SHARE

Drop a frog into a boiling pot of water and it immediately hops out. But place it into cool water and slowly heat it, and the frog won’t catch on, eventually getting cooked. This metaphor is totally wrong about frogs, but it is true of humans tweeting about unusual weather.

Unusual weather, in which temperatures deviate from long-term averages, is becoming more common as we warm the planet. Compared to pre-industrial times, we’re seeing more frequent heat waves, warm winters, and—in some places—cold snaps. But, in boiling frog fashion, we find such temperature extremes unremarkable after just a few years of repeated events, according to a new study published Monday in the Proceedings of the National Academy of Sciences.

An analysis of 2.18 billion tweets from users in the continental United States revealed this near-term baseline for thinking about weather. Basically, scientists looked at posts between March 2014 through November 2016 that used words from a predetermined weather-related list. Then, they looked at weather during that same period, identifying especially hot or cold swings compared to a 10-year reference period from 1981 to 1990. (The planet’s began warming long before then, of course, but 1981 is where PRISM, the high-resolution temperature dataset the researchers used, begins). During a week of extreme temperatures in a particular county, the scientists counted the number of tweets about weather from that location.

They found that after two years of repeated extreme temperatures, people were less likely to tweet about these events. “If you keep getting that [temperature anomaly] two years in a row, then the remarkability starts to decline,” says Frances Moore, a professor of environmental policy at the University of California, Davis and the study’s lead author. “And then after eight years it completely declines—you don’t find the events any more remarkable.”

Two to eight years seems to be people’s baseline for determining whether a particularly warm or cool week is remarkable. That’s disheartening from a climate change perspective, because such a limited memory obscures that these extremes are unprecedented in the context of, say, a century.

The researchers didn’t stop there. Moore wondered if this change in the remarkability of an extreme weather event meant that people were adapting—that they no longer found such extremes uncomfortable. So she and her team analyzed all the non-weather tweets from same time period to determine sentiment. These non-weather tweets revealed that both hot and cold anomalies correlated with expressing negative sentiments. Even after five to 10 years of anomalies, when such conditions have become unremarkable, the bouts of extreme weather still seemed uncomfortable—or at least negatively impacted the mood of Twitter users.

“If we think about how people experience climate change on a day-to-day basis, it’s really just gradual changes in the distribution of weather throughout the year,” says Moore. “So understanding how the gradual changes are perceived by people who are affected is important in understanding how people are thinking about the weather that’s happening outside their window.”

One perhaps obvious implication is that people might not think repeated heat waves are unusual, although global warming is upping their frequency. Our limited weather memory also means we might forget what a normal winter looks like. When an area is hit with a burst of cold that’s actually normal, given historic averages, we’ll find it especially cold against the backdrop of gradually warming winters.

Importantly, the study doesn’t say whether this loss of remarkablility is actually affecting how people think about climate change. The authors do express that a lack of remarkability could obscure “windows of opportunity”—crisis-type events that rally public attention around a given issue, making big policy changes possible.

Still, people tend to form their beliefs about climate change from many sources, including their political ideology. In a 2014 study, scientists found that “perceived scientific agreement; beliefs about the current onset, human cause, threat and seriousness of global warming; and political orientation” influence whether we attribute warm winter temperatures to climate change more than temperature anomalies. The authors conclude that this is not surprising given the “political polarization of climate change beliefs.”

In the future, Moore wants to study coastal flooding, which is also increasing under climate change. Do people also normalize more destructive events, too?

Moore hopes her findings help to “inform efforts around climate change communication.” Scientists and the general public might use two very-different frames of reference in thinking about weather, but “there are ways to bridge that … [by] giving people the longer-term historical context to evaluate their experiences about weather.”