pith. sign in

arxiv: 2606.21760 · v1 · pith:K6GCT2UFnew · submitted 2026-06-19 · 💻 cs.CY

AI Data Centers and the Water Use Feedback Loop

Basit A. Akinade , Amobichukwu C. Amanambu , Jonathan M. Frame , Shaolei Ren This is my paper

Pith reviewed 2026-06-26 12:21 UTC · model grok-4.3

classification 💻 cs.CY
keywords AI data centerswater consumptionfeedback loopWater Consumption Impact indexwater scarcitycoolingcommunity utility burdendata center siting
0
0 comments X

The pith

The Water Consumption Impact index shows AI data center water use burdens local utilities from 0.2% to 134% of capacity across ten US sites.

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

This review paper identifies three separate dynamics—data center water consumption for cooling, water scarcity limiting siting choices, and AI tools that can raise water system efficiency—and argues they interact as a single feedback loop. The authors formalize this loop and introduce the Water Consumption Impact index as a way to measure how much a data center strains the water utility of its host community. Applying the index to ten US locations reveals that the burden spans three orders of magnitude, from negligible shares of local capacity to more than the entire utility supply.

Core claim

The paper claims that the Water and AI Feedback Loop can be formalized by linking cooling-driven consumption, scarcity-driven siting limits, and AI-enabled efficiency gains; the new Water Consumption Impact index then quantifies the resulting community-scale utility burden and demonstrates that this burden ranges from 0.2% to 134% of host capacity at the ten studied US sites.

What carries the argument

The Water Consumption Impact index, which quantifies community-scale utility burden from data center water consumption.

If this is right

  • Data center location decisions are shaped by local water availability through the feedback loop.
  • Some host communities experience negligible added burden while others face demands exceeding their entire utility capacity.
  • AI-driven improvements in water management can close the loop by offsetting consumption increases.
  • Treating the three dynamics in isolation underestimates their combined effects on siting and resource planning.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The index could be adapted to compare burdens from other large water users such as manufacturing plants.
  • Regional water planning agencies could incorporate the index when reviewing new data center permits.
  • Extending the analysis beyond the US would test whether the three-order-of-magnitude variation holds in different climates and regulatory settings.

Load-bearing premise

The ten US sites provide a representative sample of the feedback loop dynamics and the index calculation accurately reflects real community water burden without unaccounted factors such as seasonal variation or alternative cooling technologies.

What would settle it

Water-use and capacity data from a new set of data centers outside the original ten sites that fall consistently outside the reported 0.2–134% range would falsify the claimed span of the burden.

Figures

Figures reproduced from arXiv: 2606.21760 by Amobichukwu C. Amanambu, Basit A. Akinade, Jonathan M. Frame, Shaolei Ren.

Figure 1
Figure 1. Figure 1: The Water and AI Feedback Loop. Three coupled pathways, each rooted in a distinct disciplinary domain, form a feedback structure: burden pathways (hydrology) through which AI stresses source watersheds and municipal systems, constraint pathways (infrastructure engineering) through which pipe capacity and treatment limits reshape AI development, and adaptive pathways (computer science) through which AI tool… view at source ↗
Figure 2
Figure 2. Figure 2: Conceptual watershed accounting framework for AI data centre water consumption. The schematic traces water from headwater snowmelt and precipitation through regulated storage and municipal treatment to the major demand sectors within a host community. Evaporative cooling at the data centre consumptively removes 70–90% of withdrawn water from the watershed, a consumptive loss not captured by conventional wi… view at source ↗
Figure 3
Figure 3. Figure 3: Water Consumption Impact and community water share across ten US data-centre locations. [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Three-factor decomposition of WCI and household-equivalent footprints across ten US data-centre [PITH_FULL_IMAGE:figures/full_fig_p019_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Policy lever sensitivity and projected WCI growth for ten US data-centre locations. [PITH_FULL_IMAGE:figures/full_fig_p020_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Conceptual synthesis of the Water–AI Feedback Loop framework. The figure organises eight linked components spanning: burden pathways (Steps 1–2), which show how data center cooling demand and extreme consumptive and peaking factors stress host water systems; the WCI framework (Steps 3–5), which introduces a community-scale burden index, identifies three actionable policy levers, translates burden into hous… view at source ↗
read the original abstract

AI data centres consume water for cooling, water scarcity constrains siting, and AI tools can improve water system efficiency. These dynamics are studied separately yet form a feedback loop. This review formalises the Water and AI Feedback Loop, introduces the Water Consumption Impact index to quantify community-scale utility burden, and demonstrates across ten US sites that burden spans three orders of magnitude, from 0.2% to 134% of host capacity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript formalizes the Water and AI Feedback Loop (data-center cooling water use, scarcity-driven siting constraints, and AI-enabled efficiency gains), introduces a Water Consumption Impact index to quantify community-scale utility burden, and applies the index to ten US sites to report that the burden spans three orders of magnitude (0.2 % to 134 % of host capacity).

Significance. If the index definition, data sources, and calculations are shown to be robust, the reported three-order-of-magnitude variability would provide a concrete, site-specific metric for assessing water-system stress from AI infrastructure and could usefully integrate previously separate strands of research on cooling technology, water policy, and AI optimization.

major comments (1)
  1. [Abstract] Abstract: the central empirical claim (three-order-of-magnitude span across ten sites) is presented without any definition or formula for the Water Consumption Impact index, without listing the ten sites or selection criteria, and without describing data sources, seasonal adjustments, or uncertainty quantification; these omissions make the reported range unverifiable and therefore load-bearing for the paper's main result.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for highlighting the need for greater self-containment in the abstract. We agree that the central empirical claim requires supporting details to be verifiable at a glance and will revise accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central empirical claim (three-order-of-magnitude span across ten sites) is presented without any definition or formula for the Water Consumption Impact index, without listing the ten sites or selection criteria, and without describing data sources, seasonal adjustments, or uncertainty quantification; these omissions make the reported range unverifiable and therefore load-bearing for the paper's main result.

    Authors: We agree the abstract as written omits key definitional and methodological elements. The full manuscript defines the Water Consumption Impact index (including its formula) in Section 3, enumerates the ten sites together with explicit selection criteria in Section 4.1, and details the underlying data sources, seasonal adjustments, and uncertainty quantification in Sections 4.2–4.3 and 5. To make the abstract self-contained, we will insert a concise one-sentence definition of the index, name the sites and selection rationale, and briefly note the data sources and adjustments. These additions will render the reported range verifiable from the abstract while remaining within typical length limits. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper introduces the Water Consumption Impact index as a new quantification tool and applies it directly to observed data from ten US sites to report a range in burden. No equations, derivations, fitted parameters renamed as predictions, or self-citation chains are described that would reduce any claimed result to its own inputs by construction. The abstract and summary contain no load-bearing steps matching the enumerated circularity patterns; the central demonstration is an empirical application of a defined index rather than a self-referential derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no information on free parameters, axioms, or invented entities can be extracted from the provided text.

pith-pipeline@v0.9.1-grok · 5602 in / 1155 out tokens · 30409 ms · 2026-06-26T12:21:49.399038+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

128 extracted references · 47 canonical work pages · 3 internal anchors

  1. [1]

    Shehabi, S

    A. Shehabi, S. J. Smith, A. Hubbard, A. Newkirk, N. Lei, M. A. Siddik, B. Holecek, J. G. Koomey, E. Masanet, and D. A. Sartor. 2024 United States data center energy usage report. Technical Report LBNL-2001637,LawrenceBerkeleyNationalLaboratory,2024.URLhttps://eta.lbl.gov/publications/ 2024-lbnl-data-center-energy-usage-report

    2024

  2. [2]

    Powering intelligence 2026: Updated scenarios of U.S

    Electric Power Research Institute. Powering intelligence 2026: Updated scenarios of U.S. data center electricity use and power strategies. Technical report, EPRI, February 2026. URL https: //powering-intelligence.epri.com

    2026

  3. [3]

    Energy Information Administration

    U.S. Energy Information Administration. Short-term energy outlook. Technical report, U.S. Depart- ment of Energy, April 2026. URL https://www.eia.gov/outlooks/steo/

    2026

  4. [4]

    D. Mytton. Data centre water consumption.npj Clean Water, 4:11, 2021. doi:10.1038/s41545-021- 00101-w

    work page doi:10.1038/s41545-021- 2021

  5. [5]

    M. A. Siddik, A. Shehabi, and L. Marston. The environmental footprint of data centers in the United States.Environmental Research Letters, 16(6):064017, 2021. doi:10.1088/1748-9326/abfba1

    work page doi:10.1088/1748-9326/abfba1 2021

  6. [6]

    MakingAIless“thirsty”: Uncovering and addressing the secret water footprint of AI models.Communications of the ACM, 68(7):54–63,

    PengfeiLi,JianyiYang,MohammadA.Islam,andShaoleiRen. MakingAIless“thirsty”: Uncovering and addressing the secret water footprint of AI models.Communications of the ACM, 68(7):54–63,

  7. [7]

    doi:10.1145/3724499. 38

    work page doi:10.1145/3724499

  8. [8]

    Jegham, S

    N. Jegham, S. Abdelatti, A. Abdulwahab, and M. Gupta. How hungry is AI? benchmarking energy, water, and carbon footprint of LLM inference.arXiv preprint arXiv:2505.09598, 2025. URL https: //arxiv.org/abs/2505.09598

    arXiv 2025

  9. [9]

    N. Lei, J. Lu, A. Shehabi, and E. Masanet. The water use of data center workloads: A review and assessment of key determinants.Resources, Conservation and Recycling, 211:107612, 2025. doi:10.1016/j.resconrec.2025.108310

    work page doi:10.1016/j.resconrec.2025.108310 2025

  10. [10]

    The carbon and water footprints of data centers and what this could mean for artificial intelligence.Patterns, 6(4):101177, 2025

    Alex de Vries and Yu Gao. The carbon and water footprints of data centers and what this could mean for artificial intelligence.Patterns, 6(4):101177, 2025. doi:10.1016/j.patter.2025.101177

    work page doi:10.1016/j.patter.2025.101177 2025

  11. [11]

    Principles for sustainable water use by data centers

    Water-AI Nexus Center of Excellence. Principles for sustainable water use by data centers. Insight Report, 2025. URL https://water-ai-nexus.org/insight-report/

    2025

  12. [12]

    The water footprint of artificial intelligence: Emerging solutions and gover- nance imperatives.Water Research, 299:125866, 2026

    Zohar Barnett-Itzhaki. The water footprint of artificial intelligence: Emerging solutions and gover- nance imperatives.Water Research, 299:125866, 2026. doi:10.1016/j.watres.2026.125866

    work page doi:10.1016/j.watres.2026.125866 2026

  13. [13]

    Google’s Uruguay data center approved after reformulation

    DataCenterDynamics. Google’s Uruguay data center approved after reformulation. DataCenterDy- namics news article, 2023. URL https://www.datacenterdynamics.com/en/news/googles-uruguay- data-center-approved-after-reformulation/

    2023

  14. [14]

    The cloud vs

    Mongabay. The cloud vs. drought: Water hog data centers threaten Latin America, critics say. Mongabay environmental news, 2023. URL https://news.mongabay.com/2023/11/the-cloud- vs-drought-water-hog-data-centers-threaten-latin-america-critics-say/

    2023

  15. [15]

    Divergent futures in a damaged territory: The rise of data centers and water conflicts in Santiago de Chile.Journal of Urban Technology, 32(4):51–68, 2025

    Martin Tironi and Camila Albornoz. Divergent futures in a damaged territory: The rise of data centers and water conflicts in Santiago de Chile.Journal of Urban Technology, 32(4):51–68, 2025. doi:10.1080/10630732.2025.2546784

    work page doi:10.1080/10630732.2025.2546784 2025

  16. [16]

    The ongoing impact of Amsterdam’s data center moratorium

    DataCenterDynamics. The ongoing impact of Amsterdam’s data center moratorium. DataCenterDy- namics analysis, 2022. URL https://www.datacenterdynamics.com/en/analysis/the-ongoing-impact- of-amsterdams-data-center-moratorium/

    2022

  17. [17]

    Policy decision paper, 2025

    CommissionforRegulationofUtilities.Largeenergyusersconnectionpolicydecision(CRU/202504). Policy decision paper, 2025. URL https://consult.cru.ie/en/consultation/review-large-energy-users- connection-policy

    2025

  18. [18]

    Waterandwastewateragreement(ProjectDomino/OrlaLLC/Meta)

    CityofLebanon,Indiana. Waterandwastewateragreement(ProjectDomino/OrlaLLC/Meta). Legal Agreement, 2025. URL https://lebanon.in.gov/wp-content/uploads/2025/01/Project-Domino-Water- and-Wastewater-Agreement-Final-11.03.25.pdf

    2025

  19. [19]

    Small bottle, big pipe: Quantifying and addressing the impact of data centers on public water systems.arXiv preprint arXiv:2603.02705,

    Yuelin Han, Pengfei Li, Adam Wierman, and Shaolei Ren. Small bottle, big pipe: Quantifying and addressing the impact of data centers on public water systems.arXiv preprint arXiv:2603.02705,

    arXiv

  20. [20]

    Preprint

    doi:10.48550/arXiv.2603.02705. Preprint

    work page doi:10.48550/arxiv.2603.02705

  21. [21]

    Global data center water stress analysis: 43% in high-stress areas

    S&P Global Platts. Global data center water stress analysis: 43% in high-stress areas. Technical report,S&PGlobal,2025. URLhttps://www.spglobal.com/sustainable1/en/insights/special-editorial/ beneath-the-surface-water-stress-in-data-centers

    2025

  22. [22]

    Hayslep, E

    M. Hayslep, E. Keedwell, and R. Farmani. Leakage prediction in real-world water distribution networksusingmulti-objectivemulti-genegeneticprogramming.ACMTransactionsonEvolutionary Learning and Optimization, 1(1):1–25, 2025. doi:10.1145/3729431. 39

    work page doi:10.1145/3729431 2025

  23. [23]

    A survey on intrusion de- tection system: Feature selection, model, performance measures, application perspective, challenges, and fu- ture research directions,

    R.J.L.Taloma. Machinelearningforsmartwaterdistributionsystems: Exploringapplications,chal- lenges and future perspectives.Artificial Intelligence Review, 58(3):75, 2025. doi:10.1007/s10462- 024-11093-7

    work page doi:10.1007/s10462- 2025

  24. [24]

    G. Fu, Y. Jin, S. Sun, Z. Yuan, and D. Butler. The role of deep learning in urban water management: A critical review.Water Research, 223:119098, 2022. doi:10.1016/j.watres.2022.118973

    work page doi:10.1016/j.watres.2022.118973 2022

  25. [25]

    AI-enhanced digital twin platform for smart water distribution: Integrating machine learning models with IoT-driven predictive analytics

    WaterTwin-AI. AI-enhanced digital twin platform for smart water distribution: Integrating machine learning models with IoT-driven predictive analytics. Preprint, 2025. URL https://www.preprints. org/manuscript/202510.1750

    arXiv 2025

  26. [26]

    F.Kratzert,D.Klotz,C.Brenner,K.Schulz,andM.Herrnegger. Towardslearninguniversal,regional, and local hydrological behaviors via machine learning applied to large-sample datasets.Hydrology and Earth System Sciences, 23(12):5089–5110, 2019. doi:10.5194/hess-23-5089-2019

    work page doi:10.5194/hess-23-5089-2019 2019

  27. [27]

    G. S. Nearing, F. Kratzert, A. K. Sampson, C. M. Pelissier, D. Klotz, J. M. Frame, and H. V. Gupta. Whatroledoeshydrologicalscienceplayintheageofmachinelearning?WaterResourcesResearch, 57(3):e2020WR028091, 2021. doi:10.1029/2020WR028091

    work page doi:10.1029/2020wr028091 2021

  28. [28]

    Liemberger and A

    R. Liemberger and A. Wyatt. Quantifying the global non-revenue water problem.Water Supply, 19 (3):831–837, 2019. URL https://doi.org/10.2166/ws.2018.129

    work page doi:10.2166/ws.2018.129 2019

  29. [29]

    Lei and E

    N. Lei and E. Masanet. Climate- and technology-specific PUE and WUE estimations for U.S. data centers using a hybrid statistical and thermodynamics-based approach.Resources, Conservation and Recycling, 180:106175, 2022. doi:10.1016/j.resconrec.2022.106323

    work page doi:10.1016/j.resconrec.2022.106323 2022

  30. [30]

    M. A. B. Siddik, J. R. Sanchez, and A. Shehabi. Spatially and temporally detailed water and carbon footprintsofU.S.electricitygenerationanduse.WaterResourcesResearch,50(12):WR038350,2024. doi:10.1029/2024WR038350

    work page doi:10.1029/2024wr038350 2024

  31. [31]

    Amanambu, Jonathan Frame, and Adeyemi Olusola

    Amobichukwu C. Amanambu, Jonathan Frame, and Adeyemi Olusola. Where AI runs determines water stress.ESSOAr preprint, 2026. doi:10.22541/essoar.15001296/v1

    work page doi:10.22541/essoar.15001296/v1 2026

  32. [32]

    S. N. Ahmed, K. Bencala, S. Nummer, C. L. Schultz, and A. Seck. 2025 Washington Metropolitan Areawatersupplystudy. TechnicalReportICPRBReport25-1,InterstateCommissiononthePotomac River Basin, 2025. URL https://www.potomacriver.org/wp-content/uploads/2025/12/2025_WMA_ Water_Supply_Study_ICPRB_Dec-2025.pdf

    2025

  33. [33]

    Karimi, L

    L. Karimi, L. Yacuel, J. D. Johnson, J. Ashby, M. Green, M. Renner, A. Bergman, R. Norwood, and K. L. Hickenbottom. Water-energy tradeoffs in data centers: A case study in hot-arid climates. Resources, Conservation and Recycling, 181:106194, 2022. doi:10.1016/j.resconrec.2022.106194

    work page doi:10.1016/j.resconrec.2022.106194 2022

  34. [34]

    Medalie, J

    L. Medalie, J. W. Roy, and D. Lorenz. USGS water use statistics for the nation. Technical report, US Geological Survey, 2025. URL https://pubs.usgs.gov/publication/pp1894D/full

    2025

  35. [35]

    Data centers and water use in the Potomac River basin, March 2026

    Interstate Commission on the Potomac River Basin. Data centers and water use in the Potomac River basin, March 2026. URL https://www.potomacriver.org/focus-areas/water-resources-and-drinking- water/water-resources/planning/data-centers-and-water-use-in-the-potomac-river-basin/

    2026

  36. [36]

    Datacenterwateruse: Frequentlyaskedquestions

    SusquehannaRiverBasinCommission. Datacenterwateruse: Frequentlyaskedquestions. Technical report, SRBC, 2025. URL https://www.srbc.gov/about/news/docs/srbc-data-centers-faq.pdf. 40

    2025

  37. [37]

    Zúñiga Uribe et al

    J. Zúñiga Uribe et al. Artificial intelligence in water distribution networks: A systematic review of models, input variables, databases, and output strategies for leak detection.Smart Cities, 9(3):45,

  38. [38]

    URL https://www.mdpi.com/2624-6511/9/3/45

    doi:10.3390/smartcities9030045. URL https://www.mdpi.com/2624-6511/9/3/45

    work page doi:10.3390/smartcities9030045

  39. [39]

    and LimnoTech

    Meta Platforms Inc. and LimnoTech. Volumetric water benefits assessment for data center opera- tions. Technical assessment, 2025. URL https://sustainability.atmeta.com/wp-content/uploads/2025/ 08/REVISED_FINAL_Meta_2024_Volumetric_Water_Benefit_Report_08.21.2025.pdf

    2025

  40. [40]

    Datacenterswaterfootprint: Theneedformoretransparency

    C.Privette,W.Sobczak,andJ.Zhuang. Datacenterswaterfootprint: Theneedformoretransparency. AGU Advances, 7(1):e2025AV002140, 2026. doi:10.1029/2025AV002140

    work page doi:10.1029/2025av002140 2026

  41. [41]

    The data heat island effect: quantifying the impact of AI data centers in a warming world.arXiv preprint arXiv:2603.20897,

    Andrea Marinoni, Erik Cambria, Luca Dal Zilio, Weisi Lin, Mauro Dalla Mura, Jocelyn Chanussot, Edoardo Ragusa, Chi Yan Tso, Yihao Zhu, and Benjamin Horton. The data heat island effect: quantifying the impact of AI data centers in a warming world.arXiv preprint arXiv:2603.20897,

    Pith/arXiv arXiv

  42. [42]

    doi:10.48550/arXiv.2603.20897

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2603.20897

  43. [43]

    Here comes the cloud: The effect of data centers on homeowners in Northern Virginia.SSRN preprint, 2026

    Michael Rubinovitz. Here comes the cloud: The effect of data centers on homeowners in Northern Virginia.SSRN preprint, 2026. doi:10.2139/ssrn.6136790

    work page doi:10.2139/ssrn.6136790 2026

  44. [44]

    Facility-levelenergy-drivenwaterfootprintandscarcityimplications of Chinese data centers: A bottom-up analysis and scenario-based projection.Applied Energy, 399: 126522, 2025

    F.Jiang,C.Duan,andB.Chen. Facility-levelenergy-drivenwaterfootprintandscarcityimplications of Chinese data centers: A bottom-up analysis and scenario-based projection.Applied Energy, 399: 126522, 2025. doi:10.1016/j.apenergy.2025.126522

    work page doi:10.1016/j.apenergy.2025.126522 2025

  45. [45]

    Data center electricity consumption in Ireland: 22% of na- tionalgrid

    Central Statistics Office Ireland. Data center electricity consumption in Ireland: 22% of na- tionalgrid. Statisticalreport,2025. URLhttps://www.cso.ie/en/releasesandpublications/ep/p-dcmec/ datacentresmeteredelectricityconsumption2024/

    2025

  46. [46]

    2025 corporate environmental report

    Apple Inc. 2025 corporate environmental report. Technical report, Apple Inc., 2025. URL https: //www.apple.com/environment/pdf/Apple_Environmental_Progress_Report_2025.pdf

    2025

  47. [47]

    The Dalles data center water expansion: Community and infrastructure reporting

    Oregon Public Broadcasting. The Dalles data center water expansion: Community and infrastructure reporting. NewsInvestigation,2026. URLhttps://www.opb.org/article/2026/01/15/as-googles-water- demands-grow-the-dalles-aims-to-pull-more-from-mount-hood-forest/

    2026

  48. [48]

    Microsoft data centers will use up to 8.4 million gallons of water each year, recordsshow

    Wisconsin Public Radio. Microsoft data centers will use up to 8.4 million gallons of water each year, recordsshow. WisconsinPublicRadionewsarticle,basedonpublicrecordsrequest,September2025. URL https://www.wpr.org/news/microsoft-data-centers-8-million-gallons-water-each-year

  49. [49]

    S. N. Ahmed, K. Bencala, S. Nummer, C. L. Schultz, and A. Seck. 2025 Washington Metropolitan Area water supply study: Demand and resource availability forecast for the year 2050.Interstate Commission on the Potomac River Basin (ICPRB REPORT NO. ICP-596), December 2025

    2025

  50. [50]

    Drinking water infrastructure needs survey and assessment, 7th report

    US EPA. Drinking water infrastructure needs survey and assessment, 7th report. Technical Report EPA/816-R-23-001, United States Environmental Protection Agency, 2023. URL https://www.epa. gov/dwsrf/epas-7th-drinking-water-infrastructure-needs-survey-and-assessment

    2023

  51. [51]

    Infrastructure investment benefits and water system resilience

    Value of Water Campaign. Infrastructure investment benefits and water system resilience. White Paper Report, 2025. URL https://www.thevalueofwater.org/

    2025

  52. [52]

    Alkrush, E

    A. Alkrush, E. Habib, and H. Alquobaity. Data centers cooling: A critical review of techniques, challenges, and energy saving solutions.International Journal of Refrigeration, 160:246–262, 2024. doi:10.1016/j.ijrefrig.2024.02.007. 41

    work page doi:10.1016/j.ijrefrig.2024.02.007 2024

  53. [53]

    W. E. Gnibga, A. A. Chien, A. Blavette, and A. C. Orgerie. FlexCoolDC: Datacenter cooling flexibilityforharmonizingwater,energy,carbon,andcosttrade-offs. InProceedingsofthe15thACM InternationalConferenceonFutureandSustainableEnergySystems(E-Energy’24),pages305–316. ACM, 2024. doi:10.1145/3632775.3661936

    work page doi:10.1145/3632775.3661936 2024

  54. [54]

    D. E. Sickinger, O. D. Van Geet, S. A. Belmont, T. Carter, and D. Martinez. Thermosyphon cooler hybridsystemforwatersavingsinanenergy-efficientHPCdatacenter: Resultsfrom24monthsandthe impact on water usage effectiveness. Technical Report NREL/TP-2C00-72196, National Renewable Energy Laboratory, 2018

    2018

  55. [55]

    Jager and H

    H. Jager and H. S. Yoon. Data centers as an emerging threat to freshwater biodiversity in the United States.Water Biology and Security, 5:100585, 2026. doi:10.1016/j.watbs.2026.100585

    work page doi:10.1016/j.watbs.2026.100585 2026

  56. [56]

    Data center water demand analysis and global infrastructure investment

    Roland Berger. Data center water demand analysis and global infrastructure investment. Technical report, Roland Berger GmbH, 2025. URL https://www.rolandberger.com/en/Insights/Publications/ Water-demands-of-the-data-economy.html

    2025

  57. [57]

    Lebanon, Indiana data center announcement

    Meta Platforms Inc. Lebanon, Indiana data center announcement. Press Release, 2026. URL https://about.fb.com/news/2026/02/metas-new-data-center-lebanon-indiana-marks-milestone- ai-investment/

    2026

  58. [58]

    Farfan and A

    F. Farfan and A. Lohrmann. Gone with the clouds: Estimating the electricity and water foot- print of digital data services in Europe.Energy Conversion and Management, 277:116617, 2023. doi:10.1016/j.enconman.2023.117225

    work page doi:10.1016/j.enconman.2023.117225 2023

  59. [59]

    C. Falk, S. Acharyya, B. Chou, and J. Sorensen. From FLOPs to footprints: The resource cost of artificialintelligence.arXivpreprintarXiv:2512.04142,2025. URLhttps://arxiv.org/abs/2512.04142

    arXiv 2025

  60. [60]

    Wattnet: matching electricity consumption with low-carbon, low-water footprint energy supply

    M. Castrill, P. Melguizo, and A. García. Wattnet: matching electricity consumption with low-carbon, low-water footprint energy supply.arXiv preprint arXiv:2601.11623., 2026. doi:10.48550/arXiv.2601.11623

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2601.11623 2026

  61. [61]

    Herrera, G

    M. Herrera, G. Ortiz, L. Gonzalez, and J. Perez. Sustainable AI infrastructure: A scenario-based forecast of water footprint under uncertainty.Journal of Cleaner Production, 381:135038, 2025. doi:10.1016/j.jclepro.2025.146528

    work page doi:10.1016/j.jclepro.2025.146528 2025

  62. [62]

    Environmentalimpactandnet-zeropathwaysforsustainableartificialintelligence serversintheUSA.NatureSustainability,8(12):1456–1468,2025.doi:10.1038/s41893-025-01681-y

    B.XiaoandF.You. Environmentalimpactandnet-zeropathwaysforsustainableartificialintelligence serversintheUSA.NatureSustainability,8(12):1456–1468,2025.doi:10.1038/s41893-025-01681-y

    work page doi:10.1038/s41893-025-01681-y 2025

  63. [63]

    North America data center market report: $1 trillion projected development

    JLL. North America data center market report: $1 trillion projected development. Technical report, JonesLangLaSalle,2025.URLhttps://www.jll.com/en-us/insights/market-dynamics/north-america- data-centers

    2025

  64. [64]

    US data center economic analysis: $6.7 trillion cumu- lative capex projection

    McKinsey and Company. US data center economic analysis: $6.7 trillion cumu- lative capex projection. Technical report, McKinsey and Company, 2025. URL https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/ the-cost-of-compute-a-7-trillion-dollar-race-to-scale-data-centers

    2025

  65. [65]

    Sameer H. Shah. Four water insecurity concerns about datacenters driving the AI revolution.PLOS Water, 5(1):e0000500, 2026. doi:10.1371/journal.pwat.0000500. 42

    work page doi:10.1371/journal.pwat.0000500 2026

  66. [66]

    URL https://www.nytimes.com/2025/07/14/technology/meta-data-center-water.html

    JackHealyandMarinaFang.TheirwatertapsrandrywhenMetabuiltnextdoor.TheNewYorkTimes, July 2025. URL https://www.nytimes.com/2025/07/14/technology/meta-data-center-water.html

    2025

  67. [67]

    Port Washington, Wisconsin data center water infrastructure reporting

    Spectrum News. Port Washington, Wisconsin data center water infrastructure reporting. News Coverage, 2025. URL https://spectrumnews1.com/wi/milwaukee/news/2025/09/23/data-center-port- washington

    2025

  68. [68]

    Z. J. Lyle, J. M. VanBriesen, and C. Samaras. Climate risk index for US drinking water utilities: As- sessinghazard,vulnerability,andexposurefor1,455utilities.CommunicationsEarth&Environment, 7:68, 2025. doi:10.1038/s43247-025-03044-z

    work page doi:10.1038/s43247-025-03044-z 2025

  69. [69]

    S. Solomon. Sustainable by design: Next-generation datacenters consume zero wa- ter for cooling. The Microsoft Cloud Blog, December 9, 2024, 2024. URL https://www.microsoft.com/en-us/microsoft-cloud/blog/2024/12/09/sustainable-by-design-next- generation-datacenters-consume-zero-water-for-cooling/

    2024

  70. [70]

    Commission Delegated Regulation (EU) 2024/1364 establishing a common Union rating scheme for data centres

    European Commission. Commission Delegated Regulation (EU) 2024/1364 establishing a common Union rating scheme for data centres. Official Journal of the European Union, 2024. URL https: //eur-lex.europa.eu/eli/reg_del/2024/1364/oj/eng

    2024

  71. [71]

    Data drain: The land and water impacts of the AI boom.Land Lines Maga- zine, 2025

    Jon Gorey. Data drain: The land and water impacts of the AI boom.Land Lines Maga- zine, 2025. URL https://www.lincolninst.edu/publications/land-lines-magazine/articles/land-water- impacts-data-centers/

    2025

  72. [72]

    Green Data Centre Technology Roadmap

    Infocomm Media Development Authority. Green Data Centre Technology Roadmap. Technical report, IMDA, Government of Singapore, 2024. URL https://www.imda.gov.sg/how-we-can-help/ green-dc-roadmap

    2024

  73. [73]

    Energy Efficiency Act (Energieeffizienzgesetz, EnEfG)

    Federal Republic of Germany. Energy Efficiency Act (Energieeffizienzgesetz, EnEfG). Federal Law Gazette, 2023. URL https://climate-laws.org/document/energy-efficiency-act-enefg_d64e

    2023

  74. [74]

    FOIA Friday: Professor, students find NDAs between localities and data cen- ter companies

    Virginia Mercury. FOIA Friday: Professor, students find NDAs between localities and data cen- ter companies. Virginia Mercury, 2024. URL https://virginiamercury.com/2024/10/25/foia-friday- professor-students-find-ndas-with-data-center-companies-localities-say-theyre-used-often/

    2024

  75. [75]

    SC county sued over its Google data center water-use secrecy

    Joseph Bustos. SC county sued over its Google data center water-use secrecy. The Post and Courier, 2024. URL https://www.postandcourier.com/business/dorchester-google-data-center-foia- lawsuit-water-charleston/article_13536bb8-fe45-11ee-8e2c-ab63a46ae251.html

    2024

  76. [76]

    Lawsuit seeks to force release of projected water use by Microsoft data centers

    Wisconsin Public Radio. Lawsuit seeks to force release of projected water use by Microsoft data centers. Wisconsin Public Radio, 2025. URL https://www.wpr.org/news/microsoft-data-centers- lawsuit-force-release-projected-water-use

    2025

  77. [77]

    Microsoft water and sewer infrastructure agreement

    Town of Leesburg, Virginia. Microsoft water and sewer infrastructure agreement. Capital Projects, Town of Leesburg, 2024. URL https://www.leesburgva.gov/departments/capital-projects/current- projects/utilities-projects/microsoft-water-and-sewer

    2024

  78. [78]

    Louisiana data center investment announcement: $400M water infrastructure

    Amazon News. Louisiana data center investment announcement: $400M water infrastructure. PressRelease,2026. URLhttps://www.aboutamazon.com/news/company-news/amazon-data-center- louisiana-new-jobs. 43

    2026

  79. [79]

    Tackling the trillion-gallon problem: Water loss reduction ini- tiatives benefit utilities and consumers alike

    Megan Bondar. Tackling the trillion-gallon problem: Water loss reduction ini- tiatives benefit utilities and consumers alike. Bluefield Research, December 2025. URL https://www.bluefieldresearch.com/tackling-the-trillion-gallon-problem-water-loss-reduction- initiatives-benefit-utilities-and-consumers-alike/

    2025

  80. [80]

    Jun and S

    H. Jun and S. Jung. Exploration of deep learning leak detection model across multiple smart water distribution systems.Water Resources Management, 39(4):1234–1253, 2025. doi:10.1007/s11269- 025-04272-w

    work page doi:10.1007/s11269- 2025

Showing first 80 references.