept on the hillsThe Department of Agriculture's value of a ton of soil kept out of creeks and bays: $9.18 or $15.95 a short ton, the first open choice.$16.62M to $47.89M
Clean water flowing off the landCalifornia's raw-water market price, from the Nasdaq Veles California Water Index: $336.14 an acre-foot, the past year's average in 2025 dollars.about $24.72M
Storm water held backThe yearly cost of building and keeping up a basin that holds the same water, from the Environmental Protection Agency's cost figures: $1.73 per cubic metre.$8.66M to $8.95M
Wildlife habitatStandard habitat values by land cover from the Ecosystem Services Valuation Database, a 2024 global collection of published studies: $145.54 a hectare a year for grassland, shrub and pasture, $401.79 for forest and $7,764.62 for wetland.about $7.39M
Nitrogen kept out of the creeks$24.07 a kilogram, the value of the harm nitrogen does to fresh water in a national study of the United States (Sobota and others, 2015), which took it from Europe's nitrogen assessment.$677,000 to $1.32M
Carbon dioxide taken up by forestCalifornia's carbon auction price, $32.48 a tonne, used as a reference, and counted on forest only, the one land cover with a measured yearly gain.about $413,000
PollinationMarin County's own crop values, with how much each crop depends on wild insects from a published study of United States crops; the trust is credited with half of the wild-insect share of the squash, pumpkins, melons, berries and apples grown on Marin farms.about $78,500
Cleaner airThe Forest Service's value of the air pollution plants take up in rural California: $1.62 a hectare a year.about $38,800
Cooler groundInVEST's own energy-savings formula, applied to the measured cooling: the air-conditioning electricity saved in homes and buildings within 450 metres, at California's household electricity price, counting only the share of homes that have central air conditioning. Few people live that close, so it is small.about $11,900
Figure 8 · Nine kinds of work, each with its public price. Where two figures are shown, they follow the two open choices.

SoilScope valuation of the Marin Agricultural Land Trust, method v4.5, October 2026. Every price traces to a published source, listed with its page and the date it was retrieved; the few working rules inside the formulas that have no outside source yet are named as such. A second calculation, written separately, reproduces every figure to the cent. Each combination of the two choices also carries its own likely range, which moves every other price and measurement to its low and high ends together. These are estimates built from replacement costs, avoided damage, market prices and standard values from published studies, not prices anyone pays for this land's work; carbon uses California's carbon auction price as a reference only.

How the number got here

The number has changed many times, and every change is listed with its reason in a public change log (soilscope.land/malt-methodology#changes). The first figure, $155.1 million a year in April 2026, rose to $165.7 million when a counting bug that had dropped real land was fixed, then $165.5 million after a smaller fix to how a year's rain was counted. It fell to $97.3 million when the method stopped comparing every hillside with bare, unplanted dirt it never was, and stopped treating a whole year of rain as one storm. It fell to $80.1 million when cooling was measured by satellite instead of assumed.

Then, in October 2026, a check of every price and input against its original source found that $80.1 million could be reproduced to the cent but not defended. Most of its prices traced to no study at all. Soil was worked out in tons per acre but counted as tons per hectare, which undercounted it. The rain behind the water figure came from a gauge in Calaveras County, not Marin. A file-reading bug had garbled the satellite readings for nearly a third of the easements. Smaller errors were fixed too. Fixing each of these, and taking out every price with no source, produced a range of $25.9 million to $56.9 million a year. Each step was a correction, and each is on the record.

On October 24, 2026, we changed the method itself: price every kind of work the Stanford method (InVEST) covers, using standard public values where no local price exists, and drop Los Angeles County's stormwater fee, which is not part of that method. Clean water, wildlife habitat, nutrient filtering, pollination and cooling, which had counted $0 until a price was found, now carry public prices. The range rose to $58.9 million to $90.5 million a year. That step was not a correction. It was a change of method, and it is listed like every other change, with a before and after and the reason. The figures shown so far: $155.1 million, $165.5 million, $97.3 million, $80.1 million, $25.9 to $56.9 million, and now $58.9 to $90.5 million.

Showing every change, and why, is part of the point. A measurement people can trust is one they can check, including when the number goes down.

When it can change. From here the rule is public, and since v4.5 it names changes of method too. As published with the change log, it reads: "This number changes only once a year with new data, when a documented error is found, or when the method itself changes. Every change is listed here with a before and after and the reason." If one of the two open choices is settled, the range narrows to that choice, and that change is listed the same way.

What we do not have yet

We are working with the tools and the data available today, and that is not yet enough for one accurate number. The estimate rests on public maps, satellite readings and published models. It has no lab cores, ground sensors or drone flights of its own from these ranches, and no record yet of how their soil is changing from year to year. Soil kept on the hills comes from a standard erosion equation, not from weighing what stays. The national land-cover map probably counts much of the grassland as shrub. Where no local price exists, the price is a standard value carried over from studies elsewhere, not one measured here. Outside the United States the gaps are wider still. All of it is written out on the methodology page, where anyone can find it and check it.

That is why SoilScope invites anyone with soil data, whether a farm's lab report, a researcher's survey or a land trust's monitoring records, to share it (soilscope.land/contribute). Not for SoilDAO, but for everyone: the aim is an open public ledger of soil that anyone can read, check and build on, where publishing a reading is always the owner's choice. Every real measurement replaces a piece of estimate with a fact. Soil is a public good. Its record should be part of the commons.

That range is the land's invoice, itemized line by line: the bill nature would send for the work SoilScope can see today. Proof of Soil's measurement layers sit on top of this. SoilScope shows what land does and prices what it can. Proof of Soil verifies when it gets better, and pays for the difference.

National soil indices are emerging that rank whole countries by soil health. They are complementary: country rankings show where attention is needed, while Proof of Soil works parcel by parcel and rewards the people improving each one.

Part XI

Where this stands, and questions

A proposal earns trust by saying plainly what exists and what doesn't.

This paper is a concept, offered openly. SoilDAO is not building the full system at this time. SoilScope, the valuation layer, is live and will keep growing. Everything else here is offered so that others can test it, improve it and build it.

Working today

  • SoilScope's click-anywhere map of what land does, with a public price on each kind of work (detailed in the United States, rough elsewhere)
  • A full worked portfolio (58,768 acres)
  • A versioned valuation method following InVEST: every price traced to a published source, a public change log, and a second calculation that matches to the cent
  • Public satellite layer, available everywhere

Described in this paper

  • Five-layer verification and scoring
  • Issuance and maintenance rules
  • Three-voice governance with a data test
  • Charges for measured soil loss, paid to stewards

Open for anyone to build

  • A field pilot with lab-calibrated baselines
  • Independent review of the scoring method
  • Land structures for each jurisdiction
  • An open public ledger of soil measurements
  • The first verified soil credits

This is an invitation to farmers, land trusts, soil scientists, governments, funders and builders: test it, criticize it, improve it, build on it.

Questions people ask

Is this a cryptocurrency?

It doesn't have to be. Proof of Soil is a way of measuring and rewarding regeneration. A shared digital ledger is one good way to keep the records so they can't be quietly changed, but the idea works with any honest, public record. Soil credits are not mined and not created from debt. They are issued only when soil health is verified to have improved. How they can be offered and traded depends on each country's law.

Do I have to give up my land?

Land joins only by its owner's choice. What the owner keeps (use, income, legacy) depends on the structure chosen. The one thing removed is the right to destroy the land.

Who decides what counts as healthy soil?

Three voices share it: science, practice, and the people with a stake. A change needs two of the three and must pass a test against the measured history of enrolled land. When experts disagree, both approaches run side by side and the land's data decides. Part V covers this in full.

What stops people from faking results?

Five independent layers must agree, baselines come from years of satellite history, weather is accounted for, anomalies pause payment, and penalties exceed any gain. Part V covers this in full.

Why not just use carbon credits?

Carbon is one output of healthy soil. Proof of Soil can include carbon, but it measures the base that produces carbon storage, water retention, food and biodiversity together.

Does this work for small farms?

Yes, by design. Neighboring small parcels or a cooperative enroll together and share lab sampling and calibration, so measurement costs don't eat the reward. Fees scale with size, a shared regeneration fund supports first-time stewards, and the satellite layer costs nothing to start. Pooling also protects small parcels, which are easier to attack one at a time.

Closing

Soil holds more carbon than the atmosphere and all plant life combined. It filters water, grows food and holds the land together. It is the foundation of life on land.

Humus, homo, human, humility.27 We are animated soil.

Proof of Soil doesn't ask people to be better than they are. It builds a system in which being human, self-interested, competitive and resourceful, is exactly what the soil needs.

Feed the soil, and it feeds everything.

Origins and how to cite

This framework developed over four years through a series of dated writings and builds:28

  1. Mar 20, 2022SoilDAO founded, with the stated mission "to accelerate topsoil regeneration while enabling a decentralized global community of regenerative land stewards"
  2. Apr–May 2022soildao.io launches with a plan to obtain the rights to steward land and fund regenerative agriculture
  3. May 30, 2022"It's time to change the concept of land ownership," the first public statement of the idea that became Custodia
  4. Feb 14, 2025Proof of Soil named in working notes: soil as the measure of success for any land project, "our GDP," with new units issued only for proven gains in soil health
  5. Mar 7, 2025"GDP: The Broken Compass of Progress," published on soildao.io and Medium, proposing to move "from GDP to Proof of Soil"
  6. Mar 2025In-Depth White Paper introducing a Soil Health Index
  7. Jun 2025Proof of Soil economic design drafts
  8. Mar 2, 2026SoilDAO Founding Paper, v1
  9. Mar 21, 2026Founding Paper V4, adding the honest-economy first principles
  10. Mar 30, 2026SoilScope launched (soilscope.land)
  11. Oct 24, 2026This universal edition, draft v0.9
Atanasov, H. (2026). Proof of Soil: A proposal to measure an economy by the health of its ground. Draft v0.9, October 24, 2026. SoilDAO. soilscope.land

© 2026 Hristiyan Atanasov. License: Creative Commons Attribution 4.0 (CC BY 4.0). Anyone may share, adapt and build on this work, including commercially, as long as they credit the author.

Implementation details, such as how credits are issued and kept stable, which legal structures fit each country, and exact scoring weights, are left open for those who build on this work. Working drafts on these topics from SoilDAO's four years of development are available on request to teams building on this proposal.

Notes
  1. UNCCD, Global Land Outlook, 1st edition (Bonn: UNCCD, 2017). The 24-billion-tonne figure is UNCCD's widely cited estimate of fertile soil lost each year. ↩
  2. FAO and ITPS, Status of the World's Soil Resources: Main Report (Rome: FAO, 2015). About 33% of land is moderately to highly degraded. ↩
  3. Michael McLeay, Amar Radia and Ryland Thomas, "Money creation in the modern economy," Bank of England Quarterly Bulletin, 2014 Q1. ↩
  4. Statute of Merton (1235), which let lords of the manor enclose common land provided enough pasture remained for free tenants. ↩
  5. Simon Kuznets, National Income, 1929–1932, Senate Document No. 124, 73rd Congress, 2nd Session (Washington, 1934). ↩
  6. Gross National Happiness: attributed to Bhutan's Fourth King and commonly dated to 1972; later written into Bhutan's 2008 Constitution. UNDP, Human Development Report 1990. Clifford Cobb, Ted Halstead and Jonathan Rowe, "If the GDP Is Up, Why Is America Down?", The Atlantic, October 1995 (Genuine Progress Indicator). Kate Raworth, Doughnut Economics (2017), building on her 2012 Oxfam discussion paper. ↩
  7. Pew Research Center, "Public Trust in Government: 1958–2024," June 24, 2024. About 22% of Americans said they trust the federal government to do what is right just about always or most of the time. ↩
  8. International Energy Agency, Energy and AI (Paris: IEA, April 2025). Data center electricity use is projected to more than double, to around 945 TWh, by 2030. ↩
  9. Brad Smith, "Microsoft will replenish more water than it consumes by 2030," Microsoft On the Issues, September 21, 2020. ↩
  10. Lionel Robbins, An Essay on the Nature and Significance of Economic Science (London: Macmillan, 1932). ↩
  11. N. Gregory Mankiw, Principles of Economics (Cengage; many editions), chapter 1. ↩
  12. For example, Thales A. P. West et al., "Action needed to make carbon offsets from forest conservation work for climate change mitigation," Science 381 (2023): 873–877. ↩
  13. USDA Natural Resources Conservation Service, soil biology and soil health guidance. ↩
  14. Rattan Lal, "Soil carbon sequestration impacts on global climate change and food security," Science 304 (2004): 1623–1627. The soil carbon pool is about 3.3 times the atmospheric pool and 4.5 times the pool in living things. ↩
  15. David R. Montgomery, "Soil erosion and agricultural sustainability," PNAS 104, no. 33 (2007): 13268–13272. Conventionally plowed fields erode one to two orders of magnitude faster than soil forms. The 100–1,000-years-per-inch range is a common rule of thumb that varies with climate and parent rock. ↩
  16. The top six inches of an acre (the "acre-furrow slice") weighs about 2 million pounds, so 1% is about 20,000 lb, roughly 9 tonnes, of organic matter. Organic matter is about 50–58% carbon, which gives roughly 5 tonnes of carbon. ↩
  17. The NRCS estimate as reported in Lara Bryant, "Organic Matter Can Improve Your Soil's Water Holding Capacity," NRDC, 2015. Budiman Minasny and Alex B. McBratney, "Limited effect of organic matter on soil available water capacity," European Journal of Soil Science 69, no. 1 (2018): 39–47. ↩
  18. European Space Agency, Sentinel-2 mission. With two satellites, the revisit time at the equator is five days. ↩
  19. Charles Goodhart, "Problems of Monetary Management: The U.K. Experience" (1975). The popular wording is Marilyn Strathern's (1997). ↩
  20. Organisation for Economic Co-operation and Development, Recommendation of the Council on Guiding Principles concerning International Economic Aspects of Environmental Policies, adopted 26 May 1972, Annex, paragraph 4: the polluter "should bear the expenses" of the pollution measures public authorities decide on. legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0102. The 1972 text was formally retired on 8 November 2023. The principle remains in force through the Council's Recommendation on the Implementation of the Polluter-Pays Principle, adopted 14 November 1974, which refers back to it: legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0132. ↩
  21. Clean Water Act section 404, United States Code, title 33, section 1344, "Permits for dredged or fill material": law.cornell.edu/uscode/text/33/1344. Department of the Army and Environmental Protection Agency, "Compensatory Mitigation for Losses of Aquatic Resources; Final Rule," Federal Register 73, no. 70 (April 10, 2008): 19594–19705, effective June 9, 2008 (Code of Federal Regulations, title 33, part 332, and title 40, part 230, subpart J). The rule calls compensatory mitigation a tool for meeting "the longstanding national goal of 'no net loss' of wetland acreage and function" (p. 19594), requires measurable ecological performance standards and monitoring, and defines the credits that mitigation banks sell (p. 19671): govinfo.gov/content/pkg/FR-2008-04-10/pdf/E8-6918.pdf. ↩
  22. The usus–fructus–abusus formula is a later summary of Roman ownership by medieval and early-modern jurists. It is not a phrase found in Roman texts. ↩
  23. Proverb, usually described as Greek. Its exact origin is uncertain. ↩
  24. Te Awa Tupua (Whanganui River Claims Settlement) Act 2017, New Zealand. Constitution of the Republic of Ecuador (2008), Articles 71–74. ↩
  25. Natural Capital Alliance (formerly the Natural Capital Project at Stanford University), InVEST User Guide, release 3.20.2 (2026), whose suggested citation reads "Natural Capital Alliance, 2026. InVEST 3.20.2": storage.googleapis.com/releases.naturalcapitalproject.org/invest-userguide/latest/en/index.html, retrieved October 24, 2026. InVEST is a set of separate models, one for each service. Each works out a physical amount, and where a model has a dollar step, the user supplies the price. SoilScope runs its own code, using the same methods: for soil, the universal soil-loss equation behind InVEST's sediment model, in the Department of Agriculture's revised form; for flood control, the runoff method for a single design storm used by InVEST's urban flood model, priced like InVEST's stormwater model, at what the same storage would cost to build; for clean water, the Budyko water-balance curve behind InVEST's annual water yield model, in the 2001 form by Zhang and others; and for cooling, InVEST's energy-savings formula. Standard values: L. M. Brander, R. de Groot and others, "Economic values for ecosystem services: A global synthesis and way forward," Ecosystem Services 66 (2024): 101606, the Ecosystem Services Valuation Database, Table 1: only its two habitat values (maintaining life cycles and genetic diversity), never a land cover's total, so that no service SoilScope already prices is counted twice. ↩
  26. SoilScope valuation of the Marin Agricultural Land Trust, method v4.5, adopted October 24, 2026: 90 easement records, 57,840 map cells of 64 by 64 metres, each with 12 monthly records for 2025. The four combinations of the two open choices: $58,900,948.56 a year (soil at $9.18 a short ton, steepest hills compared with heavier grazing), $70,208,978.21 ($9.18, farmland), $71,144,922.20 ($15.95, heavier grazing) and $90,526,444.81 ($15.95, farmland). Per acre: $58,900,948.56 ÷ 58,767.61 deeded acres = $1,002; $90,526,444.81 ÷ 58,767.61 = $1,540. Soil prices: LeRoy Hansen and Marc Ribaudo, Economic Measures of Soil Conservation Benefits: Regional Values for Policy Assessment, Technical Bulletin 1922 (Washington: United States Department of Agriculture, Economic Research Service, 2008), regional and county values brought to 2025 dollars, ers.usda.gov/publications/47550. Clean water: the Nasdaq Veles California Water Index, the 52 weekly values from October 29, 2025 to October 21, 2026, with the 2026 weeks brought to 2025 dollars, $336.14 an acre-foot, on the typical-year flow of 90,724,873 cubic metres (73,552 acre-feet). Wildlife habitat: Brander and others 2024 (note 25), Table 1, brought from 2020 to 2025 dollars. Nitrogen: export rates from the California Regional Water Quality Control Board, Central Coast Region, Total Maximum Daily Loads for Nitrogen Compounds and Orthophosphate for the Lower Salinas River and Reclamation Canal Basin (2013), Appendix F, Table 3; price from D. J. Sobota, J. E. Compton, M. L. McCrackin and S. Singh, "Cost of reactive nitrogen release from human activities to the environment in the United States," Environmental Research Letters 10 (2015): 025006, Table 1, freshwater row, which takes its figures from Europe's nitrogen assessment (H. J. M. van Grinsven and others, 2013), brought to 2025 dollars. Pollination: Marin County Department of Agriculture, 2024 Crop and Livestock Report; United States Department of Agriculture, 2022 Census of Agriculture, California county tables; N. W. Calderone, PLoS ONE 7 (2012): e37235, Table S12; A.-M. Klein and others, Proceedings of the Royal Society B 274 (2007): 303–313, for olives and grapes. The trust's half follows its own count of "more than half" of Marin's farmland; spreading that total evenly over the trust's natural cover is a working rule. Cooling: one-tenth of the measured ground cooling counted as cooler air (a working rule), 0.74 kilowatt-hours per square metre per degree (Santamouris and others, Energy and Buildings 166, 2018, citing Santamouris 2014), California's household electricity price for January to August 2025, 31.11 cents a kilowatt-hour (United States Energy Information Administration, Electric Power Monthly, Table 5.6.B) and the 51% of homes with central air conditioning in Pacific Gas and Electric's area (California Energy Commission, 2019 Residential Appliance Saturation Study). Run first with every v4.3 setting and then with the v4.4 settings, the same calculation reproduces the earlier $80,105,679.29 and $56,879,579.84 to the cent. Measurements, every source with its page and retrieval date, the change log and the change rule: soilscope.land/malt-methodology. Trust-wide figures (98 easements, 58,917 acres): Marin Agricultural Land Trust, malt.org/farmland-protection. ↩
  27. Latin homo (human) and humilis (humble, literally "low, near the ground") are both related to humus, earth, through an Indo-European root meaning earth. ↩
  28. @Soil_DAO account on X (joined March 20, 2022; bio as quoted). soildao.io domain registration (April 5, 2022) and Internet Archive capture of soildao.io (May 5, 2022). @Soil_DAO post of May 30, 2022. Private working notes dated February 14, 2025 (not public; available on request). Internet Archive capture of the March 7, 2025 essay on Medium. PDF metadata and dated emails for the 2026 papers. ↩