Google has agreed to buy 1 million rice methane carbon credits from climate-tech startup Mitti Labs through 2030, in what the companies describe as the largest publicly announced deal for credits generated by cutting methane emissions from rice farming, and Google’s largest superpollutant purchase to date. For a voluntary carbon market still short on big-ticket demand signals in 2026, the deal matters on two levels: it validates agricultural methane as a institutional-grade credit category, and it shows that rigorous, satellite-based MRV is becoming the price of admission for frontier buyers.
The Deal in Numbers
The four-year agreement covers rice farms across the Indian states of Karnataka, Andhra Pradesh and Telangana, reaching about 100,000 hectares at peak delivery, Mitti Labs co-founder Xavier Laguarta told TechCrunch. Financial terms were not disclosed. According to the companies, the shift in farming practice across those fields will deliver the equivalent of 1 million tonnes of CO2e when methane is measured on the conventional 100-year horizon, or 3 million tonnes on a 20-year horizon, while saving around 1.5 trillion liters of water.
The underlying practice is Alternate Wetting and Drying (AWD): instead of keeping paddies continuously flooded, farmers periodically drain and reflood them. That disrupts the oxygen-depleted soil conditions in which methane-generating microbes thrive. Mitti Labs says AWD cuts methane emissions from rice fields by about 50% and irrigation water use by close to 40%, without reducing yields. More than 70,000 smallholder farmers are expected to participate, with a majority of project revenue flowing to farming communities.
Why Rice Methane, Why Now
Rice cultivation accounts for roughly 12% of global methane emissions, making it the second-largest agricultural source of methane after livestock. Methane is a short-lived but potent greenhouse gas, and the IPCC attributes almost half of global warming to date to superpollutants as a class. That makes near-term methane cuts one of the fastest levers available to buyers trying to neutralize warming impact while longer-lived decarbonization catches up.
Google’s motivation is explicit. The company has committed to spend at least $50 million by 2030 on superpollutant elimination under its Superpollutant Action Initiative, and it is under pressure on its 2030 net-zero goal: its greenhouse gas emissions grew 18% year over year in 2025 to about 14.5 million tonnes of CO2e, driven by AI infrastructure build-out. Chief Sustainability Officer Kate Brandt framed the Mitti deal as delivering “near-term atmospheric impact” alongside water conservation and smallholder livelihoods. For Google, the purchase also extends a growing India portfolio that includes a 100,000-tonne biochar credit deal with Varaha signed in January 2025.
MRV Was the Real Due Diligence Test
The detail that should interest market participants most is how the deal was won. Laguarta said Google’s evaluation included scrutiny of Mitti Labs’ monitoring technology and physical farm visits before signature, a process that took about a year of talks. Mitti, founded in 2023 and based in New York and Bengaluru, has built a GeoAI platform that combines synthetic aperture radar imagery from commercial and public satellites, at resolutions from 50 centimeters to 10 meters, with proprietary field measurements used to train its models. The platform remotely tracks crop growth, soil moisture and flooding across thousands of smallholder plots, the exact data needed to verify AWD adoption at scale.
Credits sold to Google can be issued under either Gold Standard or Isometric, with independent third-party verification before issuance. The dual-registry optionality is itself a signal: buyers at this level now expect methodological flexibility plus auditable digital MRV, not a PDF-based annual verification cycle.
Implications for Buyers and Developers
For corporate buyers, the deal reprices expectations for agricultural methane credits. When the most diligenced buyer in the market signs a seven-figure-tonne offtake in a category, that category moves from experimental to procurable. Buyers with methane-heavy supply chains or near-term warming targets should expect more competition for high-integrity AWD supply, particularly as Mitti expands to the Philippines later this year and to Indonesia and wider Southeast Asia in 2027. About 150 million farmers grow rice across India, Southeast Asia and China, so the theoretical supply runway is long, but verified supply with this level of monitoring is not.
For project developers, the message is narrower and harder. Mitti’s edge was not the agronomy, which is well established, but the measurement stack and the operating scale: more than 100,000 farmers already engaged, 500 billion liters of water saved over two years, and existing commercial relationships with buyers including Cool Effect, Ebro Foods and Syngenta. Developers entering rice methane without comparable remote-sensing capability will struggle to clear the diligence bar that Google has now made visible.
What to Watch
Three markers will show whether this deal opens a category or remains an outlier. First, issuance: watch the first verified credit deliveries under Gold Standard or Isometric and whether issued volumes track the 100,000-hectare peak-delivery plan. Second, pricing: no terms were disclosed, but any indication of price per tonne for verified AWD methane credits would give the whole agricultural superpollutant segment its first real benchmark. Third, replication: whether other frontier buyers, particularly tech companies facing the same AI-driven emissions growth, follow Google into rice methane offtakes in Southeast Asia over the next 12 months.
