The Holiday Heat Test Cities Are Still Failing

A dangerous heat wave is building across the central and eastern United States just as the country heads into the July 4 holiday, with the National Weather Service reporting moderate to severe HeatRisk conditions for well over 130 million Americans on Sunday, June 28, according to AP¹. At the same time, Europe’s June heat wave has already been linked to more than 1,300 excess deaths since June 21, according to the World Health Organization as reported by Phys.org². In the U.S. West, heat and wind are feeding fires. Three federal firefighters were killed on June 27 in a burnover on Colorado’s border with Utah, where forecasters warned of extreme fire behavior and rapid growth, according to AP³.

That is the holiday heat test. It is not whether a city can open a cooling center, or a fire agency can issue a red flag warning, or a grid operator can make it through one hot afternoon. It is whether governments can handle heat, wildfire, power demand, mass travel, outdoor events and vulnerable populations at the same time. My answer is no, not fast enough. The systems are learning. They are not yet designed for the overlap.

The first problem is that heat is still treated too much like weather, not like a mass-casualty risk. Heat kills quietly, mostly by worsening existing conditions such as heart disease, kidney disease and respiratory illness, which is why the death toll often appears later in the statistics rather than immediately in disaster briefings. Excess mortality means deaths above an expected baseline; EuroMOMO, Europe’s mortality monitoring network, describes it as the corrected number of registered deaths minus the expected number of deaths, according to its methodology FAQ⁴. That method is useful because many heat deaths are never written down simply as heat deaths, but it also means the public usually learns the full scale after the danger has already moved on.

The retrospective numbers are brutal. A Nature Medicine study⁵ estimated 61,672 heat-related deaths in Europe during the summer of 2022. WHO’s current global heat fact sheet cites that figure and says heat stress is a leading cause of weather-related deaths, with extended hot days and warm nights placing cumulative strain on the body, according to WHO⁶. World Weather Attribution found that the June 2026 European heat wave was made far worse by fossil-fuel-driven warming and that many cities broke or were forecast to break heat-stress records, according to World Weather Attribution⁷. When a death count is already above 1,300 before the event is over, the lesson is not that warning systems do nothing. The lesson is that warning systems are being asked to compensate for housing, health care, labor and energy systems that were not built for this climate.

The U.S. has made real progress on heat intelligence. The CDC and NOAA launched HeatRisk, a seven-day forecast meant to show when heat may harm health, and the CDC says the tool is designed for state and local public-health officials as well as the public, according to the CDC⁸. The CDC’s Heat and Health Tracker also uses emergency department data from the National Syndromic Surveillance Program to show heat-related illness visits by region, according to CDC dashboard documentation⁹. Syndromic surveillance is a fancy term for near-real-time health signals, such as emergency-room visit patterns, that can show stress before final diagnoses and death certificates are complete.

Those tools matter. They also expose the gap between information and protection. A cooling center is simply an air-conditioned public place where people can go during dangerous heat, but a cooling center only works if it is open when people need it, reachable without dangerous travel, welcoming to unhoused people and known to older adults living alone. New York City’s 2026 heat plan is a serious model: it opens cooling centers, expands outreach to older adults, people with chronic illness and people experiencing homelessness, monitors health impacts in real time and coordinates with utilities, according to the NYC Mayor’s Office¹⁰. New York State has also issued cooling-center guidance, expanded free air-conditioner eligibility for some residents and added vulnerable-worker outreach, according to the state Department of Environmental Conservation¹¹.

But New York is not the country. The Congressional Research Service warned that most states do not have specific laws, standards or plans addressing extreme heat and described a governance gap in which responsibility is scattered across public health, emergency management, labor agencies and local governments, according to CRS¹². That is exactly the wrong architecture for a compound event. The elderly person in a fifth-floor apartment, the delivery driver, the farmworker, the tourist at an outdoor concert and the firefighter on a wind-shifted slope do not experience heat as separate policy silos.

The wildfire side of the test is even harsher because it punishes timing. A red flag warning is a National Weather Service alert for dangerous fire-weather conditions in the next 12 to 24 hours, according to the NWS¹³. The Utah and Colorado fires show what that means in practice. Utah declared a state of emergency and restricted fireworks ahead of July 4 as the Cottonwood Fire became the largest active wildfire in the country, while the weather service issued a rare Particularly Dangerous Situation warning and red flag warnings covered much of the state, according to AP¹⁴. Utah’s state forester said fires were spreading farther and faster under conditions that defied historical expectations, according to the same AP report¹⁴.

Then three firefighters died. They were working the Knowles and Gore fires, which merged with others into the Snyder Fire, and were killed when fast-moving flames overtook them in Mesa County; two others were burned, according to AP³. AP reported that Grand Junction hit 93 degrees on Saturday and winds gusted to 44 mph, while about 44 square miles had burned in the Snyder Fire by Sunday, according to AP³. A burnover does not prove incompetence. It proves that under the new summer conditions, the margin for error is shrinking faster than staffing, evacuation planning and fuel reduction can comfortably absorb.

The wildland-urban interface, or WUI, is the zone where homes and other development meet or intermingle with undeveloped vegetation, according to the U.S. Fire Administration¹⁵. That zone turns a vegetation fire into an evacuation, power, insurance, smoke and housing crisis. Holiday weekends add ignition sources, visitors, traffic and fireworks. The old mental model of fire season as a rural land-management problem no longer fits.

The grid is the hidden hinge in all of this. Heat drives air-conditioning demand just when power plants and transmission lines become less efficient. FERC’s 2026 summer assessment says electricity consumption is expected to be higher than in each of the previous five summers and that extreme heat can strain the grid by raising demand, reducing transfers between regions, increasing transmission losses and reducing generator capacity, according to FERC¹⁶. FERC also says all regions should have enough resources under normal conditions, but New England, western ERCOT and the Northwest face tighter availability under extreme scenarios, according to FERC¹⁶.

That is better planning than denial. NERC’s 2026 snapshot says more resources have strengthened summer readiness and that elevated-risk areas fell from six in 2025 to three regions plus one locality in 2026, according to NERC¹⁷. But NERC also says load growth has increased by 11 gigawatts since 2025 and that early summer heat overlapping with spring maintenance outages could challenge reliability, according to NERC¹⁷. That warning was not theoretical. In May, the Energy Department issued an emergency order allowing PJM to curtail data centers and other large loads with backup generation after PJM said hot weather and planned maintenance left it expecting less than 5,800 megawatts of reserves, according to Utility Dive¹⁸.

The strongest counterargument is that emergency orders, pre-positioned crews and heat dashboards are evidence of adaptation, not failure. Europe, for example, has announced its largest-ever wildfire response for summer 2026, including nearly 800 pre-positioned firefighters, aircraft, EFFIS fire-risk forecasts and Copernicus emergency mapping, according to the European Commission¹⁹. Insurers and cities are also beginning to price climate risk more explicitly: EIOPA’s natural-catastrophe dashboard uses economic losses, insured losses, risk estimates and coverage data from 30 European countries to identify climate-related insurance protection gaps, according to EIOPA²⁰, and C40 says climate budgeting is meant to make climate goals shape actual city spending decisions, according to C40²¹.

I take that evidence seriously. It shows movement. It does not prove sufficiency. If a hospital buys more surge beds while admissions rise even faster, the hospital is improving and still falling behind. That is where summer governance now sits. Wet-bulb temperature, the measure of heat plus humidity that indicates how well sweat can evaporate and cool the body, captures the human version of the same problem: the body can adapt only until the physics stops cooperating. Cities and fire crews have their own wet-bulb threshold. It arrives when forecasts are accurate, alerts go out, resources are pre-positioned, and the combined load still exceeds the design.

The test for the next few years is not whether authorities can point to better tools. They can. The test is whether excess deaths fall during comparable heat events, whether cooling centers operate like critical infrastructure rather than optional public amenities, whether grid emergency orders become rare, and whether firefighter deaths decline per day of extreme fire weather. Until then, I would not call this adaptation fast enough. I would call it a race in which the public sector has finally started running while the heat keeps accelerating.