A good forecast is only a good forecast if it’s communicated well and leads to a good decision. That’s the smart common thread linking most of the 23 recommendations of the National Weather Service’s assessment of its performance during Hurricane Sandy.
The assessment, which got off to a rocky, controversial start several months ago, was released this morning. It correctly points out that the National Weather Service accurately predicted Sandy’s path but is understandably mixed on how well it conveyed information about the hazards, particularly storm surge.
Despite the fact NWS produced reasonably accurate estimates for storm surge, the assessment team finds segments of the media, decisionmakers and the public failed to grasp the expanse of the area that would be inundated by the rapidly rising water. 147 people were killed from the effects of Sandy, 49 from drowning.
“…decision makers cited the need for high resolution graphical inundation mapping,” the assessment says. “They felt it would improve their ability to interpret NWS storm surge forecasts and their subsequent decision-making capabilities.”
Personally, I remember struggling – with great frustration – to find any good visuals of the storm surge threat when preparing blog posts as Sandy was bearing down on the Mid-Atlantic and Northeast coast.
The assessment recommends the NWS develop “explicit storm surge graphics and high-resolution mapping tools” that better illustrate the storm surge threat.
Thankfully, an effort to develop these tools is underway. An example product is below:
In addition to the lack of effective storm surge graphics, the assessment reports many decisionmakers and even some NWS personnel found storm surge terminology confusing due to different points of reference used with respect to the water level. As such, they were challenged to clearly communicate the threat. I know I was.
(The AP’s Seth Borenstein notes the word “confusing” appears in the Sandy report 88 times)
“NWS should present storm surge forecasts in a single, consistent datum and adopt a unified format and language for products describing impacts from storm surge, regardless of (tropical or extra-tropical) origin,” the assessment recommends.
It also argues for the implementation of a storm surge warning system for all types of coastal storms – another solid recommendation, which apparently is in the works.
While the assessment team finds storm surge communication as one of the areas of greatest need for improvement, it points to several other communication challenges that arose during Sandy that should be addressed.
The assessment notes that even when equipped with good forecast information, public responses can be “counterintuitive.” To help the NWS better understand how the public reacts to its forecasts and warnings, and, importantly design and deliver this information, the assessment emphasizes that the NWS engage social scientists – a point I’ve stressed as well.
“NWS needs to broaden and expand its social science and communications capacity by hiring at least one more social scientist/behavioral expert within NWS or by increasing contracts with outside experts,” it recommends. “This expanded capacity should be used to develop products, services, and communications tools (e.g., Internet, social media) to drive the appropriate public response.”
Several of its other recommendations focus on the delivery of forecast information. It calls for the NWS to:
* develop concise overview of storm impacts for media and public consumption
* establish backup systems for possible website outages
* ensure storm information is available on smartphones and tablets
* improve/enhance social media presence
All of these recommendations are sound.
Overall, while I thought the flow of the report was a bit disjointed, I found it to be a sound, common sense review of the strengths and shortcomings of NWS’s performance during Sandy.
Midway into the assessment, the assessment authors state: The best weather and water forecast can only save lives if it is communicated effectively to the at risk residents and the public officials who are charged with protecting them.
To me, this statement and its repeated emphasis on risk communication illustrates the author team “gets it”. Bridging the gap between forecast delivery and a decision should be among the National Weather Service’s top priorities.
(Note: I did not address the lack of issuance of hurricane watches and warnings during Sandy, as this matter was discussed in earlier posts, see below, and has been officially addressed by the National Weather Service.)