[Edited to correct dates typed too early in the morning]
Here’s a map I would’ve liked to see a couple years ago when I was doing some thinking about seasons:
Specifically, about how we define the endpoints of the seasons. For meteorologists, the seasons are tied to the calendar: March 1, June 1, September 1, and December 1 are the boundaries. Non meteorologists usually talk about the seasons beginning and ending on the solstices or the equinoxes. Either way, it sometimes seems kind of wacky. Consider Groundhog’s Day, for instance: If the groundhog sees its shadow there will be six more weeks of winter? And if it doesn’t? Winter will end in early February? In Syracuse, when we routinely are still shoveling snow in early April, we really wonder who came up with that idea.
Wouldn’t it make more sense to center summer around the hottest day of the year, winter around the coldest, and spring and fall around the midpoints? I was thinking so, but that prompts the question of when the hottest and coldest days are. So I started looking at climate data and found them confusing. I ended up scraping data for a couple dozen locations in the US and learning how much variation there is in the hottest and coldest days. In other words I was making a very low resolution of the map above and its coldest day counterpart.
It’s pretty shocking. There are places in Texas about a hundred or so miles apart, for instance, where the hottest day for one is in early June and for the other is in mid August. Then there’s the Pacific coast, where there are places the hottest day doesn’t arrive until September.
That pink area on the Mexican border, in Big Bend National Park in Texas, is a surprise. After all, you have glib explanations for summer temperatures like this one:
In a similar manner, in the Northern Hemisphere, the amount of heat received by the Earth from the Sun is increasing slowly towards summer solstice. As one goes towards summer, the heat received during the day is greater than the heat radiated during the night and so the average temperature slowly increases. The heat input is a maximum at solstice and decreases after solstice, but the rate of heat input is still greater then the rate of heat dissipation. Hence, the average temperature keeps increasing even after solstice, and it is only later during the year that the average temperature starts decreasing.
Which proves decisively that Big Bend, which reaches its highest temperature before the solstice, does not exist.
No, climate’s more complicated than that. What I’m guessing is that the above explanation fails because it doesn’t take either the motion of the atmosphere and sea currents or the different heating rates of atmosphere, ground, and water into account. Air that’s warmed up in one latitude, for instance, can carry that heat to a different latitude before it fully cools off, and the same for ocean currents that then can affect the air temperature. The latter is why temperatures in England, for instance, are so much warmer than Labrador at the same latitude — the Gulf Stream carries warm water from the Gulf of Mexico northeastward toward the British Isles.
The places where the hottest day is very late are on the Pacific seacoast and presumably are driven by the temperature of air carried there from over the Pacific ocean by the prevailing winds. It takes a long time to heat up an ocean, and the ocean readily exchanges heat with the atmosphere — more readily than land does. I don’t know the details but my guess is some combination of relative heating rates and heat exchange rates combined with ocean versus land wind patterns is what’s responsible for the delayed warmest day.
As for Big Bend, it isn’t very far north of the Tropic of Cancer. Below the Tropic the sun reaches its highest elevation, and the air and ground get their highest rate of heat input, before the solstice. If the prevailing winds carry that warmed air north, that could account for the early warmest day.
Nice theory, except it turns out the prevailing winds in that area actually are out of the northeast! Big Bend still doesn’t exist. Or there’s something else going on.
Anyway, our present definitions of the seasons are fine, thank you. Trying to line seasons up with warmest and coldest days will just drive you to insanity.