The Thermometer Under the Umbrella

“Why are they having fires in Europe?” a family member asked.

The question took me off-guard because I thought it was obvious. “Because it’s dry.”

“But it gets hot and dry here, and we don’t have them.”

I stared at my kin. I was tempted to say “Of course we do: that’s why we have forestry units,” but at that moment I wanted to eat in peace.

“It’s 104° over there. It gets that hot here,” my kin said.

That much and hotter, and that’s before we factor in the humidity. “They aren’t used to it. There was a British governor who thought Georgia was the hottest place on earth and hung a thermometer from his umbrella to see the temperature.”

That got a laugh and we returned to our meal.

I kept thinking about that British governor. After supper, I took an old history book and happened to open it to the page I remembered. He was Henry Ellis, Governor of Georgia from 1757 to 1760. Ellis was so impressed by the Georgia heat that he wrote a short paper, published in Philosophical Transactions in 1758 (you can read the article at Ellis wrote of experiencing temperatures of 102° and 105° F (38.9° and 40.6° C) in the summer of 1757. Ellis observed:

I think it is highly probable, that the people of this town (Savannah, Georgia) breath a hotter air than any other people on the face of the earth.

Which isn’t true, but Ellis was convinced it was.

That brought to mind one summer night at work some decades back. We were working to restore power when a company official stopped by with his brother-in-law. His bother-in-law, from London, looked shook up as he told about the tornado he’d just experienced. Except it was just an average summer thunderstorm. Well, average to us. We didn’t have the heart to tell him. On the other hand, he likely would have a good laugh at what we consider a heavy snowfall. We become accustomed to weather.

That’s an important thing to keep in mind. I’m not making fun of the residents of the UK who are suffering in their heat wave. Far from it. We in the southern US are used to such temperatures, but they aren’t, And that leads to something that’s gone unsaid: It’s possible to get used to such temperatures.

“Ah, but the southern US has air conditioning and the UK doesn’t,” some are apt to point out. That’s true for many Southern homes – now. I grew up when it wasn’t. From the first European colonies in the US until maybe the last quarter of the 20th century, when it was hot you opened your shutters or windows and hoped you wouldn’t have to close them for rain. If you had electric fans, you used them. I remember when movie theaters and some stores and doctor’s offices were air conditioned, but schools weren’t. Nor were our churches. We were fortunate enough to get central heating and cooling in our home in the early 1970s and I remember when window units became popular, but it was some time after 1976 when the local high school installed them. Even now it’s impossible to air condition the outside, and that’s where many of us are in the summer, because most of the work in rural areas is outdoors. In the summer we live with the heat. When it’s humid, it feels even hotter. And if Henry Ellis’ short observation of Savannah temperatures is any indication, the high temperatures aren’t that much hotter in a little over a quarter of a millennium. You have to take precautions with the heat, but it is survivable. And, despite what Governor Ellis thought, people do live and function in hotter places.

Dismissive? No. Realistic, yes.

This won’t sit well with those who are absolutely convinced that, in the words of Dr. Zachary Smith, “We’re doomed.” Uh, no. If summers in the southern UK begin to resemble those in the southern US, the British will survive, just as those of us with European ancestors have survived in warmer climes. They will adapt and get used to it. It’s what we did in the southern US; it’s what the Australians did, too.

How about us in the southern US? How will we fare? Let’s start by asking what’s the highest heat an average human can endure. The theoretical human limit is a heat index of 160° F / 71.1° C, and that for three hours at the most. That corresponds to a wet bulb temperature of 95° F / 35° C but note that web bulb temperature is measured in direct sunshine while the heat index is for the shade. Either way, we now have an upper limit.

Both heat index and wet bulb temperature are based on temperature and humidity. Humidity is often measured as relative humidity, or RH, which is how much water vapor is in the air divided by how much water vapor the air can hold, then multiplied by 100 to express this as a percentage. How much water vapor the air can hold depends on temperature and air pressure. As the temperature rises, the air can hold more water vapor. If the amount of water vapor in the air remains the same, the RH will drop as the temperature rises, and rise as the temperature drops. Humidity affects how well perspiration cools the body. The higher the humidity, the less perspiration can evaporate and the harder it is for the body to cool itself. The heat index is a “feels like” value based on this, which is why higher humidity raises the heat index above the actual air temperature.

A quick search shows Savannah has an average RH of 55-57% in the afternoons of June and July, with an average temperature of 89° F / 31.7° C in June and 92° F / 33.3° C in July. Just when the relative humidity is measured is very important here, since the relative humidity tends to drop as the temperature rises. This is why higher temperatures in Savannah tend to have lower humidity than 57%. Let’s say that one humid morning the RH is 90% with a low of 79° F / 26.1° C, but the temperature climbs to 100° F / 37.8° C. If the air pressure remains the same, the relative humidity will drop to about 46%. The average low in June and July is 71° F / 21.7° C and 74° F / 23.3° C, so far more likely is what occurred on June 23, 2022, where records from one site recorded an RH of 87% with a low of 72° F / 23.9° C. but an RH of 27% when the temperature reached 102° F / 38.9° C.

Governor Ellis wrote that for several nights in the summer of 1757 the low was 89° F / 31.7° C. Let’s use this as a low for the morning he observed his high temperature. Let’s also guess that the RH at the time of the low was 80%. If this was the case, the RH when Ellis observed a temperature of 105° F / 40.6° C, the RH could have been about 49%. If so, the heat index would have been about 134° F / 56.7° C. Since Ellis didn’t note the morning low and we don’t know the humidity, all this is a guess. If the RH or the low that morning were less, the heat index when Ellis measured the high would have been, too. My guess, based on skimming this year’s Savannah weather data, is that it was.

This, of course, assumes his thermometer gave him the correct reading. Most likely it did. In 1776, the Royal Society of London appointed a committee, headed by Henry Cavendish, to examine the accuracy of thermometers (you can read the report at On page 18 of the linked article we find:

Though the boiling point (of water) is placed so much higher on some of the thermometers now made than others, yet we would not have the reader think that this can make any considerable error in the observations of the weather, at least in this climate; for an error of 1¼ in the position of the boiling point will make an error of only half a degree at 92°, and not more than a quarter of a degree in the point of 62°.

The highest official recorded temperature for Savannah is also 105° F / 40.6° C, occurring on July 20, 1986. The National Weather Service records for Savannah begin in 1871, so that’s the highest observed temperature from then until July 2022. A generally accepted estimate holds that average global temperature has increased about 1° C / 1.8° F since 1750 due to the Industrial Revolution. If this is the case and if it shows up in the average annual temperature for Savannah, it’s interesting that Ellis’ observation and the peak high recorded over a quarter of a millennium apart are the same. Pointing that out is apt to get knickers in a bunch, so if anyone needs to scream “Heretic!” go ahead; we’ll wait.

Now then: Some say average global temperatures will rise by another 1° C by 2050, so let’s use that in our estimate. Let’s also assume that the peak high in Savannah will increase by that amount, even though we don’t see it between Governor Ellis’ observation and the peak official temperature. That would make it 106.8° F. Let’s call it 107° F / 41.7° C.

Now let’s consider humidity. Let’s bump up the average morning low to 76° F / 24.4° C and say the RH is 80% (to pull a number out of the, er, air). At 107° F / 41.7° C, the RH would be about 30% and the heat index would be about 118° F / 47.8° C. If we bump up the morning low to 80° F / 26.7° C and the RH to 90%, the RH at the daily high would be about 39%. That gives us a heat index of about 122° F / 48.9° C. If we apply the same conditions we used for Governor Ellis’ observation, that gives us a heat index of 142° F / 61.1° C.

Remember, this is a peak high value. Increasing the average high by 2° F to 94° F / 34.4° C and keeping the average afternoon RH of 57% gives a heat index of about 109° F / 42.8° C.

Based on this, my guess is that yes, the southern US would adapt. Taking the often cited increase of 1°C by 2050 at face value, it would be hot, but not so hot humans couldn’t live through it. That’s if we see these kinds of temperatures.

Oh, my: I said “if,” didn’t I? If anyone needs to howl “Heretic!” again, feel free to do so. Yet when I got the idea for this post, I didn’t know Governor Ellis had written a short paper on the temperature he observed in Savannah in the summer of 1757, nor that period thermometers were reasonably accurate. I certainly didn’t expect that the official peak temperature for Savannah, measured over a quarter of a millennium later, would be no higher than what he observed. Of course, Ellis only provided a data snapshot, not daily records of his time in Savannah, and thus we have no clear idea of what the average observed temperature was in Savannah during those years. Yet this raises a question beyond the most obvious: If the average global temperature has risen about 1° C, almost 2° F since 1750, does Ellis’ data snapshot from 1757 mean that average annual temperatures can rise without unusual temperatures rising as well or not rising as much? This seems counter-intuitive, since you’d think that a rise in temperatures would happen uniformly. But what if this isn’t the case? One data point does not a data set make, so it’s only a question. Yet it’s a question that needs to be asked. The answer could turn out to be important.

I wonder if anyone would do the research to answer it.