![Data center being built ](/sites/g/files/omnuum10826/files/styles/hwp_21_9__1920x825/public/2026-10/AdobeStock_1996037071%20data%20center.jpg?itok=rR13we4m) # How Cities Can Regulate Data Centers to Prioritize Air Quality City leaders can regulate data center development to protect air quality by leveraging permitting reforms, strategic coordination, and data-driven mapping to mitigate environmental impacts on their communities. October 06, 2026 As the proliferation of energy-hungry data centers sparks national debate, local leaders face a critical question: How can they address the impacts of data centers on local air quality? The answer, ultimately, is a local one; there's no one-size-fits-all approach, since the right strategy depends on a city's unique footprint, existing pollution burden, and available regulatory tools. Local control gives municipal leaders the flexibility to tailor their approach. A recent virtual session from the Community Data Health Initiative (CDHI), a Robert Wood Johnson funded project which focuses on tackling environmental challenges that directly affect people’s health, offered some answers for cities where air quality is the primary concern. Adam Freed of Bloomberg Associates spoke to the CDHI’s air quality learning cohort, walking them through key considerations surrounding data centers and air quality, and ways city halls can leverage local powers to protect residents’ health. Freed drew upon [a new playbook for city mayors and municipal leaders](https://www.climatemayors.org/data-centers), written in partnership with Freed, his Bloomberg Associates colleagues Flavio Coppola and Sheetal Shah, and [Climate Mayors](https://www.climatemayors.org/) meant to help municipal leaders leverage local powers to meet the data center boom with strategies that reinforce sustainability goals and mitigate negative impacts on communities. He noted four key levers city leaders can pull on: regulation, negotiation, coalition building and strategic leadership. “So much of the development’s happening outside of cities,” Freed said, “but the impacts are on cities and urban residents.” The playbook also shares several policy case studies to demonstrate how local leaders can influence data center development with land-use planning and other tools. While the session focused on actions to protect air quality, the key recommendations Freed shared show how cities can influence many aspects of data center development. Here are key takeaways from the meeting. ## Treat Permitting as a Lever Freed outlined principles and actions municipal leaders can apply when evaluating data center proposals in their communities. He recommended eliminating “by-right” permitting – in which data center projects face minimal review -- in favor of a special-use permit process. “One of the first things cities can do is, as quickly as possible, get rid of as-of-right permitting for data centers and make it a special or conditional use ,” explained Freed. “That is going to give city leaders a lot more leverage and power over how data centers are being built, designed, and operated.” Cities should evaluate the following areas when making data center decisions: - Public health burdens and/or air quality, especially in communities that already contend with cumulative pollution and health risks - Existing industrial activity and clustered data center development, rather than evaluating each data center in isolation - Backup power technology and strategies to curtail greenhouse-gas emissions - Clearly designated and enforceable air quality standards, including regular and transparent data collection along with clear consequences enforceable at the city level ## Understand the Purpose and Ownership Freed emphasized that not all data centers are the same in terms of community impacts. The purpose and ownership of the data center in question impacts a community and should inform how city leaders negotiate terms in the permitting and approval process. [BloombergNEF](https://about.bnef.com/), the strategic research provider, tracks the technologies driving the transition to a low-carbon economy, categorized data centers into four types based on their workloads – each with distinct implications for electricity demand, cooling requirements, and siting. From highest to lowest impact on communities, the four categories are: 1. AI training facilities: large facilities that process huge datasets for demanding computational tasks, which therefore require the highest amount of energy and cooling resources. 2. Non-AI hyperscale centers: large, continuously operating facilities used for general cloud computing, storage, and enterprise workloads, which require significant but less concentrated power needs than AI training facilities. 3. AI inference facilities: smaller facilities that run optimized, pre-trained computational models for specific tasks, which require less power than training facilities and often need to be closer to users. 4. Co-location facilities: multi-tenant sites with variable energy load and cooling needs that feature the lowest relative intensity workloads. ## Coordinate Across Leadership Levels “We recognize that cities don't have all the power, and they do need to rely on state support for a lot of these efforts,” said Freed. “So, city-state coordination, particularly around air quality, is important.” While each context is different, depending on which level of government oversees or regulates air quality or what impacts can be managed through zoning, in many cases a city leader must coordinate at the county or state level. The Climate Mayors guide recommends coordination in a few specific areas: - Mandating air quality permits for industrial backup generator NOx and PM emissions tests within 30 days of permit application or renewal - Banning generator testing and maintenance on days with high air pollution forecasts - Setting hourly limits on non-emergency use of diesel generators as well as disclosure of the use (including maintenance, testing, and demand response) Furthermore, given the regional nature of data center effects, Freed recommended cities work together to leverage their political power. Speaking about a utility regulator in one state, Freed explained: “This regulator was saying yes to different cities on different regulations. Had they banded together and said, ‘We five cities want these things’, they would have been in a much better position. Doing that city-to-city collaboration in these cases can be really powerful.” ## Use Mapping and Visualizations “There are approximately 3,000 new data centers that are under construction or planned within the U.S., many in your communities or regionally,” Freed told attendees. Understanding current community conditions before approving new data centers is essential to evaluating cumulative air quality impact. Mapping is particularly critical for environmental work; [localized spatial data](https://datasmart.hks.harvard.edu/why-spatial-data) has in some cases revealed wide-ranging [disparities in temperature and pollution levels mere blocks apart](https://www.wri.org/insights/cities-getting-hotter-heat-inequity). There are a few tools to help cities get started. For example, CDHI designed a practical framework to help city teams tackle environmental issues and [embed geographic information systems into the policymaking process](https://datasmart.hks.harvard.edu/using-spatial-analytics-address-localized-environmental-harm), with a focus on equity and health. The geo-spatial company [Esri offers environmental and climate justice data](https://www.esri.com/en-us/arcgis/products/living-atlas?rsource=https%3A%2F%2Fwww.esri.com%2Fen-us%2Farcgis%2Fproducts%2Fdata%2Fdata-portfolio%2Fenvironmental) layers, and the federal Environmental Protection Agency provides an interactive [EnviroAtlas map](https://www.epa.gov/enviroatlas). These tools, especially when paired with emerging technology like agentic AI, can help communities and policymakers better understand the potential impacts of a data center and make data-driven decisions. **Interested in joining the conversation? The CDHI Air Quality cohort brings together city leaders working on real-world, pressing environmental health issues.** [**Learn more about our cohorts and request to join here.**](https://datasmart.hks.harvard.edu/about-community-data-health-initiative) See also:- [ Artificial Intelligence ](/topics/artificial-intelligence) - [ Asset Management and Sustainability ](/topics/asset-management-and-sustainability) - [ Data Visualization ](/topics/data-visualization) - [ Environment ](/topics/environment) - [ Equity ](/topics/equity) - [ Infrastructure ](/topics/infrastructure) - [ Public Health ](/topics/public-health) - [ Regulation ](/topics/regulation) - [ Smart Grid & Utilities ](/topics/smart-grid-utilities) Share on:- [ Facebook ](#) - [ Twitter ](#) - [ Linkedin ](#)