DSR vs AWD: Which Rice Pathway Should an Indian Project Choose?
Direct Seeded Rice and Alternate Wetting and Drying both cut paddy methane, but they change the farm very differently. Here's how the two compare on agronomy, water, labour, risk and carbon — and why many Indian projects combine them.
On this page
Rice is India's largest single agricultural methane source, and it has two established routes to reducing it. They are often mentioned in the same breath, which obscures an important fact: they ask very different things of the farmer. Choosing between them — or sequencing them — is one of the more consequential design decisions in an Indian rice carbon programme.
Quick answer
What's the difference between DSR and AWD?
Both cut methane by reducing the time paddy soil spends waterlogged and anaerobic, but they intervene at different points. AWD keeps the transplanted system and changes irrigation, letting fields dry to a threshold before re-flooding. DSR changes crop establishment itself — seed is sown directly rather than transplanted into standing water, so the field is never puddled and continuously flooded in the same way. AWD is a smaller change layered onto existing practice; DSR is a more fundamental shift with larger agronomic consequences, particularly for weed management.
Why both work
Flooded paddy soil is anaerobic, and anaerobic decomposition of organic matter produces methane. Reduce the time the soil spends in that state and you reduce methane. AWD does it by interrupting flooding during the season. DSR does it by removing the puddling and continuous early flooding that transplanting requires in the first place.
We cover the mechanism, the methodology reset and realistic credit volumes in detail in our guide to AWD carbon credits. This piece is about choosing.
Head to head
| Factor | AWD | DSR |
|---|---|---|
| What changes | Irrigation schedule only | How the crop is established |
| Transplanting | Retained | Eliminated |
| Main new skill | Reading a water-level tube | Weed management without flooding |
| Labour effect | Modest change | Large reduction — no nursery or transplanting |
| Water saving | Significant | Significant, from a different point |
| Land levelling | Helpful | Close to essential |
| Adoption difficulty | Lower — incremental | Higher — systemic |
| Reversibility | Easy to revert | Harder mid-season |
| Main risk | Over-drying if unsupervised | Weed pressure; establishment failure |
The weed problem nobody should skip past
This is the single biggest determinant of DSR success
In a transplanted, continuously flooded system, standing water suppresses weeds. It is doing agronomic work that farmers rarely think about explicitly, because it has always just happened.
DSR removes that suppression. Weed pressure rises substantially, and managing it becomes the main determinant of whether DSR succeeds or fails. That means different herbicide chemistry and timing, more attention in the early weeks, and genuine advisory support.
A programme that promotes DSR for its carbon benefit without funding weed-management support is setting farmers up for a bad season — and setting itself up for the reversal and retention problems described in our piece on permanence and buffer pools.
Yield, honestly
AWD, well implemented, is intended to be yield-neutral. The "safe AWD" threshold — letting the water level fall to around 15cm below the surface before re-irrigating — is deliberately set to avoid drought stress rather than to withhold water aggressively. Most protocols also pause AWD around flowering, when rice is most water-sensitive.
DSR outcomes are more variable. Well-executed DSR with good land levelling, appropriate seed rate and effective weed control can match transplanted yields. Poorly executed DSR can lose yield materially. The variance is real and should be stated plainly to farmers rather than averaged away.
This asymmetry is why many Indian programmes lead with AWD and introduce DSR later, to farmers who have already built confidence and whose fields are level.
MRV: which is easier to prove?
This matters more than it might seem, because a practice that cannot be evidenced generates no credits.
AWD has an unusually clean MRV story. The credit-relevant variable is the water regime, and standing water is directly observable — by water-level tube on the ground and, increasingly, by satellite including radar sensors that see through monsoon cloud. As covered in digital MRV and remote sensing, this is one of the better fits between what a credit depends on and what technology can actually observe.
DSR is evidenced differently — establishment method, sowing records, absence of puddling, and the resulting flooding pattern. It is provable, but it leans more on farmer-level activity records and field verification than on a single continuously observable signal.
For a programme monitoring thousands of fragmented plots, that difference has real cost implications.
Combining them
The two are complementary and are frequently used together: DSR shortens the flooded period at establishment, AWD manages irrigation through the rest of the season. The combination generally delivers a stronger methane reduction than either alone, plus larger water savings.
The caution is human rather than agronomic. Combining them means asking a farmer to change two significant things at once, in a season where a mistake costs them a crop. Phased adoption — AWD first, DSR added once confidence and levelling are in place — is often more realistic than a simultaneous switch, even though it delivers less carbon in year one.
Choosing, in practice
- Start with AWD where farmers are new to the programme, land levelling is variable, or advisory capacity is thin. It is incremental, reversible and easy to evidence.
- Consider DSR where labour cost or availability is a real pressure, land is well levelled, and weed-management support genuinely exists.
- Combine where farmers are experienced, fields are level and the programme can support both.
- Assess the baseline honestly. As with every pathway, credits pay for change from documented existing practice — fields already under DSR cannot be credited again for adopting it.
- Do not quote firm per-hectare credits before assessing the fields. Volumes depend on baseline, soil, season and implementation quality.
Our Rice & Methane Reduction service covers pathway selection, farmer support design and MRV for rice programmes; for the aggregation and benefit-sharing side, see our FPO programme guide.
Running a rice programme, or deciding between pathways? Request a free assessment — we'll give you an honest read on which fits your farmers, not the one that maximises headline credits.
Current as of August 2026. Agronomic outcomes depend on soil, season, water availability and implementation quality. General information only — not agronomic, financial or legal advice.
Frequently asked questions
What is Direct Seeded Rice (DSR)?
DSR means sowing rice seed directly into the field rather than raising seedlings in a nursery and transplanting them into standing water. Because it removes the puddling and continuous early flooding that transplanting requires, the soil spends far less time in the waterlogged anaerobic condition that produces methane. It also cuts the labour and water associated with nursery raising, puddling and transplanting.
How is DSR different from AWD for carbon purposes?
AWD keeps the transplanted system but changes irrigation, letting fields dry to a threshold before re-flooding so the soil periodically becomes aerobic. DSR changes the establishment method itself, so the field is never puddled and continuously flooded in the same way. AWD is a water-management change layered onto existing practice; DSR is a more fundamental change to how the crop is grown, with correspondingly larger agronomic implications.
Which one generates more carbon credits?
It depends on the baseline, the soil, the season and how rigorously the practice is implemented, so no fixed ranking holds across projects. What can be said is that both work by reducing the time paddy soil spends anaerobic, and that they are frequently combined for a stronger combined reduction. Any project quoting a firm credits-per-hectare figure for either practice before assessing your fields is estimating, not measuring.
Is DSR riskier for farmers than AWD?
Generally yes, in the early years. DSR shifts weed pressure substantially, because flooding in a transplanted system suppresses weeds and DSR removes that suppression, making weed management the main determinant of success. It also demands precise land levelling, good seed placement and different herbicide timing. AWD is a smaller change to an existing system, which makes it easier to adopt and easier to reverse if it goes badly.
Can DSR and AWD be used together?
Yes, and combining them is common. DSR shortens the flooded period at establishment, and AWD manages irrigation through the rest of the season. The combination typically delivers a stronger methane reduction than either alone, along with larger water savings. It also increases the agronomic change being asked of the farmer at once, so phased adoption is often more realistic than switching to both in a single season.
Do these practices reduce rice yields?
Well-implemented AWD is generally intended to be yield-neutral, since the drying threshold is set to avoid drought stress rather than to withhold water aggressively. DSR outcomes are more variable and depend heavily on weed control, land levelling and seed rate, with poorly managed DSR capable of losing yield. Any project promoting either practice should be honest that outcomes depend on implementation quality and should support farmers accordingly.
Related reading
AWD Carbon Credits: How Rice Farmers in India Can Earn from Methane Reduction
Alternate Wetting and Drying (AWD) cuts methane from paddy fields and can generate carbon credits for Indian rice farmers. Here's how AWD works, realistic credit volumes, and what a 2024 methodology reset and a farmer-outcomes study mean for you.
What a VVB Actually Checks: Validation and Verification, Explained
No credit is issued until an accredited Validation and Verification Body signs off. Here's what a VVB examines, how validation differs from verification, what audits typically find, and how Indian FPOs and developers should prepare.
Digital MRV and Remote Sensing: What Satellites Can and Can't Measure
Digital MRV is the only affordable way to monitor thousands of fragmented Indian smallholdings. But satellites cannot see below the soil surface. Here's what remote sensing genuinely delivers, and where the overselling starts.
Dr. Anaya Rao
Head of Agronomy & MRV
Anaya leads our agronomy and measurement practice, translating regenerative and water-management science into bankable, audit-ready carbon projects for Indian cropping systems.
- PhD, Soil Science
- 15+ years in Indian agriculture
- Lead author on project MRV protocols