Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing

Inez Hua; Yanran Wu; Yi Ding

REVIEW 3 major objections 5 minor 50 references

Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Computing's water impact is not its consumption volume: a new metric multiplies consumption by local, time-varying water stress, and case studies show order-of-magnitude swings by location and season.

desk verdict Temporal water-stress discounting is a real increment, but the off-site attribution and the 'first framework' claim need fixing before the numbers mean what the abstract says. read the letter →

arxiv 2506.22773 v2 pith:CB53O7BJ submitted 2025-06-28 cs.DC cs.ARcs.CYcs.LG

Yanran Wu , Inez Hua , Yi Ding This is my paper · ORCID

classification cs.DCcs.ARcs.CYcs.LG

keywords waterstressadjustedimpactSCARFsustainablecomputingdatacenterconsumptionLLMservingsemiconductorfabricationfootprint

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to change how the computing industry measures its water footprint. Rather than counting liters consumed, it argues that the same liter is much more damaging in a stressed watershed than in an abundant one, and that water stress changes over seasons and decades. To capture this, it introduces SCARF, which computes an Adjusted Water Impact (AWI) by multiplying raw on-site and off-site water consumption by a basin-level Water Stress Factor, optionally discounted over future years. Three case studies—LLM serving, datacenters, and semiconductor fabs—show that AWI can differ by orders of magnitude across locations and months, and that rankings by raw consumption do not match rankings by adjusted impact. If the metric becomes standard, siting, scheduling, and procurement decisions could be made on real water burden rather than volume alone.

What carries the argument

The central object is the Adjusted Water Impact identity $AWI = (W_{\mathrm{on}} + W_{\mathrm{off}}) \times WSF_b$, where water stress is the ratio of local water demand to supply. On-site consumption is $W_{\mathrm{on}} = P \cdot t \cdot WUE_{\mathrm{on}}$, and off-site consumption is $W_{\mathrm{off}} = P \cdot t \cdot PUE \cdot WUE_{\mathrm{off}}$, so the framework hinges on two efficiency ratios—water per kWh on site and per kWh of purchased electricity—combined with a basin-level Water Stress Factor. $WSF_b$ is the mechanism that carries the argument: for immediate impact it is simply the basin's current water stress, and for long-term facilities it is a discount-rate-weighted average of projected stress, making the policy choice about the future explicit. Multiplying the two turns a volume metric into a burden metric.

What would settle it

Take a datacenter that buys electricity from a grid whose power plants lie in a low-stress basin, and compute AWI two ways: with the datacenter's basin stress, as SCARF does, and with the plants' actual basin stress; if the rankings of two candidate sites flip, the single-basin attribution is the point of failure.

Watch

Extended reading notes

Core claim

The paper's central claim is that water impact assessments for computing should weight consumption by where and when it occurs, and that this can be done with a single unified metric. SCARF maps each facility to its hydrological basin, obtains current and projected water stress from a global risk dataset, and forms the Water Stress Factor: current stress for short-term analyses, or a discounted sum of stress in 2030, 2050, and 2080 for long-lived infrastructure. The Adjusted Water Impact is $AWI = (W_{\mathrm{on}} + W_{\mathrm{off}}) \times WSF_b$, where on-site water comes from cooling and operations and off-site water comes from electricity generation. In the case studies, the same LLM served in a high-stress, inefficient location can have over 1,000 times the adjusted impact per request as a low-stress location; Arizona fabs outrank far-larger Oregon consumers in AWI; and changing the discount rate can reverse which datacenter looks sustainable. The paper reads these results as evidence that a stress-weighted metric reveals hidden opportunities for water-sustainable computing that volume-only accounting misses.

Load-bearing premise

The load-bearing assumption is that water consumed off-site to generate electricity feels the same water stress as the datacenter's own watershed, even if the power plant sits in a different basin; the case studies all rely on this.

Editorial extensions

If this is right

Datacenter and LLM-serving sustainability comparisons should report AWI, not raw water volume, because rankings by the two measures differ.
Workload schedulers can shift inference jobs across months to lower water impact without reducing consumption, since water stress varies seasonally.
For long-lived facilities, the discount rate is a policy parameter: a site can look better or worse depending on how much future stress is valued, so sustainability claims should state the discount rate.
Siting decisions for fabs and datacenters should weigh basin stress alongside efficiency, since high efficiency in a stressed basin can still carry a larger burden than moderate consumption in a wet basin.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper: the same AWI construction could be applied to other spatially variable burdens—such as watershed nutrient loading or local air pollution—wherever a basin-level stress factor exists.
Beyond the paper: because off-site consumption is assigned the datacenter's basin stress, extending SCARF to use the power plant's actual watershed could change rankings in regions that import electricity across basins; this is a direct test of the spatial weighting.
Beyond the paper: with sub-monthly water-stress data, the temporal axis could be pushed from seasons to hours, allowing water-aware job shifting inside a single day for LLM serving.
Beyond the paper: AWI could be paired with carbon accounting to expose trade-offs, since a low-carbon site in a stressed watershed may score worse on water than a higher-carbon site elsewhere.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

SCARF is a clear, useful step for sustainable computing: it turns water volume into stress-weighted impact with a transparent formula, folds in future water stress via a discount rate, and ships code plus three case studies. The genuinely new component is the temporal aggregation; spatial weighting by water stress already exists in the water-aware scheduling work they cite (e.g., Islam et al., WaterWise), so 'first general framework' overstates the novelty. Still, the combined spatial-temporal packaging across LLM serving, datacenters, and fabs is worth having.

The metric and code are solid enough to use with eyes open. The LLM serving measurements are real measured power data; the Google and Intel case studies use public reporting and show how rankings shift with discount rate. That sensitivity analysis is good practice.

The load-bearing caveat is the off-site water attribution. Eq (6) multiplies W_on + W_off by the facility's basin WSF, but W_off is consumed at the power plant, which can sit in a different watershed. The paper never states or justifies that assumption, and all three case studies inherit it. In an interconnected grid this is not a small parameter error; it can over- or under-state AWI by the WSF ratio between basins, and Figure 4 shows those ratios are large. The framework survives because the fix is straightforward: use the generation basin's WSF for W_off, or disclose the approximation and bound its effect. But as written, the 'where' claim in the abstract is not fully delivered for Scope 2 water.

Two smaller soft spots. The Google power capacity proxy (max reported capacity within 100 miles) is crude, and the paper gives no error bars or sensitivity around it; the ranking takeaways probably hold but the absolute AWI numbers are rough. And the novelty claim should be narrowed: refs [17,18,21,28] already weight water use by location stress; what is new is the time-discounted aggregation, not the spatial weighting itself.

Who benefits: system operators and sustainability teams wanting a single number for siting or scheduling trade-offs, and researchers comparing water impact across layers. It is a workshop-grade but useful contribution; with the off-site fix and a revised novelty claim it becomes a respectable full paper. I would send it to peer review rather than desk reject: the code ships, the metric is transparent, and the flaw is correctable. My own verdict would be 'conditional accept' pending the Scope 2 fix.

Referee Report

3 major / 5 minor

Summary. The paper introduces SCARF, a four-step framework for evaluating the water impact of computing systems in a way that accounts for spatial and temporal variation in water stress. The framework models on-site (Scope 1) and off-site (Scope 2) water consumption, maps facilities to hydrological basins, constructs a Water Stress Factor (WSF) either from current conditions or from discount-rate-weighted future projections, and computes an Adjusted Water Impact (AWI) by multiplying total raw consumption by WSF. The authors demonstrate the framework on three case studies: LLM serving in Microsoft datacenters, Google's U.S. datacenters, and Intel's U.S. semiconductor fabs. The main claims are that deployment location and time can change adjusted water impact by orders of magnitude, that high consumption in medium-stress regions can outweigh moderate consumption in high-stress regions, and that discount-rate choice can alter long-term site rankings.

Significance. If the framework's spatial accounting is corrected, SCARF is a useful and timely contribution to sustainable-computing evaluation. Its strengths include a transparent, parameter-explicit metric; the use of publicly available basin-level water-stress data (Aqueduct 4.0); publicly released code; and a sensitivity analysis for the discount rate. The three case studies span the computing stack and illustrate that raw water volume alone is insufficient. However, the paper's central claim to be 'the first general framework' that accounts for where water stress occurs is weakened by the treatment of off-site water consumption, and the datacenter case study rests on an ad hoc power-capacity proxy. These issues are fixable, but they affect the numerical results and the interpretation of the metric as spatially accurate.

major comments (3)

[§2.4, Eq. (6)] The AWI definition multiplies total raw water consumption, including Woff from Eq. (2), by the WSF of the facility's basin b. Because Woff represents water consumed at electricity generation sites, which generally lie in different watersheds from the computing facility, this assigns the wrong water-stress factor to Scope 2 water. All three case studies (Figures 3, 7(b), and 9(b)) inherit this assumption. Please revise Eq. (6) to use generation-basin WSF values, e.g., AWI = Won × WSF_b + Σ_g Woff_g × WSF_{b_g}, or clearly justify the approximation and quantify its effect using the spatially explicit WUEoff data from [41].
[§4.1, power capacity proxy] Estimating each Google site's power capacity by taking the maximum reported capacity of any datacenter within a 100-mile radius is an ad hoc approximation. It can over- or under-attribute energy and water consumption to a specific Google site, and the paper provides no sensitivity analysis for the radius choice or for the choice of the maximum rather than another aggregation. Since Figure 7 and Takeaway 3 depend on the resulting consumption volumes, the quantitative datacenter results need to be re-examined or accompanied by a robustness check.
[§2.3, Eqs. (4)–(6)] The long-term AWI multiplies a current annual water-consumption value (Won+Woff) by a normalized weighted average of future water-stress projections. As written, this is neither a standard discounted lifetime impact (which would sum discounted annual impacts over the facility lifetime) nor a purely current annual-impact metric. The discount-rate sensitivity in Figure 8(b) therefore reflects the weighting of stress projections rather than the timing of the consumption stream. Please clarify whether the long-term WSF is intended as a forward-looking siting indicator or as a discounted lifetime impact, and adjust the aggregation and notation accordingly.

minor comments (5)

[§4.1] The text says 'maximum reported power capacity P (kWh)', but power should be expressed in kilowatts (kW), not kilowatt-hours; the energy calculation E = P × 24 × 365 × 0.7 then uses power correctly, so the unit label needs to be fixed.
[§3.1, Figure 4] The source and calculation of the monthly water-stress values plotted in Figure 4 are not described; §2.2 only mentions retrieving current and projected (2030/2050/2080) stress from Aqueduct 4.0, while Eq. (5) uses annual horizons. Please state where the monthly values come from and whether they are part of Aqueduct 4.0 or another dataset.
[Figure 2 caption] The caption states that the star symbol 'refers to the location with lowest value' but does not specify which series (on-site WUE, off-site WUE, total WUE, or water stress); please clarify.
[§1 and Abstract] The phrase 'first general framework' should be qualified, since prior work [17,18,21,28,40] already integrates water stress into scheduling and siting decisions; the novelty claim could be narrowed to the specific combination of basin-level mapping, temporal discounting, and a unified AWI metric.
[§4.2.1, Figure 7] The site labels in Figure 7(a) run together (e.g., 'VA2OH3'), making them difficult to read; please add separators or a legend, and define the site abbreviations (NV2, OH2, OH3, VA2, VA3) in the caption or text.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: AWI is an externally anchored definition, case studies are illustrations, and there are no self-citations or fitted-input predictions.

full rationale

SCARF's AWI is introduced as a definition (Eq. 6: AWI = (Won+Woff) × WSF_b), with WSF_b taken from the external Aqueduct 4.0 dataset and Won/Woff computed from measured or publicly reported WUE/PUE values. No parameter is fitted to match a target outcome, and no load-bearing premise is justified by a citation to the authors' own work; the reference list contains no self-citations. The case-study takeaways, such as location and time sensitivity, are arithmetic consequences of multiplying consumption by water stress, but the paper does not claim to derive those takeaways independently of the metric; they are illustrations of the metric's properties. The off-site water basin concern (Eq. 6 applies the facility's WSF_b to Woff even though generation may occur in another watershed) is a modeling-accuracy limitation, not a circularity: it does not make any output equal to an input by construction. The derivation chain is self-contained with respect to the metric definition and uses external data throughout, so no circular step is present.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the validity of external water stress datasets, on a normative discounting choice, and on a simplifying basin attribution for off-site electricity. These are hand-chosen values that the AWI numbers depend on, but none are fitted to make the case-study conclusions match a predetermined target.

free parameters (3)

discount_rate = 0.03 (3%), with 0.014 and 0.07 in sensitivity
User-defined discount rate in Equation (5) that weights future water stress; chosen from WRDA 1974, not fitted to data, but it directly controls long-term AWI rankings in Sections 4 and 5.
power_utilization_factor = 0.7
Assumed average datacenter power utilization from Google's dataset [37], used to convert power capacity to annual energy for Google datacenters in Section 4.1.
radius_for_power_capacity_proxy = 100 miles
The maximum reported power capacity of any site within 100 miles is used as a proxy for a Google datacenter's capacity in Section 4.1.

assumptions (4)

domain assumption Aqueduct 4.0 water stress values and projections are accurate and appropriate for basin-level impact assessment.
Invoked in Steps 2 and 3 and all case studies; if the WRI dataset misrepresents local scarcity, all AWI results are off.
ad hoc to paper Off-site water consumption can be assigned the same basin-level water stress as the facility that consumes the electricity.
Equation (6) multiplies (Won + Woff) by WSF_b of the facility's basin, even though off-site consumption occurs at power plants that may lie in other watersheds.
domain assumption Discounting future water stress is an appropriate way to aggregate long-term environmental impact.
Sections 2.3 and 4.2.2 follow environmental economics, but the choice of discount rate is a normative policy decision, not a physical fact.
ad hoc to paper The maximum power capacity within 100 miles of a Google site approximates that site's power capacity.
Section 4.1 uses this proxy to estimate annual energy for all Google datacenter sites.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing." pith.science (2026). https://pith.science/paper/CB53O7BJ

@misc{pith2026250622773,
  author       = {Pith},
  title        = {Pith review of: Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CB53O7BJ}},
  note         = {Machine review of arXiv:2506.22773}
}

read the original abstract

Water consumption is an increasingly critical dimension of computing sustainability, especially as AI workloads rapidly scale. However, current water impact assessment often overlooks where and when water stress is more severe. To fill in this gap, we present SCARF, the first general framework that evaluates water impact of computing by factoring in both spatial and temporal variations in water stress. SCARF calculates an Adjusted Water Impact (AWI) metric that considers both consumption volume and local water stress over time. Through three case studies on LLM serving, datacenters, and semiconductor fabrication plants, we show the hidden opportunities for reducing water impact by optimizing location and time choices, paving the way for water-sustainable computing. The code is available at https://github.com/jojacola/SCARF.

Figures

Figures reproduced from arXiv: 2506.22773 by the authors.

**Figure 1.** Overview of the SCARF Framework. water impact from computing. The key insight is to integrate spatial and temporal variation in water stress, which is the ratio of local water demand to supply, reflecting regional scarcity. SCARF follows four steps: (1) model water consumption volume, including both on-site (Scope 1) and off-site (Scope 2) consumption; (2) map each site to its corresponding hydrological basin; (3) … view at source ↗

**Figure 2.** On-site and Off-site WUEs of Microsoft datacenter [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Adjusted Water Impact (AWI) per request for differ [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Water stress levels across various Microsoft data [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: Adjusted water impact (AWI) per request of 32B [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

**Figure 8.** Figure 8: Sensitivity Check of Discount Rate. (a) Discounting [PITH_FULL_IMAGE:figures/full_fig_p005_8.png]

**Figure 7.** Figure 7: (a) Annual water consumption of Google’s U.S. dat [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

**Figure 9.** Figure 9: (a) Annual water consumption of Intel’s fab plants, [PITH_FULL_IMAGE:figures/full_fig_p006_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 44 canonical work pages

[41]

Md Abu Bakar Siddik, Arman Shehabi, Prakash Rao, and Landon T Marston. 2024. Spatially and temporally detailed water and carbon footprints of US Electricity generation and use. Water Resources Research 60, 12 (2024), e2024WR038350

work page 2024
[1]

Water Resources Development Act of 1974

1974. Water Resources Development Act of 1974. https://www .congress.gov/ bill/93rd-congress/house-bill/10203/text. Public Law 93-251

work page 1974
[2]

Taiwan is Facing a Drought, and It Has Prioritized Its Computer Chip Business Over Farmers

2021. Taiwan is Facing a Drought, and It Has Prioritized Its Computer Chip Business Over Farmers. https://www .nytimes.com/2021/04/08/business/taiwan- is-facing-a-drought-and-it-has-prioritized-its-computer-chip-business-over- farmers.html The New York Times, Accessed: 2025-05-09

work page 2021
[3]

Intel Plans Massive Fab Expansion in Oregon

2024. Intel Plans Massive Fab Expansion in Oregon. https:// www.tomshardware.com/news/intel-plans-massive-fab-expansion-in-oregon. Accessed: 2024-06-25

work page 2024
[4]

Amazon Web Services. 2023. 2023 Amazon Sustainability Report – AWS Summary. https://sustainability.aboutamazon.com/2023-amazon-sustainability- report-aws-summary .pdf

work page 2023
[5]

Nicole T Carter, Adam C Nesbitt, et al . 2016. Discount rates in the economic evaluation of US Army Corps of Engineers projects. Congressional Research Service 44594 (2016)

work page 2016
[6]

Li Chen and Aaron P Wemhoff. 2022. Characterizing Data Center Cooling System Water Stress in the United States. In ASHRAE Annual Conference papers

work page 2022
[7]

Datacenters.com. 2025. Datacenters.com. https://www .datacenters.com/

work page 2025

Show all 50 references

[8]

Clément Desroches, Martin Chauvin, Louis Ladan, Caroline Vateau, Simon Gosset, and Philippe Cordier. 2025. Exploring the sustainable scaling of AI dilemma: A projective study of corporations’ AI environmental impacts. arXiv preprint arXiv:2501.14334 (2025)

2025 arXiv
[9]

Patrick Donovan. 2013. Fundamentals of Managing the Data Center Life Cycle for Owners. APC, Schneider Electric White Papers (2013)

2013
[10]

Equinix. 2013. Everything You Need to Know About DC11: Q&A with Data Center Guru Jim Farmer. https://blog .equinix.com/blog/2013/05/10/everything- you-need-to-know-about-dc11-qa-with-data-center-guru-jim-farmer/

2013
[11]

Wedan Emmanuel Gnibga, Andrew A Chien, Anne Blavette, and Anne Cécile Orgerie. 2024. FlexCoolDC: datacenter cooling flexibility for harmonizing water, energy, carbon, and cost trade-offs. In Proceedings of the 15th ACM International Conference on Future and Sustainable Energy ...

2024
[12]

Google LLC. 2024. Google Environmental Report 2024. https://www.gstatic.com/ gumdrop/sustainability/google-2024-environmental-report .pdf

2024
[13]

Pranjol Sen Gupta, Md Rajib Hossen, Pengfei Li, Shaolei Ren, and Mohammad A Islam. 2024. A dataset for research on water sustainability. In Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems . 442–446

2024
[14]

World Resources Institute. 2024. Aqueduct Water Risk Atlas. https:// www.wri.org/applications/aqueduct/water-risk-atlas. Accessed: 2025-05-05

2024
[15]

Intel Corporation. 2024. Explore Intel’s U.S. Semiconductor Manufacturing Expan- sion. https://www .exploreintel.com/ Accessed: 2025-05-04

2024
[16]

Intel Corporation. 2024. Intel 2023–24 Corporate Responsibility Report . https: //csrreportbuilder.intel.com/pdfbuilder/pdfs/CSR-2023-24-Full-Report .pdf Ac- cessed: 2025-05-04

2024
[17]

Mohammad A Islam, Kishwar Ahmed, Hong Xu, Nguyen H Tran, Gang Quan, and Shaolei Ren. 2016. Exploiting spatio-temporal diversity for water saving in geo-distributed data centers. IEEE Transactions on Cloud Computing 6, 3 (2016), 734–746

2016
[18]

Mohammad A Islam, Shaolei Ren, Gang Quan, Muhammad Z Shakir, and Athana- sios V Vasilakos. 2015. Water-constrained geographic load balancing in data centers. IEEE Transactions on Cloud Computing 5, 2 (2015), 208–220

2015
[19]

Ian James. 2024. Groundwater depletion is worsening worldwide. Parts of California are suffering rapid declines . https://www .latimes.com/environment/story/2024- 01-25/groundwater-depletion-is-worsening-worldwide Los Angeles Times, Ac- cessed: 2025-05-04

2024
[20]

Scott Jasechko, Hansjörg Seybold, Debra Perrone, Ying Fan, Mohammad Sham- sudduha, Richard G Taylor, Othman Fallatah, and James W Kirchner. 2024. Rapid groundwater decline and some cases of recovery in aquifers globally. Nature 625, 7996 (2024), 715–721

2024
[21]

Yankai Jiang, Rohan Basu Roy, Raghavendra Kanakagiri, and Devesh Tiwari. 2025. WaterWise: Co-optimizing Carbon- and Water-Footprint Toward Environmen- tally Sustainable Cloud Computing. In Proceedings of the 30th ACM SIGPLAN Annual Symposium on Principles and Practice of Paral...

2025
[22]

Matti Kummu, Joseph HA Guillaume, Hans De Moel, Stephanie Eisner, Martina Flörke, Miina Porkka, Stefan Siebert, Ted IE Veldkamp, and Philip J Ward. 2016. The world’s road to water scarcity: shortage and stress in the 20th century and pathways towards sustainability. Scientific...

2016
[23]

Samantha Kuzma, Marc FP Bierkens, Shivani Lakshman, Tianyi Luo, Liz Saccoccia, Edwin H Sutanudjaja, and Rens Van Beek. 2023. Aqueduct 4.0: Updated decision- relevant global water risk indicators . World Resources Institute Washington, DC, USA

2023
[24]

Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In SOSP

2023
[25]

Bernhard Lehner and Günther Grill. 2013. Global river hydrography and network routing: baseline data and new approaches to study the world’s large river systems. Hydrological Processes 27, 15 (2013), 2171–2186

2013
[26]

Nuoa Lei, Jun Lu, Arman Shehabi, and Eric Masanet. 2025. The water use of data center workloads: A review and assessment of key determinants. Resources, Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing HotCarbon’25, July 10-11th, 202...

2025
[27]

Islam, and Shaolei Ren

Pengfei Li, Jianyi Yang, Mohammad A. Islam, and Shaolei Ren. 2024. Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models. Commun. ACM (2024)

2024
[28]

Pengfei Li, Jianyi Yang, Adam Wierman, and Shaolei Ren. 2024. Towards en- vironmentally equitable AI via geographical load balancing. In Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems . 291–307

2024
[29]

Jordan Macknick, Robin Newmark, Garvin Heath, and KC Hallett. 2011. Review of operational water consumption and withdrawal factors for electricity generating technologies. (2011)

2011
[30]

Iman Mallakpour, Mojtaba Sadegh, and Amir AghaKouchak. 2018. A new normal for streamflow in California in a warming climate: Wetter wet seasons and drier dry seasons. Journal of hydrology 567 (2018), 203–211

2018
[31]

Microsoft Corporation. 2024. Efficiency and Sustainability at Microsoft Datacen- ters. https://datacenters .microsoft.com/sustainability/efficiency/

2024
[32]

Microsoft Corporation. 2024. Microsoft 2024 Environmental Sustainability Report. https://cdn-dynmedia-1 .microsoft.com/is/content/microsoftcorp/ microsoft/msc/documents/presentations/CSR/Microsoft-2024-Environmental- Sustainability-Report.pdf

2024
[33]

Jacob Morrison, Clara Na, Jared Fernandez, Tim Dettmers, Emma Strubell, and Jesse Dodge. 2025. Holistically Evaluating the Environmental Impact of Cre- ating Language Models. In The Twelfth International Conference on Learning Representations (ICLR)

2025
[34]

Leonardo Nicoletti, Michelle Ma, and Dina Bass. 2025. AI Is Draining Water From Areas That Need It Most . https://www .bloomberg.com/graphics/2025-ai- impacts-data-centers-water-data Bloomberg, Accessed: 2025-05-09

2025
[35]

Greenhouse Gas Protocol. 2011. Greenhouse gas protocol. Sector Toolsets for Iron and Steel-Guidance Document (2011)

2011
[36]

Paul Reig. 2013. What’s the difference between water use and water consump- tion? https://www .wri.org/insights/whats-difference-between-water-use-and- water-consumption

2013
[37]

Varun Sakalkar, Vasileios Kontorinis, David Landhuis, Shaohong Li, Darren De Ronde, Thomas Blooming, Anand Ramesh, James Kennedy, Christopher Malone, Jimmy Clidaras, et al. 2020. Data center power oversubscription with a medium voltage power plane and priority-aware capping. I...

2020
[38]

ShareGPT. [n. d.]. ShareGPT - Share and Save Your Conversations with AI. https://sharegpt.com/
[39]

Arman Shehabi, Alex Hubbard, Alex Newkirk, Nuoa Lei, Md Abu Bakkar Siddik, Billie Holecek, Jonathan Koomey, Eric Masanet, Dale Sartor, et al . 2024. 2024 United States Data Center Energy Usage Report. (2024)

2024
[40]

Md Abu Bakar Siddik, Arman Shehabi, and Landon Marston. 2021. The environ- mental footprint of data centers in the United States. Environmental Research Letters 16, 6 (2021), 064017

2021
[42]

Nicholas Stern. 2007. The economics of climate change: the Stern review. cambridge University press

2007
[43]

Richard SJ Tol. 2011. The social cost of carbon. Annu. Rev. Resour. Econ. 3, 1 (2011), 419–443

2011
[44]

environmental

Martin L Weitzman. 1994. On the" environmental" discount rate. Journal of Environmental Economics and Management 26, 2 (1994), 200–209

1994
[45]

An Yang, Baosong Yang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Zhou, Cheng- peng Li, Chengyuan Li, Dayiheng Liu, Fei Huang, Guanting Dong, Haoran Wei, Huan Lin, Jialong Tang, Jialin Wang, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Ma, Jin Xu, Jingren Zhou, Jinze Bai, Jinzhe...

2024 arXiv
[46]

An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, Kai Dang, Keming Lu, Keqin Bao, Kexin Yang, Le Yu, Me...

2024 arXiv
[47]

Lindsey MW Yasarer, Jason M Taylor, James R Rigby, and Martin A Locke. 2020. Trends in land use, irrigation, and streamflow alteration in the Mississippi River Alluvial Plain. Frontiers in Environmental Science 8 (2020), 66

2020
[48]

Adam Zewe. 2025. Explained: Generative AI’s Environmental Impact . https: //news.mit.edu/2025/explained-generative-ai-environmental-impact-0117 MIT News, Accessed: 2025-05-04

2025
[49]

2021.Drought in Taiwan Pits Chip Makers Against Farmers

Raymond Zhong and Amy Chang Chien. 2021.Drought in Taiwan Pits Chip Makers Against Farmers . https://www .nytimes.com/2021/04/08/technology/taiwan- drought-tsmc-semiconductors .html The New York Times, Accessed: 2025-05-09

2021
[50]

Guido Zuccon, Harrisen Scells, and Shengyao Zhuang. 2023. Beyond CO2 Emis- sions: The Overlooked Impact of Water Consumption of Information Retrieval Models. In Proceedings of the 2023 ACM SIGIR International Conference on Theory of Information Retrieval (Taipei, Taiwan) (ICTI...

2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [41]

Md Abu Bakar Siddik, Arman Shehabi, Prakash Rao, and Landon T Marston. 2024. Spatially and temporally detailed water and carbon footprints of US Electricity generation and use. Water Resources Research 60, 12 (2024), e2024WR038350

work page 2024

[2] [1]

Water Resources Development Act of 1974

1974. Water Resources Development Act of 1974. https://www .congress.gov/ bill/93rd-congress/house-bill/10203/text. Public Law 93-251

work page 1974

[3] [2]

Taiwan is Facing a Drought, and It Has Prioritized Its Computer Chip Business Over Farmers

2021. Taiwan is Facing a Drought, and It Has Prioritized Its Computer Chip Business Over Farmers. https://www .nytimes.com/2021/04/08/business/taiwan- is-facing-a-drought-and-it-has-prioritized-its-computer-chip-business-over- farmers.html The New York Times, Accessed: 2025-05-09

work page 2021

[4] [3]

Intel Plans Massive Fab Expansion in Oregon

2024. Intel Plans Massive Fab Expansion in Oregon. https:// www.tomshardware.com/news/intel-plans-massive-fab-expansion-in-oregon. Accessed: 2024-06-25

work page 2024

[5] [4]

Amazon Web Services. 2023. 2023 Amazon Sustainability Report – AWS Summary. https://sustainability.aboutamazon.com/2023-amazon-sustainability- report-aws-summary .pdf

work page 2023

[6] [5]

Nicole T Carter, Adam C Nesbitt, et al . 2016. Discount rates in the economic evaluation of US Army Corps of Engineers projects. Congressional Research Service 44594 (2016)

work page 2016

[7] [6]

Li Chen and Aaron P Wemhoff. 2022. Characterizing Data Center Cooling System Water Stress in the United States. In ASHRAE Annual Conference papers

work page 2022

[8] [7]

Datacenters.com. 2025. Datacenters.com. https://www .datacenters.com/

work page 2025