Beware of faulty, flaky Facebook weather forecasts

Anyone can play meteorologist on Facebook. Build a page with a credible-sounding name, post some weather maps, get some friends and family to “like” it, and you’re off and running.

In the last few years, the number of Facebook weather pages has exploded, many maintained by curators without clear meteorological credentials.  Unfortunately this has led to misleading or flat-out inaccurate information spreading virally, especially around high-interest weather events, like winter storms (or possible winter storms).

For example, the page “Mid-Atlantic Storm Watch” posted an update on Tuesday (October 15) headlined “*Possible winter storm October 24th??!!“, then adding the storm “might become something of historic proportions.”   This post has been shared over 8,000 times!

Yet its basis was a single computer model run simulation (from the GFS model) which, in reality, showed a storm forming north of the Mid-Atlantic and barely grazing New England. Other computer models showed no storm at all.

In any event, forecasts of storms beyond 5-8 days into the future have little to no reliability. If you see a computer model forecast of a week or more into the future, that should sound off an immediate warning bell to question the legitimacy of the source posting it!

As another example of shoddy weather information sharing, the Facebook page “PlanaWeather” is over-hyping cold in the Mid-Atlantic next week.

“Check out the GFS model showing temperatures in the 30s for highs around Oct 25,” its page says. It uses the map below as illustration of this.


Forecast minimum temperatures between 2 a.m. and 2 p.m. on Friday, October 25 by the GFS model (WeatherBell.com)

Except the map above is showing low temps, not highs. Specifically, it’s illustrating the minimum or lowest temperature in the 12-hour period between 2 a.m. and 2 p.m. that day (Oct. 25), which happens to be in the 30s.  For the Mid-Atlantic, the actual highs that day are mostly forecast to be in the upper 40s and 50s (below).


Forecast high temperatures between 2 a.m. and 2 p.m. October 25  (WeatherBell.com)

I don’t mean to disparage the efforts of well-intended weather enthusiasts.  Many of these pages are maintained by passionate amateur weather wonks as well as eager high school and college students, some of whom have professional forecasting aspirations.  Posting updates is good way for them to gain experience in forecasting and communicating weather information through social network interaction.

But, when there are potential high impact weather stories, it’s easy to get carried away and post sensational, hype-filled narrative, especially given the draw for recognition (“likes” and shares) as being “first” in getting the word out.

“You’d be surprised at how many people sound an alarm at the first sign of a possible long range storm then claim to have “called it” in the rare event it materializes,” writes ABC7 meteorologist Adam Caskey. “Yet, they take no responsibility for the vast majority of systems that don’t materialize. There’s no accountability.”

In order to confront the spread of bad weather updates on Facebook, ask these questions before buying into them or, worse, sharing them:

* Do you know/trust the source? If you’ve never heard of the source before, disregard it and/or check information sources you trust for a second opinion.

* Is the source legitimate? Try to figure out who is maintaining the page that information or update is coming from.  Is it a reputable news or weather organization like a TV station, newspaper, established private sector weather company (like AccuWeather or Weather Underground), or government agency (like the National Weather Service)?  Is there available information about the meteorological credentials of the person/people managing the page?

* Is the page regularly updated? Lack of regular updates suggest the page managers are not serious and committed to sustained engagement with readers

* Are the updates filled with typos and spelling errors? That might suggest the operation is less than professional (although I do know one excellent forecaster whose updates are sometimes filled with typos, so there are exceptions).

* Are the updates hype-filled, focusing on low likelihood far-off threats (e.g. relying on specific computer model forecasts a week or longer into the future)? As mentioned above, this is a screaming warning sign to be skeptical of the information.

* Do you have friends and family following the page? Facebook provides this information and if others are in your network trust the page, it’s a decent indicator you can too.

Hopefully, by asking these questions, you can be a more knowledgeable consumer of weather information on Facebook and help stop the propagation of bad information.

Jason is currently the Washington Post’s weather editor. A native Washingtonian, Jason has been a weather enthusiast since age 10.
Comments
Show Comments
Most Read Local
Next Story
Jason Samenow · October 17, 2013