Energy Consumption and CO₂ Emissions for Cryptocurrency Networks

Still need help? 

New regulations like the EU’s Markets in Crypto-Assets (MiCA) require crypto platforms to disclose the environmental impact of each token they offer. In practice, this means you as an end user will see figures for electricity consumption (in kWh) and associated CO₂ emissions for tokens. This transparency helps you understand the climate impact of different cryptocurrencies and comply with sustainability expectations. Below, we explain our methodology in simple terms, so you know exactly what these numbers mean and how we arrive at them.

Under MiCA, any crypto service provider in the EU must prominently display the annual energy usage of every crypto-asset they support. The goal is to let investors and users easily compare the environmental footprint of different coins. For example, you’ll be able to see how much electricity a proof-of-work token like Bitcoin Cash consumes versus a low-energy token like Stellar Lumens. This matters for anyone concerned about climate impact or complying with ESG standards. In short, MiCA empowers you with data – similar to nutrition labels on food – showing the “energy diet” of your crypto investments. Since the end of 2024, this disclosure is mandatory for EU-facing platforms, reflecting a broader push to highlight climate and environmental impacts in finance.

For tokens exceeding a certain energy consumption threshold (500 kWh/year), MiCA further mandates the disclosure of five specific sustainability indicators, including energy and emissions intensity, and the share of renewable energy used

Not all blockchains work the same way – and their consensus mechanism (how they secure the network) largely determines energy consumption.

> Proof-of-Work (PoW): Used by Bitcoin and Bitcoin Cash (BCH), PoW relies on miners racing to solve puzzles, which demands enormous computing power (and thus electricity). PoW networks tend to consume gigawatts of power. For instance, Bitcoin is estimated to use around 149 TWh (terawatt-hours) of electricity per year  – on the order of a small country’s energy usage. BCH, which uses similar mining, consumes a fraction of Bitcoin’s amount (since BCH has much less mining activity) but is still in the range of hundreds of gigawatt-hours annually (see comparison table below). All that energy use also means higher carbon emissions if the electricity isn’t green.

> Proof-of-Stake (PoS) and other Low-Energy Systems: Networks like Ethereum (post-Merge), Avalanche (AVAX), Stellar (XLM), and Toncoin (TON) use non-PoW methods (stake or voting-based consensus) that are far more energy-efficient. Instead of competitive mining, a limited set of nodes (validators) confirm transactions, so electricity use stays very low. For example, Ethereum’s switch from PoW to PoS in 2022 slashed its energy draw by ~99.9% – from an estimated ~21 TWh/yr down to only ~0.0026 TWh (2,600 MWh) per year . In real terms, the entire Ethereum PoS network now uses roughly the same energy per year as 2–3 thousand U.S. homes, instead of millions previously. Likewise, Stellar’s network consumes on the order of 0.48 GWh (480,000 kWh) per year  – a tiny footprint – thanks to its efficient consensus protocol. Studies have found that modern PoS chains use only a minute fraction of the energy of PoW chains; for example, Avalanche uses just 0.00034% of the energy Bitcoin uses. In short, consensus type is the biggest factor: PoW tokens are energy-intensive, whereas PoS and similar tokens are comparatively “green” in their energy usage.

For crypto-assets identified as consuming more than 500 kWh of electricity per year, the EU’s Markets in Crypto-Assets (MiCA) regulation, guided by ESMA’s technical standards, mandates the disclosure of specific sustainability indicators. Our methodology for deriving these indicators is as follows:

1. Total Annual Energy Consumption (kWh/year)

Methodology: This is calculated as described in the "How We Calculate Energy & Emissions" section, differentiating between Proof-of-Work (PoW), Proof-of-Stake (PoS), and other consensus mechanisms, as well as tokens on existing chains.
Data Sources: As detailed previously (e.g., network hash rates, mining hardware efficiency, validator node counts, energy use per node from studies like CCRI, project-specific reports).
Threshold Trigger: This full set of indicators is reported for any token whose estimated Total Annual Energy Consumption exceeds 500 kWh.

2. Total Annual Greenhouse Gas Emissions (tCO₂e/year)

Methodology: Calculated by multiplying the Total Annual Energy Consumption (kWh/year) by our standard carbon intensity factor (currently ~0.475 kg CO₂ per kWh, representing a global average electricity grid mix), as detailed in the "Converting Electricity to CO₂" subsection. This yields emissions in metric tons of CO₂ equivalent per year.
Data Sources: Derived from the energy consumption estimate and the selected carbon intensity factor (based on IEA or similar global averages, unless more specific data is available and verifiable for a particular network).

3. Energy Consumption Intensity

This indicator provides context to the total energy consumption relative to the scale of the crypto-asset. We will calculate this in two ways, where feasible:
* (a) Per Unit of Crypto-Asset (kWh / unit):
* Methodology: Total Annual Energy Consumption (kWh/year) / Total Circulating Supply of the Crypto-Asset.
* Data Sources for Supply: Circulating supply data will be sourced from reputable public data aggregators (e.g., CoinGecko, CoinMarketCap APIs) or directly from the crypto-asset project's official explorers or APIs, cross-referenced where possible. The "as-of" date for the supply figure will be recorded.
* (b) Per Unit of Market Value (kWh / EUR):
* Methodology: Total Annual Energy Consumption (kWh/year) / Total Market Capitalization of the Crypto-Asset (in EUR).
* Data Sources for Market Value: Market capitalization data will be sourced from reputable public data aggregators, using an average value over a defined recent period (e.g., 30-day average) or a snapshot concurrent with the supply data. The "as-of" date and currency conversion methodology will be noted.

Presentation: We may present one or both intensity metrics, clarifying the denominator used.

4. Greenhouse Gas Emissions Intensity

Similar to energy intensity, this contextualizes the total emissions.
* (a) Per Unit of Crypto-Asset (tCO₂e / unit):
* Methodology: Total Annual Greenhouse Gas Emissions (tCO₂e/year) / Total Circulating Supply of the Crypto-Asset.
* Data Sources for Supply: Same as for Energy Consumption Intensity.
* (b) Per Unit of Market Value (tCO₂e / EUR):
* Methodology: Total Annual Greenhouse Gas Emissions (tCO₂e/year) / Total Market Capitalization of the Crypto-Asset (in EUR).
* Data Sources for Market Value: Same as for Energy Consumption Intensity.

Presentation: We may present one or both intensity metrics, clarifying the denominator used.

5. Percentage of Energy from Renewable Sources (%)

This is the most challenging indicator to estimate accurately from solely public information without direct disclosures from network operators (miners/validators).

Methodology & Approach:

Primary Approach (Project-Specific Data): We will prioritize and report figures on renewable energy usage that are publicly disclosed by the crypto-asset project foundation, development team, or through verifiable third-party audits specifically addressing the energy mix of their network's transaction validation/mining. Such data must be clearly sourced and recent. (e.g., if Stellar Development Foundation reports X% renewables based on a study of their validators, this would be used).

Secondary Approach (Miner/Pool Disclosures for PoW): For PoW crypto-assets, if significant mining pools or publicly known large-scale miners disclose their renewable energy usage and their share of the total network hash rate is substantial and verifiable, this data may be used to estimate a portion of the network's renewable energy use. This will be clearly caveated due to the dynamic and often opaque nature of mining operations.

Tertiary Approach (Geographic Distribution Estimates - Exploratory, High Uncertainty):
If reliable data on the geographic distribution of a significant portion of PoW hash rate or PoS validators becomes available (e.g., through network analysis tools, academic research, or aggregated self-reported data), we may attempt to estimate the renewable energy percentage by correlating these locations with the known renewable energy generation percentage in the electricity grid mix of those regions (using IEA or national energy statistics).

Caveat: This approach carries high uncertainty due to:
Difficulty in accurately pinpointing validator/miner locations.
Miners/validators in a region do not necessarily use that region's average grid mix (they might have specific Power Purchase Agreements for renewables, or use off-grid sources).
Data on geographic distribution is often partial or quickly outdated.

Default Position (If No Verifiable Data): If no sufficiently reliable or verifiable data from the above approaches is available, we will state: "The percentage of renewable energy used by this crypto-asset network is not determinable from publicly available information."
In some limited cases, for illustrative purposes or if required to provide a figure, we might reference the global average share of renewables in electricity generation (as per IEA, currently around 29-30%) as a baseline, but will explicitly state this is a global average and not specific to the crypto-asset, carrying a very low degree of confidence for the specific network. This is a last resort and will be clearly marked as a general reference rather than a network-specific estimate.

Data Sources: Project sustainability reports, third-party audits, reputable mining pool disclosures, academic research on validator/miner geography, IEA reports on regional/global grid mixes.

Transparency: We will be highly transparent about the source and confidence level for any renewable energy percentage reported.

Our methodology combines industry best practices and the latest research to estimate each token’s annual electricity consumption and its carbon footprint. We tailor the approach based on the token’s technology:

> Proof-of-Work tokens (e.g. BCH): For PoW assets, we track the network’s total hash rate (computing power) and use it to estimate electricity use. Essentially, we consider what kinds of mining machines are likely in use and their efficiency (joules per hash). By multiplying the total hashes computed by the energy per hash of typical hardware, we get the network’s yearly electricity consumption. This approach is similar to the Cambridge Bitcoin Electricity Consumption Index model , adjusted for each token’s mining network. For Bitcoin Cash, which shares its mining algorithm with Bitcoin, we start from Bitcoin’s known energy draw and scale it down proportional to BCH’s much lower hash rate (only a tiny fraction of Bitcoin’s)  . This yields BCH’s estimated annual energy usage (shown in the table below). We continually update these estimates as mining power or efficiency changes.

> Proof-of-Stake and other non-PoW tokens (e.g. AVAX, XLM, TON): For PoS networks, we estimate energy use by counting active validator nodes and estimating the power each node uses. Most validators are run on standard servers or cloud instances, which typically draw on the order of tens of watts of power per node. We use data from studies (for example, CCRI’s research) that measured node power consumption for networks like Ethereum and Avalanche  . As a rough guide, a single validator might use ~50–100 watts continuously, which translates to ~0.4–0.9 MWh (megawatt-hours) per year. We multiply a per-node estimate by the number of validators to get the network total. For instance, Avalanche has around 1,200 validators and consumes roughly 470 MWh/year in total , which aligns with measured results. Toncoin (TON), with ~349 validators , will have a lower total (we estimate on the order of 100–150 MWh per year for TON). And Stellar, which uses a federated consensus with a few dozen core nodes plus support nodes, has been externally audited to use about 481 MWh/year . Whenever possible, we incorporate such published figures from foundations or third-party audits to improve accuracy.

> Tokens on existing chains (e.g. Chainlink’s LINK on Ethereum): If a token doesn’t run its own blockchain but uses another network, we attribute the underlying network’s energy usage to it (as per MiCA’s requirement to base the calculation on the distributed ledger it transacts on). For example, Chainlink is an ERC-20 token relying on Ethereum. So, we report Chainlink’s footprint based on Ethereum’s network consumption. (The Chainlink oracle nodes do consume some additional energy off-chain, but it’s relatively negligible compared to the Ethereum network itself.) In Ethereum’s case, that’s roughly 2,600 MWh of electricity per year for the whole network, which corresponds to about 870 tonnes of CO₂ emissions annually . We apply the same principle for any token built on a parent chain.

> Converting Electricity to CO₂: Once we have the annual electricity consumption (in kWh or MWh), we estimate carbon emissions by multiplying by a carbon intensity factor (kg CO₂ per kWh). We use a global average emissions factor for electricity. For transparency, our default factor is around 0.475 kg CO₂ per kWh (475 g/kWh) , which is in line with studies and roughly represents the world’s average grid mix. In practice, this means 1 MWh (1,000 kWh) of electricity is estimated to produce 0.475 tonnes of CO₂. If there’s specific data indicating a network’s miners or validators use greener energy (or more carbon-intensive energy), we can adjust the factor accordingly. But lacking specific info, an average factor ensures we don’t underestimate the footprint. All CO₂ figures we present are annual and measured in metric tons of CO₂ equivalent (tCO₂/yr).

Data updates:

We update the energy and CO₂ numbers at least annually (as MiCA mandates ) or more frequently if major changes occur. For example, if a network undergoes a big upgrade (like Ethereum’s Merge) or if its activity spikes or drops significantly, we recalc the estimates to keep information current. We also note the sources of all data points (hash rates, node counts, etc.) so you can see where the numbers come from.

When you see a token’s energy use and CO₂ emissions on our site, it’s telling you the scale of that network’s environmental impact per year. These figures are for the entire network supporting the token:

> Electricity Use (kWh/year): This is how much electrical energy the network consumes in total to process transactions and secure itself over a year. A higher number means the network draws more power from the grid. For context, a typical European household might use around 3,500–4,500 kWh in a year. So if a network uses 350,000 kWh (350 MWh), that’s roughly equivalent to 100 homes’ annual electricity. In our list, you’ll notice huge differences: for example, Bitcoin Cash consumes on the order of hundreds of millions of kWh (comparable to tens of thousands of homes worth of electricity), whereas Stellar or Avalanche use only a few hundred thousand kWh (more like tens of homes worth). In fact, Avalanche’s entire network consumption (~469.8 MWh/year) is about the electricity usage of 44 U.S. households for a year ! Lower energy usage generally implies a smaller environmental footprint.

> CO₂ Emissions (tCO₂/year): This indicates the estimated carbon footprint associated with that electricity use over a year. It depends on how the electricity is generated. We express it in metric tons of CO₂. For a mental picture, 1 tonne of CO₂ is roughly the emissions from driving an average gasoline car about 4,000 km. So if a network emits 50 tCO₂/yr, that’s like 200,000 km driven by cars in total. When comparing tokens, the one with higher tCO₂/year is contributing more to climate change (unless offset by renewables or carbon credits). For example, Stellar’s yearly emissions (~173 tCO₂ ) are extremely low in the crypto world – that’s about the footprint of 30 average Americans for one year, which is minor for a global network. Meanwhile, Bitcoin Cash’s emissions are estimated in the hundreds of thousands of tonnes range – a figure more on par with a small industrial sector. Such a stark contrast highlights how much more carbon-intensive proof-of-work mining is versus efficient networks.

How to use this info: If you’re an individual investor, these numbers can guide you toward more sustainable choices (e.g. preferring a low-energy network for certain applications). If you’re a professional or institution, you might use the data for ESG reporting or to ensure compliance with regulations. Keep in mind that a lower energy/CO₂ number doesn’t necessarily reflect the project’s quality or utility – it’s purely an environmental metric. But it does show, objectively, the environmental cost of running that blockchain. In combination with other factors (like performance, security, decentralization), energy and carbon data help form a holistic view of a crypto asset’s impact.

Our methodology strives for accuracy and transparency, but there are important limitations to be aware of:

> Estimates, not direct measurements: The figures we provide are modeled estimates. No one has a giant electricity meter on a blockchain, so we rely on proxies (hash rate, node counts, hardware specs, etc.). Real-world usage can fluctuate. For example, miners might switch hardware or change locations, and node operators might upgrade servers. These changes could alter energy use. We do our best to keep up, but there’s always a margin of error. Think of our numbers as the best order-of-magnitude estimates based on available data, rather than exact readings.

> Carbon intensity varies by region: We use an average emissions factor (as noted, ~0.475 kg CO₂ per kWh) to calculate CO₂, but in reality the carbon emitted per kWh can range widely. If mining is done in a coal-heavy region, the true CO₂ per kWh could be higher (for instance, coal-dominated grids can emit ~700+ g CO₂ per kWh ). Conversely, if validators use mostly hydro or solar power, the real emissions could be much lower (some grids are as low as 25 g CO₂/kWh). Our use of a general factor means actual emissions might be over- or under-estimated for a specific network. Unfortunately, data on the exact energy sourcing for miners/validators is limited. We assume a typical mix, but the reality could be greener (or dirtier). Interpret the CO₂ numbers as rough indicators, not precise totals.

> Scope of energy use: We focus on the direct network operations. That is, the electricity used by miners and nodes to run the network. We do not currently include indirect or upstream energy costs (like manufacturing mining rigs, cooling data centers, etc.). Some studies consider those factors in an overall “lifecycle” assessment, but such data is highly variable and beyond the scope of MiCA’s disclosure requirements. By keeping to operational energy, we ensure we’re comparing like-for-like across different projects. However, it’s good to know that the true environmental impact includes more than just running the nodes (much like an electric car’s footprint includes manufacturing the car, not just the electricity it uses).

> Data sources and updates: We rely on third-party data sources for many inputs. Hash rates come from public blockchain explorers or indices; node counts might come from project websites or network scanners and public data aggregators (e.g., CoinGecko, CoinMarketCap) for metrics like circulating supply and market capitalization. We cite sources (see next section) and choose reputable data, but if those sources have inaccuracies, our results will inherit them. We also update our figures periodically (at least annually). There may be times when the network has changed (grew or shrank) and our displayed number is a bit out-of-date – please check the “last updated” timestamp we provide. We encourage readers to refer to our sources or reach out with new data. Transparency is a priority: all our assumptions (e.g. hardware efficiency used, carbon factor, etc.) are documented either on this page or in linked references. 

> Comparability: Be careful when comparing our numbers with figures from other reports. Methodologies can differ. For example, one report might claim “Network X uses Y kWh per transaction” – that’s a per-transaction metric, which can be misleading because it averages the total energy over transaction count (which can fluctuate). Our data is total annual energy, as required by regulation . It gives a sense of scale of the network, but it’s not the same as an efficiency per transaction or per user. If you come across other estimates, check if they cover the same scope (whole network vs. per txn vs. per market cap). In general, our approach aligns with the forthcoming standard format regulators want: total yearly consumption in kWh, plus related indicators.

In summary, while the numbers we provide are grounded in the best available research, they are still approximations. They are very useful for relative comparisons (you can trust who’s larger or smaller in footprint), but don’t take a difference of say 5% too strictly given the uncertainties. We intend this data to foster understanding and responsible choices, and we continually refine the methodology as better information comes to light.

We base our methodology on established research and data from experts. Here are some key sources that inform our calculations and ensure factual accuracy:

> EU MiCA Regulation and ESMA Reports: The requirement for sustainability disclosures comes from MiCA Article 6 and related provisions. ESMA’s final report on crypto sustainability indicators outlines the standards we follow. These documents specify what needs to be reported (e.g. annual kWh) and how often to update it. We adhere closely to these regulatory guidelines.

> Cambridge Centre for Alternative Finance (CCAF): CCAF provides widely-cited data on Bitcoin’s energy consumption and methodology for PoW estimation. For instance, Cambridge’s index shows Bitcoin consuming on the order of 100–150 TWh/year . We use similar techniques (hash rate and efficiency modeling) for PoW tokens. Cambridge’s research (and tools like the Cambridge Bitcoin Electricity Consumption Index) set a benchmark for transparent crypto energy analysis.

> Crypto Carbon Ratings Institute (CCRI): CCRI is a research institute that has published detailed studies on the energy use of various blockchains, especially proof-of-stake networks . Their 2023 benchmarking report measured networks like Avalanche, Algorand, Solana, etc., giving us solid reference points for how much power a typical PoS validator uses and the total footprint of those networks. For example, CCRI reported Avalanche’s network uses ~469.8 MWh and 178 tCO₂ annually , which we cite and cross-verify with our own calculations. We consider CCRI’s methodology (bottom-up node measurement) as a validation for our PoS estimates.

> Project Sustainability Reports: Where available, we incorporate data published by the projects themselves or independent audits. A great example is the Stellar Development Foundation’s environmental footprint assessment (conducted with PwC)  . Stellar publicly shared that their network used about 481,324 kWh and ~173,000 kg CO₂ per year  . We used those figures directly for XLM since they were carefully measured. Similarly, Ethereum’s foundation has published energy and emissions estimates post-Merge , which we use as the basis for Ethereum (and Ethereum-based tokens like LINK). Whenever a network publishes vetted sustainability data, we prefer to use it (and we’ll cite it) instead of reinventing the wheel.

> Other Academic and Industry Research: We stay up-to-date with studies from universities, think tanks, and industry groups. This includes papers in journals (for example, the MDPI Energies journal review on cryptocurrency energy use) and analyses by organizations like the International Energy Agency (IEA) or the U.S. DOE. These help us understand trends like the average carbon intensity of electricity  or the range of hardware efficiency. They also provide context (e.g., comparisons of crypto energy use to other industries) that ensure our methodology is grounded in a broad perspective.

Every number we present is backed by one or more of the above sources. We include inline citations (like those you’ve seen in this text) so you can click and verify the info. For further reading or to dive deeper, we encourage you to check out those references. Transparency and accuracy are core to our approach – we want you to trust the data, and that’s why we show you where it comes from.