Ethereum Merge Carbon Footprint Reduction
The Merge reduced Ethereum's energy use and CO2e by over 99%, from ~20–24 TWh/10–13M tons to under 3,000 tons annually.
The short answer: Ethereum’s Merge cut its carbon footprint by more than 99%. In September 2022, Ethereum moved from proof-of-work mining to proof-of-stake validation. That shift took annual electricity use from about 20–24 TWh down to a tiny fraction of that, and annual emissions from about 10–13.6 million metric tons of CO2e to roughly 870 to 2,800 tons.
If you just want the main points, here they are:
- Before the Merge: Ethereum used about 20–24 TWh per year in stricter estimates
- Before the Merge emissions: about 10–13.6 million metric tons of CO2e
- After the Merge: electricity use and CO2e fell by more than 99%
- Post-Merge emissions: about 870 to 2,800 tons of CO2e per year
- Why estimates differ: hardware assumptions, node counts, time windows, and grid carbon models
- Main takeaway: Ethereum still uses power, but its footprint is now far smaller than it was under mining
Here’s the quick way I’d frame it: before the Merge, Ethereum used power on the scale of about 2 million to 2.3 million U.S. homes. After the Merge, that mining-heavy model was gone. That’s why most studies land on the same answer even when their exact numbers differ.
| Period | Electricity use | CO2e emissions | Main system |
|---|---|---|---|
| Pre-Merge | ~20–24 TWh/year | ~10–13.6 million metric tons/year | Proof of Work |
| Post-Merge | 99%+ lower | ~870–2,800 tons/year | Proof of Stake |
So if you want the plain-English version, it’s this: the Merge did not make Ethereum zero-carbon, but it cut its emissions by almost all of their old level.
Ethereum Energy Use Before vs After The Merge: 99%+ Carbon Reduction
Pre-Merge Baseline: How Much Energy Ethereum Used
Before the Merge, two sources set the main baseline: CCRI and Cambridge.
CCRI and Cambridge Baseline Estimates
The Crypto Carbon Ratings Institute (CCRI) built a bottom-up model using mining hardware, hash rate, and machine efficiency. Based on the final pre-Merge month, it estimated 22.9 TWh of electricity use per year and about 11 million metric tons of CO2e on an annualized basis. Across the full period from Aug. 1, 2021, to July 31, 2022, CCRI's estimate was 24.10 TWh and 13.64 million metric tons.
Cambridge put Ethereum's pre-Merge power demand at about 2.4 GW, which works out to roughly 20–21 TWh per year. Its emissions estimate came in near 10.3 million metric tons of CO2e.
Those numbers don't match line for line, but they're in the same ballpark. And that's the key point.
Higher Tracker Estimates and Why They Vary
Some public trackers and reports placed Ethereum's pre-Merge electricity use much higher, around 70–80 TWh.
Why such a big gap? It came down to the inputs. Higher estimates often leaned on assumptions like:
- older hardware
- device power close to peak levels
- broad grid emissions factors
- extrapolation from peak hash rate
Without detailed data on hardware mix, uptime, and miner location, those models pushed Ethereum's use too high. That's why the stricter estimates from CCRI and Cambridge sit well below the biggest public tracker numbers.
U.S. Scale Comparisons for Context
Using CCRI's 22.9 TWh figure and the U.S. Energy Information Administration's estimate that a typical household uses about 10–11 MWh per year, Ethereum's pre-Merge network used about as much electricity as 2 million to 2.3 million average U.S. homes in a year.
Its emissions, at roughly 10–11 million metric tons of CO2e, were also on the scale of the annual emissions from more than 2 million gasoline cars.
These figures give you the baseline for the post-Merge drop that comes next.
sbb-itb-0796ce6
Post-Merge Results: Studies Show a 99%+ Drop
After that pre-Merge baseline, the picture changed fast. Post-Merge studies show a steep drop in both electricity use and emissions. Across the research, the takeaway is the same: Ethereum's energy use and emissions fell by more than 99%.
CCRI Findings on Electricity and CO2e
CCRI's post-Merge analysis found that Ethereum's electricity demand and CO2e emissions fell by more than 99% compared with the pre-Merge network.
Cambridge and Other Confirming Reports
Cambridge and other reports reached the same conclusion.
What the New Footprint Looks Like in Practice
Post-Merge Ethereum now runs with a much lower continuous power demand and no longer depends on industrial-scale electricity. That's the big shift. Ethereum is now validated by staked nodes instead of miners, so its footprint is tiny next to the pre-Merge network.
That also explains why per-transaction comparisons can be misleading. They miss the network-level footprint, which is what researchers focused on when measuring the change.
The next section breaks down how researchers measured that reduction.
How Researchers Calculated the Reduction
CCRI and CCAF both used a bottom-up method to estimate Ethereum's post-Merge footprint. They looked at node and hardware power use first, then built up from there. That's why their estimates line up well enough to compare.
Bottom-Up Node and Validator Measurements
Researchers tested actual hardware running Ethereum client software. CCRI found that paired execution and consensus clients, such as Prysm and Geth, used about 9% less power than adding up separate measurements would suggest. That matters because it shows the setup behaves a bit more efficiently in practice than it does on paper.
CCAF took a slightly different route. It measured about 20 client-hardware combinations, found an average of 105 W per full node, and then scaled that figure across 8,522 reachable full nodes.
After that, the job shifted from power use to emissions.
Network-Level Models and Grid Assumptions
Electricity use doesn't tell the whole story by itself. To turn it into CO2e, researchers mapped nodes to regions and applied local grid carbon factors. In plain English: a node running in one place can lead to different emissions than the same node running somewhere else, because local power grids vary.
CCAF used node peer-to-peer communication data to estimate where nodes were located. CCRI used carbon-intensity benchmarks from earlier research to convert MWh into tons of CO2e.
That led to a small spread in the final numbers. CCRI estimated annual emissions at about 870 tons of CO2e, while CCAF estimated 2,370-2,800 tons. Even with that gap, both estimates point to the same big takeaway: emissions fell by more than 99% from the pre-Merge baseline.
Why the difference? Mostly these factors:
- Node counts
- Measurement windows
- Grid assumptions
Limits of Per-Transaction Comparisons
Post-Merge Ethereum power use is based on the network itself, not on each transaction. A node keeps using roughly the same amount of power whether activity is busy or quiet. So if you divide annual energy use by annual transaction count, you get a number that looks neat and exact, but it can give the wrong impression.
That's also why institutional summaries caution against leaning on per-transaction metrics in ESG reporting.
These measurement choices explain why the Merge's carbon change is treated as a network-level shift, not a per-transaction one.
What the Merge Means for Ethereum's Carbon Profile
For readers, the main point is simple: the Merge cut Ethereum's carbon footprint by more than 99%.
Why These Findings Matter for New Crypto Users
That drop matters because it changes the sustainability trade-off for everyday Ethereum use. This makes it easier to buy crypto without the heavy environmental cost. Ethereum now runs on a much smaller, validator-based footprint.
Before the Merge, the network relied on miners. After the Merge, it moved to validators. That change is why Ethereum's carbon profile fell so sharply.
Ethereum's footprint is not zero, though. Validators still use electricity. What remains depends mostly on where validators are located and what kind of power grid they use.
Key Numbers to Remember
More than 99%, with some summaries landing near 99.95%.
The bottom line: the Merge left Ethereum with a tiny ongoing footprint, not a zero one. That makes it far lower-emission than the proof-of-work network it replaced.
FAQs
Why did the Merge cut emissions so sharply?
The Merge in September 2022 cut Ethereum’s carbon emissions in a dramatic way by moving the network from Proof of Work to Proof of Stake.
That shift got rid of energy-hungry mining and the heavy computing load that came with it. As a result, Ethereum’s energy use fell by an estimated 99%.
Why do post-Merge carbon estimates still differ?
Post-Merge carbon estimates still vary because researchers don’t all measure indirect emissions the same way in a decentralized system.
The final numbers can shift based on a few things: hardware efficiency, the carbon intensity of local power grids, and whether off-chain Layer 2 activity is counted. So yes, Proof of Stake cut Ethereum’s energy use by over 99%. But the exact carbon footprint can still look a bit different from one study to another, depending on how that study does the math.
Is Ethereum now carbon-neutral?
No. The Ethereum Merge cut the network's energy use by more than 99% by moving to a proof-of-stake consensus mechanism, but Ethereum still uses energy.
So while its impact on the planet is far lower than it was before, its carbon footprint didn't drop to zero.