Environmental, social and economic justice of data centers on urban sustainability
This study evaluates the environmental, social, and economic impacts of data centers on urban sustainability in the Los Angeles metropolitan area, revealing significant challenges such as noise pollution, exacerbated urban heat islands, and high energy consumption that threaten livability while offering policy recommendations for balancing digital growth with sustainable urban futures.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The digital world we rely on for everything from streaming movies to running artificial intelligence is not just a cloud of invisible code; it is anchored by massive physical buildings called data centers. These facilities house thousands of computers that never sleep, processing information at a speed that reshapes how cities function. However, these buildings are not neutral structures. They consume vast amounts of electricity, generate significant heat, and produce constant noise from their cooling systems. As cities like Los Angeles become the hubs for this digital infrastructure, a critical question arises: who bears the cost of this progress? The conversation is shifting from simply counting how many servers exist to understanding how their presence changes the air, the temperature, the noise levels, and the value of homes in the neighborhoods surrounding them.
A new study by researchers from University College London and Monash University takes a close look at this issue within the Los Angeles metropolitan area. The team recognized that not all data centers are the same, and their impact on a city depends entirely on where they are built and what they look like. They categorized these facilities into three distinct types: small to medium-sized centers tucked into or right next to residential neighborhoods, massive industrial campuses located in suburban industrial zones, and tall, vertical hubs packed into dense downtown office towers. By treating these three types as separate entities, the researchers could measure their specific effects on the environment, the local community, and the economy, rather than lumping them all together as a single problem.
The investigation began by examining the social and economic tension in residential areas. In the Long Beach community, researchers analyzed how the presence of a data center influenced nearby housing prices. They found that homes closer to the facility tended to have lower price values compared to those further away, suggesting that the noise and industrial nature of the building might be a deterrent for buyers. However, this effect was not overwhelmingly strong, and the data showed that other factors like job opportunities or the general location of the neighborhood often played a bigger role in determining a home's value. The study also looked at the social vulnerability of these neighborhoods, checking if data centers were being built in areas with high poverty or large populations of elderly and young residents. The findings indicated that developers often chose areas with existing industrial or commercial foundations, seemingly avoiding the most fragile communities, though the potential for conflict remained high.
Noise emerged as a primary source of friction in these residential zones. In the El Segundo community, where several data centers operate near homes, the researchers calculated exactly how far the sound travels. They determined that the constant hum of cooling systems and generators creates an unbearable environment for residents within a radius of approximately 162 meters during the day. At night, when the city is quieter, this zone of disturbance expands to about 248 meters, and on summer nights, it can reach as far as 565 meters. This spatial reality explains the fierce opposition seen in local public hearings, where residents describe the low-frequency noise as a constant intrusion that disrupts sleep and daily life.
The story changes significantly when moving to the city center. In downtown Los Angeles, data centers are often housed in towering office buildings, such as the famous One Wilshire building, which serves as a major gateway for internet traffic crossing the Pacific Ocean. Here, the primary concern is not noise, but heat. Because these facilities are packed into high-rise structures in a dense urban canyon, the waste heat they generate cannot escape easily. The study measured that the One Wilshire building alone raises the temperature of the surrounding area by 2 degrees Celsius within a radius of 0.46 kilometers, by 1 degree within 1.38 kilometers, and by 0.5 degrees out to 2.31 kilometers. This localized warming contributes to the urban heat island effect, making the city center significantly hotter than it would be otherwise. Interestingly, while this heat is a burden, the economic activity driven by these hubs also increases the value of the surrounding real estate, creating a complex mix of environmental cost and financial gain.
On a broader scale, the researchers tallied the total energy and carbon footprint of the 71 data centers in the Los Angeles metropolitan area. Collectively, these facilities consume 3,364 gigawatt-hours of electricity every year, which is more than 5.54 percent of the total electricity generated by the entire metropolitan region. This massive demand places a heavy strain on the power grid and challenges the city's ability to transition to clean energy. The carbon emissions from this energy use amount to approximately 1,549,824 tons of carbon dioxide annually. To put this in perspective, this single industry produces nearly 0.88 percent of the total carbon that the vegetation in the Los Angeles metropolitan area is capable of absorbing, and that number is growing as the demand for artificial intelligence increases.
The researchers concluded that the unchecked growth of data centers poses a genuine risk to the livability of cities like Los Angeles. They argued that a "one-size-fits-all" approach to regulation is ineffective because the problems caused by a small center in a neighborhood are fundamentally different from those caused by a skyscraper in the downtown core. Instead, they proposed a new way to measure sustainability that accounts for these differences, weighing noise and housing impacts for residential sites, and heat and energy use for industrial and vertical sites. The study suggests that cities need stricter rules on where these buildings can be located, mandatory standards to reduce noise and heat, and a greater integration of green infrastructure to offset the carbon impact. Without these targeted measures, the digital transformation of the city could come at the expense of the very people who live there.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.