As temperatures in the Midwest plunged thanks to the polar vortex last week, Australia was experiencing record heat. It made us wonder: What was the widest temperature gap between the world's hottest and coldest places on a given day? Is global warming affecting that spread? After processing 63 years of data from 90,000-plus weather stations, we have an answer: It's not clear.

Well, that's not entirely true. On June 13, 1966, the recorded difference between the world's hottest and coldest places was 134.4º Celsius. At a weather station on the border of Senegal and Mauritania, the temperature maxed out at 52.2º Celsius (126º Fahrenheit). The coldest spot was -82.2º Celsius, or -116º Fahrenheit — in the Antarctic.

The locations of the recording stations are the root of our uncertainty. The Wire compiled this data using the records of the National Climatic Data Center, which lets you download a huge (two-plus gig) file containing all of the weather readings from its Global Historical Climatology Network (the GHCN-Daily). After running scripts to pull out the maximum and minimum temperatures around the world for each day between January 1, 1950, and December 31, 2013, we got the following graph.

That 1966 date, you'll notice falls into the weird period between those two orange lines, a period in which the fluctuations in the spread are varied by season, and in which the spread is larger than in the periods before 1957 (or so) and after 1996 (or so). Then we plotted the daily maximum and minimum temperatures, giving us this graph.

Clearly, something changed in the low temperature data for a few decades. We sent our graphs to Matthew Menne of the National Oceanic and Atmospheric Administration, who confirmed our suspicion: The change was due to the system using different tracking stations. "[T]he low variability period you pointed out in your graph," he said via email, "coincides with the period of record in GHCN-Daily for the South Pole station. This would explain the lower variability and the lower annual extremes during that period."

Similar data access explains why the spread goes up and down so much with the seasons, despite it being winter in the Southern Hemisphere when it's summer in the Northern, and vice versa. "There are many more high latitude Northern Hemisphere stations than high latitude Southern Hemisphere stations," Menne pointed out. "There is also much more high latitude land area in the Northern Hemisphere." So there are more stations, higher north, which gives lower temperatures when it's winter in the Northern Hemisphere than in the Southern.

What this means is that our assessment is dependent on the data we have available. No one is tracking temperatures everywhere all the time; or, at least, no one was doing that in 1950. If you exclude the period when the NCDC tracked temperatures near the South Pole, the widest single-day spread between the world's hottest and coldest recorded temperatures was on January 16, 1952, when there was a 109º Celsius spread between the hottest and coldest places.

The limitations on the available data also means that some fascinating details emerge. On July 7, 1951, for example, the hottest place recorded was Death Valley, where it was about 122º Fahrenheit. The coldest place was near Lake Tahoe, where temperatures dropped below freezing. But that means that the hottest and coldest places on Earth were both in California — only some 263 miles from each other. You could, theoretically, have been in both the hottest and coldest places recorded on Earth on the same day.

The inclusion of the geographic locations of the weather stations allows us to do some interesting analysis. Below, we compare all 365 of the hottest and coldest places from each day in 1950 and 2013.

1950

2013

Can you guess what continent saw its hottest temperatures on record in 2013? The answer is Australia, which is home to the stations that recorded the hottest days of the year. Compare that with 1950, in which the United States and Mexico had more hot days — and in which the American Rockies recorded a number of the coldest temperatures. Some of this is related to where weather stations are placed, of course, but not all of it.

We also made this tool that lets you see the hottest and coldest recorded temperatures on any of the days from 1950 through 2013. (Some of the coldest temperatures were recorded in the Antarctic for an extended period, remember, so they won't always show up on the map.)

What does all of this tell us about climate change? Not a whole lot. The spread has remained relatively consistent between the hottest and coldest places over time. The graph at right shows how the average hottest temperatures for each day in each decade has increased. In the 1950s, the hottest places on each day were, on average, 44.68º Celsius. In the 2000s, that was up to 45.08º. Which reinforces a key point about climate change: even long-term weather trends are subtle, even if isolated incidents — like a polar vortex or a heat wave — are not.