Shortly before 10 a.m. on October 22, 2014, Cpl. Nathan Cirillo, standing guard at the steps of the National War Memorial in downtown Ottawa, tragically lost his life when a gunman opened fire. It was the start of a shooting rampage that would eventually spread to the halls of Parliament. Almost immediately, news of the shooting lit up social media as first responders, witnesses, journalists and ordinary citizens (in Ottawa and around the world) took to online platforms to post situational updates, express grief and concern for the wounded and opine about the political causes and consequences of the attack.

An intense period of uncertainty, anxiety and fear gripped the country. In part this was due to a rapid proliferation of rumour and misinformation, an all-too-common feature of emergency events. Reports later determined to be untrue circulated on Twitter, Reddit and Facebook that multiple attackers (some allegedly at the Rideau Centre shopping mall and others perched on nearby rooftops) were shooting at innocent pedestrians.

Shared Services Canada, the federal agency responsible for e-mail, data centre and network support to ensure delivery of federal programs and services, issued warning messages asking federal employees to refrain from using their wireless devices for nonemergency purposes, and to avoid e-mail and other Internet activities for nonessential work to reduce network congestion. The high level of network traffic contributed to broader communications challenges that created delays and intensified confusion as the crisis unfolded. An internal government incident review would later reveal that many alerts sent by e-mail never reached their destination; federal employees deleted e-mails alerting them to the shooting because they mistook them for spam; and unclear and inconsistent messaging meant that some government employees did not follow established safety protocols.

Thus, when Public Safety Minister Ralph Goodale announced the launch of the smartphone-based National Public Alerting System — also called Alert Ready — in April, those who spend their time worrying about the next such emergency breathed a deep sigh of relief. As the Minister declared, “By adding smartphones to our emergency alert system, we will be able to reach more Canadians, faster, in times of crisis. Wireless alerts will also allow us to issue more geographically targeted alerts. The safety of Canadians is our highest priority.”

Indeed, forward-thinking governments are increasingly embracing the capabilities of digital and mobile media to help them fulfill a core responsibility — protecting citizens from threat and harm — and this announcement positioned Canada in a positive way.

A smartphone screen capture shows a false incoming ballistic missile emergency alert sent from the Hawaii Emergency Management Agency system in January. (AP Photo/Jennifer Kelleher, File)

Alas, the gap between policy concept and policy impact is lamentably large. Just talk to the Hawaii officials who embarrassingly engaged their own alert system earlier this year, causing widespread public panic and setting off an international conversation about the importance of trust and credibility to emergency communications. But while the idea of a modernized National Public Alerting System in Canada with mobile capability has been a long time coming, the hard work is just beginning.

A study published by University of Alberta professor Gordon Gow provides some context for the policy history of this recent announcement. Gow showed that in 2007, the Canadian Radio-television and Telecommunications Commission (CRTC) released a set of decisions that constituted a new policy framework for a national all-hazards, all-media public alerting system in Canada. As Gow noted, the need for a modernized emergency broadcast framework was very much a response to a combination of factors, including changing geopolitical realities (such as the end of the Cold War) and a transformed environment of converged media that called for more effective regulatory oversight.

In 2009, the CRTC mandated the creation of the National Alert Aggregation and Dissemination (NAAD) System. Yet participation by media organizations in this new system was strictly voluntary. In 2014, owing to a lack of voluntary compliance, the CRTC mandated participation of all FM radio, AM radio and over-the-air television stations. Despite these improvements, gaps in Canada’s national emergency response system persisted, and the Parliament Hill shooting helped bring these to the surface.

As of April 6, 2018, the CRTC requires all wireless service providers in every province and territory to distribute public emergency alerts on their LTE networks. Ian Scott, chair of the CRTC, described the recent change to the NAAD as “an important evolution of the National Public Alerting System.” Whether the threat is a terrorist attack, a forest fire, a flash flood, a radiological incident or another civil emergency, all Canadians with a wireless device attached to an LTE network will receive a mandatory SMS alert specific to their geolocation.

Officials will send a localized notification that emits an alarm, much like the high-pitched beep of the traditional emergency broadcast system, and displays a bilingual warning message. While these notifications have the look and feel of a text message, they are sent via cell broadcast distribution, thus allowing for simultaneous delivery to vast numbers of devices without encountering network congestion: unlike conventional voice or data traffic, emergency alerts use a dedicated distribution path in the LTE network. Most of the country’s wireless providers called for an opt-out choice for consumers, but the CRTC wisely ignored those requests. As one government official remarked, “When you’re getting those alerts, your life is at risk. So it’s not there’s potentially a danger. There is a danger.”

There is much to like about these changes to the NPAS. The shift to a mandatory emergency notification model reflects what other jurisdictions are already doing. Involving a broad range of first response partners will increase communication impact, and the full range of threats it intends to warn the public about will increase the likelihood that those receiving the information will understand its intent and respond accordingly.

The introduction of mandatory wireless notifications is also consistent with current practices in other jurisdictions; on the surface, this is an obvious improvement over the traditional broadcast-only model, to the extent that it leverages current technologies to reach Canadians where and when they require emergency notification. But as we will describe, those same advantages come with some potential pitfalls and may raise new challenges for policy-makers. Four stand out as particularly important.

Scope creep and alert fatigue

Government officials developed a list of 21 alerts for use in specific situations across eight categories that would qualify for immediate broadcast on television, radio and wireless devices. The vast range of these possible scenarios for using alerts raises the potential for intensifying both scope creep and alert fatigue. As officials broaden the scope of what could be considered an emergency, it becomes more difficult to decide which actual threats should trigger the alert system and when these alerts should happen.

The empirical evidence regarding AMBER Alerts provides a case in point: while we want to believe that these alerts are vital to recovering abducted children, their life-saving potential is low, and optimizing these alerts for mobile devices may have little impact on the safe return of an abducted child. Worse, they could complicate investigations by adding unwanted “noise” into the response environment, such as fake alerts or the recirculation of outdated ones.

Clearly, authorities want the public to be familiar with the emergency system, but alert fatigue can occur if civilians become overexposed to the alerts or find them uninformative or otherwise irrelevant to their immediate needs, or if those alerts are found to be ineffective. Emergency responders and policy-makers will need to continue monitoring closely how frequently the new system is used, and whether it is used appropriately and to good effect.

Message design and efficacy

One of the most important limitations of emergency alert messages is their brevity. A 2014 study conducted for the US Department of Homeland Security identified the features that make wireless alerts about imminent threats most effective. If the goal is to encourage people to take measures that will minimize their exposure to a given hazard, how long should these messages be and what should they look like? While shorter messages (90 to 140 characters) are more likely to be rapidly distributed and shared, they are less likely to lead to changes in individual behaviour. Messages that include visuals, hyperlinks and other multimedia content are also more likely to get the point across in a manner that both aids in message dissemination and leads to changes in behaviour. Canada’s system is set up to accommodate alerts of up to 600 characters, but it does not support the use of audiovisual or other multimedia content. Alerts that have a similar look and feel to a breaking news story, complete with appropriate multimedia content, may be more effective in mobilizing the ideal public response. However, more research and testing is needed here.

Racial profiling

In September 2016, New York police enlisted the support of millions of New York City and New Jersey residents to capture Ahmad Khan Rahimi, the so-called Chelsea Bomber. Officials sent a text looking like a wanted poster through the Wireless Emergency Alerts system via cell towers to all mobile devices in a geographically restricted area. It was the first time such a system had been used.

While the system contributed to the arrest of Rahimi (following a brief shootout with police), it also has the potential for abuse. Critics worry that short wireless alerts may unnecessarily induce public fear and contribute to the overpolicing of already racialized communities. These concerns are not unreasonable in light of rising rates of Islamophobia in Canada and other countries. Ultimately, authorities will need to balance the benefits of immediate, targeted mass notification against the potential for misuse or abuse; they should set up appropriate accountability frameworks to evaluate and assess impact on minorities and racialized groups.


A federal, provincial, territorial or municipal alerting authority issues a notification to the NAAD System in an actual emergency. If that alert matches predefined criteria for urgency, severity and certainty, the authority then issues an immediate broadcast alert, triggering the dissemination of an emergency message to the public via traditional broadcasters and, now, wireless service providers. This multijurisdictional model embraces the complex Canadian reality, which is that no one agency is ever really in charge.

Collaboration among various orders of government lays the foundation for effectively coordinated communication; yet it also presents a series of communication challenges. Chief among those challenges is the positive effort to ensure partner approval: coordination is vital but can slow down information dissemination. In the pursuit of communication consistency, adherence to management processes that emphasize coordination can present a threat to transparency and openness.

Canada’s updated National Public Alerting System represents a significant and positive step toward improving preparedness for the kinds of high-risk events that  we dread and that we always hope will not happen again. But it’s a step that comes with risks of its own and potentially with some disappointment. Authorities should be commended for finally getting this done, but they should also realize that the effort to build a truly efficient and effective system is only just beginning.

Photo: OTTAWA – A police officer patrols a street near Parliament Hill on Oct. 22, 2014, shortly after Cpl. Nathan Cirillo was fatally shot at the National War Memorial. THE CANADIAN PRESS/Adrian Wyld.

Do you have something to say about the article you just read? Be part of the Policy Options discussion, and send in your own submission. Here is a link on how to do it. | Souhaitez-vous réagir à cet article ? Joignez-vous aux débats d’Options politiques et soumettez-nous votre texte en suivant ces directives.

Joshua Greenberg
Josh Greenberg, PhD, is a professor of communication and media studies in the School of Journalism and Communication at Carleton University. His research expertise is in the area of crisis and health risk communication. He has collaborated with and provided expert guidance and advice to the World Health Organization, Transportation Research Board (US), Council of Canadian Academies, and Public Health Agency of Canada, among others.
John Rainford
John Rainford is the Director of The Warning Project. He is the former Director, Emergency and Risk Communications for Health Canada and Global Project Lead, Risk Communication Capacity Building for the World Health Organization.

Vous pouvez reproduire cet article d’Options politiques en ligne ou dans un périodique imprimé, sous licence Creative Commons Attribution.

Creative Commons License

More like this