Key Findings
- Major cities across the United States are experiencing heat waves with increasing frequency. From an average of two heat waves annually in the 1960s, the figure has risen to six per year in the 2010s and 2020s, highlighting a significant escalation (Figure 1).
- The duration of the average heat wave in major U.S. urban centers has extended to approximately four days in recent years. This represents an increase of about one day compared to the average heat wave duration in the 1960s (Figure 1).
- The heat wave season has expanded considerably. Across the 50 cities analyzed, the average heat wave season is now about 46 days longer than it was in the 1960s (Figure 1). This shift in timing can heighten health risks as people may be less prepared for early or late-season heat waves.
- Heat waves are not only lasting longer and occurring more often, but they are also becoming more intense. In the 1960s, the average heat wave intensity was 2.0°F above the local 85th percentile threshold. By the 2020s, this average intensity had increased to 2.5°F above the same threshold (Figure 1).
- A significant majority, 46 out of 50, of the metropolitan areas studied have seen a statistically significant rise in heat wave frequency between the 1960s and 2020s. Increases in heat wave duration were significant in 28 locations, season length in 46, and intensity in 20 (Figure 2).
- Historical records indicate that the heat waves of the 1930s remain the most severe in U.S. history (Figure 3). The peak observed in Figure 3 corresponds to the extreme and prolonged heat waves in the Great Plains during the “Dust Bowl” era. Land mismanagement and prolonged drought exacerbated these heat waves by depleting soil moisture and diminishing evaporative cooling effects.5
Background on Heat Waves
An extreme heat event, commonly known as a heat wave, is defined as a sustained period of unusually hot weather. These events are more than just uncomfortable; they pose serious health risks, potentially leading to illness and death, particularly among vulnerable groups such as older adults, young children, and individuals with pre-existing conditions (refer to the Heat-Related Deaths and Heat-Related Illnesses indicators).1 Beyond human health, extended periods of extreme heat can damage crops, harm livestock, and elevate the risk of wildfires. The increased demand for air conditioning during prolonged heat waves can also strain power grids, potentially leading to power outages.
While hot days and heat waves are natural variations in weather patterns, the global climate is warming, leading to more frequent and intense hotter-than-usual days and nights (see the High and Low Temperatures indicator). Heat waves are projected to become longer, more frequent, and more intense in the future.2 This increase in extreme heat events can result in a greater incidence of heat-related illnesses and deaths, especially if adaptive measures are not implemented by individuals and communities.3 Even minor increases in extreme heat can significantly elevate mortality and morbidity rates.4
Understanding the Heat Wave Indicator
This indicator analyzes trends in four key characteristics of heat waves in the United States over time:
- Frequency: This refers to the annual number of heat wave occurrences.
- Duration: Measured in days, this indicates the length of each heat wave event.
- Season Length: This is the total number of days from the first heat wave of the year to the last, representing the span of the heat wave season.
- Intensity: This measures how much hotter a heat wave is relative to typical local temperatures.
The definition of a heat wave can vary. For consistency across the nation, Figures 1 and 2 define a heat wave as a period of two or more consecutive days where the daily minimum apparent temperature (temperature adjusted for humidity) in a specific city exceeds the 85th percentile of historical July and August temperatures (1981–2010) for that city. The EPA has chosen this definition based on several important factors:
- Nighttime temperatures are critical for health. The most severe health impacts of heat waves are often linked to high nighttime temperatures, which are represented by the daily minimum.4 The human body requires cooler nighttime temperatures to recover from daytime heat. Elevated nighttime temperatures place additional strain on the cardiovascular system as the body works harder to regulate its temperature.
- Humidity is a crucial factor. Adjusting for humidity is essential because high humidity levels impede evaporation, making it harder for the body to cool down through sweating. This is why heat advisories often rely on the “heat index,” which combines temperature and humidity.
- The 85th percentile threshold reflects unusually hot conditions. The 85th percentile of July and August temperatures corresponds to the nine hottest days during the hottest two months of the year. Temperatures at or above this level are statistically rare and are generally perceived as unusually hot by most people.
- Local context is important. By using the 85th percentile for each city individually, Figures 1 and 2 define “unusual” heat in relation to local climate norms. A temperature of 95°F, for example, might be considered extreme in one city but normal in another. Furthermore, populations in warmer regions, like the Southwest, may be more acclimatized and adapted to hotter weather.
The National Oceanic and Atmospheric Administration (NOAA) calculated apparent temperature for this indicator using temperature and humidity data from long-term weather stations, typically located at airports. Figures 1 and 2 focus on the 50 most populous U.S. metropolitan areas with consistent weather data from a stable location and minimal missing data throughout the study period. The year 1961 was selected as the starting point because consistent data collection for most major cities dates back to at least that year.
Figure 3 offers another perspective on the magnitude and frequency of prolonged heat wave events. It presents the U.S. Annual Heat Wave Index, which tracks the occurrence of heat wave conditions across the contiguous 48 states from 1895 to 2021. This index defines a heat wave as a period of at least four days with an average temperature that historically occurs for four or more days only once every 10 years. The index value for a given year is determined by the frequency and spatial extent of such severe heat waves.
Data Insights
Indicator Considerations
Urban development and the urban heat island effect can influence heat wave trends. As cities expand, vegetation is replaced by paved surfaces and buildings. This development contributes to higher temperatures, particularly at night, creating an “urban heat island” effect.8 The urban growth experienced since 1961 may partially explain the increase in heat waves observed in Figures 1 and 2 for certain cities. This indicator does not adjust for urban development effects, as it focuses on the actual temperatures experienced by people, regardless of the underlying causes of these trends.
Figures 1 and 2 include the 50 most populous metropolitan areas that had complete data from 1961 to 2023. Several large metropolitan areas, such as New York City, Houston, Minneapolis–St. Paul, and Denver, were excluded due to insufficient data. In some cases, this was because the primary long-term weather station was relocated during the 1961–2023 period, for example, with the opening of a new airport.
As mentioned earlier, Figures 1 and 2 prioritize daily minimum temperatures due to their direct link to human health. For additional information, EPA’s technical documentation for this indicator presents a similar analysis based on daily maximum temperatures.
Temperature data from the early 20th century has greater uncertainty due to fewer operational weather stations at that time. Additionally, measurement instruments and methodologies have evolved, and some stations have been relocated. While the data in Figure 3 have been adjusted to minimize these influences and biases, some uncertainties remain. However, these uncertainties are not significant enough to alter the fundamental trends observed.
Data Sources
Figures 1 and 2 are adapted from an analysis conducted by Habeeb et al. (2015).9 They are based on temperature and humidity data from weather stations managed by NOAA’s National Weather Service. NOAA’s National Centers for Environmental Information compiled and provided the data.
Figure 3 is based on measurements from weather stations within the National Weather Service Cooperative Observer Network. The data are publicly available online at: www.ncei.noaa.gov. Components of this indicator can also be found at: www.globalchange.gov/indicators.
Technical Documentation
References
1 Hayden, M. H., Schramm, P. J., Beard, C. B., Bell, J. E., Bernstein, A. S., Bieniek-Tobasco, A., Cooley, N., Diuk-Wasser, M., Dorsey, M. K., Ebi, K., Ernst, K. C., Gorris, M. E., Howe, P. D., Khan, A. S., Lefthand-Begay, C., Maldonado, J., Saha, S., Shafiei, F., Vaidyanathan, A., & Wilhelmi, O. V. (2023). Chapter 15: Human health. In USGCRP (U.S. Global Change Research Program), Fifth National Climate Assessment. https://doi.org/10.7930/NCA5.2023.CH15
2 Marvel, K., Su, W., Delgado, R., Aarons, S., Chatterjee, A., Garcia, M. E., Hausfather, Z., Hayhoe, K., Hence, D. A., Jewett, E. B., Robel, A., Singh, D., Tripati, A., & Vose, R. S. (2023). Chapter 2: Climate trends. In USGCRP (U.S. Global Change Research Program), Fifth National Climate Assessment. https://doi.org/10.7930/NCA5.2023.CH2
3 U.S. EPA (Environmental Protection Agency). (2015). Climate change in the United States: Benefits of global action (EPA 430-R-15-001). www.epa.gov/sites/default/files/2015-06/documents/cirareport.pdf
4 Sarofim, M. C., Saha, S., Hawkins, M. D., Mills, D. M., Hess, J., Horton, R., Kinney, P., Schwartz, J., & St. Juliana, A. (2016). Chapter 2: Temperature-related death and illness. In USGCRP (U.S. Global Change Research Program), The impacts of climate change on human health in the United States: A scientific assessment (pp. 43–69). http://dx.doi.org/10.7930/J0MG7MDX
5 CCSP (U.S. Climate Change Science Program). (2008). Synthesis and Assessment Product 3.3: Weather and climate extremes in a changing climate. www.globalchange.gov/browse/reports/sap-33-weather-and-climate-extremes-changing-climate
6 NOAA (National Oceanic and Atmospheric Administration). (2024). Heat stress datasets and documentation (provided to EPA by NOAA in April 2024) [Data set].
7 Kunkel, K. (2022). Updated version of Figure 2.3 in: CCSP (U.S. Climate Change Science Program). (2008). Synthesis and Assessment Product 3.3: Weather and climate extremes in a changing climate. www.globalchange.gov/browse/reports/sap-33-weather-and-climate-extremes-changing-climate
8 U.S. EPA (Environmental Protection Agency). (2016). Heat island effect. www.epa.gov/heat-islands
9 Habeeb, D., Vargo, J., & Stone, B., Jr. (2015). Rising heat wave trends in large US cities. Natural Hazards, 76(3), 1651–1665. https://doi.org/10.1007/s11069-014-1563-z
