Data–water symbiosis: Coupling data centres with wastewater treatment plants could cut 84 million tonnes of CO₂ eq and save 1,300 million m3 of freshwater annually

As artificial intelligence (AI) continues its exponential growth, the data centres powering it are consuming electricity and freshwater at an unprecedented rate. Now, a new global analysis reveals an unexpectedly elegant solution hiding in plain sight: the world's wastewater treatment plants (WWTPs) already sit close enough to data centres to form a powerful symbiotic network. By routing treated effluent to cool servers and capturing the returned heat to power sludge processing, this data–water symbiosis could eliminate approximately 84 million tonnes of CO₂ equivalent (Mt CO₂ eq) per year, conserve roughly 1,300 million m3 of freshwater, and generate net economic savings of approximately US$95.4 billion annually—all without waiting for breakthrough technologies.

Data centres currently account for roughly 40% of their total power consumption on cooling alone, a figure that is climbing as AI-driven computing demands intensify. WWTPs , by contrast, discharge enormous volumes of treated effluent carrying substantial recoverable cooling energy—yet this resource has remained almost entirely untapped. Previous efforts to green data centre operations have focused on hardware upgrades, renewable energy integration, and workload migration between facilities, while wastewater reuse for industrial cooling has been explored only in isolated regional cases. No prior study has attempted a globally coordinated framework for pairing these two infrastructure systems. Based on these challenges, there is a need for in-depth research into globally optimised data–water symbiosis as a scalable, cross-sectoral strategy for decoupling AI expansion from its environmental toll.

Researchers at the Harbin Institute of Technology, in collaboration with scientists from KWR Water Research Institute (Netherlands), the University of Exeter (UK), and the University of Michigan (USA), reported (DOI: 10.1016/j.ese.2026.100702) on 2 May 2026 in the journal Environmental Science and Ecotechnology that a globally coordinated infrastructure symbiosis between data centres and municipal WWTPs can substantially reduce greenhouse gas (GHG) emissions, freshwater consumption, and operational costs simultaneously. The study constructed a geodatabase encompassing more than 4,775 data centres and 57,547 WWTPs across 98 countries to model the proposed linkages at national scale.

The team combined geospatial analysis, optimisation algorithms, and life-cycle assessment to evaluate the environmental and economic performance of data–water symbiosis under real-world constraints. They found that data centres and WWTPs exhibit a strong global spatial co-occurrence (Moran's I = 0.67, p = 0.001), meaning they are clustered together far more often than chance would predict—making large-scale pairing not just theoretically possible but practically feasible.

Under the optimised symbiosis scenario, 3,362 WWTPs were paired with 4,725 data centres, processing 18.2 billion tonnes of treated wastewater per year with a combined cooling energy potential of 593 million MWh. The model showed that these linkages could reduce global GHG emissions by 84.3 Mt CO₂ eq annually—a 16.30% reduction relative to total emissions from the global wastewater treatment sector and an 11.4% reduction relative to total information and communications technology (ICT) industry emissions. Freshwater savings reached 1,300 million m³, equivalent to 798 times Google's global freshwater use in 2021. The greatest benefits concentrated in the United States (24.14% of total GHG mitigation), Japan (18.14%), China (16.15%), the Netherlands (9.96%), and the United Kingdom (5.92%), which together account for over 74% of global savings. Critically, 87.8% of the economically viable linkages fall within 40 km of each other, making near-term urban deployment realistic without long-distance infrastructure.

The authors said the symbiosis works because data centres and WWTPs are natural thermodynamic complements: One produces treated water that the other can use for cooling, while the other generates waste heat that the former needs. "The beauty of this approach is that it requires no new technology—we already have the engineering tools to make this work," they told EurekAlert. "What we needed was a global systems-level analysis to show policymakers and investors that the economics and the environmental benefits are both compelling at scale. The numbers speak clearly: meaningful GHG mitigation, genuine water conservation, and billions in savings—all from infrastructure that already exists."

The findings position WWTPs as far more than disposal facilities: they can evolve into active urban energy-coupling hubs that exchange heat and water across industrial systems. The researchers demonstrated the concept in a simulated Shenzhen case study, where linking a data centre to a local WWTP reduced cooling energy consumption by 4.30% and improved anaerobic digestion efficiency by 6.81%, while server-room temperatures dropped by 5.35°C. The framework is directly applicable to cities worldwide where both facilities are already present in close proximity, requiring primarily regulatory coordination and third-party heat-supply agreements rather than entirely new infrastructure. The authors argue this represents one of the most readily scalable climate solutions available to the AI sector today, and that policies incentivising spatial integration between data centres and WWTPs could unlock substantial progress toward United Nations Sustainable Development Goals 6 (Clean Water and Sanitation), 9 (Industry, Innovation and Infrastructure), and 13 (Climate Action).

###

References

DOI

10.1016/j.ese.2026.100702

Original Source URL

https://doi.org/10.1016/j.ese.2026.100702

Funding information

This work was supported by the National Natural Science Foundation of China (No. 52321005, No. 52293443, and No. 52230004), Shenzhen Science and Technology Program (No. KQTD20190929172630447), Shenzhen Key Research Project (No. GXWD20220817145054002), Shenzhen Natural Science Foundation (No. JCYJ20240813104812017), and Talent Recruitment Project of Guangdong (No. 2021QN020106). Dragan Savic has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (No. 951424).

About Environmental Science and Ecotechnology

Environmental Science and Ecotechnology (ISSN 2666-4984) is an international, peer-reviewed, and open-access journal published by Elsevier. The journal publishes significant views and research across the full spectrum of ecology and environmental sciences, such as climate change, sustainability, biodiversity conservation, environment & health, green catalysis/processing for pollution control, and AI-driven environmental engineering. The latest impact factor of ESE is 14.3, according to the Journal Citation ReportsTM 2024.