Climate change is a complex, slowly shifting thing, measured over the course of decades, not days. Since it first rose to the attention of the international scientific community, there has been a regular struggle to both predict what was to come and to explain those likely outcomes to the public. Record downpours from massive hurricanes bore the fingerprints of a warming planet, but it was often tricky to explain how and to place blame for the storm squarely on changing climate.

This summer, though, has been different. Unprecedented heat waves in the Pacific Northwest, record temperatures in California and on the East Coast and a staggering drought desiccating much of the west are not difficult to explain in terms of global warming. When the world gets hotter, the weather gets hotter. The line is direct.

The New York Times’s David Leonhardt had an interesting bit of data earlier this week looking at how temperatures have shifted in American cities since the 1970s. The average high temperatures in a number of metropolitan areas has increased by several degrees over that period, a mark of how warmth has crept upward, making possible sudden record-breaking surges. Shift the baseline of the curve and the outliers become more likely.

Leonhardt’s analysis made me curious about what that shift looks like universally. So I looked it up.

The National Oceanic and Atmospheric Administration has an enormous amount of data from thousands of weather stations around the world, stretching back more than a century in some cases. Here, for example, you can download weather data for 1763 or any year since. In more recent years there are more data available in a broader range of places, but there’s still more than enough to give a sense of patterns.

I took every point of data from 1961 to 1970 and from 2011 to 2020 — periods a half-century apart — and assigned the daily maximum temperatures in January and July during those periods from each weather station to the nearest tenth of a degree latitude and longitude. In places where there was data from both decades, I compared current and historic averages to generate a picture of how temperatures have changed.

The result? This map.

There’s a lot of data there, obviously. Notice first that there is a lot more red — meaning more average increases — in January than in July. Winters in the Northern Hemisphere have gotten warmer more uniformly than the summers. But the main thing that is immediately obvious is the lopsidedness of the heating.

Of the nearly 9,200 latitude-longitude combinations for which I had usable data, the average temperature increased in January for nearly 7,800, or about 85 percent. The average increase was about 1.5 degrees Celsius, or about 2.7 degrees Fahrenheit. In July, just over 6,600 locations saw increased temperatures, or 72 percent. The average increase was about half as large as in January.

Distributing the frequency of those changes by latitude gives us the following chart.

Here, notice that more northern locations saw bigger temperature increases in January. (You can also see here, as in the first map, how heavily weighted the number of stations are in the Northern Hemisphere, a function in part of population distribution but also of American-European overrepresentation in the data.)

Having this data in hand allowed me to go a little further, creating an interactive showing not only what the change looks like in every United States location for which the data were available but also showing how the temperature in the 2010s compared to the temperatures in other places in the 1960s.

If you click the map below, it will find the closest latitude-longitude combination and show you places that were about as warm 50 years ago as the selected location is now. Often, those locations will be further south; occasionally, much further south.

Show values for

Your browser cannot display this map.

And, of course, you should flip from July to January. The change is remarkable.

None of this tells us anything particularly new, admittedly. We know that the world is getting warmer, and we know that this has significant effects. Perhaps seeing the shift just over the past 50 years presented in this way, though, will help make the scale of that change more tangible.

Well, not more tangible than walking out into 100-degree heat for a week straight. But you get the point.