# Ethereum Emissions: A Bottom-up Estimate

**Authors:** Kyle McDonald  
**arXiv:** [2112.01238](https://arxiv.org/abs/2112.01238)  
**Date:** 2021-11-19  
**Summary page:** https://www.summarizepaper.com/en/paper/2112.01238v3/

## Key points

- The Ethereum network is a distributed global network of computers that requires massive amounts of computational power to maintain.
- Previous studies on estimating the energy use and emissions of the Ethereum network have relied on top-down economic analysis and rough estimates of hardware efficiency and emissions factors.
- This study provides a bottom-up analysis that works from hashrate to an energy usage estimate, and from mining locations to an emissions factor estimate, combining these for an overall emissions estimate.
- The authors analyze the entire history of Proof-of-Work (PoW) Ethereum, from creation to the merge.
- The authors estimate that the Ethereum network has used between 42 TW h and 89 TW h with a best guess of 60 TW h since its inception in 2015.
- They also estimate emissions factors for popular regions where Ethereum is mined by guessing which region a block was mined in based on block metadata.
- The estimated emissions have increased over time due to an increase in hashing power and electricity consumption.
- Limitations include only estimating direct operational emissions from mining activities and not considering indirect or embodied emissions from manufacturing hardware or infrastructure.
- Comparing their results with previous studies, they found that their estimates were similar but slightly higher due to accounting for more factors such as datacenter overheads, hardware overheads, grid loss, etc.
- High prices may not be significantly changing the hardware mix as hashing efficiency has increased since 2020-04 at a similar rate to their trendline.
- Overall, this study provides valuable insights into understanding the environmental impact of PoW-based blockchain networks like Ethereum and highlights areas where further research can be done to better understand their environmental impact.

## Summary

The Ethereum network is a distributed global network of computers that requires massive amounts of computational power to maintain. Previous studies on estimating the energy use and emissions of the Ethereum network have relied on top-down economic analysis and rough estimates of hardware efficiency and emissions factors. However, in this study, the authors provide a bottom-up analysis that works from hashrate to an energy usage estimate, and from mining locations to an emissions factor estimate, combining these for an overall emissions estimate. The study analyzes the entire history of Proof-of-Work (PoW) Ethereum, from creation to the merge. The authors estimate that the Ethereum network has used between 42 TW h and 89 TW h with a best guess of 60 TW h since its inception in 2015. They also estimate emissions factors for popular regions where Ethereum is mined by guessing which region a block was mined in based on block metadata. The daily energy usage is then multiplied by daily emissions factors to estimate the emissions of the Ethereum network. The results show that the estimated emissions have increased over time due to an increase in hashing power and electricity consumption. However, there are some limitations to this study as it only estimates direct operational emissions from mining activities and does not consider indirect or embodied emissions from manufacturing hardware or infrastructure. Comparing their results with previous studies, they found that their estimates were similar but slightly higher due to accounting for more factors such as datacenter overheads, hardware overheads, grid loss, etc. Additionally, they found that high prices may not be significantly changing the hardware mix as hashing efficiency has increased since 2020-04 at a similar rate to their trendline. Overall, this study provides valuable insights into understanding the environmental impact of PoW-based blockchain networks like Ethereum and highlights areas where further research can be done to better understand their environmental impact.