Materializing the Cloud: Data Center Siting and Environmental Justice in the United States
Name
gapihan-marine22-mcp-dusp-2026-thesis.pdf
Size
2.03 MB
Format
Adobe PDF
Checksum (MD5)
7235f52d7ec9f6ffe5b6c14d1741c217
Author(s)
Gapihan, Marine
Advisor(s)
Hsu, David
Date Issued
May 2026
Publisher
Massachusetts Institute of Technology
Abstract
Recent years have been marked by an unprecedented expansion of digital technologies and services, driven in large part by the rapid rise of artificial intelligence. At the core of this digital ecosystem are data centers, the facilities that house the computational power and storage systems of artificial intelligence. While data centers are not new and have supported the digital economy for decades, their scale, intensity and spatial footprint have changed drastically, bringing them to the forefront of public and political debate. In particular, hyperscale data centers, designed to support cloud computing and large-scale AI workloads, have come under heightened scrutiny due to their substantial demands for electricity, water, and land. These facilities often require continuous, energy-intensive operations and extensive cooling systems, leading to significant resource consumption and localized environmental impacts. Concerns have emerged around their contribution to grid stress, water scarcity, air pollution from backup generation, and broader cumulative environmental burdens in host communities.
As a result, a growing body of research has begun to examine the intersection of data center infrastructure, land use, and environmental justice. This emerging literature raises critical questions about where these facilities are sited, which communities bear their environmental and infrastructural costs, and how existing planning, zoning, and regulatory frameworks shape these outcomes. In doing so, it situates data centers not only as digital infrastructure, but also as a new form of industrial land use with potential uneven spatial and social consequences.
Using a mixed-methods approach, this thesis draws on quantitative and geospatial analysis, as well as semi-structured interviews to determine whether data center facilities are disproportionately located in places where environmental harms and structural vulnerabilities intersect, and to generate insights that can inform more equitable infrastructure governance. It seeks to understand not only where these facilities are located and being proposed, but also how data center siting decisions are made. As the development of new data centers accelerates, examining the patterns that shape data center siting offers critical insight into the emerging geography of AI infrastructure.
MIT Department
Massachusetts Institute of Technology. Department of Urban Studies and Planning
Terms of Use
In Copyright - Educational Use Permitted
Copyright retained by author(s)
Persistent DSpace Link
