La Bourse / Volume IV / Nº IV.03 / Quiz, reflection, essays

A watercolour of two trees side by side, one green and one turning gold, their roots in the same earth. — Plate IV.03 · Quiz, reflection, essaysThe Core, at Eleven in the Morning.Every argument in this chapter is about that line: how far down it goes, how long it took to get there, and how quickly it can be put back.

ASSESSMENT · Chapter IV.03 — Agriculture as the Model Case

Three instruments: a ten-point quiz, eight reflection questions, five essay prompts. The quiz checks comprehension rather than recall. The reflections are private and first-person. The essays are arguable from more than one side.

This chapter's assessment is stricter than most in the edition, deliberately. Agriculture is where the strongest evidence and the most inflated claims live in the same paragraph, and a student who cannot separate them has not learned the chapter.

THE QUIZ — ten points

Four on recall.

1. How much carbon is held in the world's soils, to 30 cm and to one metre, and how does that compare with what human land use has removed?

About 700 Gt C in the top 30 cm and about 1,500 Gt C in the top metre (Batjes 1996). Human land use has removed roughly 133 Gt C over twelve thousand years (Sanderman et al. 2017). One mark for the stocks, one for recognising that the debt is real, large, and smaller than the remaining stock — which is why the hole exists and also why refilling it is not unlimited.

2. Write the first-order saturation model for soil carbon and say, in one sentence, what it implies about the annual rate.

C(t) = C∞ − (C∞ − C₀)e^(−kt), with rate(t) = k(C∞ − C(t)). The rate is proportional to the remaining gap, so it decays toward zero as the soil fills. Full marks require the second half: the sink closes, and no management repeals that.

3. State the three published meta-analytic estimates of the organic yield gap with their sources.

Seufert, Ramankutty & Foley (2012): 25 percent. De Ponti, Rijk & van Ittersum (2012): 20 percent across 362 paired comparisons. Ponisio et al. (2015): 19.2 ± 3.7 percent, falling to 8–9 percent with diversification. Credit any answer that also gives Mäder et al.'s 20 percent from the DOK trial.

4. Under the US National Organic Program, how long must land be free of prohibited substances before an organic crop can be harvested, and what does that period do to a farm's cash position?

Thirty-six months (7 CFR §205.202(b)). During that period the farm takes organic yields at conventional prices, which is the transition hole — modelled in the chapter at $810 per hectare over three years.

Four on application.

5. A developer offers your co-operative a soil carbon contract priced on the total soil carbon stock beneath the enrolled fields. Diagnose it.

It is not a carbon contract; it is a payment for carbon that is already there and was never at risk. A contract may only pay for the difference between the practice and its counterfactual. At Rodale that difference is 0.439 t C/ha/yr, against 0.634 for the organic-animal system taken alone — so about 31 percent of a stock-based contract would be paying for carbon the conventional control was building anyway. The stronger answer also names permanence and leakage as the other two tests.

6. A colleague says: "4 per 1000 shows that soil can offset human emissions. This is settled." What is right and what is missing?

Right: 0.4 percent of the top metre is 6.0 Gt C/yr and of the top 30 cm is 2.8 Gt C/yr, against an atmospheric increase near 5.10 Gt C/yr — so the arithmetic of the target is sound. Missing: the achievable rate. At 0.30 t C/ha/yr across 1.6 Gha of cropland the world delivers 0.48 Gt C/yr — 17 percent of the 2.8 Gt target and 4.8 percent of global fossil CO₂. Full marks require holding both: the target is arithmetically real and agronomically out of reach, and the reachable contribution is still worth having.

7. Your trial reports "no significant change in soil carbon" after four years of cover cropping on a 40-hectare field, from 60 unpaired cores. What do you conclude?

Nothing about the soil. You conclude that the study could not have detected the effect: an unpaired design at a 15 percent spatial CV needs about 614 cores per group to resolve 1.20 t C/ha at 80 percent power. Sixty cores is roughly a tenth of that. The stronger answer gives the fix rather than the complaint — re-sample fixed points, which drops the requirement to about 50 — and notes that "no significant change" and "no change" are different statements.

8. Why does the chapter's facility cancel reversals from a buffer pool rather than clawing back from the farmer?

Because a farmer facing a personal clawback in a bad year will till, and the scheme then loses the carbon and the member. The buffer converts a reversal from a dispute into an accounting entry. Credit any answer that notes the buffer is sized off the hazard — a 1 percent annual hazard gives a 9.6 percent ten-year reversal probability, so a 22 percent buffer carries real margin.

Two that require the arithmetic to be done.

9. A 120-hectare farm, one soil stratum, signs a four-year soil carbon contract at 0.30 t C/ha/yr. Verification is by paired re-sampling: 50 georeferenced points, cored twice, at $25 a sample. CO₂ sells at $30 a tonne and the scheme holds a 22 percent buffer. Compute the verification cost per tonne and the farm's net over the four years. Show your working.

Samples: 50 × 2 × $25 = $2,500. Carbon: 0.30 × 4 = 1.20 t C/ha, × 3.6667 = 4.40 t CO₂e/ha, × 120 ha = 528.0 t CO₂e. Verification cost per tonne: 2,500 / 528.0 = $4.73/t. Net per tonne: 30 × (1 − 0.22) − 4.73 = $18.67/t. Whole-farm net over four years: 528.0 × 23.40 − 2,500 = $9,855.
The point of the question is the last comparison the student should make unprompted: run the same farm on an unpaired design at 614 cores and the verification costs $30,700 — more than three times the entire gross carbon value. The design choice, not the agronomy, decides whether this farm has an asset.

10. A field carries 50 t C/ha. Twenty-five years of changed management add 0.30 t C/ha/yr. The farm is then sold and returned to continuous tillage, losing 30 percent of its stock over twenty years. How much carbon is lost, how does the field compare with where it started, and how many years of work does the reversal erase?

Built: 25 × 0.30 = 7.5 t C/ha. Stock at reversion: 50 + 7.5 = 57.5 t C/ha. Lost: 57.5 × 0.30 = 17.25 t C/ha. Remaining: 40.25 t C/ha — nearly 10 tonnes below where the field started. Years of work erased: 17.25 / 0.30 = 57.5 years, which is 2.30 times everything the twenty-five years built.
Full marks require the conclusion in the right register: a reversal does not merely undo the project, it takes inherited carbon with it. That is why the covenant must run with the land and not with the owner, and why the credit is a tenancy rather than a sale.

REFLECTION — eight questions, for one person and a pen

These are not for a room. Write the answers by hand if you can; the slowness is the point.

Think of a piece of ground, a team, or a body of work that you have been drawing on for years. What is its regeneration rate, honestly — and when did you last measure rather than assume it?

Where have you accepted a number because it was in the direction you wanted? Name one. What would it have cost you to check it, and what did not checking it cost instead?

The chapter argues that the strongest case must be stated with its four honest negatives attached. Where in your own advocacy do you currently leave the negatives off — and what do you fear would happen if you put them in?

What in your life or work is saturating? Something that responded well to effort for years and now responds less. What would a fresh sink look like?

Recall something you built slowly and lost quickly. What was the ratio, and what would have had to exist for it to be protected while nobody was watching?

Where are you paying for a measurement you cannot afford at your current scale? Who else is in the same position, and what would it take to ask them?

What have you substituted a purchased input for, in your own practice, that used to be done by a process you no longer have time for? What would putting the time back cost, and what would it return?

If someone you respect read your strongest claim and checked one figure, which figure would you want them to check first — and which one are you hoping they pick last? Go and do that one now.

ESSAY PROMPTS — five

Each is arguable from more than one side. Each requires at least one source the chapter cites and at least one it does not.

1. Is soil carbon a commodity at all? The chapter shows that soil carbon saturates, reverses about four times faster than it accumulates, and costs more to verify than it is worth below an aggregation threshold. Argue either that these properties are ordinary risks that contract design handles — buffers, covenants, vintages — or that they make soil carbon categorically unsuitable for a credit market and better addressed through practice-based payments that never pretend to be tonnes. Use Smith et al. (2020) on measurement and verification, and one source on carbon market design or additionality that the chapter does not cite.

2. The land-use claim inside the yield gap. A 19.2 percent yield gap implies 23.8 percent more land for the same output, and land converted to agriculture is usually carbon released. Argue either that the efficiency and soil gains outweigh the land claim, or that a system requiring a quarter more land cannot be called regenerative at global scale without accounting for what is grown on the extra quarter. Engage Seufert et al. (2012) and Ponisio et al. (2015), and one source on land sparing versus land sharing that the chapter does not cite.

3. Marsden against the certificate. The Marsden Farm results deliver higher yields, 86 percent less synthetic nitrogen and comparable profit with no organic certification and no premium. Argue whether the certification and premium model has therefore outlived its usefulness as the route to regenerative agriculture, or whether the premium performs a market-making function that rotation lengthening alone cannot. Use Davis et al. (2012) and Crowder & Reganold (2015), and one source on certification economics or eco-label price dynamics that the chapter does not cite.

4. Whose evidence counts? The chapter treats the Rothamsted, Rodale and DOK long-term trials as the strongest evidence available, and presents Gabe Brown's farm record as a different kind of evidence explicitly labelled as such. Argue for or against the proposition that replicated long-term trials systematically under-detect the benefits of integrated whole-farm systems, because trials must hold most variables constant and whole-farm systems work through their interactions. Use Mäder et al. (2002) and Brown (2018), and one source on the methodology of agricultural systems research that the chapter does not cite.

5. The measurement that creates the market. The chapter's central inversion is that verification cost scales with soil types rather than hectares, so aggregation — not agronomy — is the binding constraint. Argue either that this makes co-operative institutions the decisive innovation in regenerative agriculture, or that it simply relocates power to whoever owns the aggregation, reproducing the concentration that regenerative agriculture claims to escape. Use Ostrom (1990) on commons design principles and the chapter's measurement arithmetic, and one source on agricultural co-operative governance or supply-chain concentration that the chapter does not cite.