lessons learned in communicating weather and climate uncertainty - - PowerPoint PPT Presentation

Challenges and lessons learned in communicating weather and climate uncertainty

Jason Samenow, Capital Weather Gang – The Washington Post

April 15, 2016

• Washington Post’s weather team
• One full-time managing editor, one deputy editor
• 20 outside freelance contributors
• Content spans print, web, mobile devices,

includes radio hits & video

• Broad mix of content

– Local forecasts and commentary – National and international weather stories – Weather photography, history, astronomy, space weather, weather policy, climate change & more

• Emphasis

– Helping people make weather-based decisions – Engagement http://www.washingtonpost.com/capitalweathergang

About the Capital Weather Gang

@capitalweather

• n Twitter

Lesson learned: Qualities of successful science communication

• Credibility
• Legitimacy
• Salience
• Good story telling

Challenge: Capturing reader attention

• Headline and lead in to any discussion needs to

be strongest, most compelling

– Newspaper readers read 56% of the headlines, but only 13% of the stories are at least half-read. (Joe Romm, Language Intelligence)

• Stories w/ boring headlines become irrelevant;

nuance doesn’t sell and not enough characters for it

• Perceived click bait, sensationalism can backfire
• Objective: Find the right balance

Some lessons learned:

Effective weather communication

• Confidence levels in forecasts
• Scenarios for complex, high impact events
• Compelling visuals
• Specifics on timing, location
• Accountability
• Learn by doing, seek feedback

• High: Bank on it.
• Medium-High: Overall forecast is sound,

but minor variations are possible.

• Medium: We think we’re on the right track,

but the forecast details are still taking shape.

• Low-Medium: This is our best guess, but

don’t hold us to it.

• Low: Crapshoot.

Confidence levels

Snow case

• Learning from a bad forecast: The

Snowquester, March 6, 2013

Snowquester (March 6, 2013):

The forecast we issued Reality

A bad forecast for a high impact storm can be devastating for consumer trust

“Never has there been an industry which spends so much time churning out so much bull” “ I can tell you that weather predictions have always been a standing joke. It's all a scam to increase ratings on the weather shows, and to reward advertisers” “As far as I am concerned, all of our local meteorologist should be fired for cause. Why? Because they are wrong more often than right and costs us big money.”

From: Snowquester: when forecast information fails http://www.washingtonpost.com/blogs/capital-weather-gang/post/snowquester-when-forecast-information- fails/2013/03/07/5d0d77ae-873b-11e2-9d71-f0feafdd1394_blog.html

But readers also told us they’d welcome more honesty, uncertainty, probabilistic information

Brad Panovich, WCNC

The best forecast for Snowquester was one we could not issue with a straight face, and

• ne most Washingtonians would

have ridiculed: Rain, sleet, and/or snow likely - heavy at times - with snow accumulations of 0-14 inches.

“Yes. Why not? The rain/snow line will be wobbling all over so that's all you can say. If it's the best, it's the best. Say that, and then give your probabilities. ” “What does ridicule have to do with it? I support the CWG, but this business about how meteorologists have to depart from truth and accuracy because they're worried about their popularity ratings I don't understand. Are they scientists or entertainers?”

From: Snowquester: when forecast information fails http://www.washingtonpost.com/blogs/cap ital-weather-gang/post/snowquester- when-forecast-information- fails/2013/03/07/5d0d77ae-873b-11e2- 9d71-f0feafdd1394_blog.html

The winter after Snowquester

Sandy

A successful scenario approach

8 days out: a heads-up

5-7 days out Provide general level sense of what could happen

“Although a historic storm is a possibility, the storm could deliver just a glancing blow or even miss the East Coast entirely. And for residents

• f the mid-Atlantic (including Washington, D.C.

and points further south), a direct hit is not particularly likely although it cannot be ruled

• ut.

“Residents of the Northeast, perhaps, should be most concerned. The European model - which did the best job simulating hurricane Isaac’s track in late August - delivers a devastating blow from central New Jersey to southern New England (including New York City)”

(http://www.washingtonpost.com/blogs/capital-weather- gang/post/tropical-storm-sandy-scenarios-northeast-smash-or-out-to- sea-most-likely/2012/10/23/29e6f7ce-1d41-11e2-9cd5- b55c38388962_blog.html)

4 days out: Communicate 4 scenarios

SCENARIO 1 - NJ TO Long Island landfall: Indirect hit, major impacts [for DC] (45 percent chance) SCENARIO 2 - Mid-Atlantic landfall: Worst case – direct hit, severe impacts (30 percent chance) SCENARIO 3 – New England landfall: Glancing blow, minor impacts (20 percent chance) SCENARIO 4 – Out to sea: FEW IMPACTS (5 percent chance)

(http://www.washingtonpost.com/blogs/capital-weather- gang/post/hurricane-sandy-scenarios-for-washington- dc/2012/10/25/0ebdb0f2-1ec5-11e2-9cd5-b55c38388962_blog.html)

3 days out Communicate 3 scenarios

SCENARIO 1 - Landfall from Virginia Beach to the Delmarva: Worst case (1 in 3 chance) SCENARIO 2 – Landfall over south Jersey: Almost direct hit (1 in 3 chance) SCENARIO 3 - Landfall between northern New Jersey and southern New England: Indirect hit (1 in 3 chance)

(http://www.washingtonpost.com/blogs/capital- weather-gang/post/washington-dc-will-not- escape-hurricane-sandy-latest-storm- scenarios/2012/10/26/19a6c5b8-1f8c-11e2- 9cd5-b55c38388962_blog.html)

2 days out Communicate 2 scenarios

SCENARIO 1- Landfall from northern Delmarva to northern New Jersey: Near direct hit (70 percent chance, very bad case) SCENARIO 2 - Landfall from southern to central Delmarva: Direct hit (30 percent chance, worst case)

http://www.washingtonpost.com/blogs/capital-weather- gang/post/washington-dc-braces-for-hurricane-sandy-the- bad-and-the-ugly-storm-scenarios/2012/10/27/f7f9abde-2052- 11e2-9cd5-b55c38388962_blog.html

The day before Detailed deterministic forecast

http://www.washingtonpost.com/blogs/capital-weather-gang/post/hurricane-sandy-and-washington-dc-

Challenge: The armchair meteorologist problem

• Social media gives everyone a voice, irrespective of

credentials

• High school kids, weather amateurs, (pros too) can

misuse, misinterpret model and push out unreliable info that spreads virally

What should digital and broadcast meteorologists do?

Confront? Ignore? Engage constructively w/ bad actors? “pointless to expose and shame…It’s a never- ending

and unwinnable game of whack-a-mole. ” “Focus on educating audience on limitations of weather forecast”

(http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2014/08/20/hurricane-hype-is-here-to-stay-forecasters-must-adapt/)

National Hurricane Center speaks out…

“Now that we have entered the heart of the hurricane season, there is an increase in the Internet hype around disturbances that NHC is monitoring. Given the long lead times involved, the wide range of possible outcomes, and the historically poor and erratic performance of guidance models with weak disturbances, there is no reliable science to forecast potential impacts to specific locations that would be more than a week away. . . . “…The bottom line really is: be alert, be prepared, but also be wary of long- range projections that go beyond what the science can offer. And make the NOAA National Hurricane Center www.hurricanes.gov your calm, clear, and trusted source for official forecast and warning information on tropical cyclones.”

https://www.facebook.com/NWSNHC/photos/a.126275484104607.2253 0.112957945436361/734769166588566/?type=1

Challenge: Some for-profit companies providing unreliable forecasts

• Weather forecast “skill” is no more than 7-

10 days into the future AMS: “Presently, forecasts of daily or specific weather conditions do not exhibit useful skill beyond eight days, meaning that their accuracy is low.”

• Yet some companies provide such

forecasts 90 (or more) days out with no uncertainty information

• Misleading? Ethical issues?

Lessons learned: Climate change communication

• Don’t cherry pick and hype – discuss full range of results and

credible views; nuance builds credibility

• Avoid labels “deniers”, “alarmists”, “warmists”…better

“convinced”, “unconvinced” (doubters or skeptics ok, depending on context)

• Consider alternative views, but avoid “false balance” (equal

time)

• Be respectful of different voices in the crowd. Don’t demonize

people, perspectives.

Thank you!

Contact Jason Samenow, Weather Editor - Washington Post samenowj@washpost.com - 202.334.9937 Blog: http://www.washingtonpost.com/capitalweathergang Twitter: @capitalweather Facebook: http://www.facebook.com/capitalweather