Content uploaded by Alex de Vries
Author content
All content in this area was uploaded by Alex de Vries on Dec 04, 2023
Content may be subject to copyright.
Commentary
Bitcoin’s growing water footprint
Alex de Vries
1,2,3,
*
1
School of Business and Economics, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands
2
Founder of Digiconomist, Almere, the Netherlands
3
De Nederlandsche Bank, Amsterdam, the Netherlands
*Correspondence: alex@digiconomist.net
https://doi.org/10.1016/j.crsus.2023.100004
SUMMARY
Amid growing concerns over the impacts of climate change on worldwide water security, Bitcoin’s water
footprint has rapidly escalated in recent years. The water footprint of Bitcoin in 2021 significantly increased
by 166% compared with 2020, from 591.2 to 1,573.7 GL. The water footprint per transaction processed on the
Bitcoin blockchain for those years amounted to 5,231 and 16,279 L, respectively. As of 2023, Bitcoin’s annual
water footprint may equal 2,237 GL. To address this increasing water footprint, miners could apply immersion
cooling and consider using power sources that do not require freshwater. A change in the Bitcoin software
could also significantly reduce the network’s water footprint.
INTRODUCTION
In 2021, Greenidge Generation, a Bitcoin-
mining and power-generation company
gained global attention for allegedly dis-
charging large volumes of hot water into
New York’s Seneca Lake. The company
had repurposed a once-abandoned po-
wer plant for Bitcoin mining in 2019, lead-
ing to increased carbon emissions and
water usage. Residents raised concerns
that the warm water discharges were
heating Seneca Lake to beyond state
water-quality standards. In 2023, environ-
mental groups sued Greenidge Genera-
tion for violating the Clean Water Act and
other environmental regulations, though
this lawsuit was dismissed in the same
year.
1
Despite this high-profile case and the
looming threat of a global water crisis,
2
the water usage of the Bitcoin-mining
network has remained relatively underre-
ported. Academic research on the envi-
ronmental impact of Bitcoin mining has
primarily focused on the network’s energy
consumption, carbon footprint, and elec-
tronic waste generation.
3
Between 2 and
3 billion people worldwide already experi-
ence water shortages, which is expected
to worsen in the coming decades. Due to
growing international concerns around
drinking water availability, it is crucial to
understand the water footprint of Bitcoin
mining and its potential impact. Siddik
et al.
4
estimate that Bitcoin mining
was responsible for consuming 1,572.3
gigaliters (GL) of water in 2021. A better
understanding of the Bitcoin water foot-
print will help facilitate the development
of a responsible approach to manage
the limited freshwater supply. This com-
mentary analyzes the time period of
September 2019 to March 2023 to provide
insights into the evolving water footprint of
Bitcoin mining and the different ways in
which Bitcoin miners consume water.
Quantifying Bitcoin’s water
footprint
Bitcoin mining is an energy-intensive pro-
cess that contributes to the creation of
new blocks of transactions in Bitcoin’s
underlying blockchain. Mining resembles
a numeric guessing game in which the
first miner who guesses a certain winning
number gets to create the next block for
the blockchain. The only way to obtain
this winning number is through a process
of trial and error, with the whole Bitcoin
network generating approximately 350
quintillion guesses every second of the
day as of May 2023.
5
Even so, it typically
takes 10 min, on average, to create a
new block. The Bitcoin software has
a built-in mining difficulty adjustment
mechanism that keeps the issuance
rate steady. Each newly created block
pays out a reward of 6.25 newly min-
ted Bitcoins, valued at approximately
$180,000 USD as of May 2023,
5
which
acts as an incentive to participate in the
block-creation process. Because only
the winners of the guessing game get re-
warded, this process is highly competi-
tive. The Bitcoin-mining network consists
of millions of specialized devices world-
wide, primarily located in the US, China,
and Kazakhstan.
6
These devices all
compete for the available rewards and
consume substantial amounts of elec-
tricity, with an estimated power demand
of 16.2 gigawatts (GW) as of March 2023.
7
In addition to electricity, Bitcoin miners
also require water, which is utilized in two
main ways. The first involves onsite
(direct) water use for cooling systems
and air humidification. Water usage de-
pends on cooling system types and local
climate conditions. It is important to
differentiate between water withdrawal
and water consumption in terms of this
usage. Water withdrawal pertains to
the water taken from surface water or
groundwater sources, while water con-
sumption refers to the portion of water
that becomes unavailable for reuse after
withdrawal, primarily due to evaporation
in cooling systems. Water consumption
is not extensively studied in Bitcoin mining
or generic data center research, as reli-
able data on water consumption factors
are challenging to obtain.
The second way in which miners use
water relates to the (indirect) water con-
sumption associated with generating the
electricity necessary to power their de-
vices. Thermoelectric power generation
Cell Reports Sustainability 1, January 5, 2024 ª2023 The Author(s). 1
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
ll
OPEN ACCESS
Please cite this article in press as: de Vries, Bitcoin’s growing water footprint, Cell Reports Sustainability (2024), https://doi.org/10.1016/
j.crsus.2023.100004
plays a major role in water consumption,
as a portion of the withdrawn water for
cooling purposes evaporates (unless dry
cooling technologies utilizing air are em-
ployed). These systems can utilize both
freshwater and non-freshwater sources.
This commentary, however, exclusively
focuses on freshwater consumption. Hy-
dropower generation generally consumes
even more water per kilowatt-hour (kWh)
generated, as water evaporates from the
hydropower reservoirs. However, these
reservoirs may serve purposes beyond
electricity generation; thus, this commen-
tary considers only the portion of water
loss attributable to power generation.
The total water footprint of Bitcoin exam-
ined in this commentary encompasses
the freshwater consumed due to both
direct and indirect water consumption dur-
ing the operational stage of Bitcoin mining
devices. The water consumption relating
to construction and manufacturing is not
considered, as it likely contributes little to
the full life-cycle footprint.
4
Indirect water footprint
The indirect water footprint of Bitcoin min-
ing is relatively straightforward to assess
as it relies on similar inputs as the analysis
of Bitcoin’s carbon footprint.
8
To deter-
mine the indirect water footprint, it is
important to establish the power demand
of the network and the power sources.
The Cambridge Centre for Alternative
Finance (CCAF) provides data on the po-
wer demand, spatial distribution of Bit-
coin mining, and electricity mix, which
are used to generate daily estimates of
the network’s carbon footprint.
6
To
calculate the water footprint instead of
the carbon footprint, these inputs are
combined with data on the water intensity
of electricity generation on a specific grid.
The water intensity is expressed in L of
freshwater consumption per kWh gener-
ated, which is also referred to as the
water consumption factor of electricity
generation.
By using the CCAF’s estimates on po-
wer demand and spatial distribution of
Bitcoin mining following the approach by
Siddik et al.
4
and using the same esti-
mated water consumption factors for
electricity generation per country, the
water footprint of Bitcoin in 2021 was
determined to be 1,573.7 GL (Data S1).
This estimate closely aligns with the esti-
mate of 1,572.3 GL by Siddik et al., with
the minor discrepancy attributed to the
exclusion of certain countries with a small
contribution to the total water footprint in
Siddik et al.’s study.
Because the CCAF has provided data
on the spatial distribution of Bitcoin miners
since late 2019, we can determine that the
water footprint of Bitcoin in 2021 signifi-
cantly increased by 166% compared
with 2020. In 2020, the network consumed
591.2 GL of water. Furthermore, the water
intensity of electricity consumed rose sh-
arply, from 8.63 L per kWh in 2020 to
15.0 in 2021, signifying a 74% increase.
Notably, the majority of this growth can
be attributed to increased mining activities
in Kazakhstan since late 2020 following
China’s ban on cryptocurrency mining in
spring 2021, which led to miners relocat-
ing. The water consumption by Bitcoin
miners in Kazakhstan alone amounted to
260.6 GL in 2020 and rose to 997.9 GL in
2021, a 283% increase. Consequently,
Kazakhstan accounted for 63% of the net-
work’s estimated water footprint in 2021.
Despite representing a limited share of
the total computational power in the Bit-
coin network (around 14% by the end of
2021) and having hydropower only play a
minor role in electricity generation, Ka-
zakhstan’s high water intensity in elec-
tricity generation amplifies its impact on
the network’s water footprint. Notably,
the spatial distribution of Bitcoin miners
provided by the CCAF has some uncer-
tainty as the underlying sample represents
only 44% of the global total computational
power in the Bitcoin network. However, an
internet outage in Kazakhstan in January
2022 served to confirm the nation’s share
of the network.
8
Figure 1 presents the estimated monthly
and annual (indirect) water consumption
of Bitcoin miners since late 2019 along
with their spatial distribution in 2021.
Although estimates on the spatial distribu-
tion are available until January 2022, the
water footprint of Bitcoin may have
continued to rise beyond that date. The
estimated power demand of the Bitcoin
network reached a new all-time high in
March 2023,
7
suggesting an annualized
total electricity consumption of 141.9 tera-
watt-hours (TWh). This represents a 35%
increase compared with the total esti-
mated electricity consumption of 104.9
TWh in 2021 (Data S1, sheet 4). Conse-
quently, the water footprint of Bitcoin
may have increased by a similar magni-
tude. Assuming a constant water intensity
of electricity consumption (15.76 L per
kWh) since January 2022, 141.9 TWh of
electrical energy consumption could
result in a water footprint of 2,237 GL
(Data S1, sheet 4). With the network
handling 113 million transactions in 2020
and 96.7 million in 2021, the water foot-
print per transaction processed on the Bit-
coin blockchain for those years amounted
to 5,231 and 16,279 L, respectively.
Direct water footprint
To determine the direct water footprint of
Bitcoin miners, it is required to assess the
ratio between the on-site water consump-
tion of data centers (for cooling systems
and air humidification) and the energy
consumption of their equipment. This ra-
tio is expressed in L per kWh of electricity
consumed and known as the water usage
effectiveness (WUE). However, there is
almost no public information available
that can help quantify the WUE of crypto-
currency-mining facilities.
To obtain an estimate of the WUE of Bit-
coin miners, it is possible to use research
by Lei and Masanet,
9
who investigated
the WUE of various sizes of data centers
in different climate zones in the US. This
information can be combined with a list
of large-scale miners within the US, iden-
tified by the New York Times and pub-
lished in April 2023.
10
This list includes
34 Bitcoin mines, with the power require-
ment for each mine ranging from 38 to 450
megawatts (MW) as of March 2023.
Together, these 34 mines are responsible
for 3.91 GW of power demand, represent-
ing roughly a quarter of Bitcoin’s total esti-
mated power demand in the same month
(i.e., 16.2 GW
7
) and a majority of the share
that can be attributed to the US according
to the CCAF (i.e., 37.84% as of January
2022
6
).
By assuming the WUE of large-scale
US miners identified by the New York
Times corresponds to that of large-scale
data centers in their respective climate
zones defined by Lei and Masanet, it
can be established that the weighted
average WUE for Bitcoin miners in the
US ranges from 0.25 to 1.03 L per kWh
(Data S1, sheet 3). Given this range, the
total direct water footprint for US miners
could be 8.6–35.1 GL.
2Cell Reports Sustainability 1, January 5, 2024
Please cite this article in press as: de Vries, Bitcoin’s growing water footprint, Cell Reports Sustainability (2024), https://doi.org/10.1016/
j.crsus.2023.100004
ll
OPEN ACCESS Commentary
Total water footprint
This list of identified miners by the
New York Times can be combined with
regional water intensity factors of gener-
ating electricity in the US (Lee et al.
11
)to
find that the total capacity of these miners
could result in an annual indirect water
footprint of at least 84.9 GL in 2023 (not
including smaller and/or unidentified
miners), up from an estimated 70.9 GL in
2021 and 12.9 GL in 2020 (Data S1,
sheets 3 and 4). This means the total wa-
ter footprint of US Bitcoin miners, after
adding their direct water consumption,
could be an annual 93–120 GL, which
is 10%–41% more than the estimated
indirect water consumption of 84.9 GL
per year. It also means the total water
footprint of US Bitcoin miners could be
equivalent to the average annual water
consumption of around 300,000 US
households, comparable with a city such
as Washington, DC. Figure 2 shows a
breakdown of the spatial distribution of
Bitcoin’s total water footprint in the US.
It is important to note that the estimated
range for the WUE of Bitcoin miners in
this commentary is based solely on a list
of identified large-scale miners, whereas
even the smallest mining facility on the
New York Times list occupies approxi-
mately 100,000 square feet. Lei and Ma-
sanet show that smaller data centers,
occupying less than 20,000 square feet,
tend to have a higher WUE. It is not un-
likely that this could also apply to smaller
cryptocurrency-mining facilities, which
means the estimated WUE in this com-
mentary may be overly optimistic due
to the inclusion of only large-scale fa-
cilities.
There are upcoming developments that
may impact the WUE of Bitcoin mining.
Riot Platforms is reportedly building a fa-
cility in Texas that will require 1.4 million
gallons (5.3 million L) of water daily while
operating on 1 GW of electricity.
12
Once
completed, this facility would become
the largest Bitcoin mine in the US, with a
power demand more than double that of
the largest Bitcoin mine.
10
The WUE of
this facility would be just 0.22 L per kWh,
aligning with the best-case scenario for
the estimated WUE of Bitcoin miners. It
remains to be seen whether this WUE
can be achieved, but it has the potential
to lower the average WUE of US Bitcoin
miners from 0.25–1.03 to 0.24–0.86 L
per kWh. It should be noted that the
completion of this project would still in-
crease both the direct and indirect water
footprint of Bitcoin miners in the US. The
indirect water footprint of US Bitcoin
miners would increase by 30%, from
84.9 to 110.7 GL of annual water con-
sumption, while the direct water footprint
would increase from an annual range of
8.6–35.1 to 10.5–37.1 GL (Data S1, sheet
5). Consequently, the total water footprint
of US Bitcoin miners would range from
121.2–147.8 GL compared with an esti-
mated annual range of 93–120 GL prior
to the completion of this project.
IMPACT AND SOLUTION
Bitcoin’s expanding water footprint must
be considered in the context of escalating
water scarcity. Central Asia, including
Kazakhstan, is already facing a severe
water crisis.
13
The estimated water foot-
print of Bitcoin mining in Kazakhstan
alone was 997.9 GL in 2021, while the na-
tion’s capital could face a water shortage
of 75 GL per year by 2030.
14
Without con-
straints, Bitcoin mining could worsen con-
flicts over freshwater resources that the
region is already experiencing. Kazakh-
stan’s president signed legislation in
early 2023 to restrict the power demand
of cryptocurrency miners due to strain
on the local grid. However, renewable
energy usage remains exempt from these
limitations.
Figure 1. Development of Bitcoin’s indirect water footprint from September 2019 to January
2022
(A) Indirect water consumption of worldwide Bitcoin miners from September 2019 to January 2022,
including the cumulative total indirect water consumption for 2020 and 2021 and the estimated annualized
total indirect water consumption as of March 2023. These totals are shown including and excluding
Kazakhstan (KZ).
(B) Indirect water consumption of Bitcoin miners by country in 2021.
Cell Reports Sustainability 1, January 5, 2024 3
Please cite this article in press as: de Vries, Bitcoin’s growing water footprint, Cell Reports Sustainability (2024), https://doi.org/10.1016/
j.crsus.2023.100004
ll
OPEN ACCESS
Commentary
The US is also grappling with water
scarcity, particularly in the western states,
where a water crisis has emerged. In
2022, southern California urged residents
to reduce water consumption, and as hu-
man-induced climate change intensifies,
water shortages may further impact drink-
ing water access and agricultural yields.
In this context, Bitcoin’s growing
water footprint is a concerning develop-
ment. However, there are various ways
to reduce the water consumption of
Bitcoin miners. Miners can choose loca-
tions with favorable climate conditions to
minimize direct water consumption.
Immersing mining devices in a dielectric
fluid can also provide cooling without
relying on water. Furthermore, indirect
water consumption can be reduced by
using power sources that do not require
freshwater. Wind and solar power genera-
tion, as well as thermoelectric power
generation with dry cooling technologies,
can be utilized. Non-freshwater sources
can also be used for cooling systems,
as demonstrated by the extensive use
of once-through cooling systems with
non-freshwater along the US coastline.
11
However, while these options can miti-
gate freshwater consumption, they may
not address concerns such as electronic
waste generation and carbon emissions.
Ethereum, the second-largest crypto-
currency by market capitalization, demon-
strated in 2022 that the resource intensity
of cryptocurrencies can be reduced by
modifying the underlying software. Prior
to September 2022, Ethereum used the
same energy-intensive mining process as
Bitcoin. The total power demand of the
Ethereum mining network at this time
may have been responsible for an environ-
mental impact equivalent to half of Bit-
coin’s. However, a new upgrade was intro-
duced in Ethereum in September 2022
that replaced mining with an alternative
mechanism known as proof of stake
(PoS), which meant the right to create
new blocks for the underlying blockchain
was no longer earned through computa-
tional effort. Implementing PoS reduced
Ethereum’s power demand by at least
99.84%.
15
Implementing PoS in Bitcoin
could have a similar impact on the net-
work’s energy and water footprint. How-
ever, the Bitcoin community is reluctant
to make such changes to the software as
they affect the network beyond resource
intensity.
15
In the US, Bitcoin miners are facing
increased pressure to disclose more
information about their environmental
impact. US lawmakers have introduced
the Crypto-Asset Environmental Trans-
parency Act, which seeks to mandate
emissions disclosure by cryptocurrency-
mining operations.
16
Similar disclosure
requirements could be extended to water
usage by cryptocurrency miners, aiding in
the understanding of the water footprint of
the industry. Meta, a technology com-
pany, already discloses the water require-
ments of its data centers, serving as a po-
tential blueprint for US Bitcoin miners.
Other nations may need to consider
similar legislation, as the majority of Bit-
coin mining worldwide still takes place
outside of the US.
SUPPLEMENTAL INFORMATION
Supplemental information can be found online at
https://doi.org/10.1016/j.crsus.2023.100004.
DECLARATION OF INTERESTS
The authors declare no competing interests.
REFERENCES
1. Weny News (2023). Lawsuit filed against
Greenidge Generation on behalf of local
non-profits dismissed. https://www.weny.
com/story/49490746/lawsuit-filed-against-
greenidge-generation-on-behalf-of-local-
nonprofits-dismissed.
2. UNESCO (2023). Imminent risk of a global wa-
ter crisis, warns the UN World Water Develop-
ment Report 2023. https://www.unesco.org/
en/articles/imminent-risk-global-water-crisis-
warns-un-world-water-development-report-
2023.
3. Badea, L., and Mungiu-Pupazan, M.C. (2021).
The Economicand Environmental Impactof Bit-
coin. IEEE Access 9, 48091–48104. https://doi.
org/10.1109/ACCESS.2021.3068636.
4. Siddik, M.A.B., Amaya, M., and Marston, L.T.
(2023). The water and carbon footprint of cryp-
tocurrencies and conventional currencies.
J. Clean. Prod. 411, 137268. https://doi.org/
10.1016/j.jclepro.2023.137268.
5. blockchain.com. (2023). Hash Rate. https://
www.blockchain.com/charts/hash-rate?days
AverageString=7.
Figure 2. Regional water footprint of Bitcoin miners in the United States
(A) Annualized water footprint of large-scale Bitcoin miners in the United States by state as of March 2023.
(B) Relative share of direct versus indirect water consumption of large-scale Bitcoin miners in the United
States by state as of March 2023.
4Cell Reports Sustainability 1, January 5, 2024
Please cite this article in press as: de Vries, Bitcoin’s growing water footprint, Cell Reports Sustainability (2024), https://doi.org/10.1016/
j.crsus.2023.100004
ll
OPEN ACCESS Commentary
6. Cambridge Centre for Alternative Finance
(2023). Cambridge Bitcoin Electricity Con-
sumption Index. https://ccaf.io/cbeci/index.
7. Barron’s (2023). Bitcoin Is Consuming More
Energy Than Ever. Rising Crypto Prices Aren’t
All Good. https://www.barrons.com/articles/
bitcoin-energy-crypto-prices-1eb8f97f.
8. de Vries, A., Gallersdo
¨rfer, U., Klaaßen, L., and
Stoll, C. (2022). Revisiting Bitcoin’s carbon
footprint. Joule 6, 498–502. https://doi.org/
10.1016/j.joule.2022.02.005.
9. Lei, N., and Masanet, E. (2022). Climate- and
technology-specific PUE and WUE estimations
for U.S. data centers using a hybrid statistical
and thermodynamics-basedapproach. Resour.
Conserv. Recycl. 182, 106323. https://doi.org/
10.1016/j.resconrec.2022.106323.
10. New York Times (2023). The Real-World Costs
of the Digital Race for Bitcoin. https://www.
nytimes.com/2023/04/09/business/bitcoin-
mining-electricity-pollution.html.
11. Lee, U., Han, J., Elgowainy, A., and Wang, M.
(2018). Regional water consumption for hydro
and thermal electricity generation in the United
States. Appl. Energy 210, 661–672. https://doi.
org/10.1016/j.apenergy.2017.05.025.
12. NBC News (2022). Can the fragile Texas power
grid handle a cryptomining gold rush?. https://
www.nbcnews.com/tech/crypto/can-fragile-
texas-power-grid-handle-cryptomining-gold-
rush-rcna15726.
13. Wang, X., Chen, Y., Fang, G., Li, Z., and Liu, Y.
(2022). The growing water crisis in Central Asia
and the driving forces behind it. J. Clean. Prod.
378, 134574. https://doi.org/10.1016/j.jclepro.
2022.134574.
14. UNDP (2021). The climate change impact on
water resources in Kazakhstan. https://www.
undp.org/kazakhstan/stories/climate-change-
impact-water-resources-kazakhstan.
15. De Vries, A. (2023). Cryptocurrencies on the
road to sustainability: Ethereum paving the
way for Bitcoin. Patterns 4, 100633. https://
doi.org/10.1016/j.patter.2022.100633.
16. Senate.gov (2023). Lawmakers reintroduce
Crypto-Asset Environmental Transparency
Act. https://www.markey.senate.gov/news/
press-releases/sen-markey-rep-huffman-
announce-legislation-to-crack-down-on-
crypto-energy-consumption-as-markey-
holds-first-ever-senate-hearing-focused-
on-industrys-environmental-impacts.
Cell Reports Sustainability 1, January 5, 2024 5
Please cite this article in press as: de Vries, Bitcoin’s growing water footprint, Cell Reports Sustainability (2024), https://doi.org/10.1016/
j.crsus.2023.100004
ll
OPEN ACCESS
Commentary
