You've felt forty degrees in the shade. You've probably never felt fifty-seven. Daniel wants to know why. He wrote in with a whole stack of questions — why forty-plus is routine but fifty-plus is vanishingly rare, why the planet seems to have a temperature speed limit, and whether there's an actual survivability threshold where heat stops being uncomfortable and starts being lethal even for healthy people. He points out that car thermometers lie constantly, that the Death Valley record from 1913 sits at nearly fifty-seven degrees and nothing has ever officially crossed sixty, and that heatwaves reliably spike mortality — especially for older people and diabetics. So we've got a physics question, a physiology question, and a public health question, all wrapped around the same number on a thermometer.
And the number most people think they know is wrong. That 1913 Death Valley reading — 56.7 Celsius, 134 Fahrenheit — is under active review right now. The American Meteorological Society published an article in late 2025 questioning whether it should be decertified entirely. The instrumentation was questionable, there was a sandstorm that may have contaminated the readings, and the observer who recorded it had a pattern of... let's call it enthusiastic reporting.
Enthusiastic reporting. I'm going to remember that phrase.
The point is, if that record falls, the new official maximum drops to fifty-four degrees — recorded in Death Valley in 2020 and 2021, and 53.9 in Kuwait and Pakistan. Suddenly sixty degrees isn't just unobserved — it's almost physically impossible given how the atmosphere works. And that's where the real story starts.
So let's start with the physics. Why does the atmosphere stall out? Forty degrees is common near the Dead Sea — I've felt it, you've felt it, it's unpleasant but unremarkable. Fifty is a once-in-a-generation event. Sixty has never happened. What's the ceiling made of?
It's an energy balance problem. The atmosphere heats up because the ground absorbs solar radiation and re-radiates it as infrared. The air near the surface warms, rises, and convection carries heat upward. At the same time, any moisture in the soil or vegetation evaporates, and that phase change pulls enormous amounts of heat out of the system. You've got radiation coming in, radiation going out, convection shunting heat upward, and evaporation acting as a giant air conditioner. Those four processes set a thermodynamic speed limit.
And that limit is somewhere around fifty-four degrees.
Roughly. The math isn't a single clean number — it depends on latitude, surface type, atmospheric pressure, moisture content — but the theoretical maximum surface temperature on Earth, given our solar input and atmospheric composition, clusters around fifty-four degrees Celsius. You can push slightly higher in a perfect setup — a deep desert basin below sea level, bone-dry air, no vegetation, intense high-pressure system sitting overhead for days — and that's exactly what Death Valley is. It's below sea level, surrounded by mountains, and it acts like a solar oven. Even then, the highest credible readings top out at fifty-four.
Which makes the 1913 number — 56.7 — genuinely suspicious.
It does. The 2025 AMS analysis points out several problems. The thermometer used was a standard mercury maximum thermometer, but the observer at Furnace Creek that day noted a sandstorm had passed through. Sand and dust can coat the instrument housing, trap heat, and produce readings several degrees above true air temperature. The same observer had recorded other implausible highs earlier that summer that didn't match any surrounding station. And the weather pattern that day doesn't support a reading nearly three degrees above anything recorded before or since.
So you've got a dubious instrument, a sandstorm, and a guy whose notebook reads like he was gunning for the record.
I'm not saying he was gunning for it. But the AMS article is careful to note that the observer, a man named Oscar Denton, was under significant pressure to produce notable results for the fledgling weather station. The Furnace Creek station had been established partly to attract tourism. A world record was good for business.
Of course it was.
If the record is decertified, the new official maximum becomes the fifty-four from 2020 and 2021 — both from Death Valley, both using modern automated equipment with rigorous calibration. And fifty-four fits the physics. It's right at the ceiling of what the energy balance allows.
Let me push on that ceiling for a second. The global average surface temperature is about fifteen degrees. Fifty degrees is a thirty-five degree anomaly. What has to line up to produce a thirty-five degree departure from average?
You need a confluence of factors that almost never occur together. First, you need a subtropical desert basin — low elevation, because the air compresses and warms as it descends, and being below sea level gives you a thicker column of atmosphere to trap heat. Death Valley is eighty-six meters below sea level. Second, you need absolutely dry air — no moisture means no evaporative cooling, and water vapor is a greenhouse gas, but dry air actually heats faster in direct sunlight because there's nothing absorbing and redistributing the energy. Third, you need a persistent high-pressure system — a ridge that sits over the region for days, suppressing cloud formation and preventing any cooler air from moving in. Fourth, you need the right surface — bare rock or dry soil with low albedo, something that absorbs nearly all the solar radiation and re-radiates it as heat.
Low albedo meaning dark, non-reflective.
Right. Fresh snow has an albedo of about 0.9 — it reflects ninety percent of incoming solar radiation. Dark desert rock has an albedo of 0.2 or lower. It absorbs eighty percent or more. And when you've got all four of those factors lined up, the surface temperature can climb toward fifty degrees. But here's the thing — even then, the atmosphere fights back. As the surface heats, it radiates more infrared. Radiation scales with the fourth power of temperature — a surface at fifty Celsius radiates far more energy back to space than one at twenty. The hotter it gets, the faster it loses heat. That's the negative feedback that creates the ceiling.
So it's a system that accelerates its own braking.
Every degree above about forty-five requires an exponentially larger energy input to sustain. You're fighting the fourth-power law. That's why forty is routine and fifty is extraordinary — the energy required to push from forty-five to fifty isn't just twenty-five percent more, it's a fundamentally different regime.
Which brings us to car thermometers.
I knew you were going there.
Because everyone has seen their dashboard read fifty-five or sixty and thought, well, there it is, the record's nonsense. Daniel mentioned it explicitly. What's actually happening?
Car thermometers measure air temperature near the road surface, typically from a sensor mounted behind the front grille or under the side mirror. The road surface — asphalt — absorbs solar radiation and can reach temperatures of sixty-five to eighty degrees Celsius in direct sun. The air immediately above the asphalt is heated by conduction and radiation from the road, not by the ambient air mass. Plus the car itself is a heat source — the engine, the exhaust, the dark body panels. The sensor is sitting in a microclimate that's ten to fifteen degrees hotter than the true air temperature measured by a properly sited weather station.
A properly sited weather station meaning what, exactly?
A Stevenson screen — a white, louvered box mounted at 1.25 to two meters above a grass surface, with free airflow on all sides. The louvers block direct solar radiation but allow wind to pass through. The thermometer inside is measuring the temperature of the free air, not the temperature of a superheated surface. That's the standard. Every official record on Earth comes from an instrument in a Stevenson screen or an equivalent modern automated system. Your car's sensor is in none of those things.
So when someone's dashboard says sixty degrees, the real air temperature might be forty-six.
Easily. I've seen a fifteen-degree discrepancy myself, driving through the Jordan Valley in August. Dashboard said fifty-two. The nearest weather station reported thirty-eight. And that's not a broken sensor — that's physics. The asphalt was hot, the car was hot, and the sensor was reading the air in that specific six-inch pocket behind the grille.
You know what I love about this? It's the same error every time and nobody learns. Every summer, social media fills up with dashboard photos and someone saying the official records are fake.
The records are fine. The dashboard is lying. And the distinction matters because when we talk about survivability thresholds, we're talking about the true air temperature and the humidity that comes with it — not what your Kia thinks it is.
Alright. Let's move to the body. What actually kills people in a heatwave?
It's rarely heatstroke. Heatstroke is the dramatic one — core temperature above forty degrees, confusion, loss of consciousness, the body's thermoregulation completely collapses. But that's a small fraction of heatwave deaths. The majority are cardiovascular.
Explain that mechanism.
When your core temperature rises, your body does two things. It dilates blood vessels near the skin — vasodilation — to shunt warm blood to the surface where it can radiate heat. And it activates sweat glands. Both of those processes require the heart to work much harder. Cardiac output can increase by four or five times during severe heat stress. For a healthy twenty-five-year-old, that's uncomfortable but manageable. For a seventy-five-year-old with even mild coronary artery disease, that increased cardiac demand can trigger a myocardial infarction. The heart simply can't keep up with the demand for increased blood flow.
So the heat doesn't cook the person. The heat makes the heart work until it fails.
That's the mechanism for most heatwave deaths, yes. The second big category is renal failure. Sweating pulls water and electrolytes out of the blood. If you're not replacing those fast enough, blood volume drops, the kidneys receive less flow, and in severe cases you get acute kidney injury. Older adults often have reduced thirst sensation — they don't feel thirsty until they're already significantly dehydrated. And many common medications make it worse. Diuretics for blood pressure reduce blood volume further. Beta-blockers limit the heart's ability to increase output. Some antidepressants impair sweat response.
And diabetes specifically — Daniel asked about that. Why are diabetics so vulnerable?
Two reasons. Diabetic neuropathy damages the small nerve fibers that control sweat gland activation and blood vessel dilation. A diabetic person may have impaired sweating — they literally can't cool themselves as effectively. Second, diabetes damages blood vessels throughout the body, including the microvasculature in the skin that handles heat exchange. The combination means they both generate more internal heat from metabolic dysfunction and have a harder time shedding it. Add in the fact that diabetes often comes with obesity, which adds insulation and increases metabolic heat production, and you've got a perfect storm.
I'm remembering something you said once about heat index assumptions.
The heat index — the "feels like" temperature — assumes a standard person in the shade with a light breeze. It's calculated for a healthy adult of average height and weight, wearing light clothing, walking at about one point four meters per second. It doesn't account for direct sun exposure, which can add eight to ten degrees to the apparent temperature. It doesn't account for higher body mass, or age, or medication use, or any of the chronic conditions that change thermoregulation. It's a useful public communication tool, but it's calibrated for the easiest case, not the hardest.
So the official heat index might say forty-two degrees, and a diabetic seventy-year-old in direct sun is experiencing something closer to a physiological load of fifty-plus.
And that's before we even get to humidity. Which is where the wet-bulb temperature comes in.
Define wet-bulb for the conversation.
Take a standard thermometer, wrap the bulb in a wet cloth, and swing it through the air or put it in front of a fan. As water evaporates from the cloth, it cools the thermometer — that's evaporative cooling, same thing sweat does for your skin. The wet-bulb temperature is the lowest temperature that evaporative cooling can achieve under current conditions. If the air is dry, evaporation is rapid and the wet-bulb is much lower than the dry-bulb — the regular air temperature. If the air is saturated — a hundred percent humidity — no evaporation occurs, and the wet-bulb equals the dry-bulb.
And this matters because sweat works the same way.
Your skin is the wet cloth. If the wet-bulb temperature exceeds thirty-five degrees Celsius, sweat cannot evaporate fast enough to cool you — regardless of how much you produce. You just drip. And if you can't shed metabolic heat, your core temperature rises inexorably. At a wet-bulb of thirty-five, a healthy young adult at rest will reach a core temperature of forty degrees — the threshold for heatstroke — within about six hours.
That's the theoretical survivability limit Daniel was asking about.
It is. And it's important to say: thirty-five wet-bulb is the absolute limit for a healthy person at rest, in the shade, with unlimited drinking water. Any activity, any sun exposure, any health condition, and the limit drops. At a wet-bulb of thirty-two — which is about forty degrees Celsius at fifty percent humidity, or thirty-five degrees at seventy percent — a healthy person doing moderate outdoor work is in serious trouble within a couple of hours.
Has anywhere on Earth actually hit a wet-bulb of thirty-five?
Not sustained. There have been brief excursions — a few hours — recorded in parts of the Persian Gulf and the Indus Valley in Pakistan. A weather station in Jacobabad, Pakistan, has come close multiple times. The Persian Gulf, because it's a shallow, warm body of water, produces extremely high humidity combined with air temperatures in the high forties. That combination pushes wet-bulb into the thirty-three, thirty-four range. But a sustained thirty-five — six hours or more — hasn't been documented yet.
Yet.
Yet. And that's where the climate trajectory gets alarming. There's research projecting that parts of South Asia and the Middle East could regularly exceed wet-bulb thirty by 2050, and approach thirty-five during extreme events. That shifts heat from "dangerous for vulnerable people" to "dangerous for everyone." When you can't be outside for more than an hour without risking heatstroke, you can't do outdoor labor. You can't farm. You can't do construction. The economic and social implications are enormous before you even get to the mortality numbers.
Let's talk about those mortality numbers. Daniel mentioned the spike during heatwaves.
The 2003 European heatwave killed approximately seventy thousand people. Most were over seventy-five. Many were in Paris, in upper-floor apartments without air conditioning, during a heatwave that caught the continent unprepared. The 2022 European heatwave killed about sixty thousand. These are not small numbers — heatwaves kill more people than floods, hurricanes, and tornadoes combined, year after year. They're the deadliest weather phenomenon on Earth, and they don't leave dramatic footage, so they're under-covered.
Seventy thousand deaths and most people remember a photo of a dried-up river and a story about nuclear plants shutting down.
The river photo is the visual. The seventy thousand deaths are a spreadsheet. It's a visibility problem. And it's getting worse because the populations most vulnerable to heat — older adults, people with chronic conditions — are growing as a share of the total population in almost every country. More vulnerable people plus more frequent extreme heat equals a rising mortality curve that we're not really talking about.
What's the mechanism for the Paris deaths specifically? You mentioned upper-floor apartments.
Urban heat island effect. Paris is a dense city with lots of stone and concrete — materials with high thermal mass that absorb heat during the day and release it at night. During the 2003 event, nighttime temperatures in Paris didn't drop below twenty-five degrees for over a week. Normally, the body gets a reprieve at night — core temperature drops, the cardiovascular system recovers. Without that nighttime cooling, the heat stress is cumulative. Each day you start from a slightly higher baseline. By day four or five, people who were stable on day one are in crisis. And the upper-floor apartments in old Parisian buildings — they're often directly under the roof, with no insulation, and they turn into convection ovens by late afternoon.
No air conditioning, no nighttime cooling, cumulative stress, and a population that wasn't acclimatized.
And a public health system that didn't recognize what was happening until the morgues started filling up. France's funeral directors ran out of space. They had to use refrigerated warehouses. It was a wake-up call that led to major changes in European heatwave preparedness — early warning systems, cooling centers, check-in programs for elderly residents. The 2022 heatwave was comparable in intensity but the death toll was lower, partly because those systems were in place. Sixty thousand is still catastrophic, but it would have been worse without the reforms.
So the survivability question Daniel asked — is there a threshold for healthy people — the answer is yes, it's wet-bulb thirty-five, but the practical threshold for populations is much lower, and we're already crossing it regularly.
We're crossing the threshold for vulnerable populations every summer. The 2003 event had a maximum wet-bulb around twenty-eight in Paris. Twenty-eight. That's seven degrees below the theoretical limit for a healthy twenty-year-old, and it still killed seventy thousand people. The gap between "theoretically survivable" and "actually survivable for a real population" is enormous, and it's filled with chronic conditions, medications, poverty, isolation, and buildings that weren't designed for the heat they're now experiencing.
That's the part that keeps me up. We talk about the physics of fifty-degree air as this fascinating ceiling, and it is fascinating. But the mortality starts well before you get anywhere near the record. Forty degrees with humidity and no nighttime cooling will do it.
And that's the public health reality. The number on the thermometer is only part of the story. The humidity, the duration, the nighttime minimum, the built environment, the age structure of the population, the prevalence of air conditioning, the strength of the social safety net — all of those determine how many people die. A forty-degree heatwave in Phoenix kills far fewer people than a forty-degree heatwave in Paris, because Phoenix is built for it. Every building has air conditioning. The infrastructure assumes extreme heat. Paris wasn't, and isn't, and retrofitting a thousand-year-old city is slow and expensive.
Alright. Before we move to Hilbert, let me try to pull these threads together. The atmosphere has a thermodynamic ceiling around fifty-four degrees, which is why sixty has never been recorded and probably never will be. Fifty is rare because it requires a perfect alignment of geography, atmospheric pressure, and surface conditions. The 1913 Death Valley record is probably wrong, and if it falls, the new record will be fifty-four — which fits the physics. Car thermometers are not weather instruments and their readings are inflated by ten to fifteen degrees. And the survivability threshold is not a dry-bulb number — it's wet-bulb thirty-five for the absolute healthy-young-adult limit, but the real danger zone for populations starts at wet-bulb twenty-eight to thirty, and we're already hitting that.
That's the episode.
Not quite. Hilbert's been sitting at the board for twenty minutes and I can see he's got something.
Hilbert: I used to take weather readings in the Mojave. 1984, 1985. We had a Stevenson screen at a remote station about forty miles from Furnace Creek. Not a proper weather service installation — it was a contract job for a mining company. They wanted to know if they could work the crews through August.
What kind of temperatures were you seeing?
Hilbert: Official logbook had us at 51.2 on the hottest day. That's what got reported. What I saw on the uncalibrated max thermometer before I swapped it out was 58.
Fifty-eight Celsius.
Hilbert: The screen had a warped louver. The ventilation was wrong. I wrote it up, they sent a replacement screen, and the readings dropped four degrees the next day. But the logbook only shows the calibrated numbers. The 58 never existed, officially.
So you're saying the station was reading high because of faulty equipment, and the real temperature was lower.
Hilbert: I'm saying the station was reading what it read. The equipment was faulty, so the number got adjusted down. That's how it works. But I've always wondered how many other stations did the same thing and nobody caught it. Or caught it and adjusted the other direction.
You think there are records out there that should be higher?
Hilbert: I think the difference between 54 and 57 is three degrees, and three degrees is well within the margin of a poorly sited Stevenson screen, a lazy calibration schedule, or a supervisor who wants the numbers to match the regional average. I'm not saying the 1913 record is real. I'm saying the system is designed to smooth out outliers, and sometimes the outlier is the truth.
That's a fair point. Weather services do apply quality control algorithms that flag and adjust extreme readings. It's necessary because instruments fail, but it also means the true maximum might get erased.
Hilbert: I've still got the logbook from that station. The unofficial one. It's in a box in my garage, next to my thermometers.
Your thermometers. Plural.
Hilbert: I collect them. Mercury, alcohol, bimetallic, a few early digital ones from the nineties. Got about forty. The mercury ones are the best. You can feel the quality.
You have forty thermometers in your garage.
Hilbert: Forty-two. I picked up two more at an estate sale in Barstow last spring. One of them's from a decommissioned weather station in Nevada. Still holds calibration within half a degree.
Do you test them?
Hilbert: Ice bath once a year. Boiling water at sea level. If they're off by more than half a degree I adjust them or retire them to the display shelf.
The display shelf.
Hilbert: Anyway, the logbook's there if anyone wants to look at it. 58 degrees, July 1985. Unofficial. Doesn't change anything, but I saw what I saw.
I'd like to see that logbook sometime.
Hilbert: It's not going anywhere.
The misconception that sticks with me from all of this is the idea that if your car dashboard says fifty-five degrees, the official records are wrong. They're not. Your car is measuring the temperature of a superheated microclimate six inches from asphalt that's been baking in the sun. The official record comes from a calibrated instrument in a ventilated white box on a grass field. Those two numbers have almost nothing to do with each other.
And the corollary — if the 1913 Death Valley record falls, it won't be because the planet got cooler. It'll be because we got better at measuring. The physics of the atmosphere hasn't changed. Our instruments have.
Which leaves the open question: if fifty-four is the real ceiling, and we're pushing more regions toward wet-bulb thresholds that make outdoor life impossible for parts of the year, does the exact record even matter? The number that kills you isn't the one in the record book. It's the one on the thermometer outside your window, combined with the humidity, combined with whether you can get cool at night.
That's the thing to watch. Not whether Death Valley hits fifty-seven again, but whether the Persian Gulf and the Indus Valley start seeing wet-bulb thirty-five — even briefly. Because once that happens, the question stops being academic.
Thanks to our producer Hilbert Flumingtop, who apparently has a museum-grade thermometer collection in his garage and a logbook that might rewrite meteorological history. Or might not. He's not saying.
This has been My Weird Prompts. If you want to send us your own questions about physics, physiology, or anything else that keeps you up at night, email the show at show at my weird prompts dot com.
We'll be back soon. Stay cool.
