Twitter's new limits may block your next tornado warning

A thunderstorm moves south of Lakin, Kan., on Thursday evening. When storms strike, or threaten to, the National Weather Service and broadcast meteorologists warn their followers using automated tweets, among other methods of dissemination. (Matthew Cappucci/The Washington Post)
  • Oops!
    Something went wrong.
    Please try again later.

Among the recent changes to Elon Musk-helmed Twitter is a policy meteorologists say could limit their ability to quickly alert the public to extreme weather dangers.

When tornadoes, flash floods and thunderstorms strike, or threaten to, the National Weather Service and broadcast meteorologists warn their followers using automated tweets, among other methods of dissemination. Twitter has long allowed users free access to systems that allows them to program such posts in response to external sources of information, be it the Weather Service's alerts, the movement of Musk's private jet or entries in the dictionary.

Subscribe to The Post Most newsletter for the most important and interesting stories from The Washington Post.

But as Musk evolves the social network toward paid features, Twitter warned users that it would limit the automated tweets, allowing 1,500 of them per month free - or, for a $100-a-month fee, as many as 50,000. Weather Service officials said that they expect a limit of 50 automated tweets during any 24-hour period - a threshold its accounts routinely surpass during the most dangerous weather events - and that Twitter officials told them no exceptions would be made.

The Weather Service began warning the public this month that the change means its automated tweets sharing severe weather watches, warnings and advisories "may not be posted" once the policy goes into effect, scheduled for April 29.

"Have multiple ways to receive weather information and alerts," regional Weather Service accounts urged Twitter users this month.

The change is heightening concern that, under Musk, Twitter is losing what has made it an essential source of news. The site has become a vital platform for communication between meteorologists, first responders and public officials during natural disasters and other emergencies.

As meteorologists grapple with how to use the social network going forward, Twitter has faced backlash for removing the New York Times' verified badge and labeling NPR as "government funded Media." NPR announced Wednesday that its Twitter accounts would no longer be active out of concern that the social network was "undermining our credibility."

Brad Panovich, a meteorologist at WCNC-TV in Charlotte, said automated weather alerts he shares on Twitter don't just go to his 114,000 followers, but also to his station's website and app. The loss of those tweets could further damage Twitter's utility.

"It just becomes an unreliable platform in the long run," Panovich said.

Twitter did not respond to questions about the impact the changes could have on public safety. It responded to an email from The Washington Post with a single character, as it has with all recent media requests: a poop emoji.

Weather Service officials said in a statement that they have used the automated tweets since 2014. If they go away, manually preparing such tweets "would take minutes for forecasters" and, for every single warning, "seconds could make the difference between life and death," the statement said.

Meteorologists acknowledged that a relatively small slice of the populace uses Twitter, and the fraction who come to the site for local weather warnings is smaller still. The company reported 237.8 million active daily users as of June 30, but more recent data is not publicly available. Musk took over Twitter in October.

But the automated weather tweets nonetheless alert many people indirectly. Twitter is one of many important tools for meteorologists to amplify messages about extreme weather risks, said James Spann, chief meteorologist for ABC 33/40 in Birmingham, Ala. During a recent tornado outbreak there, "the biggest issue we had is people not getting the warning," Spann said.

While Spann doesn't use automated tweets on his own account, with nearly 500,000 followers, he directs followers to an automated account run by ABC 33/40. Any channel to reach people in harm's way can help, he said.

"It's an important tool, and I don't want it to go away," Spann said. "I want it to be robust, and I want it to be reliable."

It's not yet clear how meteorologists will adapt to the new limits on automated tweets. Some questioned whether the Weather Service's 122 forecast offices across the country have the personnel to manually tweet every severe weather watch and warning.

During a severe weather and tornado outbreak April 5-6, for example, the Weather Service's Jackson, Miss., forecast office sent 64 automated tweets in rapid succession. At the St. Louis forecast office, automated tweets - both in English and Spanish - also numbered more than 60 on April 5.

Along with each warning comes information on potential wind speeds and hail size, as well as about what's in harm's way. A tornado warning issued in Mississippi via Twitter this month stated 16,000 people, seven schools and one hospital were threatened.

"If you're going to do it manually, you've got to have somebody working on that account constantly," Spann said. "Time is of the essence. Five or 10 minutes makes a huge difference."

Daryl Herzmann, a systems analyst at the University of Iowa, said the burden of manually sharing so many weather alerts is the reason he has developed a stable of accounts sharing them via automated tweets. Herzmann operates the Iowa Environmental Mesonet, a repository for environmental weather data that is widely used by meteorologists nationwide and supports many automated Twitter feeds.

He said Twitter has not made clear how the new tweet limits will work, including whether the company will apply them to each individual Weather Service account or to a single application that runs all of the Weather Service's automated tweets. Given the frequency of weather warnings at different times a year, the agency could face significant cost even at Twitter's proposed $100-a-month tier.

"It's not clear from reading what sparse documentation there is if that's even enough for them," Herzmann said.

- - -

The Washington Post's Jeremy Merrill and Rachel Lerman contributed to this report.

          

Related Content

Texas is home. The Briggle family and their trans child are fighting for it.

How progressive Denmark became the face of the anti-migration left

Did Roy Lichtenstein create art or copy it? A new film stokes the controversy