Anthony Watts is not a scientist, but instead someone who uses his online blog to attack real climate science. One of his biggest efforts was the Surface Stations project, where he assembled “a grass roots network of volunteers” to second-guess the thermometer-based temperature stations he believed were showing too much warming because of urban development. He also once tried to start a (now-defunct) climate journal of his own, which was open to “anyone who has an interest in atmospheric sciences.” The most generous description of Watts would be that he, and his work, falls into the “citizen science” genre of non-academics who collect data and advance science from outside the ivory tower.
With the image of Watts as an online, amateur citizen (pseudo)scientist in mind, let’s take a look at his latest post. In it, he attacks a Nature Communications paper that uses citizen science and the internet to show that nuisance flooding, which are floods arising from sea level rise pushing high tides higher, may be more widespread than standard measures capture, based on spikes in Twitter activity of people complaining about nuisance flooding. Because there are only 132 tide gauges across 3,700 miles of the East and Gulf coasts, the study used tweets to fill in between the gauges, showing where flooding is happening without being recorded by official measurements.
Watts, unsurprisingly, either didn’t understand the research or deliberately mischaracterized it in his post calling it a “very, very bad idea.” While the study was designed to show the difference between what’s reported by tide gauges and what people are actually experiencing, Watts criticized it for ignoring extreme rainfall and the short window of time the study looked at. But he considers the “biggest flaw” in the study to be that they didn’t adjust for Twitter account inflation; since Twitter grew during the study period, he argues, the study’s metric of a spike in tweets is flawed. He concludes with the melodramatic statement that “I weep for science, and I especially weep for climate science.”
But again, the study isn’t about using tweets to prove climate change is causing more flooding, the hypothesis which Watts “debunks.” And even if it were, the authors did factor for population in comparing counties with a lot of Twitter users with those with few, finding that there is “no evidence that the number of Twitter users in a county systematically biases” the results.
What the study does do is show that flooding is happening, frequently, in places that the tidal gauges aren’t capturing. It’s a clever way to make use of a giant free dataset, using social media to turn Tweeters into citizen scientists who can refine and improve our understanding of the world. It highlights a place where our observation network isn’t capturing the whole picture, and uses people’s real-world experiences to improve on what we know.
It is everything that Anthony’s SurfaceStation project could have been, if it were a legitimate scientific endeavour instead of an exercise in denial.
So instead of weeping for science, Tony, maybe trying doing some for a change.
Top Climate and Clean Energy Stories: