Researchers Are Using Social Media to Monitor How Californians Experience Extreme Heat in Real Time
Researchers at the University of California, Irvine have developed a new and surprisingly practical way to understand how people experience extreme heat — by analyzing what they post on social media. The study shows how online conversations can be transformed into a real-time heat monitoring system, helping public health agencies identify which communities are struggling the most during heat waves and where interventions are urgently needed.
The research was led by health communication experts from UC Irvine’s Joe C. Wen School of Population & Public Health and was published in the academic journal Weather, Climate, and Society. At its core, the study connects social media behavior, machine learning, and health equity data to address one of California’s most underestimated public health threats: extreme heat.
Why Extreme Heat Is a Growing Public Health Issue in California
California has been experiencing more frequent, longer, and more intense heat waves, driven largely by climate change. While heat affects everyone, it does not affect everyone equally. The state’s diverse geography, housing patterns, and socioeconomic inequalities mean that some communities are far more vulnerable than others.
One striking example highlighted in the research is access to air conditioning. In Los Angeles County, about 20% of households lack air conditioning, while in Riverside County, fewer than 7% of households are without it. This difference is rooted in historical development patterns. Riverside’s hotter inland climate led to widespread adoption of cooling systems decades ago. Meanwhile, many older homes in Los Angeles — particularly in coastal and lower-income neighborhoods — were built at a time when air conditioning was considered unnecessary.
The result is a patchwork of cooling access across California, where some urban neighborhoods are far more exposed to dangerous heat than others. Traditional heat monitoring systems rely heavily on weather data, but they often fail to capture how people actually experience heat in their daily lives. This is the gap the UC Irvine researchers aimed to fill.
How Social Media Became a Heat Monitoring Tool
To understand real-world heat experiences, the research team analyzed over 7,000 heat-related posts on X (formerly Twitter) made by Californians. These posts spanned seven years of data, allowing researchers to examine patterns across multiple heat events and seasons.
Using a machine learning–based approach, the team categorized posts based on sentiment and content. They then linked each post to specific geographic locations using California census tract data. To add a health equity lens, the researchers incorporated the Healthy Places Index (HPI) — a tool developed by the Public Health Alliance of Southern California that combines 23 social, economic, and environmental indicators such as income, education, housing quality, healthcare access, and pollution exposure.
By connecting what people were saying online with where they lived and the resources available in their communities, the researchers created a detailed picture of how heat affects Californians differently depending on their circumstances.
What the Analysis Revealed About Heat Inequality
One of the clearest findings was the contrast between low-HPI and high-HPI communities.
People living in low Healthy Places Index areas, which are typically characterized by fewer economic resources and limited access to cooling infrastructure, were more likely to post messages expressing negative emotions. These posts often involved venting about the heat, warning others about dangerous conditions, or describing how unbearable the temperatures felt.
In contrast, residents of high-HPI areas tended to post more about coping strategies, such as staying indoors, using air conditioning, or adjusting daily routines. They were also more likely to express relief and positive emotions when temperatures dropped.
Importantly, the researchers found that discussions about climate change and community-level actions appeared across both low- and high-HPI areas at similar rates. This suggests that awareness of climate issues is widespread, but the ability to cope with extreme heat is not evenly distributed.
Turning Online Conversations Into Real-Time Action
By combining social media analytics, health equity data, and geospatial mapping, the research team developed what they describe as a powerful AI-based monitoring tool. This system could allow public health agencies, emergency responders, and fire departments to track heat-related distress as it unfolds, rather than relying solely on delayed reports or hospital data.
The potential applications are significant. Agencies could use this information to issue targeted heat warnings, deploy cooling resources, open cooling centers in specific neighborhoods, or conduct outreach in areas showing signs of distress. Instead of broad, one-size-fits-all messaging, responses could be tailored to the communities most at risk.
According to the researchers, extreme heat is often underestimated compared to other natural hazards, despite being one of the leading causes of weather-related deaths. Integrating social media monitoring with tools like the Healthy Places Index offers a new way to identify vulnerabilities before they escalate into emergencies.
Why Social Media Data Matters for Public Health
Social media platforms capture immediate, human responses to environmental conditions. People post when they are uncomfortable, frustrated, scared, or relieved — emotions that are not visible in temperature charts alone. When analyzed responsibly and at scale, these posts can serve as a real-time feedback loop between communities and public health systems.
The study demonstrates that social media data does not replace traditional climate or health data. Instead, it adds context and urgency, revealing how environmental stressors are experienced at the neighborhood level.
Understanding the Healthy Places Index
The Healthy Places Index played a central role in this research. It combines 23 indicators across areas such as education, housing, employment, transportation, clean air and water, and healthcare access. The index is widely used in California to guide funding decisions and public health planning.
By linking HPI scores with social media sentiment, the researchers showed that structural inequalities directly shape how people experience extreme heat. Communities with fewer resources are not just hotter — they are also more stressed, less protected, and more vocal about the risks they face.
The Bigger Picture: Climate Change and Urban Heat
As climate change continues to intensify heat waves, cities across the world are grappling with the urban heat island effect, where dense development and limited green space trap heat. California’s experience reflects a global challenge: how to protect vulnerable populations from rising temperatures.
Tools like the one developed by the UC Irvine team could become increasingly valuable as governments search for data-driven, equitable climate adaptation strategies.
Funding and Research Team
The study was supported by the National Science Foundation, underscoring its scientific and societal importance. The lead author was Suellen Hopfer, an associate professor of health, society, and behavior at UC Irvine’s Wen Public Health. The research team included scholars from public health, data science, and environmental disciplines, highlighting the study’s interdisciplinary approach.
Research Reference
How Californians Tweet about Extreme Heat Events on Social Media: A Health Equity Perspective
Weather, Climate, and Society (2025)
https://journals.ametsoc.org/view/journals/wcas/17/4/WCAS-D-24-0147.1.xml
