Rice is both a vital food crop and a significant methane emitter, responsible for around 10% of global methane emissions. The majority of the world's rice — grown by over 200 million smallholder farmers — is cultivated in flooded paddies, creating oxygen-poor conditions that produce methane. Projects that reduce the duration of field flooding can significantly lower methane emissions, generating credits through eligible improved rice cultivation practices.

Until recently, rice methane avoidance credits accounted for nearly one-third of all agricultural credits in the voluntary carbon market (VCM). However, in 2024, 99.9% of rice credits issued were invalidated, primarily due to additionality concerns. In early 2025, the standard-setting body Verra released an updated, more rigorous methodology that addresses previous shortcomings, rejoining the list of standards with active methodologies, alongside Climate Action Reserve (CAR) and Gold Standard, whose methodologies were last updated in 2013 and 2023, respectively. As of spring 2025, no credits have yet been issued under the new methodology, but early signs point toward stronger quality assurance.

However, projects of this credit type still carry quality concerns regarding baseline setting, additionality, and monitoring, reporting, and verification (MRV). There is little to no risk associated with leakage, permanence, or greenhouse gas (GHG) accounting, assuming projects follow the updated methodologies.

To mitigate identified risks, seek projects that:

• Use the improved and active methodologies

• Include methane (CH₄) emissions factors calibrated with site- or region-specific data when possible, rather than those with broader country-level statistics

• Include digital MRV components, especially within irrigation-related practices (like using satellite imagery to verify alternate wetting and drying, or AWD, practices)

• Clearly articulate how credit revenue enables activity adoption — particularly in regions that have high activity adoption rates and existing reliable irrigation infrastructure without carbon projects

• Include verified third-party attestation or surveys, especially with additionality tests or MRV reporting

As market demand for high-integrity carbon credits grows — and as global demand for rice as a food staple expands — this credit type has potential to deliver real, high-quality methane reduction projects to the VCM. Supporting these projects offers outsize short-term climate benefits and promotes water savings and rural development for some of the world's most vulnerable farming communities.

Nine out of ten grains of rice are grown in Asia , with China and India accounting for more than half of the global harvest. Unlike traditional commodity crops, such as barley, wheat, and soy, which are primarily grown across swaths of land at highly industrial scales, 90% of the world's rice is produced by more than 200 million farmers working on an average of less than 1 hectare of land each. Often, rice farmers continuously flood fields (also known as paddies) because it is the simplest and least labor-intensive way to effectively enhance weed control, minimizing the need for costly herbicides and fertilizers. However, perpetually waterlogged fields create an oxygen-poor environment in which bacteria emit large volumes of methane. The longer the flooding lasts, the more methane-producing bacteria thrive.

While rice is a vital food staple, flooded paddies are responsible for 10% of global methane emissions. Improved rice cultivation projects generate carbon credits by decreasing the time that rice fields spend flooded, thus reducing soil methane production. Prior to 2024, rice carbon credits accounted for roughly 30% of all agricultural credits in the VCM, with 99% of rice projects occurring in China . In August 2024, the standard-setting organization Verra began invalidating 37 rice methane projects responsible for 4.5 million credits (99.9% of all rice credits issued), which had been issued under a Clean Development Mechanism (CDM) methodology, due to significant additionality concerns.

As of spring 2025, only 597 credits remain on the market, and all major carbon standards have inactivated the CDM methodology. Verra has since released a new methodology with significant improvements, holding the potential to generate higher-quality credits. Given the steadily increasing demand for rice globally, and the current emissions intensity of producing the crop, this project type has the potential to generate millions of credits annually — if developers can address quality concerns.

Rice carbon projects incentivize farmers to reduce the time their fields spend flooded, primarily by seeding in dry conditions, draining their paddies early, or alternating between flooded and dry conditions (see Exhibit 1).

Exhibit 1: Primary Practices Eligible for Crediting

Primary Practices	Description
Early drainage/delayed flooding of rice paddies	Shortening the flooding period of the paddies
Alternate wetting and drying (AWD)	Periodically flooding paddies and allowing them to dry has been shown to reduce water applications by roughly 20% and methane emissions by 53%
Direct seeding, or dry seeding, of rice (DSR)	Sowing seeds directly in the paddy under dry conditions, avoiding the flooding required for conventional rice transplanting

Depending on the methodology, projects could be eligible to generate credits from secondary practices (see Exhibit 2). Such changes to farming practices can provide desirable co-benefits to rice farmers, such as reducing the need for fertilizers, saving water, and, in some cases, increasing rice yields.

Exhibit 2: Additional Activities Eligible for Crediting

Secondary Practices	Description
Use of methanotrophs	Applying agricultural inputs with added bacteria that consume methane
Short-duration rice varieties (SDRV) or low-emissions rice cultivars (LERC)	Growing SDRV or LERC, which mature faster than conventional cultivars, reduces emissions by shortening the growing season, meaning less time spent in methane-producing flooded fields
Avoided biomass burning of rice residues	Using rice husks for animal feed, erosion control, or other purposes instead of burning them
Reduction in fossil fuel use	Using manual or renewable energy instead of fossil fuels for farm operations
Improved nitrogen management	Reducing total nitrogen applied to paddies and/or using nitrification inhibitors or slow-release nitrogen fertilizers

One project can implement a combination of practices for combined climate benefit, even when following a single methodology. For a list of eligible activities by methodology, see Appendix.

Verra's large-scale invalidation of rice methane projects began in 2024, after the organization concluded that the projects were not additional, among other concerns. Verra stated that the decision, "was based on serious issues identified during the quality control review related to insufficient demonstration of additionality, projects being designated as small scale, project areas being overstated, and the lack of sufficient evidence to confirm the baseline and project scenario implementation." One independent news source put it more starkly, suggesting that project activities had never occurred at all.

The invalidated projects, which qualified for credits under the CDM AMS-III.AU methodology, claimed to be small-scale, when later reviews showed they were most likely large-scale projects that used the small-scale classification to game the system. The small-scale label allowed individual projects to use simpler additionality tests that they may not have passed if they were properly classified. Verra has since sought compensation for the issued credits and sanctioned the four validation and verification bodies that audited the projects. As of spring 2025, only 597 credits from two US-based projects remain on the market.

The CDM AMS-III.AU methodology has since been inactivated by all major standard bodies. Gold Standard released an updated methodology in 2023, addressing many concerns, but retains the same eligible activities as the original CDM methodology. In turn, Verra released an updated rice methodology in 2025 with more rigorous requirements, which include:

• more robust additionality requirements,

• broader lists of eligible activities to better account for pre- and post-harvest emissions,

• improved guidance for how project plots must be arranged, and

• more dynamic baseline-setting.

Newer methodologies are projected to issue four to six credits per hectare, a figure within the reasonable range of estimates from the academic literature , and substantially lower than the 12 to 15 credits allowed under the CDM AMS-III.AU methodology. The turbulence of available methodologies has stalled the flow of new credits entering the market: as of June 2025, only one project has registered using an updated methodology (VM0051), and it has not yet generated credits. There is no widespread consensus on the delayed adoption of these methodologies, but projects that were previously under development with now-invalidated methodologies are likely transitioning to the updating methodologies and are expected to enter the market once that transition is complete.

This transition coincides with additional competition from recently released holistic agricultural methodologies. These methodologies cover agricultural land management (ALM) practices broadly and include improved rice cultivation practices as activities eligible for crediting. By using ALM methodologies that focus primarily on soil organic carbon (SOC) — but allow reductions from improved rice practices — developers can receive credits for SOC removals and methane reductions, appealing to a growing market appetite for removal projects. Specifically, certain rice programs may gravitate toward a holistic methodology if the rice is part of a mixed rotation. For example, in India and China, where a rice-wheat rotation is common, projects are expected to use a holistic methodology to quantify the SOC associated with wheat. The updated rice methodologies listed below are a better fit for double or triple rice rotations where water management (e.g., AWD) is the core intervention .

*Note: One project is registered under development under the new VM0051, but the project has not yet issued credits.

Please note this technical explainer examines methodologies that focus solely on rice production. Holistic methodologies such as the CAR US Soil Enrichment Protocol, Gold Standard SOC Framework, and Verra's VM0042 Improved Agricultural Land Management allow for the generation of credits from a variety of eligible activities, including rice farming. For example, a project registered under the CAR US Soil Enrichment Protocol has issued credits that are currently on the market from rice farming interventions, which are not reflected above. This project and related credits will be analyzed in a forthcoming technical explainer on regenerative agriculture.

Since nearly all rice credits have been removed from the market, we analyzed the risk scores based on how the new, active methodologies addressed the quality issues that materialized in the invalidated methodologies (see Exhibit 4). Improved rice cultivation projects primarily carry risks regarding baseline setting, additionality, and MRV. There is little to no risk associated with leakage, permanence, or GHG accounting, assuming projects follow updated guidance. (For more information on how we evaluate risk severity and prevalence, see Appendix.)

Exhibit 4: Carbon Credit Risk Matrix for Improved Rice Cultivation

Quality Criteria	Risk Severity	Prevalence
Additionality/Baselines	Medium concern	Common
Leakage	Low concern	Very common
Permanence/Durability	Negligible concern	Rare
GHG Accounting	Low concern	Common
MRV	Medium concern	Common
Socio-Environmental Safeguards	Low concern	Uncommon

Additionality/Baselines : Medium Concern, Common Prevalence

Inaccurate baseline assumptions can lead to overcrediting. Since rice projects credit reductions of emissions from the baseline scenario, accurately estimating that baseline is central to credit quality. CDM's AMS-III.AU methodology allowed projects to assume that rice paddies would have been continuously flooded in the baseline scenario, regardless of the local climate or land management regime. This was an oversimplification that did not reflect real-world environmental variability especially as weather patterns have become harder to predict. Newer, active methodologies mitigate this risk by applying more rigorous baseline-setting approaches, requiring project developers to use models or direct measurements that better reflect the land management and environmental conditions at the project site. These requirements help ensure that rice emissions reductions are based on accurate baselines.

Rice projects have additionality risks, especially in regions with other incentives for conservation practices aside from carbon financing. Even without carbon credits, improved rice paddy management is gaining traction in key rice-growing regions in Southern and Eastern Asia due to increased government support, but adoption of these activities varies. For example, Asia's agriculture sector has long incorporated the strategy of dry seeing, which is currently part of China's 14th Five-Year Plan . However, high rates of AWD and other types of early drainage have not been observed across Asian regions, partially due to:

Since rice agriculture is dominated by small, dispersed producers, adoption trends of practices tend to vary between and within countries.

Projects from regions with high activity adoption rates are at higher risk for nonadditionality. Buyers should take steps to avoid projects from areas where eligible irrigation practices are prevalent unless strong evidence links project activities to carbon credit finance. Because many irrigation practices rely on a system of communal water infrastructure, these water-conservation practices (like AWD or early drainage) are more likely to be implemented without carbon credit finance if supportive infrastructure already exists within the region. Projects implementing dry seeding and complementary practices such as avoided biomass burning can also risk nonadditionality but tend to have fewer barriers to adoption.

Leakage : Low Concern, Very Common Prevalence

In some cases, project activities can lower rice yields, which risks triggering market leakage if other farmers increase production outside the project boundary. This is addressed by all active methodologies, which stipulate that project activities resulting in reduced yields are not eligible for crediting. Projects must have plausible data confirming consistent rice crop yields, and exact yield data is then factored into leakage deduction equations for an added layer of caution.

Activity-shifting leakage is primarily a risk where rice hulls, which some agrarian communities use as a fuel source, are diverted to alternative uses. Using rice husks for purposes other than fuel (e.g., erosion control or livestock feed) can trigger an increasing demand for other fuel sources (e.g., wood or diesel), causing emissions outside the project boundary to increase. Buyers can mitigate leakage risks by investing in credits that require a deduction for activity-shifting leakage.

Permanence/Durability: Negligible Concern, Rare Prevalence

Methane emissions reduced by rice carbon projects are permanent. These projects credit emissions reductions from the business-as-usual scenario, so once eligible irrigation management practices are instituted, emissions from the baseline flooded scenario remain permanently out of the atmosphere.

In terms of durability, farmers may revert to conventional farming activities after carbon crediting ends due to lack of buy-in, changing environmental conditions, market shifts, or an inability to maintain equipment and/or infrastructure. Discontinuation of project activities doesn't impact credits already generated, but it does imperil future reductions.

GHG Accounting: Low Concern, Common Prevalence

The improved rice cultivation credit type has come a long way since the accounting practices in the CDM AMS-III.AU methodology. Several activity types known for their impact on overall rice agricultural emissions (e.g., improved nitrogen management or avoided burning of rice hulls) have been added to eligible crediting practices, leading to more comprehensive emissions accounting. Additionally, methodologies now indicate protocols for nitrous oxide (N₂O) accounting and provide updated guidance on how to omit values if they are insignificant.

However, with these updates, projects still risk inaccurate methane estimates when emissions factors for CH₄ calculations are accepted at the most basic level and lack local activity data. Newer methodologies attempt to balance the benefits of direct measurement and modeling, accepting the use of high-quality models and either of the following, or a combination of the two, for calibration:

The scientific literature suggests that region-specific values are necessary to accurately calculate emissions factors and to calibrate models. When direct measurement is used, it is considered a best practice to compare the collected values to published regional factors (where available) to ensure that field measurements possess the right order of magnitude. Using factors that are as regionally precise as possible is essential to rice projects, as emissions can vary depending on altitude and precipitation levels.

Buyers should prioritize the inclusion of site-level data where possible but understand that the absence of such data does not necessarily mean that a project is low-quality. In some well-researched regions, such as China and India, geographic-specific emissions factors may be easily corroborated by literature. In rice-growing countries where less data is available, such as Laos, reliable regional-level emissions factors can be hard to find, so on-site direct measurement can be more effective in calibrating models.

MRV: Medium Concern, Common Prevalence

Without a fine balance between models and measurement, projects risk inaccurately monitoring methane emissions from rice paddies. As referenced above, it is costly and difficult to continuously measure methane emissions from rice paddies, so methodologies allow for the use of emissions factors. These emissions factors tend to be calculated in one of two ways:

Project-specific emissions factors provide greater granularity but require appropriate sampling and measurement. Conducting closed-chamber measurements of CH₄ requires technical knowledge, which not all projects have access to. Operating a closed chamber without proper expertise can lead to an overestimation of CH₄ reductions, posing a risk for future projects. To address this, the active methodologies have updated their guidelines to chamber collection and often recommend that an external consultant conduct these measurements to ensure accuracy.

Reliance on farmer-reported data without digital MRV, remote sensing, or double-checking typical data ranges increases the risk of inaccurate or unverifiable activity data . Another risk is validating and verifying project activities reported by farmers across the small, scattered paddies where farming occurs. Without digital MRV, projects risk outdated or misleading data management practices and must rely on farmer testimony. For activity data, remote sensing allows field flooding patterns to be validated with greater certainty compared with farmer logs and attestations alone.

Where possible, projects should use digital MRV and remote sensing technologies — such as satellite monitoring, digital logbooks, and automatic sensors — to bolster efficiency. While methodologies are converging to require remote sensing of flooding patterns, farmer logs are still accepted as evidence in some cases, including for projects where no satellite imagery is available to verify when a field was flooded or drained. Methodologies outline that primary data can be captured using a farmer logbook, but the best practice would be to compare such data against a typical range for the given parameter and substitute any out-of-range values with the most conservative value within the predetermined range.

Social and Environmental Safeguards: Low Concern, Uncommon Prevalence

Successful projects proactively plan for and address the points below to help ensure community support, equitable participation opportunities, and skill-building sessions that are well-designed and well-executed:

• Projects that operate with shared cropping and processing systems need mechanisms to ensure community buy-in and trust to perform at their potential.

• Projects should conduct thorough pre-project stakeholder planning to ensure that all community members can participate, rather than defaulting to those who already have larger, more productive farms. A case study from India's Haryana and Madhya Pradesh regions showed that marginalized farmers were underrepresented in carbon farming projects.

• Addressing issues of land tenure, ownership, and socio-economic class increases the likelihood of project success. Including both older and younger farmers in a project aids knowledge transfer and helps overcome cultural barriers to the adoption of more sustainable practices.

Environmentally, midseason and early end-of-season drainage provide benefits in terms of water availability and overall use of pesticides. AWD is primarily known for its water-saving capabilities, reducing irrigation water by up to 20% compared with continuous flooding . This can also contribute to fertilizer savings, as less water may mean fewer losses of water-soluble nutrients.

Catalyzed by market scrutiny, improved rice cultivation credits are making significant strides to increase quality. Rice projects offer a real climate mitigation opportunity because rice paddies are a significant source of global emissions and reducing methane yields outsize near-term climate benefits. Active rice methodologies have taken steps to address prior quality challenges, with more robust additionality requirements, updated accounting and MRV guidance, and a more expansive list of eligible activities, helping to reduce overall risk.

With these advancements, rice projects have the potential to generate quality credits that can help cut emissions from the most methane-intensive food crop globally. They can also enhance water conservation and the livelihoods of rural communities, particularly where farmers and their families play an active role in project design and receive a fair share of project revenues.

ACR – American Carbon Registry
ALM – Agricultural land management
AWD – Alternate wetting and drying
CAR – Climate Action Reserve
CDM – Clean Development Mechanism
CH₄ – Methane
CMI – Carbon Markets Initiative
DSR – Dry seeding of rice, or direct seeding of rice
GHG – Greenhouse gas
IPCC – Intergovernmental Panel on Climate Change
LERC – Low-emissions rice cultivars
MRV – Monitoring, reporting, and verification
N₂O – Nitrous oxide
SDRV – Short-duration rice varieties
SOC – Soil organic carbon
VCM – Voluntary carbon market
VVB – Validation and verification body

Eligible Activities by Methodology

Methodology	Status	Early Drainage/ Delayed Flooding	AWD	DSR	Avoided Burning/ Bailing of Rice Residues	Methanotroph Use	SDRV/LERC	Reduction in Fossil Fuel Use	Improved Nitrogen Management
VM0051: Improved Management in Rice Production Systems	Active (2025)	X	X	X	X	X	X	X	X
Gold Standard: Methane Emission Reduction by Adjusted Water Management Practice in Rice Cultivation	Active (2023)	X	X	X
CAR: US Rice Cultivation Protocol	Active (2013)			X	X
CDM AMS-III.AU: Methane Emission Reduction by Adjusted Water Management Practice in Rice Cultivation	Inactive	X	X	X

Note that standards often allow for biochar application to be accounted for within carbon boundary accounting, but it is not an eligible crediting activity on its own.

CMI reviewed resources from across the VCM and each credit type's sector to develop neutral, succinct risk profiles. We focused on the most crucial risks to quality, their drivers, how standards require projects to mitigate these risks, and how projects could mitigate these risks beyond the standard requirements. We mapped these factors to help stakeholders understand how to think about risk management within a particular credit type.

We looked at both risk severity and prevalence , giving an average score for each based on all identified risks and their mitigation measures.

We define risk severity as the extent to which the identified risks threaten the integrity of the credit type. We reduce risk severity to match the effectiveness and feasibility of the current available and required mitigation measures in the literature. Our risk severity scoring options are as follows:

1. High concern means the identified risks seriously impact project credibility. These risks either lack mitigation measures, or the available mitigation measures are ineffective, or the mitigation measures are difficult for the project developer to access or apply.

2. Medium concern means the risks significantly impact project credibility, and mitigating the risks requires effort — above and beyond basic methodology or standard requirements — on the part of the project developer.

3. Low concern means that the projects have some risk, but these risks are easily mitigated by the project, generally through methodology or standard requirements.

4. Negligible concern means there is little or no risk to a given criterion, and mitigation is not necessary.

5. Not enough information means there is no credible analysis or opinion in the literature on the relevant risks and mitigation measures that would enable an informed decision.

We define risk prevalence as the likelihood that projects within the credit type encounter the risks we identified. To evaluate prevalence, we assessed the frequency with which credits would encounter each risk, regardless of whether the risk was mitigated. While the mitigation measures are incorporated into the severity score, mitigation feasibility and efficacy are not incorporated into the prevalence score. Our risk prevalence scoring options are as follows:

1. Very common means all or most credits in this type will encounter the risks identified (approximately 75% or more of credits in the type).

2. Common means many credits (approximately 30%–75% of credits in the type) will encounter the risks identified.

3. Uncommon means fewer than half of all credits under the credit type will encounter the risks identified (approximately 10%–30% of credits in the type).

4. Rare means none or almost none of the projects/credits in the credit type will encounter the risks identified (approximately 10% or less of credits in the type).

5. Not enough information means there is no credible analysis or opinion in the literature on how often projects encounter these risks that would enable us to make an informed decision.

Within each credit type, there may be multiple methodologies and activity types. We've generated risk profiles that provide the average risk severity and prevalence for the credit type broadly.

Example of Risk Scoring

Let's put this into perspective with an example: many nature-based credit types, like reforestation or grassland management credits, face challenges with nonpermanence, because it is hard to guarantee that carbon will remain stored in trees or soil over time. Factors like wildfires, extreme weather, and pests are often outside a project developer's control, and because these risks are inherent to nature-based projects, a typical assessment might flag permanence as "High Concern" and "Very Common."

Our analysis goes a level deeper, looking at how active methodologies and project developers work to address risks — for example, through buffer pool contributions required by registries, or through additional tools like credit insurance and community-centered project design. If these mitigation measures are strong and widely adopted, we might downgrade risk severity to "Medium Concern," meaning the risk is still serious, but there are effective ways to manage it.