Weather and Society Watch
Communicating Severe Weather to Mobile Stakeholders: Challenges and Future Trends
Momentous weather events have occurred over the past year. From the tragic and devastating tornados in Alabama and Missouri that set an all-time record in April, to the record heat and drought in the southwest, to the torrential rains caused by Tropical Storm Lee and Hurricane Irene, 2011 ranks as one of the most severe weather event years in recent memory. In Joplin, Mo. alone, the National Weather Service (NWS) Central Region service assessment team found that the tornadoes on May 22, 2011, were responsible for 159 deaths and over 1,000 injuries.  Further, a Dun & Bradstreet report estimated a loss of $3.3 billion in business sales. Further, EQECAT, which is a risk management company, estimated between $1 and $3 billion in insured loses. Given the potential of severe weather events to affect both life and property, it is imperative that the NWS continue its critical charge to inform the public in a timely manner about possible threats.
Historically, the NWS has been charged to disseminate watch and warning information both directly to the public (through weather radio and Web postings) and to their media and commercial partners, who then redistribute the information over a variety of media channels such as TV, radio, and the Web. Further, some commercial Web sites allow users who sign up to have the watch and warning information “pushed” to their cell phones using short-message service (SMS) text messaging. The advent of sending warning information to cell phones has been especially helpful to educate an increasingly mobile public. For individuals who are not near a TV, radio, or computer, the message is, instead, pushed to their cell phone. On the surface, at least, this appears to be a reasonable, and fairly low cost, method to get the word out.
Although SMS appears to be a reasonable and practical solution to keep the public informed, it has a number of shortcomings that make it a less than ideal solution. As many readers probably know, wireless communications can become bogged down during crises (as occurred during the East Coast earthquake in September), and sending many SMS messages only exacerbates this problem. Additionally, messages that also include graphics or hyperlinks can make a bad problem even worse, given the network bandwidth required to send graphics or to browse the Web. SMS can also be notoriously unreliable, with tests showing that up to 5% of messages are never received during heavy load events.  Finally, long messages get concatenated (spread over multiple screens). As a result, some message screens might appear out of order or be lost completely.
Within the past few years, a monumental change has occurred in the potential to distribute weather information more widely and rapidly, based on rapid advances in mobile technology. Cell phones are now ubiquitous in the United States, with the most recent statistics from the Cellular Telecommunications and Internet Association (CTIA, the non-profit organization supporting the wireless industry) showing a saturation rate over 100% (some people own more than one cell phone). Further, it is estimated that over 70% of these phones are “smart” phones that are capable of browsing the Web. Of course, with the advent of smartphones (as well as connected tablets) has come the ability to download weather-specific mobile applications (better known as apps). While almost all of these apps have the ability to show current weather and radar information for the user's location, some apps offer greatly expanded capabilities, such as event-specific information (golfing, beach outings, marine weather, and so forth). Many of these apps are also either free or low cost, making the access to weather information literally at one's fingertips.
Apps offer a great solution to those individuals who have a smartphone. Of course, not everyone owns a smartphone. Additionally, another consideration is that using the information available through the app requires the user to take the time to launch the app to get the information. This might seem a minor concern, but considering that most people feel that there is not enough time in a day to get everything done, it is not a stretch to imagine that many people do not check their weather apps regularly.
A critical technological innovation that is just around the corner (April 2012) may minimize some of the shortcomings associated with these existing technologies. In a unique collaborative approach, the federal government is partnering with the wireless industry to roll out an alerting system known as the Commercial Mobile Alert System (CMAS), which is a direct outgrowth of the Integrated Public Alert and Warning System (IPAWS), overseen by the Federal Emergency Management Agency (FEMA). CMAS is also sometimes referred to as Personal Localized Alerting Network (PLAN). CMAS/PLAN (hereafter referred to as CMAS) is unique in that it allows officials to send text-based emergency messages to wireless providers, who will then deliver those messages to their subscribers' phones. The phones do not need to be smartphones, although they do need recent software to receive the alert. There will be no charge to the subscriber to receive these messages since the wireless companies have absorbed these costs.
CMAS alerts will be one of three types: Presidential alerts, Amber warnings, and imminent threats, including those posed by severe weather. End users can opt out of the Amber and imminent threat alerts, but cannot opt out of receiving Presidential alerts. Although the message will look like a traditional text message, CMAS warnings are sent through cell broadcast technology (rather than the point-to-point technology used by SMS messages) and are distributed to only those cell users who are in range of a cell tower within the affected warning area. The messages do not require GPS and, as the messages are only sent one-way (from cell tower to phone), there are no privacy concerns about tracking individuals' phones. The geographically targeted nature of the CMAS message ensures that only those individuals within the warning area receive the alert. This will be especially helpful for those individuals who may be traveling or who otherwise are unfamiliar with their geographic location. Anyone who receives a CMAS message will know that he or she is in the affected area, and, one hopes, take protective action. CMAS messages will also be distinct from SMS messages by having a special distinct ring tone and vibration cadence.
From a social science perspective, there are two interesting aspects of CMAS messages that raise questions about their effectiveness for the end user. First, based on recommendations from an advisory group, CMAS messages (at least in the initial iteration of these guidelines) will be limited to 90 characters, though cell phones are capable of handling up to 160 characters per screen. Information in the message will include the following elements: type of event, location of event, time frame and expiration of event, and a suggested action plan. An interesting (and presently unknown) question is how effectively a 90-character message will be at prompting the public to take appropriate and protective action. A second characteristic of CMAS messages is that such messages cannot include hyperlinks or clickable phone numbers. This prohibition was adopted for fear that many users would simultaneously access the information upon receiving the warning, thereby potentially crippling the wireless networks. This raises the question, of course, of whether the lack of a hyperlink will truly prevent those with smartphones from immediately launching a Web browser to search for additional information. These are not minor issues, and some have claimed that any federal mobile alerting system that does not include the newest technologies (hyperlinks, maps, multimedia content) is actually regressive. A rejoinder to this argument is that those individuals who might be the least likely to own a smartphone might be those also least likely to be tuned in to currently unfolding weather events and hence, most in need of at least a basic text alert. This is an advantage of CMAS in that it will work with any cell phone that has the updated software.
It is laudable that the federal government is trying to update their alerting technologies. Clearly, the goal here is to develop a messaging strategy that is designed to reach the most people. A number of questions remain, however, from both a technical and a social science perspective. We believe the most pressing questions to address concern the utility of the 90-character message itself; will it be sufficient at both informing the public and prompting appropriate protective actions ? Also, will the public be receptive to alerts that provide nothing more than very basic information that begs for additional information? Indeed, these are empirical research questions that beg for an answer.
* Mark A. Casteel (firstname.lastname@example.org) is an Associate Professor of Psychology at Penn State York.
** Joe R. Downing (email@example.com) is an Assistant Professor of Communication Arts and Sciences at Penn State University, York Campus
 Meng, X., Zerfos, P., Samanta, V., Wong, S. H. Y., & Lu, S. (2007). Analysis of the reliability of a nationwide short message service. Retrieved from http://www.cs.ucla.edu/wing/publication/papers/Meng.INFOCOM07.pdf