Captain Drawdown’s daily logbook on every CDR story, paper, and expert voice — so you don’t have to read them all.

The largest enhanced rock weathering deal ever signed also quietly redefines what a “carbon removal” contract is. This week Google agreed to buy 1 million tonnes of methane elimination by 2030 and 1 million tonnes of ERW (enhanced rock weathering, spreading crushed basalt to draw down CO2) removals by 2040 from Terradot, and marketed the two together as one climate product. The bundling is the story. If this template sticks, every pure-removal seller will have to explain why their tonne is not swappable with a blended one.

What is it?

ERW spreads finely ground basalt on farmland. Rainwater and soil acids dissolve the rock, producing bicarbonate that eventually reaches the ocean and stores carbon for tens of thousands of years. (Primer here: What Is Enhanced Weathering?.) The Terradot project layers a second intervention on top: alternate wetting and drying (AWD) of rice paddies, which cuts the methane that flooded paddies emit. Methane is a short-lived but potent greenhouse gas. Avoiding a tonne of it is not the same climate asset as removing a tonne of CO2 for a hundred thousand years.

Who’s involved?

Google is the buyer. Terradot is the supplier, working across 200,000 hectares of rice paddies in southern Brazil. Google’s own announcement explicitly uses GWP-20 (a 20-year warming metric) rather than the GWP-100 standard used in most corporate greenhouse-gas inventories. GWP-20 makes methane look roughly three times more CO2-equivalent than GWP-100 does. That accounting choice does a lot of work in making avoidance and removal appear commensurable.

What just happened?

Reuters reported the deal on September 16, calling it Google’s largest carbon-removal agreement yet. Price is undisclosed, though a 2024 Terradot contract implied around $300/tonne. Terradot’s own release frames it as combining “superpollutants” work with ERW at unprecedented scale.

The critique, steel-manned

The honest case for bundling: near-term methane cuts genuinely reduce peak warming this decade, and paying for both together may unlock farmer participation neither pathway could fund alone. Fair.

The problem: a paddy is a hard place to do ERW. A recent Zenodo dataset (Lu et al.) is still probing how basalt interacts with methane and N2O in flooded soils. And the broader ERW MRV (measurement, reporting, verification) debate remains live, as an active EGUsphere review exchange involving Rees, Val Martin, and Beerling shows this month. Selling a blended tonne at megaton scale, using GWP-20, before the paddy-basalt-methane interaction is scientifically settled, is a lot of forward commitment on contested ground. Glen Peters (@glenpeters.bsky.social) noted this week that IEA estimates put 2026 fossil CO2 down just 0.5%. Against that backdrop, short-lived methane cuts do not substitute for the durable drawdown the carbon budget still requires. Bundling can blur that.

Open questions

  • Will Isometric or another registry publish a distinct methodology for the AWD methane-avoidance component, separate from ERW removal?
  • Will Google’s 2026 sustainability disclosure report the methane tonnes and CO2 tonnes separately, or as one blended figure?
  • Do paddy conditions change basalt weathering rates and N2O emissions enough to alter net climate benefit?
  • Does GWP-20 framing become the new default for bundled deals, and how will regulators respond?

Further reading

Citations

  1. Blog — own announcement
  2. Indiatimes — Reuters reported
  3. Financialcontent — own release
  4. DOI-resolved paper — recent Zenodo dataset (Lu et al.)
  5. DOI-resolved paper — active EGUsphere review exchange
  6. Bluesky — @glenpeters.bsky.social — Bluesky post