The US National Weather Service considers changing warning language.

Developments in the US National Weather Service might influence the way weather warnings are issued in Australia. In any event discussion about the use of language can inform Australia’s hazard managers as they seek to improve warnings.

The US National Weather service is considering changing it’s warnings, in line with its aim to ensure more recognition for impact to be described when issuing forecasts.

This approach was taken by Australia’s bushfire and emergency experts in creating the National Bushfire Warning Framework. It’s interesting to consider how it might be included in weather warnings.

The NWS is gathering opinions from the public and weather product users to help guide the process, and in a promising move their approach highlights it’s relationship with the public – and its belief that people are capable of receiving and processing large amounts of data to make their own decisions.

It’s trialling a new system side by side with the present system, despite the obvious concern about confusion.

In its own words, it says it wants to have “a conversation” with users:

“In support of our Weather-Ready Nation initiative, NWS wants to start a conversation on how we might simplify and clarify our products. For this demonstration, we are proposing an alternative way to express headlines within our hazard messages, with winter hazard messages as a focus. If you have other ideas for simplifying and clarifying these messages, we want to hear them.”

The details are here:

The language used by the NWS in the US is very close to the language used in Australia. The principal enhancements are to headlines:

For all current, official winter weather hazard messages that lead with the phrase:

“THE NWS HAS ISSUED A _____ WATCH” it will say: “THE NWS FORECASTS THE POTENTIAL FOR ____” with the hazard type (snow, ice, wind, etc.), level of certainty, timing and expected impact(s) clearly stated in the blank space.

For all official winter weather hazard messages which lead with the phrase: “THE NWS HAS ISSUED A(N) ______ ADVISORY” it will say instead: “THE NWS ADVISES CAUTION FOR ____”  and for messages which lead with the phrase: “THE NWS HAS ISSUED A(N)______ WARNING” it will change to: “THE NWS HAS ISSUED A WARNING FOR A DANGEROUS______”

“We retain the term “Warning” for the demonstration because of its direct connection to protection of life and property, but this alternative approach would eliminate the individual hazard products within the Warning category.”

The NWS  will receive feedback early this year, and is proposing that it will revisit its approach to water and wind warnings if this proposal receives public endorsement.



US National Weather Service Partnerships and warnings

The National Weather Service in the US has a symbiotic working relationship with the broadcast media, which far surpasses anything in Australia.

Although The ABC has wonderful and often personal relationships with the Bureau of Meteorology, the US NWS systemises the process. (Mind you the idea that the ABC conducts hourly interviews with BoM staff in most of Australia was something that really excited the NWS people I spoke too – but that’s an aside).

The central pillar behind the success of the NWS relationship is the Integrated Warning Team, but there are other very important elements.

Warnings officers are stationed in all regional forecasting centres who have an emergency and community education role

Direct access between warnings officers, forecasters and broadcasters. A very important component of this approach is NWSchat – the “real-time interactive communications system” which would be well worth considering in Australia for all emergency agencies and the Bureau of Meteorology.

It allows direct communications with experts “off line” seeking more information, more urgency, raising questions, and providing feedback.

To quote the NWSchat home page:

“NWS partners can use NWSChat as an efficient means of seeking clarifications and enhancements to the communication stream originating from the NWS during a fast-paced significant weather or hydrologic event. NWSChat is an Instant Messaging program used by NWS operational personnel to share critical warning decision expertise and other types of significant weather information essential to the NWS’s mission of saving lives and property.

Mike Hudson, the warning specialist at Kansas City, Missouri, says it helps to ensure that the messages being broadcast over multiple platforms is consistent: “If people receive more than one message at a time it can lead to paralysis.

“Inconsistent messaging leads people to “shop” for information, taking up valuable time to see if other radio and TV stations are carrying the same message.”

Real time chatroom content between the duty warnings forecaster and all media (or between emergency agency duty officers and the warnings media ) would enhance understanding at critical times.

NWSchat is linked directly to the local warnings officers. The use of Instant Messaging (IM) and chatrooms have proved to be valuable for this type of communication internally at The ABC and in many businesses, but to open them to various partners, like the NWS has done, is a bold step, which reflects the relationship between broadcasters and the NWS. The technical details are online and the following information is all provided by their site,

NWSchat is maintained by the National Weather Service and is situated behind a firewall. NWSChat is comprised of a pair of servers configured in a resilient primary/backup configuration, and receives auto-updates for all operating system patches and bug-fixes. The systems are scanned quarterly to identify and correct IT security vulnerabilities as required by NOAA IT Computer Security policy.

Individual user accounts are required for NWSChat; shared or group accounts are disallowed. A standardized account naming syntax is also enforced for manageability.

To register with NWSChat, users must submit an online form. Once submitted, the selected primary office receives an email of the request, and will approve or deny authorization for each user. Once approved, the requesting user is notified via email and then must complete online training for NWSChat.

Most multi-user chatrooms on NWSChat are open to NWS partners once they are authorized by the NWS. However, certain rooms are restricted for “members-only” access. This is necessary to secure information in specific chatrooms intended for certain partners only. For example, some information may be required by emergency managers that is not appropriate for media partners due to the sensitivity of and timeliness of emergency operations.

As a result, a members-only chatroom would be provided limiting access to NWS and authorized emergency managers exclusively, for a given location. In order to participate in NWSChat, you must meet at least one of the following standards:

• Be a member of the emergency management (EM) community: Members of the EM community includes public safety officials who serve as employees or contract agents of a government agency at the federal, state, local, or tribal level and are charged with protecting the public from hazards that are influenced by weather or weather-related events. Other members of this community include: safety and emergency personnel, from universities or other large entities with large populations, whose roles are functionally equivalent to the public safety officials described above, and Amateur Radio Emergency Services.

• Be a government partner of a NWS office: This includes Government partners who have missions that require close coordination with the NWS. Government partners include (but are not limited to) the FAA, and water and land management officials.

• Be a member of the electronic media: Members of the electronic media are parties, and contract agents of parties who:  Have a need to actively participate in discussions with NWS Forecast Offices on imminent weather or other hazards, and Operate systems that routinely and rapidly relay weather and water watches, advisories, warnings and forecast information to a significant part of the population served by an NWS office; via electronic information distribution such as radio, television, internet, cellular, and other wireless means. Note: Individuals, companies, or other entities involved in ‘chasing’ weather events and posting or streaming video or pictures of the event, but do not otherwise have a need to communicate with NWS do not meet the qualifications for this Service.


editors note: Typograp[hical erroir updated Dece,mber 3, 2012: Third para should read:

It allows direct communications with experts “off line” seeking more information, more urgency, raising questions, and providing feedback.


Broadcast community partnerships in the US

The media and broadcast in the US are highly valued partners in disseminating information and warnings about all hazards in a way that isn’t apparent in Australia.

There are three elements on which the partnerships are based which serve the US well – legislation, obligation and co-operation. All three are, to some extent, missing from the landscape in Australia.


The US set up CONtrol of ELectromagnetic RADiation, or CONELRAD (1951 – 1963), and the Emergency Broadcast System, or EBS (1963 – 1997) to deal with the threat posed by the Cold War. Legislation was passed which compelled radio stations to participate in the program at their own cost, and to ensure they could carry all Presidential level warnings and addresses.

“It is a condition of their licence agreement now that TV and radio broadcast companies serve the public interest,” says Manny Centeno, the Program Manager of the Integrated Public Alert and Warnings system with the Federal Emergency Management Agency. “They are using public airwaves, the public owns them.”

Jim Bremer, the program director at KQMS Newstalk Radio backs up the observation, ”We’re licensed to serve the public.”

Jim Bremer, EP at KQMS Reading

In 1997, the EBS modernised into the Emergency Alert System, or EAS, and in 2006 President Bush issued Executive Order 13047, which required that an Integrated Public Alert and Warning System be built. The legislation was changed to enable the system to be used by state and county agencies to alert their community to potential hazards. In all counties now the system is used frequently to alert the community to weather hazards, and “AMBER alerts“ – notifications of recent child abductions.

 The Commercial Mobile Alert System is one of several systems within IPAWS. The Commercial Mobile Service Providers’ decision to broadcast alerts and warnings is voluntary. “We have 400 telecommunications providers in the US, and within a few months of the process being set up, we already have over 100 providers who’ve opted in,” said Manny Centeno.”

The cost to send these alerts: “Nothing at all, they do it voluntarily.”

Phone companies will receive alerts via the IPAWS Open Platform for Emergency Networks, or IPAWS-OPEN, and then broadcast them to the public in areas which have been geo-targeted by emergency alerting authorities.

To ensure the process is smooth, the Federal Communications Commission and FEMA set standards and protocols, developed a common operating system for emergency agencies and provide training. The FCC requires a monthly test of the system.

In addition, the National Weather Service was provided with funding to set up its own radio network – the National Weather Radio. With more than 1100 transmitters, it’s a robust and comprehensive system, which has come to be relied upon by a sizeable proportion of the US community.


Legislation alone cannot compel organisations to ensure that all aspects of their content meet the requirements of Federal Government or the needs of their communities. Only the “Presidential Address” is obligatory, all other warning content is carried voluntarily. But the competitive nature of the private sector has ensured that all participants feel obliged to participate. The telecommunications companies created demand for phone handsets that are warning compatible.

The radio and TV stations know that in an emergency their competitors will be broadcasting warnings, and they can’t afford to ignore the possible impact on their audience reach.

TV stations play the most important role, with weather presenters being the go-to people when an emergency begins. TV watching seems to be universal. Homes, clubs, bars, restaurants, schools, public gathering places like airports all seem to have TV networks switched on 24/7.

Jim Bremer from KQMS estimates that his company spends $20,000 on the hardware to receive EAS content, and there are numerous stations in his comparatively small company.


“Tremendous private sector involvement is the key to the success of CMAS and EAS,” says Wade Witmer, FEMA’s Deputy Director of the Integrated Public Alert and Warning System.

There is no doubt that emergency managers are grateful for the support they are receiving from broadcasters.

But to ensure the process is integrated, there has to be high level and frequent co-operation. FEMA and the Federal Communication Commission have a monthly meeting with national broadcasters and telecommunications companies, to ensure the systems remain technically robust and to create a standard for sharing of information between emergency agencies and the disseminators.

“You have to engage the broadcasters. A lot of them don’t trust each other” says Manny Centeno, in a phrase that will resonate with broadcasters and emergency agencies everywhere. “But some broadcasters do call themselves first responders, which shows they are fully engaged.”

In addition to the nationwide meetings, each state has a meeting of State broadcasters and their emergency management agencies each month.

The National Weather Service also reflects the expectation that there will be leadership and co-operation. Warnings’ broadcasters meet their local and regional broadcasters frequently to extend understanding. Additionally, the NWS set up NWSchat, to talk directly to broadcasters to ensure their message is well understood and useful.


Howard Price, the emergency manager at the American Broadcast Corporation, which runs radio and TV stations and networks across the US, sums up the three elements nicely: “It’s about public interest, convenience and necessity.” The public interest is obvious but why convenience? “Because FEMA, NWS and FCC created the platforms which we plug into to receive all the warnings and content.” Necessity? “That’s just competition to retain listenership and audience.”

The partnership model was started by the private sector. “Richard Rudman, the Director of Engineering at CBS started the “Partnership for Public Warning” says Howard Price. “We also started the Media Security and Advisory Council after 9/11. During Hurricane Katrina, Entercom, which owns one of their local stations never went off air. They set the gold standard for emergency broadcasting.”

Howard Price is proud of the role the broadcasters play in issuing warnings and information in the US. “EAS is a good system, but it has to be defended, and the local stations have to practice using it. The total cost for ABC to implement the mandatory system on its own network runs to five figures, and no-one here has ever suggested the Government should pay.”

Nevertheless the companies retain editorial control. “We put warnings on every station. It’s good for the warnings to be on during kids watching time as their parents are often watching with them. But the ABC never puts an unverified warning to air. No talk station in their right mind will do that.”

Charles W. McCobb

Program Manager

IPAWS Program Office /NCP


Joplin tornado – Warnings

 The Joplin tornado neatly encapsulates all the problems inherent with multiple, simultaneous, complex emergencies. The US National Weather Service (NWS) was widely praised for its work. The relationship between the warning providers and the disseminators is crucial.  

 It is fair to say there is much soul searching at the NWS in the US when people die in weather related disasters. It’s the same in Australia, and no doubt everywhere else. Weather forecasters are scientists but they understand their work, at its best, will save lives.

The problems confronting the forecasters at Joplin were the same as those wherever  multiple simultaneous complex weather events occur. (I call these MSC events) For most of us that’s thunderstorms and hail forecasts, but the same problems were experienced by fire fighters during the Black Saturday Bushfires in Victoria in 2009 and to a lesser extent in the flash flooding of Grantham in Qld in 2011.

In the case of the Joplin tornado the public had to contend with the following bewildering array of events – and many in the disaster community will have an understanding of exactly this type of scenario:

– No big tornado for a generation, even though the town is in :”Tornado alley.”

– Thunderstorm forecasts consistently for two days

– Thunderstorm and hail forecasts all day

– Tornado watches issued a couple of hours before he main event

– Multiple tornadoes

– Two blasts on the town tornado warning sirens. 

These events create massive uncertainty, and warnings, unless very carefully compiled and placed in context and with a sense of priority, will lead to confusion and the public quite naturally will hesitate to act.

Remains of school, 16 months after the tornado Pic Ianm

This is how the NWS reported on warnings from the Joplin tornado – taken from its review. (Items in barckets are mine)

“A series of complex meteorological events and interactions took place during the afternoon hours of May 22 that eventually resulted in the devastating EF-5 tornado. A Tornado Watch  was issued at 1.30 pm for all of southwest Missouri.

A routine Area Forecast Discussion (a  type of weather service “heads up”)  was issued at 2.37 pm as well as  at 3.47 pm. Forecaster focus remained on very large hail as the main severe weather threat, but isolated tornadoes were also deemed a possibility due to the very unstable air mass in place and sufficient low level wind structure. The first thunderstorms of the day developed between 2 pm and 3 pm  over southeast

As severe storms moved east, forecasters became increasingly concerned about their tornado potential and issued the first Tornado Warnings of the day at 4.25 pm and 4.51 pm for counties west of Joplin. At 4.33 pm forecasters briefed the Jasper County (which encompasses Joplin) Emergency Manager on the severe storms to the west.

A tornado Warning was issued at 5.09 pm for western Jasper County, including the northeast part of the city of Joplin but was for a different storm than the one that eventually hit the city.

This alert was followed by a new Tornado Warning  at 5.17 pm CDT for the next storm to the south for southwestern Jasper County and portions of neighboring counties which included all of Joplin. (Another coordination call was made to the Jasper County Emergency Manager at 5.25 pm  to update him on the Tornado Warning and latest information concerning the storm.)

At this point, the severe thunderstorm west of Joplin had become the dominant thunderstorm in the region and was poised to produce a violent tornado. Based on storm surveys and radar imagery, it was estimated that initial tornado touchdown occurred just west of Joplin at 5.34 pm, moved into western portions of Joplin around 5.36 pm. The town had 17 minutes of lead time for touchdown and 19 minutes lead time before the tornado entered Joplin.

The first indication of a confirmed tornado was issued via another Severe Weather Statement at 5.39 pm that stated,

“At 534 pm CDT…trained weather spotters reported a tornado near Galena” and that “This storm is moving into the city of Joplin.” The tornado eventually dissipated around 6.12 pm.

Unfortunately, the tornado developed rapidly on the southwestern outskirts of a densely populated area and had moved through much of the city before the size and violence of the tornado was apparent to warning forecasters. They could not not issue a Severe Weather Statement with a “Tornado Emergency.”

Amongst the general public, the majority of residents had little idea there was a threat of severe weather prior to Sunday, May 22. About half of those interviewed, reported learning of the possibility of severe weather in the hours leading up to the tornado. Just less than half reported their first indication of a severe weather threat was in the moments just prior to the tornado.

 According to the (Joplin Emergency Manager, IM) the first 3-minute siren activation, at 5.11 pm resulted primarily from funnel cloud reports to the west of Joplin in southeastern Kansas (and was not based on NSW information! IM)

Residents heard the initial siren activation and then the warning details were provided by the emergency telephone system (called Reverse 911 – IM), and assumed the activation was for the area to the north. In one example, a man was clearly confused by the string of warning information he received and processed from various sources.

1. Heard first sirens at 511 pm CDT (estimated 30-35 minutes before tornado hit).

2. Went to the TV and heard NWR warning from TV override that indicated tornado near airport drive 7 miles north of his location.

3. Went on porch with family and had a cigar. Looked like a regular thunderstorm.

4. Heard second sirens (estimated 27 minutes later).

5. Thought something wasn‘t right so went inside and turned local TV stations on.

6. Saw on TV several colored counties for tornado warnings, but regular programming was still on and thought the threat was still to the north.

7. Heard his wife yell “basement,” Grabbed the cat and told son to put his shoes on.

8. Tornado hit as they reached the top of the basement stairs, destroying their home.

The Joplin (Missouri, US, May 2011) tornado

The tornado which hit this town in 2011 brings with it many insights for emergency managers trying to ensure that warnings change people’s behaviour when they are confronted with a disaster with very little or no warning.

Joplin’s a town of about 50,000 people on the south eastern corner of Missouri, US. It’s in “tornado alley” and many people in the region have experienced tornadoes before.

At 5.30 pm on  Sunday May 22, 2011, a tornado about a kilometre wide, travelled across the residential areas of the town, just missing the CBD, but destroying thousands of homes, a nine storey hospital, a Walmart, big sports stores, schools and municipal buildings. The damage path was 9 km long.

158 people were killed and more than one thousand injured. According to a report by the US National Weather Service this was the “single deadliest tornado in U.S. history since modern record-keeping began in 1950.”  It traveled 35 km on the ground and the maximum wind speed was estimated at 321 km/h. Compare this to reports from the BoM on Tropical Cyclone Yasi of March 2011, which hit North Queensland with maximum wind gusts of 285km/h.

Joplin High School, Franklin Technology Center and Irving Elementary School were destroyed. The roofs were blown off East Middle School and Cecil Floyd Elementary.

St. John’s Regional Medical Center took a direct hit from the tornado and was eventually knocked down. The local newspaper said the town’s other major hospital was quickly “overwhelmed” by injured people. “People were being delivered in pickup trucks, lying on doors and pieces of plywood that served as makeshift stretchers. Also overwhelmed was an emergency medical center that was set up at Memorial Hall.”

A Joplin street in September 2012, showing what the suburbs would have been like before the tornado struck. Pic: Ianm


Damaged homes and buildings after the tornado: Pic NOAA

As with many catastrophic emergencies, the death toll is not a good measure of how the community responded to this event.

The NWS report says: “Some people took shelter in appropriate locations, but did not survive. Others mistakenly drove their vehicles into the tornado path, but somehow lived to tell of it.”

The  NWS comprehensive survey receommended many changes to the warning system: the rest of this post is from the report. 

The team determined that a number of factors contributed to the high death toll. Through interviews with more than 100 Joplin residents, the team found that societal response to warnings is highly complex and involves a number of factors, such as risk perception, overall credibility of warnings and warning communications.

The report includes a number of key recommendations:

  • Improve warning communications to convey a sense of urgency for extreme events. This will compel people to take immediate life-saving action;
  • Collaborate with partners who communicate weather warnings to develop GPS-based warning communications, including the use of text messaging, smart phone apps, mobile communications technologies, in addition to upgrades to the Emergency Alert System and NOAA Weather Radio;
  • Collaborate more throughout the weather enterprise to ensure that weather warning messages sent via television, radio, NOAA Weather Radio, local warning systems such as sirens – are consistent to reduce confusion and stress the seriousness of the threat; and
  • Continue to increase community preparedness