Stay informed with free updates
Simply sign up to the Visual and data journalism myFT Digest — delivered directly to your inbox.
April is renowned for its unpredictable weather in many parts of the world. In German lore, the moody weather is captured in the saying April, April — der macht was er will, or “April does what it wants”.
A “false spring” in the northern hemisphere can be followed by snowfall and temperatures below freezing — all within a matter of days.
This phenomenon of rapid swings between warm and cold extremes is also often described as “temperature whiplash” and according to the latest research — peer-reviewed, unlike weather lore — it’s becoming more common as our planet warms.
Research published in Nature Communications this week found that more than 60 per cent of the world experienced more frequent, intense and more rapid flips between warm and cold extremes between 1961 and 2023.
The study identified stark examples such as the warm-to-cold flip in the Rocky Mountains in September 2020, where areas suffering a severe heatwave were blanketed in snow within hours as temperatures dropped by more than 20C.
The researchers analysed the data over more than six decades to track the warm-to-cold flip events when temperatures swung from one standard deviation (s.d.) above the mean temperature to one s.d. below the mean within five days — and vice versa.
They found broad areas in the mid-latitude regions of west Europe, south and south-east Asia, South America and the southernmost region of Africa experienced the greatest effects.
The physics behind the effect were “challenging”, a scientist who was not involved in the study, which attributed the increased wild swings to a combination of an increased “waviness” of the jet stream which encircles the globe as well as changing evaporation pattern as the planet warms.
How we made it
To map flip event frequency, I downloaded the geospatial data provided by the authors here. They used temperature data from sources including the fifth generation European ERA5, Berkeley and the US National Centers for Environmental Prediction in their analysis (all three processed sets are available at the link). I only downloaded the ERA5 data for mapping purposes.
The file contains all the variables they computed: flip event frequency, intensity and flip transition duration for both cold-to-warm and warm-to-cold swings for each of the four seasons as well as the annual average.
I was primarily interested in visualising the most affected spots for cold-to-warm and warm-to-cold events, and so used the programming language R to open the netcdf file and extract the appropriate data layers (finding my way around the file took a minute — too many dimensions!)
[Nerd note: NetCDF, or network Common Data Form, is a format that supports the creation, access and sharing of array-oriented scientific data.]
Some quick small multiples maps (using the open source ggplot package) allowed for a good overview of the data, but it was obvious that trends would be quite difficult to make out at this scale.
So instead, I calculated the average event frequency across all years (1961-2023). After some final tweaks to the data format, I styled the final data in the open-source geospatial software Qgis.
The scientists also modelled what would happen to flip events under various emissions pathways.
On average, flip events are forecast to increase in frequency and intensity and become more rapid in their swing transition in future years. In other words — even bigger temperature drops and spikes, more of them, and faster swings.
The magnitude of change is largely dependent on which emissions scenario is used as the multi-line chart shows.
And finally, I created a simplified version of the explanatory diagram that was included in the paper to show which swings qualified as a flip event.
In a scenario of high greenhouse gas emissions which are the cause of climate change, the research found that the global population exposed to rapid temperature flips would more than double by the end of the century — with low-income countries the most affected.
Wei Zhang, professor at Utah State University and co-author of the study warned that the increased risk of flips can lead to “cascading effects on both natural and built environments such as croplands, ecological systems, and cities.”
As the term “temperature whiplash” suggests, this is largely due to the very short amount of time for humans and ecosystems to adapt.
Join us for the Climate & Impact Summit on 21—22 May, held in-person in London as well as online, bringing together business, finance, and policy leaders dedicated to progress towards the UN Sustainable Development Goals through collaboration, innovation and investment. As a valued newsletter subscriber enjoy a 10% discount on your pass here.