Digest

Rethinking Extreme Heat and Urban Resilience through Scales Connectivity 

Extreme heat is becoming more frequent, intense, and synchronized across cities. This digest explores how climate science can guide proactive resource allocation, strengthen urban resilience, and better protect vulnerable communities.

At A Glance

Key Challenge

We are approaching a critical inflection point. Extreme heat events are becoming more frequent, intense, and synchronized, but emergency planning models have yet to catch up. We must rethink how emergency resources are allocated, shared, and deployed.

Policy Insight

Extreme heat is no longer a localized anomaly—it is a converging, compound threat. Addressing this challenge demands a science-informed, equity-focused overhaul of how emergency resources are allocated and coordinated across jurisdictions.

Rethinking the Rules of Urban Resilience

The climate is shifting faster than our response infrastructure can adapt. Extreme heat, unlike hurricanes or wildfires, is a silent, invisible killer and the leading cause of weather-related deaths across the U.S., responsible for 10,527 fatalities between 2004 and 2018. There were even higher tolls in recent years; for example, 1,600 deaths in 2021 alone (National Oceanic and Atmospheric Administration 2023; Vaidyanathan 2020). Its impacts fall disproportionately on vulnerable populations: the elderly, low-income households, and outdoor workers often lack access to adequate protection or reliable cooling (Bell, Gasparrini, and Benjamin 2024). But the risks extend far beyond public health.

Extreme heat places significant strain on energy systems, disrupts transportation infrastructure, and reduces labor productivity. These effects are especially acute in urban environments, where the urban heat island (UHI) effect—stemming from the dense concentrations of heat-retaining surfaces—exacerbates both temperatures and exposure durations (Tuholske et al. 2021).

What was once considered a rare and isolated threat is now emerging as a widespread and synchronized phenomenon (National Oceanic and Atmospheric Administration’s National Centers for Environmental Information (NCEI 2025). In the U.S., the past four consecutive summers (2021–2024) have shattered historical temperature records and expanded the geographic and temporal footprint of heat-related disruptions (Fischer et al. 2025). This growing convergence of heat extremes has exposed deep structural limitations in traditional emergency management systems, which were largely designed to respond to isolated, short-term hazards, rather than compounding, climate-driven events unfolding across multiple jurisdictions at once.

An effective response to extreme heat requires rapid mobilization of cooling resources, emergency personnel, medical aid, and public communication systems. Yet, given the relative rarity of such events in the past, it has been neither practical nor cost-effective for most cities to maintain dedicated reserves of heat-specific emergency resources. Instead, many of these assets, such as mobile cooling units, emergency medical teams, and federal disaster assistance, are housed at the national level and deployed reactively when cities issue requests for aid (Doan and Shaw 2019).

Historically, all requests for extreme heat disaster declarations under the Robert T. Stafford Disaster Relief and Emergency Assistance Act (the Stafford Act) have been denied, as the act has traditionally prioritized sudden-onset hazards that cause visible structural damage. However, in the context of a changing climate, this framework is beginning to evolve toward a broader definition of eligible events—those that exceed state and local capacity to respond and recover (Congress.gov 2024b).

When multiple cities are struck by extreme heat simultaneously—as was the case during the 2021 Pacific Northwest heatwave and the nationwide record-breaking heat domes of 2023 and 2024—demand for shared emergency resources can quickly outstrip available capacity (Congress.gov 2024a). In these situations, municipalities are often forced to compete for limited pools of support, stretching national response systems thin. This pressure is further compounded when extreme heat co-occurs with other climate-driven hazards, such as droughts, wildfires, and floods, or amid public health threats (Hyde et al. 2006).

This possibility raises a central policy question: How can climate science-based early warning signals be used to allocate scarce emergency resources more equitably and efficiently when multiple cities facing extreme weather simultaneously? One promising avenue lies in the emerging understanding of atmospheric resonance, a large-scale climate phenomenon capable of synchronizing weather extremes across distant regions.

This digest draws on our prior research on quasi-resonant planetary wave amplification (QRA), which demonstrates a robust link between resonant atmospheric circulation patterns and the occurrence of simultaneous extreme weather events (Li et al. 2024; Li et al. 2025; Mann et al. 2017; Mann et al. 2018). Planetary waves, or Rossby waves, are large-scale meanders in the jet stream. Under certain conditions, these waves can become effectively trapped and amplified within mid-latitude “waveguides.” When this occurs, the jet stream can slow and deform in ways that stall weather systems, locking regional patterns in place for days or even weeks (Mann et al. 2017; Mann et al. 2018), an atmospheric state referred to here as QRA.

By integrating expertise in climate science, sustainability, and energy policy, we propose an actionable, science-based tool to guide resource allocation decisions in the face of extreme heat. Such a tool could enhance urban resilience by informing coordinated emergency response planning at both local and federal levels, thereby:

  1. Reducing competition for resources;
  2. Improving equity in distribution; and
  3. Mitigating the cascading impacts of extreme heat.

In the sections that follow, we summarize key findings from this work and outline the policy implications for building resilience across scales in an increasingly warming climate.

Planetary Waves and Urban Heat Extremes

Extreme heat events are no longer isolated or localized in nature. Increasingly, they are driven by large-scale atmospheric circulation patterns that can synchronize weather extremes across distant regions. QRA is one such mechanism: during resonant atmospheric states, phase-aligned planetary waves can produce simultaneous heat extremes across multiple metropolitan areas (Figure 1). 

Map of the contiguous United States showing anomalies in maximum 2-meter air temperature, with colors ranging from dark blue (about −3 °C, cooler than average) to dark red (about +3 °C, warmer than average). The strongest warm anomalies form a broad band across the southern United States, extending from southern Arizona and New Mexico through Texas and the Gulf Coast to southern Florida. Cooler-than-average areas appear in parts of the central Rockies, northern Plains, Great Lakes, and the Northeast.

Gray contour lines labeled with values such as 5792, 5872, 5905, and 5918 indicate atmospheric height contours. Colored circles and labels mark selected cities grouped by color: black (Los Angeles, Las Vegas, Atlanta), lime green (Phoenix, Tucson, New Orleans, Miami), yellow (El Paso, Houston), purple (Portland, Minneapolis, Chicago, Philadelphia, New York). Curved gray lines connect cities within each group, suggesting related atmospheric pathways or travel routes. A vertical color bar on the right is labeled "Maximum 2-meter Air Temperature (°C)" and spans from −3 to +3 °C.
Figure 1: Planetary Wave-Driven Urban Heat Extremes Across U.S. cities. This map illustrates the spatial footprint of planetary wave pattern during the summer of 2023. Shading represents anomalies in daily maximum 2-meter air temperature (expressed in standard deviations above the historical average), while contours indicate 500 hectopascal (hPa) geopotential height fields, highlighting the structure of the wave pattern. Colored circles mark major U.S. cities affected by concurrent heat extremes, which were dynamically connected through the phase-aligned atmospheric wave.

During these resonant states, areas beneath persistent high-pressure ridges experience prolonged and intense heatwaves, while regions under low-pressure troughs may endure sustained rainfall or flooding. These quasi-stationary wave patterns act as atmospheric teleconnections, dynamically linking geographically distant cities so that extremes occur in tandem rather than in isolation.

For policymakers, the significance of QRA lies in its operational potential. Monitoring these atmospheric patterns can provide early signals that multiple regions may experience extreme conditions simultaneously. Such insights can help decision-makers anticipate surges in demand for cooling resources, medical personnel, and emergency response assets, allowing for earlier and more coordinated mobilization of support, especially when resource needs are likely to exceed supply across jurisdictions.

Resonant Waves as Early Warnings

Over the last four summers (2021–2024), QRA events have consistently coincided with record-breaking extreme maximum temperatures (EMXT) across a broad network of U.S. cities (Figure 2). These amplified planetary wave patterns repeatedly aligned with clusters of heat extremes, spanning not only traditionally hot cities but also cooler, northern metros that are historically less acclimated, and therefore less prepared, for prolonged heat. 

Four scatter plots arranged in a 2 × 2 grid summarize annual extreme maximum temperatures (EMXT, °F) for selected U.S. cities from June through August in 2021, 2022, 2023, and 2024. The x-axis in each panel shows dates, and the y-axis shows EMXT (approximately 94–120 °F). Each colored point represents one city, with the legend identifying Philadelphia (PHL), New York City (NYC), Chicago (CHI), Minneapolis (MSP), Seattle (SEA), Portland (PDX), Los Angeles (LA), Las Vegas, Atlanta (ATL), Phoenix (PHX), New Orleans (NOLA), Miami (MIA), Houston (HOU), and Dallas (DAL). Vertical gray shaded bands indicate periods of notable atmospheric conditions or heat events.

Across all four years, Phoenix, Las Vegas, and other southwestern cities consistently record the highest temperatures (roughly 112–120 °F), while northern and eastern cities generally range from the mid-90s to low 100s °F. Southern cities such as Houston and Dallas typically fall between about 102 and 110 °F. Most extreme temperature observations occur during or near the shaded intervals, although the timing and duration of these periods vary by year. The 2023 panel contains several separate shaded periods throughout the summer, whereas the 2021, 2022, and 2024 panels each feature one or two longer shaded intervals.
Figure 2: The role of QRA in driving urban heat extremes in the U.S. across four consecutive summers (2021–2024). Each panel shows the annual extreme maximum temperature (EMXT) observed during the summer for the major U.S. cities shown in Figure 1. Colored dots indicate EMXT values by city, based on long-term observational records. Shaded vertical bands mark periods associated with QRA events, identified from upper-atmosphere diagnostics. Across each year, clusters of high-temperature extremes align with these QRA windows.

What sets QRA apart from other meteorological phenomena is its potential as an early warning indicator. When diagnosed in advance, these amplified planetary wave patterns can offer one to two weeks of lead time. This early warning provides a critical window for action, enabling:

  • Public health systems to brace for spikes in heat-related illnesses;
  • Utilities to scale power reserves; and
  • Emergency response teams to pre-position mobile cooling and medical resources.

As climate extremes become increasingly synchronous across cities, QRA-based diagnostics offer a scalable, science-informed tool for proactive, cross jurisdictional resource planning. By transitioning from reactive crisis response to anticipatory coordination, we can better strengthen urban resilience, safeguard vulnerable populations, and reduce cascading impacts across interconnected urban systems.

Cooling Demand on Overdrive

Climate change is fundamentally altering the rhythm of the atmosphere. Over the past 75 years, the number of summer weather extremes linked to QRA has tripled (Figure 3). Once rare, these amplified jet stream patterns now occur with increasing regularity—rising from roughly one per summer in the 1950s to nearly three per summer today (Li et al. 2025). This sharp increase means more cities are likely to be hit by extreme heat at the same time, transforming what was once a localized hazard into a multi-regional challenge. As these atmospheric patterns become more persistent, the stresses they place on infrastructure, public health, and energy systems are becoming harder to manage.

Energy demand is particularly sensitive to these changes. In the U.S., space cooling already accounts for about 10% of total electricity use, and that share is rapidly growing. During a severe heatwave in July 2011, for example, space cooling alone accounted for up to 75% of Philadelphia’s peak electricity demand (Waite et al. 2017). Under QRA-driven heatwaves, multiple metropolitan areas can experience peak cooling demand simultaneously, placing unprecedented stress on regional power systems. Such synchronized demand increases the likelihood of grid instability, rolling blackouts, and cascading infrastructure failures as indicated by the Texas grid emergency in 2022 and blackout incidents during the 2023 Southwest heatwave.

Line graph showing annual counts of QRA events from 1950 to 2023. The x-axis spans years from 1950 to 2023, and the y-axis shows annual QRA counts (unitless), ranging from 0 to 6. A solid black line represents the yearly count, while a gray dashed linear trend line indicates a gradual increase over time. The legend identifies the series as "QRA (0.024)," where 0.024 denotes the slope of the trend.

Annual QRA counts fluctuate substantially from year to year, with values ranging from 0 to 6. Counts are generally lower during the 1950s through the 1970s, when most years record between 0 and 3 events. Beginning in the early 1980s, higher annual counts become more frequent, including several peaks of 5 to 6 events. Notable maxima of 6 events occur in the early 1980s, early 2000s, and the early 2020s. Despite the strong year-to-year variability, the upward-sloping trend line indicates that QRA events have become modestly more common over the observational period.
Figure 3: Time series of the boreal summer QRA event counts from 1950 to 2024. Dashed trend lines (with slope values shown in the legend) indicate a significant upward trend in QRA.

A widely used metric to estimate cooling energy demand is the cooling degree days (CLDD) index, which measures the cumulative intensity and duration of daily temperatures above 65 °F (18.3 °C). The greater the number of CLDDs, the more energy is typically required to maintain indoor thermal comfort through air conditioning or fans. Across six northern and mid-latitude U.S. cities, statistically significant increases in CLDDs since 1950 point to steadily rising seasonal cooling demands (Figure 4).

Particularly in cities like Portland, Oregon, which historically have been considered temperate and less heat-prone, the combination of rising annual extreme temperatures and more frequent days exceeding 90 °F (32.2 °C) (Appendix Figures S1 and S2) underscores a widening mismatch between climate risk and existing infrastructure design.

Crucially, these surging demands are not borne equally. Higher-income households are more likely to have reliable access to air conditioning, well-insulated housing, and backup power during outages (Proussaloglou, Kane, and Tomer 2022). In contrast, low-income neighborhoods face greater exposure to heat-related health risks and energy insecurity, due to higher population densities, aging housing stock, and limited access to cooling technologies. In these communities, residents may be forced to choose between cooling their homes and affording other essentials, exacerbating preexisting health and economic disparities in the face of escalating heat stress. 

Six time series plots arranged in a 3 × 2 grid show annual cumulative cooling degree days (CLDD, °F) from 1950 to 2023 for six U.S. cities: Philadelphia (PHL), New York City (NYC), Chicago (CHI), Minneapolis (MSP), Seattle (SEA), and Portland (PDX). The x-axis spans 1950–2023, and the y-axis shows CLDD, with values ranging from 0 to approximately 1,800 °F. Each panel contains a solid colored line representing annual CLDD and a dashed linear trend line. The city name is followed by the slope of the linear trend in parentheses.

All six cities exhibit substantial year-to-year variability but show positive long-term trends, indicating increasing cumulative cooling degree days over the study period. Philadelphia (7.052 °F per year) and New York City (6.142 °F per year) have the largest increases, with CLDD generally rising from about 1,000–1,200 °F in the 1950s to around 1,400–1,600 °F in recent decades. Chicago (3.541) and Minneapolis (4.254) also show steady increases, though with lower overall CLDD than the eastern cities. Seattle (2.921) has the lowest CLDD values throughout the record, increasing gradually from below 100 °F to roughly 300–400 °F. Portland (5.965) displays a pronounced upward trend, with CLDD increasing from approximately 200–300 °F in the early record to 600–800 °F in recent years. Overall, the figure shows increasing summertime cooling demand across all six cities despite considerable interannual variability.
Figure 4: Annual cooling degree days (CLDD) from 1950 to 2024 across major U.S. cities:
Philadelphia (PHL), New York City (NYC), Chicago (CHI), Minneapolis–St. Paul (MSP), Seattle (SEA), and Portland (PDX) CLDD is calculated by summing daily values when the average temperature exceeds 65 °F (18.3 °C). Dashed trend lines (with slope values shown in the legend) indicate a significant upward trend in cooling energy demand. Additional cities referenced in Figure 1 are presented in Figure S3 of the Appendix.  

Rethinking Resource Allocation across Scales

As extreme heat events grow more intense and increasingly synchronized across regions, the existing emergency response architecture for extreme heat in the U.S. remains largely reactive and locally driven. The current response mechanisms are built to manage episodic, discrete hazards, such as hurricanes, rather than linked, slow-onset heat threats that can persist for days or weeks and increasingly unfolds across multiple regions at once.

In practice, most heat-risk management still occurs at the state and local levels. Local governments develop heat-health action plans, issue public risk communications, expand emergency medical staffing, distribute water and other essential supplies to vulnerable populations, provide emergency energy assistance, and open cooling centers (Congress.gov 2024a). These measures are critical, but their effectiveness depend heavily on local capacity.

Federal assistance typically enters only after an event exceeds state and local response capabilities. In such cases, the Federal Emergency Management Agency (FEMA) may reimburse certain heat-related response costs—such as air conditioning units or electric fans—through programs like the Individuals and Households Program (IHP), provided that the President issues an emergency or major disaster declaration under the Stafford Act. However, extreme heat is not explicitly listed as a qualifying incident type for a major disaster declaration, and historically, heat events have not received Stafford Act declarations, even though the law could theoretically be invoked if circumstances warranted.

Even when federal support becomes available, FEMA’s role remains primarily toward post-disaster recovery rather than proactive preparedness and response to extreme heat (Federal Emergency Management Agency 2025; U.S. Department of the Interior 2025; Keith et al. 2021). This current orientation reflects legacy assumptions about disasters as sudden, localized events. Yet climate-induced heat is fundamentally different: it unfolds gradually, spans large geographic areas, and requires sustained, proactive measures to prevent fatalities and protect public health.

As a result, states and local governments bear the brunt of frontline responsibilities during extreme heat events, often while operating with constrained budgets, outdated infrastructure, strained public health systems, and limited surge capacity. They must keep cooling centers open, deploy emergency services, and communicate risks to the public—all while managing rising electricity demand and mounting pressure on energy systems.

The challenge becomes even greater when heatwaves strike multiple cities or states simultaneously. In these cases, demand for federal support can quickly outpaces available capacity. At the same time, ongoing political and institutional uncertainty surrounding federal disaster management—including proposals to eliminate FEMA (The Guardian 2025) and recent legal disputes over disaster relief funding allocations (Frazin 2025)—underscores just how fragile the current support framework can be.

The result is a widening policy gap: the nation’s emergency management model remains largely reactive and fragmented, poorly suited to address the compounding, cross-regional nature of modern heat extremes. Without more forward-looking coordination—guided by climate science and centered on equity—the nation risks entering an era of shared scarcity, where jurisdictions are forced to compete for limited resources just as climate hazards are becoming more widespread, more connected, and more difficult to manage. 

Actionable Pathways Forward

Addressing the growing threat of synchronized urban heat extremes requires moving from reactive crisis management toward a proactive, risk-informed system for allocating emergency resources. This transition must operate across all levels of government and be grounded in science-based early warning signals—such as atmospheric resonance diagnostics—that offer actionable lead times and clarify where extreme heat risks are likely to cluster. To improve resilience and equity, key recommendations include:

  • Develop a tiered heat risk trigger system that activates shared protocols. This framework would translate early warning signals, like atmospheric resonance, paired with surface forecasts into predefined readiness actions across agencies and jurisdictions once thresholds for severity and/or duration are met (e.g., Tier 1–3). A tiered approach can reduce “late activation” in places that still rely on damage assessments or reimbursement triggers, and it enables earlier, equity-centered escalation when vulnerable communities face elevated risk.
  • Authorize an emergency fund that provides advance assistance when Tier 2–3 thresholds are triggered. A standing, pre-approved funding mechanism would allow jurisdictions to act on forecasted risk rather than waiting for impacts to materialize. This action creates a practical decision shift: from “first come, first served” reactive support to rules-based, cross-scale coordination that prioritizes risk, vulnerability, and anticipated need.
  • Establish heat-specific mutual aid mechanisms for jurisdictions forecasted to be affected simultaneously. While many mutual aid agreements already exist, few are designed for multi-region heat events that unfold in parallel. We recommend building “heat-capable” mutual aid frameworks tailored to cities likely to be phase-aligned within planetary wave patterns. Such agreements would facilitate rapid sharing of personnel, equipment, supplies, and operational expertise without triggering competition among jurisdictions. The tiered trigger system could also identify “aid corridors” and staging locations in advance to support timely mobilization during Tier 2–3 events.
  • Treat power loss and critical infrastructure failure as first-order heat emergency risks. QRA-driven heat patterns increase the likelihood of simultaneous peak electricity demand across metropolitan areas, elevating grid stress and the risk that outages become multi-regional rather than local. Early warning signals should therefore trigger targeted protective actions before the peak—such as verifying backup power readiness at designated “critical cooling continuity sites” (i.e., clinics, shelters, senior facilities), hardening communications and transit access, and aligning public health operations with utility contingency planning. This focus is especially urgent given the limited federal assistance typically available for extreme heat emergencies.

By embedding science-based tools like QRA diagnostics into emergency planning and connecting resource systems across scales, we can begin to build a more agile, anticipatory, and equitable infrastructure for navigating an era of intensifying heat extremes. From the resonant patterns that signal risk to the policy levers that shape resilience, the path forward depends on bridging science, governance, and social equity. Only by doing so can we ensure that our cities—and the people who live in them—are not just surviving extreme heat but adapting, coordinating, and thriving in a warming world.

Xueke Li

Former Research Associate, EES and PCSSM

Xueke Li is a former research associate in the Department of Earth and Environmental Science and the Penn Center for Science, Sustainability, and the Media (PCSSM). She is currently an assistant professor in the School of Energy and Environment at City University of Hong Kong.

Michael E. Mann

Presidential Distinguished Professor

Michael E. Mann is the Presidential Distinguished Professor in the Department of Earth and Environmental Science. He is a faculty fellow with the Kleinman Center and the director of the Penn Center for Science, Sustainability, and the Media (PCSSM).

Shannon Christiansen

Former Senior Research Coordinator, EES

Shannon Christiansen is the former senior research coordinator for the Mann Research Group at Penn. She is currently a research associate at the Gulf Coast Repository at Texas A&M University.

Heather Kostick

Associate Director, PCSSM

Heather Kostick is the associate director at PCSSM. She manages the research and programming at PCSSM to further the mission of communicating climate science to the public and policymakers. Her research interests include climate misinformation and disinformation, urban ecology, and biodiversity.

Acknowledgements 

We thank Iman Ali (School of Engineering and Applied Science, University of Pennsylvania) for her contributions to the preliminary data analysis presented in Figure 2.

Six time series plots arranged in a 3 × 2 grid show annual extreme maximum temperature (EMXT, °F) from 1950 to 2023 for six U.S. cities: Philadelphia (PHL), New York City (NYC), Chicago (CHI), Minneapolis (MSP), Seattle (SEA), and Portland (PDX). The x-axis spans 1950–2023, and the y-axis shows EMXT, ranging from approximately 85 to 115 °F. Each panel includes a solid colored line representing annual EMXT and a dashed linear trend line. The panel titles list each city along with the slope of the linear trend; "NS" indicates that the trend is not statistically significant.

Philadelphia (0.02 °F per year), Chicago (−0.002 °F per year), and Minneapolis (0.01 °F per year) exhibit little long-term change, with substantial year-to-year variability and nearly flat trend lines. Philadelphia and Minneapolis fluctuate mainly between about 92 and 103 °F, while Chicago varies between roughly 92 and 106 °F. New York City shows a modest positive trend (0.03 °F per year), with annual maxima generally ranging from the low 90s to just above 105 °F. Seattle (0.068 °F per year) and Portland (0.075 °F per year) display the strongest warming trends, with annual extreme temperatures increasing gradually over the study period despite considerable interannual variability. Portland reaches the highest individual value in the record, exceeding 110 °F in the early 2020s. Overall, the figure shows weak or insignificant changes in annual extreme maximum temperatures for the eastern and Midwestern cities and stronger upward trends for the Pacific Northwest cities.
Figure S1: Same as Figure 4, but for the time series of annual extreme maximum temperature (EMXT) from 1950 to 2024.
Six time series plots arranged in a 3 × 2 grid show the annual number of days exceeding the 90th percentile of daily maximum temperature (Dx90, days) from 1950 to 2023 for six U.S. cities: Philadelphia (PHL), New York City (NYC), Chicago (CHI), Minneapolis (MSP), Seattle (SEA), and Portland (PDX). The x-axis spans 1950–2023, and the y-axis shows the annual count of Dx90 days, ranging from 0 to 60 days. Each panel includes a solid colored line representing annual values and a dashed linear trend line. The panel titles list the city and the slope of the linear trend in days per year; "NS" indicates that the trend is not statistically significant.

Philadelphia (0.181 days per year) and New York City (0.15 days per year) exhibit the strongest increases, with the annual number of hot days generally rising from about 15–20 days in the 1950s to 25–35 days in recent decades, despite substantial year-to-year variability. Portland (0.173) also shows a pronounced upward trend, increasing from roughly 5–10 days per year early in the record to around 20–30 days in the most recent years. Seattle (0.054) displays a smaller but positive increase, with most years recording fewer than 10 hot days. Chicago (−0.033, not significant) and Minneapolis (0.002, not significant) show little long-term change, with highly variable annual counts and nearly flat trend lines. Overall, the figure indicates increasing frequency of unusually hot summer days in the eastern United States and Pacific Northwest, while long-term changes are minimal in the Midwest.
Figure S2: Same as Figure 4, but for the time series of the number of days with maximum temperature greater than or equal to 90 °F (32.2 °C), referred to as DX90, from 1950 to 2024.
Eight time series plots arranged in a 4 × 2 grid show annual cumulative cooling degree days (CLDD, °F) from 1950 to 2023 for eight U.S. cities: Los Angeles (LA), Las Vegas, Atlanta (ATL), Phoenix (PHX), New Orleans (NOLA), Miami (MIA), Houston (HOU), and Dallas (DAL). The x-axis spans 1950–2023, and the y-axis shows CLDD, ranging from approximately 800 to 5,500 °F. Each panel includes a solid colored line representing annual CLDD and a dashed linear trend line. The panel titles list the city and the slope of the linear trend in parentheses.

All eight cities exhibit positive long-term trends in cumulative cooling degree days, indicating increasing summertime cooling demand despite considerable year-to-year variability. Phoenix shows the largest increase (25.529 °F per year), with CLDD rising from roughly 3,500–4,000 °F in the 1950s to more than 5,000 °F in recent decades. Las Vegas (16.751), Miami (15.818), and Los Angeles (13.616) also display strong upward trends, with progressively higher CLDD over the observational period. New Orleans (11.952) and Houston (10.666) show moderate increases, with values generally rising from about 2,500–3,000 °F to more than 3,500 °F by the end of the record. Atlanta (7.52) and Dallas (7.188) exhibit the smallest positive trends, although both still show gradual increases over time. Overall, the figure demonstrates increasing cumulative cooling demand across a range of warm-climate U.S. cities, with the largest increases occurring in the desert Southwest.
Figure S3: Same as Figure 4, but for annual cooling degree days (CLDD) from 1950 to 2024 for the remaining cities outlined in Figure 1. 

Bell, M.L., A. Gasparrini, and G.C. Benjamin. 2024. “Climate Change, Extreme Heat, and Health.” New England Journal of Medicine. 390(19), 1793-1801. https://doi.org/doi:10.1056/NEJMra2210769.

Congress.gov. 2024a. “Emergency Response to Extreme Heat: Federal Financial Assistance and Considerations for Congress. https://www.congress.gov/crsproduct/R46873.

Congress.gov. 2024b. “Stafford Act Declarations for Extreme Heat.” https://www.congress.gov/crsproduct/IN12384.

Doan, X.V. and D. Shaw. 2019. “Resource Allocation When Planning for Simultaneous Disasters.” European Journal of Operational Research. 274 (2): 687-709. https://doi.org/10.1016/j.ejor.2018.10.015.

Federal Emergency Management Agency. 2025. “National Disaster Recovery Framework.” https://www.fema.gov/emergencymanagers/nationalpreparedness/frameworks/recovery.

Frazin, R. 2025. “Court Temporarily Blocks Trump from Reallocating FEMA Disaster Preparedness Funds.” The Hill. August 5, 2025. https://thehill.com/policy/energyenvironment/5437562trumpfemadisasterpreparednessfundsbric/.

Fischer, E.M., M. Bador, R. Huser, E.J. Kendon, A. Robinson, and S. Sippel. 2025. “Record-Breaking Extremes in a Warming Climate.” Nature Reviews Earth & Environment. 6(7), 456-470. https://doi.org/10.1038/s4301702500681y. https://doi.org/10.1038/s4301702500681y.

Hyde, J., B. Kim, L. Sprague Martinez, M. Clark, and K. Hacker. 2006. “Better Prepared But Spread Too Thin: The Impact of Emergency Preparedness Funding on Local Public Health.” Disaster Management & Response. 4(4), 106-113. https://doi.org/10.1016/j.dmr.2006.08.002.

Ladd, K., S. Meerow, D.M Hondula, V.K.Turner, and J.C Arnott. 2021. “Deploy Heat Officers, Policies and Metrics.” Nature. 598(7879), 29-31. https://doi.org/10.1038/d41586-021-02677-2.

Li, X., M.E. Mann, M.F. Wehner, and S. Christiansen. 2025. “Increased Frequency of Planetary Wave Resonance Events Over the Past Half-Century.” Proceedings of the National Academy of Sciences. 122(25), e2504482122. https://doi.org/10.1073/pnas.2504482122.

Li, X., M.E. Mann, M.F. Wehner, S. Rahmstorf, S. Petri, S. Christiansen, and J. Carrillo. 2024. “Role of Atmospheric Resonance and Land-Atmosphere Feedbacks As A Precursor to the June 2021 Pacific Northwest Heat Dome Event.” Proceedings of the National Academy of Sciences. 121(4), e2315330121. https://doi.org/10.1073/pnas.2315330121.

Mann, M.E., S. Rahmstorf, K. Kornhuber, B.A. Steinman, S.K. Miller, and D. Coumou. 2017. “Influence of Anthropogenic Climate Change on Planetary Wave Resonance and Extreme Weather Events.” Scientific Reports. 7(1), 45242. https://doi.org/10.1038/srep45242.

Mann, M.E., S. Rahmstorf, K. Kornhuber, B.A. Steinman, S.K. Miller, S. Petri, and D. Coumou. 2018. “Projected Changes in Persistent Extreme Summer Weather Events: The Role of Quasi-Resonant Amplification.” Science Advances. 4(10), eaat3272. https://doi.org/10.1126/sciadv.aat3272.

National Oceanic and Atmospheric Administration. 2023. “Weather Related Fatality and Injury Statistics.” https://www.weather.gov/hazstat.

National Oceanic and Atmospheric Administration’s National Centers for Environmental Information (NCEI). 2025. “Global Climate Report.” https://www.ncei.noaa.gov/access/monitoring/monthlyreport/global.

Proussaloglou, E., J. Kane, and A. Tomer. 2022. “Data Shows 23 Million Americans Live in Places Most At Risk of Extreme Heat.” Brookings Institution. https://www.brookings.edu/articles/datashows23millionamericansliveinplacesmostatriskofextremeheat/.

The Guardian. 2025. “Trump Plans on ‘Phasing out’ FEMA Disaster Agency After Hurricane Season.” June 11, 2025. https://www.theguardian.com/usnews/2025/jun/11/trumpfemaphaseouthurricaneseason.

Tuholske, C., K. Caylor, C. Funk, A. Verdin, S. Sweeney, K. Grace, P. Peterson, and T. Evans. 2021. “Global Urban Population Exposure to Extreme Heat.” Proceedings of the National Academy of Sciences. 118(41), e2024792118. https://doi.org/10.1073/pnas.2024792118.

U.S. Department of the Interior. 2025. “Natural Disaster Recovery.” https://www.doi.gov/recovery.

Vaidyanathan, A. 2020. “Heat-Related Deaths—United States, 2004-2018.” Morbidity and Mortality Weekly Report. 69(24), 729-734. https://www.cdc.gov/mmwr/volumes/69/wr/mm6924a1.htm.

Waite, M., E. Cohen, H. Torbey, M. Piccirilli, Y. Tian, and V. Modi. 2017. “Global Trends in Urban Electricity Demands for Cooling and Heating.” Energy. 127, 786-802. https://doi.org/10.1016/j.energy.2017.03.095.