The Commercial Mobile Alert System uses modern technology to deliver emergency alerts to targeted audiences.
Methods for the government to alert citizens in case of emergency once consisted of warning sirens and messages broadcast over the radio and TV. But the emergence of mobile technology brought with it the promise of delivering more information to targeted audiences.
“The future of public notification is through cell broadcasting,” said Joseph Bruno, commissioner of the New York City Office of Emergency Management.
In recent years, emergency managers have been able to alert people through phone calls to land lines in particular locations. But the alerting system hasn’t quite kept up with mobile technology — and now, some people are giving up their land lines altogether.
The latest development in the government’s efforts to alert citizens about emergencies is the Commercial Mobile Alert System, or CMAS. It is meant to bring emergency alerts up-to-date with the latest technology. And although the system still faces challenges, officials say it’s off to a promising start.
“As technology changed and our mobility changed tremendously, there have been new challenges,” said Lorin Bristow, managing partner of Galain Solutions, an alerts and warning consultancy based in Franklin, Tenn.
CMAS is part of the Integrated Public Alert and Warning System, which is an office within FEMA that was created by a 2006 executive order to be sure the president could communicate with the American public during an emergency. “It really sought to take what has been the Emergency Alert System to another level,” said Bristow, who also blogs for Emergency Management magazine.
The system addresses several problems: the growth of mobile devices, the fact that people are abandoning land lines and the difficulty of getting people to subscribe to mobile alerts, Bristow said.
“Our real emphasis is on being able to reach as many people as we can by multiple means,” said Damon Penn, assistant administrator for national continuity programs at FEMA.
CMAS can send three types of alerts: presidential alerts, Amber Alerts and notifications of an imminent life-threatening situation. The alerts show up as text messages, and CMAS will be known by the public as Wireless Emergency Alerts (WEA). Individuals can check their phones to see if it’s WEA-enabled. Users can opt out of receiving all but the presidential alerts.
The system uses the same technology that cellphones use to connect to cellphone towers. It does not rely on the regular text message network, so sending the messages won’t overload regular communications. “This uses connections between your phone and the cell tower that are already going on,” Bristow said.
Alerts can be sent by the president, state or local emergency managers, or the National Weather Service. Local emergency managers need approval to get the authority to send the messages. Approval comes from the state emergency agency and requires training plus the technological capability to connect to the FEMA system to actually send the message, Bristow said.
When an emergency management office has an alert to send about a local imminent threat to life and safety, it would create a message and send it to the aggregator, the federal gatekeeper for messages created by FEMA and the FCC, Bruno said.
According to Penn, messages need to meet three requirements to be sent: They must be urgent, severe and certain.
For a local government to be certified requires numerous steps, including completing training on how to use the system. The state emergency manager decides if a local authority has qualified.
FEMA has a website for local agencies to help with the process of becoming certified: www.fema.gov/emergency/ipaws/alerting_authorities.shtm. The training is online and done through FEMA, Penn said.
One of the issues is that only newer cellphones can receive CMAS messages. It’s expected that it will take 12 to 18 months for a majority of the population to have phones that will receive the alerts, Bristow said. And even then, there will still be people with older phones who will not receive the messages.
The official rollout for CMAS was in April, so it hasn’t yet been tested during a major emergency. In New York City, a live pilot started on Jan. 1, after a test in December.
New York City is on the leading edge of implementing the system. “It gives them a capability that they didn’t have before,” Penn said.
Four main carriers — Verizon, AT&T, Sprint and T-Mobile — were engaged in the test, Bruno said.
“We were most interested to see if the system would work on a timely basis: Would they get the message? Would it go through? How long did it take?” Bruno said. New York City officials also wanted to know how granular the carriers could be: Could they send a message to just one tower or just one area outlined on a map?
Overall, the test went well, Bruno said. The city did discover that in some cases, more work was needed to make the messages as granular as it wanted, and this work has
An earlier test took place in San Diego in late 2010. County employees and volunteers fanned out around the county with dozens of phones, said Stephen Rea, assistant director of the Office of Emergency Services in San Diego County. The test, though more limited than New York’s, was basically successful, he said.
CMAS promises to extend the reach of emergency messaging beyond the people who have currently signed up for notifications or who can be reached via a notification call to a land line.
“Since it works on the cell tower that you’re affiliated with, it gives us regional geo-targeting,” Penn said.
The system allows you to “get alerts from where you are, not where you live,” said Leslie Luke, group program manager for the San Diego County Office of Emergency Services. This means, for example, that San Diego emergency officials will be able to reach tourists and business travelers who are in the area temporarily.
The system won’t replace other notification systems, but will complement them.
“We want to make sure we can reach the people without a land line,” Rea said.
Because CMAS is automatic on new phones, the system’s effectiveness doesn’t depend on people signing up.
“Registration systems are great, but they require people to take an action, and a lot of people don’t take those actions,” Bruno said.
One of the big challenges in using the system is that messages are limited to 90 characters — and unlike other short messages such as those on Twitter, they cannot include a link to a website for more information.
Although researchers are still determining what types of messages will be most effective in this medium, Jon Eisenberg, director of the Computer Science and Telecommunications Board of the National Academies, said that in general, effective warnings are accurate, use authoritative language, are consistent, explain what actions should be taken and when, specify where the hazard is, and explain why action is necessary and what the consequences could be of not acting.
While it’s not likely that a 90-character CMAS message will be able to fit all of that information, many of the items may be applicable.
Rea said that during San Diego’s test, the focus was on making sure the messages explained the type of disaster, the area affected and what action to take. Including the affected area is important because CMAS messages will reach a wider audience than, for example, alerts to land lines, which can be targeted to specific homes.
For example, one of the messages said: “TEST: Toxic air quality near Mission Bay. Remain indoors. Turn off AC. Monitor local news.”
“We were surprised — it seemed to work pretty well,” Rea said of the 90-character limit. “By giving them this short, 90-character message, they would go and get the rest of the information from other sources.”
The reason for the prohibition on URLs is a concern that recipients would all immediately check the link, overloading the data network.
It’s not yet clear whether this strategy will work, Eisenberg said. Research shows that people will generally seek more information when they receive an emergency alert.
“What’s the trade-off? If you send an alert that’s very short and doesn’t have a URL, will people now flood the cellphone network calling people?” Eisenberg said. Or will they use their smartphones to start checking websites even without a URL? It could turn out that it would be better to give people one URL to get more information. “That’s an open question — we don’t know the answer to that,” Eisenberg said.
Perfecting the messages will take some time after the rollout.
“It’s going to take a few alerts for the researchers to get in and glean how people are responding, and what the appropriate language is for the messages,” said Virginia Bacon Talati, associate program officer with the Computer Science and Telecommunications Board.
Emergency managers also are concerned about making the messages as targeted as possible.
The original goal for CMAS was to target messages at the county level. However, local officials who have tested the system have been working with carriers to be able to send messages to a more targeted area.
“San Diego County is bigger than some states,” Rea said. “We wanted to see if we could narrow that scope down a little bit.”
If there is a tsunami warning for the Southern California coast, only people in coastal San Diego County need to be notified, Luke said. “People out in the desert do not need to know about the impending tsunami,” he said. “That for us will be critical.”
This narrowing of the scope — activating just some cell towers — was part of San Diego’s test in 2010. Although the county was able to narrow it, the alerts still sometimes covered a larger area than desired.
In New York City, the issue is not the area of the counties but the population. “We don’t see the benefit of sending them to an entire county,” Bruno said. That could mean alerting 2.5 million people in Brooklyn when only 150,000 were actually affected.
During New York’s recent test, some messages did go out beyond the borders of the area they were meant for, he said. “You try to make it as efficient as possible,” Bruno said.
Bristow agreed on the importance of alerting only those who may be affected by the emergency. “Most disasters are fairly localized or regionalized,” he said. “You certainly don’t want to alert a much broader region and cause confusion.”
A further challenge is making sure the public knows about the system.
“As far as we’ve seen, there hasn’t been public education on this and what it means,” Eisenberg said.
If people start to receive messages on their cellphones without knowing what they are, there is a risk that many will call 911 to ask questions, Bristow said. “One hurdle is just the public being comfortable,” he said, referring to the CMAS messages.
Experts also worry that if users don’t fully understand how the system could benefit them, they will try to stop the alerts. “The last thing we would like is for it to be perceived as spam and for users to attempt to opt out,” Talati said.
Rea said that in San Diego, for example, there are many Amber Alerts due to the proximity to the Mexican border. “Will that cause people maybe to dig down into their phones and opt out?”
However, a massive public education program could also cause confusion, since older phones won’t receive the messages.
Bruno estimated that it will take several years before most cellphones are compatible with the system. “We are looking at a little bit of time before we have things in full gear,” he said.
The success of CMAS depends on getting cellphone carriers and local governments to participate. All the major cell carriers and many smaller ones have signed on.
The program appears to be off to a promising start, according to Penn at FEMA. “We’re really excited about the program and how it’s panning out,” Penn said.
However, the system has not been rolled out in all areas by all carriers yet. This, combined with the fact that only newer phones can receive the messages, means coverage will be spotty for a time.
An AT&T spokesperson said, “Our deployment is progressing. The specific number of markets changes frequently.”
As for local agencies, Bristow said they are just beginning to be exposed to the system. In California, for example, only three county-level emergency offices have been authorized to use the system. Some states don’t have any localities authorized. “There’s some adoption there that has to happen,” Bristow said.
Still, the system is an important new communication tool for emergency managers.
“You want to have as many channels as feasible to try to communicate in a crisis,” Bristow said.