Original article: https://www.sciencedirect.com/science/article/pii/S2666188825003570

Executive Summary

The article’s claim that Bitcoin mining has a massive, growing carbon footprint is undermined by multiple flaws. It relies on outdated or discredited sources, uses obsolete data, and considers only the negative side (CO₂ emissions) without any balancing benefits. For example, the paper assumes Kazakhstan still accounts for ~13% of global mining (based on 2021 figures), yet by mid-2023 its share had fallen to ~4%. Cambridge Centre for Alternative Finance (CCAF) data show updated emissions are much lower than older estimates: about 39.8 MtCO₂e/year (~0.08% of world emissions), 40–45% below earlier model projections. Crucially, the analysis omits positive externalities (see below), making its conclusion one-sided. In sum, using outdated inputs and a narrow scope yields a significantly overstated view of Bitcoin’s carbon impact. For this reason we recommend policy makers defer to Cambridge’s 4th mining report, which does not have these methodology defects and uses contemporary data, for guidance on bitcoin carbon  impact.

One-Sided Methodology

The authors count only gross CO₂ emissions as the “environmental cost,” ignoring any offsets or benefits (Bashari et al., 2025). This skewed approach all but guarantees a negative result.  A balanced analysis would consider both emissions and mitigations. For instance, if miners power operations with electricity that would otherwise be wasted or flared, net emissions can be much lower or even zero. Cambridge (2025)notes that accounting for the use of flared methane (instead of venting it) could reduce Bitcoin’s emissions estimate by ~25%. By contrast, the article leaves out such offsets entirely. In effect, it asks “how bad is Bitcoin mining if we count no benefits?”, an obviously biased question that overstates harm by its nature.

Errors in Literature Referencing

Many sources cited in the article are now obsolete or refuted. For example, it repeatedly cites Mora et al. (2018), a commentary claiming Bitcoin growth could warm the planet >2°C. That projection was immediately debunked in Nature Climate Change (2019) as based on impossible adoption scenarios. Similarly, the article quotes Alex de Vries’s Digiconomist energy-use and e-waste figures, which are now contradicted by real data (Sai & Vranken, 2023). De Vries had predicted ~30,000 tonnes of annual ASIC e-waste, but the Cambridge 2025 survey finds only ~2.3 kilotonnes in 2024. This large gap (2.3 kt vs. 30 kt) arises because modern miners reuse and recycle old rigs and have longer hardware lifespans (Collins & Dewhurst, 2024). By treating these discredited claims as authoritative, the paper misleads readers. In contrast, more recent studies emphasise caution with metrics like “per transaction” energy, (Carter, 2021) others have shifted to updated models showing lower footprints (WooCharts, 2025). In short, the article recycles outdated worst-case figures while overlooking current evidence.

Data Inaccuracies

The paper’s emissions numbers rely on outdated data. It assumes country-level mining shares and grid mixes from the 2019–2021 era. For instance, Kazakhstan is treated as a coal-heavy 13% of global hashrate, but by May 2023 its share was only about 4% (Mellerud, 2023). (China’s 2021 mining ban similarly invalidated any high Chinese share.) Cambridge’s researchers themselves warn that using a 2022 mining map (the latest available data) likely overestimates emissions by roughly 43%. In short, outdated inputs (like old energy mixes or hardware efficiency) produce “garbage in, garbage out.” The result is an inflated CO₂ total that does not reflect today’s mining landscape.

Omission of Positive Environmental Externalities

The analysis completely omits documented ways that Bitcoin mining can reduce emissions:

Methane mitigation: Miners can use otherwise-flared natural gas to generate power, capturing methane. Incorporating this effect significantly cuts net emissions (Rudd et al., 2024). (Cambridge estimates that 888 GWh of mining load was deliberately curtailed in 2023 for grid balancing, some of which involved using wasted gas.)

Grid balancing: Bitcoin miners are highly flexible electricity consumers. When they throttle down during peak demand or storms, they relieve grid strain. For example, Texas miners cut about 1.4 GW during a 2022 winter storm, stabilizing the grid. Cambridge (2025) reports miners voluntarily curtailed 888 GWh in 2023 to help balance supply.

Renewable development: By creating off-take revenue, mining can make wind/solar projects viable. A Cornell study found that mining revenue during pre-commercial phases can provide millions in profit for new wind and solar farms (Cornell Chronicle, 2023). This incentive can spur additional renewable capacity.

Ignoring these factors renders the analysis incomplete. It is analogous to critiquing solar energy by measuring only the pollution from panel manufacturing while ignoring decades of clean power. Without accounting for offsets and benefits, the conclusion that “mining = net emissions” is trivial and misleading.

Comparison with Updated Industry Data

Recent data paint a far less alarming picture. Cambridge’s April 2025 industry survey (covering ~48% of global mining) finds 52.4% of Bitcoin mining energy is from sustainable sources (42.6% renewables, 9.8% nuclear (Cambridge Centre for Alternative Finance, 2025). Coal use has collapsed (to just ~8.9%) while natural gas is now ~38.2% . These numbers contrast sharply with older models based on 2018 conditions (when ~60–70% of mining was coal). In absolute terms, the Cambridge report estimates Bitcoin’s annual emissions at ~39.8 MtCO₂e (~0.08% of global) – far below the ~65–70 Mt figures cited by outdated models. (Even that 39.8 Mt could be as low as ~33 Mt if flared gas use is counted.) Independent analyses find the same trends: over half of mining power is now clean and carbon intensity is falling. In short, multiple sources now show Bitcoin’s total emissions have remained roughly flat since 2019 despite a large hashrate increase, thanks to efficiency gains and cleaner energy (WooCharts, 2025). This up-to-date evidence contradicts the article’s portrayal of a runaway, coal-driven footprint.

Methodology Critique

Beyond one-sided scope, the article’s methodology has other defects:

No net-accounting: It calculates only gross emissions (energy consumption × average grid CO₂) without any offsets. All mining electricity is treated as “average” grid power, even though many miners use excess renewables or wasted gas (Rudd & Porter, 2024). This ignores the very effects discussed above.

Static assumptions: The model appears to fix mining locations and hardware in time. In reality, mining is dynamic: hardware efficiency improves ~20–30% per year, and miners relocate or adjust load constantly (Collins & Dewhurst, 2024). A robust analysis would test these variables. Instead, the paper essentially freezes 2019–2021 conditions, skewing results high.

Averages and proxies: Using national-average emission factors overstates emissions if miners actually tap cleaner local sources. (For example, labelling a certain % of hashpower in “coal-heavy” countries doesn’t mean miners themselves use coal plants.) The paper makes no such refinements. It also ignores demand-response arrangements: grid operators often treat mining as adjustable load, but the study’s annual averages miss this effect.

Each of these choices tends to inflate emissions. In sum, while the arithmetic (energy × factor) is simple, the authors’ selective framing (worst-case inputs, no offsets, no dynamics) violates best practices. Recent reviews urge exactly the opposite: to incorporate energy source differences and system response. By failing to do so, the methodology yields little insight into Bitcoin’s actual net impact.

Policy Implications

Given these flaws, the paper’s findings are unsuitable for policy guidance. Relying on such one-sided, outdated analysis could lead to misguided regulation – for example, blanket bans that accidentally stifle climate-beneficial innovations (like methane flaring projects). Policy experts warn of “runaway” citation chains where old data distort narratives. Sound policy requires the latest, balanced data on both costs and benefits. We therefore urge policymakers to consult current, peer-reviewed sources and industry reports (e.g. Cambridge 2025, IEA/OECD data) for a nuanced view. These show Bitcoin mining is evolving toward higher efficiency and more clean energy use (Naueihed, 2023). In short, Bitcoin’s environmental impact is complex; simplistic analyses based on outdated assumptions risk confusion and poor decisions, whereas up-to-date, comprehensive studies provide a far more accurate basis for policy.

References:

Bashari, M., Ghavidel Doostkouei, S., Fathabadi, M., & Soufimajidpour, M. (2025). The environmental cost of cryptocurrency: Analyzing CO2 emissions in the 9 leading mining countries. Sustainable Futures, 10, 100792. https://doi.org/10.1016/j.sftr.2025.100792

Cambridge Centre for Alternative Finance. (2025, April). Cambridge Digital Mining Industry Report: Global operations, sentiment, and energy use. Cambridge Judge Business School. https://www.jbs.cam.ac.uk/wp-content/uploads/2025/04/2025-04-cambridge-digital-mining-industry-report.pdf

Carter, N. (2021). How Much Energy Does Bitcoin Actually Consume? Harvard Business Review. https://hbr.org/2021/05/how-much-energy-does-bitcoin-actually-consume

Collins, S., & Dewhurst, Rian. D. (2024, September 20). Runaway Citations and the Persistence of Bitcoin Misinformation.

Cornell Chronicle. (2023, November). Bitcoin mining helps finance renewable energy projects, study shows. https://news.cornell.edu/stories/2023/11/bitcoin-mining-helps-finance-renewable-energy-projects-study-shows

Mellerud, J. (2023, May 2). Bitcoin Mining Around the World: Kazakhstan. Hashrate Index. https://hashrateindex.com/blog/bitcoin-mining-around-the-world-kazakhstan/

Mora, C., Rollins, R. L., Taladay, K., Kantar, M. B., Chock, M. K., Shimada, M., & Franklin, E. C. (2018). Bitcoin emissions alone could push global warming above 2 C. Nature Climate Change, 8(11), 931–933.

Naueihed, S. (2023, October 13). Revolutionizing Bitcoin Mining: Cooling Methods for Enhanced Efficiency. UNLOCK Blockchain. https://www.unlock-bc.com/110085/revolutionizing-bitcoin-mining-cooling-methods-for-enhanced-efficiency/

Rudd, M. A., Jones, M., Sechrest, D., Batten, D., & Porter, D. (2024). An integrated landfill-gas-to-energy and Bitcoin mining model (SSRN Scholarly Paper 4810964). https://doi.org/10.1016/j.jclepro.2024.143516

Rudd, M. A., & Porter, D. (2024). Economic integration of Bitcoin mining in renewable energy and grid management (SSRN Scholarly Paper 4899244). https://doi.org/10.2139/ssrn.4899244

Sai, A. R., & Vranken, H. (2023). Promoting rigor in blockchain energy and environmental footprint research: A systematic literature review. Blockchain: Research and Applications, 100169.

WooCharts. (2025). Bitcoin sustainable energy use and emissions dashboard. https://woocharts.com/bitcoin-energy

‍