Part of the Eastern Oregon University Colloquium Series
The Emergency Alert System
© 2005 By Dr. M. Mustoe Geographer

Brief Introduction to the Full Scope of this Research
This paper is derived from a broader study that considered the implications of the new localism of commercial radio on the Emergency Alert System. Using a naturalistic design (Lincoln and Guba 1985 ) the research began with a survey of the train derailment in Minot, North Dakota 18 January 2002. The Minot event became the model for this research and, as the study emerged, the role of the EAS was evaluated in additional derailments/accidents. These accidents now include Macdona, Texas in June 2004, Graniteville, South Carolina in January 2005 and most recently, Texarkana, Arkansas. The use of the EAS in these additional events was holistically evaluated and compared to its activation in Minot.

Minot has emerged as a touchstone model for several reasons. It was one of the first incidents to profoundly bring the activation and use of the EAS to the public forefront since issues of localism, and de-regulation have collectively made the news. Indeed the event became tied with these issues. Furthermore, scientific studies have been conducted of this event and have established a hypothesis open to further testing. In addition to the fact that this research only evaluates radio's role with the EAS, an unequivocal limitation to this study is that there have been many more than three train accidents since Minot. However, those chosen to be evaluated for this study match closely to the Minot scenario in that they included fatalities and property damage, they all have had a measurable release and plume of hazardous gas, and they all necessitated the call for the evacuation of a community. These factors subsequently set up the contingency for a public alert of one kind or another. Thus, this (ongoing) research attempts to answer the questions: how is the EAS used under these conditions and what circumstances prevail to facilitate or inhibit its use as a primary alert system?

What Happened At Minot?
On 18 January 2002 at 1:45 a.m. 21 cars of anhydrous ammonia on a Canadian National Railway freight train derailed about one mile west of Minot, North Dakota (North Dakota Emergency Management 2003a). At least one car ruptured and another developed a leak (North Dakota Emergency Management 2003b). The event killed one and sent hundreds to area hospitals.
Shortly after the derailment the Minot 911 dispatch center became deluged with calls which jammed the phone system and created dispatch problems. Power went off in some parts of the city, and in some cases, warning sirens did not work. "We weren't prepared for a disaster of this magnitude," said Minot Fire Chief Harold Haugstad. "I don't think anybody is" (Wagner 2002). In addition, radio and television stations that would have carried the emergency alert system could not be accessed. In their railroad accident report adopted in March 2004, the National Transportation Safety Board reported:

"The Minot Police Department attempted to contact the designated local emergency broadcast radio and television stations. At the time of the accident, only one person was working at the designated local emergency broadcast radio station (KCJB-AM), and the police department's calls to the station went unanswered. The designated local emergency broadcast television station (KMOT) did not have an overnight crew at the station. To arrange emergency broadcasts, the police department had to contact the KMOT news director at his home" (National Transportation Safety Board 2004 p. 17-18).

Most academic analysis of the apparent breakdown of the EAS system in Minot has focused its criticism toward the de-regulated broadcasting environment and, in particular, corporate radio as the cause of the failure of Minot's radio warning system. In a study conducted by Cornell University dealing with corporate broadcasting and specifically the operations of Clear Channel Broadcasting, (the largest corporate broadcaster) it was stated of Clear Channel's operation in Minot:

"In at least one instance, Clear Channel's cost cutting practices have undermined public safety. In Minot ND. the lack of staffing at Clear Channel's radio stations impeded the activation of the emergency radio response system following a train derailment and a hazardous spill" (Figueroa, Damone, and Whitefield 2004).

In another study conducted by the Department of Media Studies University of San Francisco, Kidd states:

"Then in January 2002, there was a chemical spill in Minot, North Dakota, where Clear Channel owns all six radio stations, including the designated emergency broadcaster, KCJB. Yet no one responded to the call from Emergency Services because the station was on automatic, piping out a satellite feed. This was not unusual as Clear Channel only employs one full-time news employee in Minot, who rips and reads the newscasts from state and national wire services (Kidd 2004).

To activate the digital EAS system at KCJB required access and endec that could activate the system. At the time (KCJB) was considered to be the local primary one (LP1) station that would feed this initial signal to other stations within its coverage area. However, the procedure to initiate an alert should not require contact with station personnel. In addition, even if personnel at KCJB could have been contacted that morning, according to the state plan of North Dakota (and the inherent technology), an EAS message cannot be originated from KCJB (Stensby 2004). An alert must start with an official from the emergency management side of the system, and that official must employ an endec in order to activate the system. John Funk, the area engineer in charge of KCJB at the time of the accident, also states:

"This system (the EAS system at KCJB) was in place at the time of the Minot event. We would do thunderstorms and tornado warnings all the time and it would work consistently. It was the fact that the police department didn't send anything. The system did not work because KCJB did not receive anything" (Funk 2005).

In contrast, Nislow (2003) reports that Minot police Chief, Fred Debowey, reasons that, "The signal (that they sent out) was not received by the radio station, thus the alert could not be sent out automatically. Even if it had worked, said Debowey, the station would have not received it because their radio had the wrong crystal." Funk confirms that KCJB's receiver had the wrong crystal to receive the voice transmissions from the Minot Police Department. However, the receiver referred to in this statement is not part of the EAS, but rather, part of an aural, two-way radio system which had a communications link to KCJB's main studio. Funk states, "They were using it like it was still the EBS system. They weren't using an endec, they were just getting on the radio and talking" (2005). The EAS system does not operate on voice commands, - it requires an electronically encoded message from an endec for activation.

"The EAS uses state-of-the-art digital technology to distribute messages. The system provides state and local officials with a method to quickly send out important local emergency information targeted to a specific area. Also, the EAS digital signal is the same signal that the National Weather Service (NWS) uses on the National Oceanic and Atmospheric Administration's Weather Radio (NWR). This allows NWR signals to be decoded by the EAS equipment at broadcast stations and cable systems" (Federal Communications Commission 2002).

What Went Wrong? The analog-based Emergency Broadcast System, which was exchanged at KCJB for the digitally operated Emergency Alert System in 1997, was not being activated by the Minot police department the morning of the derailment. "At the time of the accident, Minot still had what it believed was an Operational Emergency Broadcast System (EBS) that would allow police to call up the local radio station and have it issue an emergency warning. The EBS was replaced in 1994 with the Emergency Alert System. Minot had both, Debowey told Law Enforcement News" (2003).

Could Minot Authorities Have Activated An EAS That Morning? At the time of the derailment, Minot dispatch had access to the National Warning System (NAWAS) dispatch line. This aurally-operated, dedicated telephone line (Oak Ridge National Laboratories 2005) (Partnership for Public Awareness 2003) was housed at the Minot Warning Point and 9 -11 dispatch center and tied to state EAS access points such as the North Dakota state primary entry point for state radio in Bismarck and the National Weather Service in Bismarck. However, as Lt. Fred Debowey told me in an interview, they chose not to utilize this hot line because of the local nature of the event:

"We did have a hot line at the dispatch center, but our decision not to use the National Warning System that morning was based on the idea that this was a local emergency. The anhydrous cloud was affecting Minot and some of the areas around it like Burlington and Surrey and we felt that by contacting the local radio station and putting out a warning through that means we could provide adequate warning. But nothing worked" (2005).

National Weather Warning Meteorologist, John Paul Martin, in an interview with me concurs that the NAWAS line would have been a viable alternative for Minot but adds:
"But a lot of times I think it comes down to in the heat of the battle, the average person will kind of forget about the resource in the national weather service that they have. I think a good thing that can come of all of this is dispatchers, so to speak, at the warning points do not realize that they can go through the weather service, and should go through the weather service, to help in these matters. What I think we need to do is get on their procedures list, and you have this happen, this is what you do. Somewhere on there has to be the option of calling the weather service and letting them know that they can send a message" (2005).

In their report of 2002, The National Transportation board states that: "Minot Central Dispatch now has a contact number with the primary state emergency television station (KFYR) in Bismarck, ND. If the local emergency television station (KMOT) is off the air, the dispatcher can call Bismarck and request an emergency message be aired through this network (KFYR). The emergency message can also be transmitted through the National Weather Service located in Bismarck, ND using the National Warning System (NAWAS) phone or the Emergency Alert System (EAS)" (Kris et al. 2002 ).

This statement reconfirms that Minot Central Dispatch had alternative options as EAS access points available the morning of the accident. Primarily a civil emergency message (CEM) could have been issued through the National Weather Service in Bismarck, North Dakota. Minot Police Department Lt. Fred Debowey suggests that training also played a role in the decision not to activate the system with the use of the NAWAS line that morning:

"But at the time of the derailment we were not aware that we could activate the National Weather Service with the hot line. No one had informed us that this was an option. Now we can contact the state primary entry point in Bismarck with the hot line and they can issue an alert to KFYR television in Bismarck which will activate all the stations in the Minot area" (2005).

It is important to understand the issues surrounding the Minot incident so as to understands the implications of why the EAS system was either not utilized or activated late in the following events.

Macdona, Texas Texas On 28 June 2004 at Macdona, Texas a plume of chlorine and ammonium nitrate gas leaked from cars derailed by a collision involving the Burlington Northern and Union Pacific Railways (Mason 2004). The event impacted areas on the southern edge of San Antonio about fourteen miles to the northeast of Macdona. At Sea World amusement park "...six people were treated for minor respiratory irritation" (CNN 2004). Three people died in this incident. I, conducted interviews with individuals at radio stations within the San Antonio (EAS) area and the originating source for a civil emergency message (CEM), the National Weather Service in New Braunfels, Texas, and found that the EAS system during this event was never activated. This was also confirmed through interviews with officials of the Environmental Protection Agency in Dallas, Texas, (Harris 2004). Immediately following the incident, I contacted the National Weather Service in New Braunfels, Texas. In a phone interview with Bill Runyon, data acquisition program manager, and John Zeiter, science and operations officer, I was informed that the NWS did not issue a CEM and could only do so if authorized by an emergency manager (Runyon 2004). In contacting the local LP1 WOAI and LP2 KKYX in San Antonio as well as a series of other stations in the area, it was found that no EAS was issued to any of these stations. However, in almost every case, any radio station that had a news department could hear emergency scanner traffic. With this information, emergency message broadcasting was delivered by Clear Channel Broadcasting WOAI and Cox Broadcastings KKYX-AM. The problem at Macdona, Texas was found to be similar in nature to what happened in Minot, North Dakota in that an EAS was never issued, either at the local, county, or state level. Ironically, by default, robotics stations in the San Antonio market that could have delivered an EAS message electronically, did not do so because the initiation of the process did not begin at the local emergency management level (Becerra 2004). In addition to no EAS activation, the
Environmental Protection Agency reports that a community alert telephone network was never activated and "The decision not to go forward with the alert was never relayed back to the Incident Commander, who continued to operate under the assumption that the notification was in progress" (Mason 2004). In this case, perhaps, radio delivered the only immediate information to the public about this event. In particular, it was the news departments of corporate owned radio stations, Clear Channel broadcasting WOAI and Cox Broadcasting KKYX, San Antonio, Texas that broadcast the first alerts of the emergency based on information derived from scanner reports.

Graniteville South Carolina At about 2:40 in the morning on 6 January 2005 a rail accident involving the Norfolk and Southern railroad occurred near Graniteville, South Carolina. The derailment caused the release of chlorine gas from tank cars that were breached in the accident. The incident injured approximately 250 people, caused the evacuation of over 5000 and killed nine people (National Transportation and Safety Board 2005). Although this incident occurred at 2:40 a.m. it was not until about four hours later at 6:40 a.m. when an EAS was delivered. This research found the delay in delivery of the EAS to be due to human error. The local LP1 and LP2 stations for this region are just across the state line in Augusta, Georgia. Both of these stations, WBBQ and WYZA are controlled by Clear Channel Broadcasting.

It is important to note that although these stations had not received an EAS actuation, they were already on the air broadcasting emergency information to those people impacted by the event in Graniteville. It was Clear Channel's engineer, Earl Welsh, and his assistant who determined why there was no EAS activation for the Augusta/Aiken area during the event. Welsh stated in an interview with me that the NWS transmitter that WBBQ was monitoring (WXK-54 Wrens, GA) was not the transmitter to which the EAS actuation from the NWS was being sent. The initial alert went from the NWS to the Aiken transmitter WNG 627. Once the signal was transferred from the other NWS transmitter to the local transmitter at Wrens, Georgia, (the one that WBBQ and
WYZA were monitoring) the EAS came through (2005). National Weather Service Warning Coordinator Meteorologist, Steve Naglic, stated that:

"It (the EAS) was re-sent to the Wrens transmitter WXK-54. We continued to transmit the EAS on the Aiken Transmitter WNG 627. The only reason it went to Wrens was that's what BBQ monitored. Aiken transmitter is the primary transmitter for Aiken County, but because of the EAS activation we forgot to link Aiken County on the Wrens transmitter for EAS activation. Because of that, actually there were two counties that had to be fixed that were not linked to our LP1 stations" (2005).

Bernard Palmer, Meteorologist in Charge at the NWS in Columbia, South Carolina, confirms this stating:

"We had a call from SLED South Carolina Emergency State Law Enforcement Division at 5:25 AM requesting a briefing to aviation type operations over there (Graniteville). That was when we officially heard of the incident. The request for us to broadcast a CEM didn't occur until around 6 o'clock. We also phoned the LP1 to verify they got the product. I have it noted in the log that they called us at 6:40 AM from BBQ stating that they only have the capability of monitoring Wrens transmitter. So, it was at that time that they re-broadcast it onto the Wrens transmitter" (2005)

In the Graniteville case, it was human error that produced the minutiae that lent to the delay in the delivery of this alert, not the ability of the technology to perform. In contrast, it was a corporate radio station cluster in Augusta, Georgia that supplemented the local service lost by the late delivery of an EAS. News reports and live emergency message programming from these corporate radio broadcasters filled in the public service gap. In addition, it was National Weather Service standard protocol that produced a quick correction to the problem once the problem was discovered. Additionally, it was the simplicity of the technology of the NWS weather radio system that quickly facilitated the correct feed point for the EAS while, at the same time, provided public access to the same alert through its all hazards radio network

Texarkana, Arkansas On 15 October 2005 at 5 o'clock a.m., a rear end collision caused the derailment and subsequent explosion of a tank car of propylene located in the rail yard near Texarkana. About 2000 households were evacuated, seven people were sent to hospitals and one person living near the railroad tracks died in the incident. The EAS was not used as a public alert system. This was confirmed for me through interviews with the Miller County Emergency manager, the National Weather Service in Shreveport, LA, and the operations manager of the LP1 Clear Channel Radio KKYR.

Like Graniteville, Texarkana is located near a state line. The state emergency warning plan calls for radio stations to monitor the National Weather Service as well as Arkansas Public Radio and, as in the case of Texarkana near the Texas line, the Texas Broadcasting Network. In my research I found that none of these systems were activated during this event. The alerting methods used in the evacuation were sirens on police cars and door- to-door contact with local residents and police officers. In addition a 9-11 call-back system was employed and contacted residents within a one mile radius of the explosion. Miller County Emergency Manager Dave Hall explained the rationale for not using the EAS system was based on the fact that,

"The emergency alert system has to go through your contact point at the radio station. One of those was the EAS station and there was no one there. They would have to call someone in to do it. If it had been a little bit later in the day.... I'll be totally honest with you, at nights and on the weekends in the past is woefully unacceptable. We have a plethora of radio stations here and after five o'clock and on the weekends, one or two of them have a person there. So, quite honestly, in the middle of the night it is of little use to us" (Hall 2005).

The local area plan for Texarkana allows for emergency managers to directly contact station personnel at KKYR and, in turn, they can manually activate the EAS from the station. Wes Spicer, operations manager for KKYR Clear Channel radio, stated that they began broadcasting live emergency programming of the event within 15 minutes of the event. He confirmed that there was no authorization for an EAS activation, however he added, "An EAS authorization needs to come from the emergency manager we can activate an EAS from the station, manually, but it needs an authorization. We are automated at night but we have someone at the station all the time. If Dave Hall sends a proper authorization to our city public information officer, who is Sean Vaughn, he does have the responsibility and the authorization to activate it. But it never went through those channels, because we could not contact Dave Hall. In this instance, of the derailment, our disaster protocol that was set up a few months ago was not followed" (Spicer 2005). It should also be noted that officials in the Texarkana system also chose not to use an available cable television interrupt system stating that, " We did not access the cable override because at five o'clock in the morning not that many people are listening to television." Furthermore, the decision to not use weather radio for the event stemmed from having a "fairly good handle on the evacuation and to be quite honest with you in that neighborhood, I don't know how many people have a weather radio" (Hall 2005). In all three of these events, like in Minot, the EAS system was not used. In the case of Graniteville it was activated well after the emergency began, and thus, as an early warning system was useless. From the standpoint of types of radio stations employed during these events to provide emergency service, they were (as in Minot) all corporate broadcasters. In all these events stations from Clear Channel and Cox broadcasting provided supplemental emergency service when emergency managers chose not to activate the EAS or when it was late. In pragmatic terms these stations did not provide the EAS because of their association with corporate radio. Given a critical review of the corrections suggested by the National Transportation Safety Board in the Minot incident it is clear that is has nothing to do with the complexities of multiple ownership and the de-regulation of the FCC. Rather, it has everything to do with the communications interface, both technical and human, between two very different entities involved with the dissemination of the emergency information.

Radio's Role and The EAS
Eventually, even in the case of Minot, emergency programming about the disaster was aired. This use of radio as a means to reach a collective audience with current, "live", emergency messages, is its forte. The EAS is essentially a warning whistle, not much unlike the alarm of a prairie dog. Today radio's role in this initial warning phase of a disaster is not diminishing but it is being shared. Some public preparedness organizations are calling for the integration of the EAS into a variety of media: cell phones, televisions, the internet etc. No doubt this will come, but as I write this, cell phones and telephones are not working in New Orleans, likewise the internet and e-mail are down. For that matter major radio stations in the hardest hit areas of Katrina are off the air. But in this
post-initial stage of this disaster... in Houma, LA, just outside New Orleans, KJIN is broadcasting live emergency programming (not just the EAS) around the clock to anyone...with a battery operated radio. A battery radio is still considered a ubiquitous source of information.

Disasters come in stages (Ebert 1997). At every stage public communications is vital. Essentially the highly dependable digital EAS can act as an initial warning system. The National Weather Service Radio (now called All Hazards Radio) is fast becoming a popular EAS monitoring activation source for radio stations as well as a listening source for public radio. However, officials at the NWS must be notified by emergency managers to trigger the system and the NWS is not in the journalism business. Additionally, messages need to be timed to accommodate computer space. In some instances such as Texarkana, the controlling stations for the NWS are hundreds of miles away or even in another state. Thus, local radio still plays a vital role in not only the initial warning of an event but also in live programming as the event unfolds through its stages. Consider a simple triangle of communications. On one leg are emergency managers. On an opposing leg are public communication suppliers, including radio stations, and at the base of the triangle is the public. Without coordination (human and technical) between each leg of this triangle, communications through all the stages of a disaster breaks down.

Conclusions
If the EAS is an essential component of the public warning system, then analysis of what
happened in Minot and the subsequent similar rail accidents matching that model, that does not seriously consider the documented breakdown of human interaction with a digital electronic system, is seriously missing the point. The anomalies in the dissemination of the EAS in all four of these derailments point to issues surrounding the preparedness of emergency managers and their ability to interact with other agencies as well as with radio stations that have an unfounded government mandate to maintain EAS equipment at their stations. Additionally, the ability for emergency managers to have access to EAS technology such as a local endec, and the training required to operate it are critical issues surrounding these incidents. All of these issues exist aside from the environment of deregulation. Minot Police Lt. Fred Debowey suggests:

"As far as the communication system at the radio station goes I'm not blaming Clear Channel. There were responsibilities on both sides of this issue that were just not met, and when a breakdown in communications occurs like it did then there is going to be a real problem. All I want is assurances that even without anyone in that station we can still run an alert. Everybody believes that emergency services are the police and the fire response, but it's not. There is a third, and that is the radio media. We all provide an equal share of the responsibility of transmitting communications to the public. When somebody says, "it's the police department or the dispatch centers responsibility to get the word out," I don't agree with that. What I do agree with is that we all share an equal burden in ensuring that the public is notified when an emergency happens. The only way to do that is by everybody getting together and saying we have an emergency and this is how we can get the message out. It's emergency services, police, fire and the emergency management office, but an integral part of this communication's chain is the media" (2005).

From these discussions a key question emerges regarding local access by emergency managers to radio stations. In another component of this research, a small survey was conducted of radio stations found in towns having a similar population base to Minot that were located near rail or transportation lines. Given that the technology of an endec is critical in accessing the EAS at the local level, this survey asked station managers and engineers if their local emergency managers had endec capabilities into their stations? Eighty percent of all the respondents said there were no local activation capabilities at their stations. Six of the stations sampled were LP1s, and of these stations, three had endec capabilities. However, some of these managers cited problems with the delivery of the endec signal as well as with the training of personnel activating the system. More study is needed to understand the relationship of this data. However, if local access to the EAS for local emergencies is important, then perhaps endec systems, and especially training in the use of these systems, should become more available and appropriately integrated into local and state plans.

Historically, radio-based warning systems (CONELRAD, EBS) have all had a national-network-character. However, the EAS has never been activated at the national level and,
by default, there has never been a disaster that was "national in scope" (The EAS was not activated during 911). Stations are only required to broadcast EAS tests and in an actual event, an official announcement from the president. . On the other hand, the EAS has been activated many times (voluntarily) for local severe weather or other local disasters. Stations are required to deliver the EAS by an FCC mandate. Any emergency programming (news, messages, etc.) beyond that of an EAS alert is public service programming and delivered at the discretion of the station's management. However, in addition to station funding issues, studies have shown that the effective delivery of the EAS is complicated by a compendium of problems: convoluted state emergency management plans, complicated audio networks and poor links, bad reception from primary stations, stations not linked to the system, stress on volunteers heading up emergency communications committees, untrained personnel, poor communications between public officials and station managers, and a lack of government support.

EAS What Is It Good For? One might ask what value is the EAS? The simple control for an experiment attempting to answer that question might be 7 December 1941, when, in the absence of any national warning system, all the news about Pearl Harbor was handled by (radio) networks. In contrast, on 9-11, the existing warning system was not activated. One reason for non-activation came from Richard Rudman, head of the FCC's Emergency Alert System National Advisory Council's Coordinating Council for Local Stations who stated, "There was no identifiable major threat to the entire country," (Like Pearl Harbor?) Likewise another rationale given for non-activation came from then FCC Chairman, Michael Powell, who said that the "unbelievable ubiquity of coverage (from the networks) in some ways has proven to supplant those original conceptions of a senior leader's need to talk to the people." A "senior leader's need to talk to the people" is the whole essence of what these warning systems have been about historically. Required monthly and weekly tests of the system are performed to accommodate an objective
which has never been realized and indeed is being questioned. If these kinds of events continue to retain their local character (by definition of those in charge of the system), then perhaps a bottom up approach to re evaluating the EAS is needed in really getting a handle on the challenges faced by this system.

Problems with the EAS have existed long before de-regulation. Although the system as a technology is gaining stability with time, this research suggests it is the human administration of the system that presents the challenge to its use. Complicating this matter is research that suggest that programming in the post deregulated world of radio is moving toward a new localism that depends on remote satellite feeds and automation run stations. Although this new localism might exist, it has little to do with the activation of a digital EAS alert. On the other hand perhaps it may have bearing on a station's ability to cover a local event. Signs along a freeway suggesting you tune to station XYZ for weather reports.... when station XYZ is being fed with satellite programming from thousands of miles from the freeway you are on and has no idea of the weather in your area....are false advertising at best. On the other hand, there are many station owners (corporate clusters included) that fully recognize radio's potential to serve the public in times of need. They do this gratis by providing emergency programming to their communities as well as by buying into the unfunded mandate of the EAS by the FCC.

Although NOAA and National Weather Service radio will now carry terrorist alerts, it will continue to be the responsibility of broadcast radio stations to carry the brunt of the delivery of information to the general public; information such as evacuation routes and how to survive (Topping 2005). Managers in smaller market radio stations and emergency managers with proclivities toward the perception that "it can't happen here so who cares" should seriously re-consider their paradigm. Terrorist activity focused toward rural areas, smaller communities and "soft targets" is a potential reality (Jordan 2005) (Council of Foreign Relations 2005) (CNN 2002). In addition to terrorist strikes, other natural disasters require that the public be provided a viable communication system. Radio can and has assisted in this need.

As for the technology of the digital emergency alert system, it is as much at home in a small town's unattended automated studio, as it is in a low powered FM station, or bolted to the communications rack of a corporate cluster broadcaster. By itself the digital EAS is part of the new localism. However, the radio side of this system can only function if all involved with the emergency process establish sound goals and directives and seriously work together to provide a viable radio-based public warning system.

This study looked only at the application of the EAS through the use of commercial radio in a select set of circumstances. However, given radio's ubiquity amongst the public, its simplicity of use, and its ability to provide immediate information, the power of this medium to deliver the EAS should not be overlooked.

Kris, J., R. Wetzler, P. Marbut, and C. Craveb. 2002. Survival factors group chairman's factual report. Washington D.C.: National Transporation Safety Board.

Lincoln, Y., and E. Guba. 1985. Naturalistic Inquiry. Beverly Hills, CA: Sage publications Ltd.

Martin, J. 2005. Interview with National Weather service warning coordinator John Paul Martin Bismarck, North Dakota. Bismarck, 22 April.

Mason, S. 2004. Final draft of emergency response review: Union Pacific / Burlington Northern train derailment, Macdona, Texas: final report August 18, 2004. Environmental Protection Agency 2004 [cited 26 August 2004 2004]. Available from http://epaosc.net/sites\726444\files\union_pacific_macdona_texas_response_review_final.pdf.

Naglic, S. 2005. Interview with Steve Naglic, NWS Warning Coordinator Meteorologist Columbia South Carolina. Columbia, 24 April.
National Transportation and Safety Board. 2005. NTSB launches team to freight train collision in South Carolina. National Transportation and Safety Board 2005 [cited 31 March 2005].

National Transportation Safety Board. 2004. Derailment of Canadian Pacific Railway Freight Train 292-16 and Subsequent Release of Anhydrous Ammonia Near Minot, North Dakota, January 18, 2002 RAR-04-01. National Transportation Safety Board 2004 [cited 25 October 2004]. Available from http://www.ntsb.gov/publictn/2004/RAR0401.pdf.

Nislow, J. 2004. The wrong time to find out that emergency alert system doesn't work. PSComm LLC., Law Enforcement News March 2003 2003 [cited 12 November 2004]. Available from http://www.pscommllc.com/news/len_wrong_time.html.

North Dakota Emergency Management. 2003. Situation report: 1 No. 2002-007 Date January 18, 2002 time 12:30 p.m. North Dakota Emergency Management 2003a [cited 11 April 2003 2003]. Available from http://www.state.nd.us/dem//Sitrep/2002/sr02-007.pdf.
---. 2003. Situation report: 2 no. 2002-007 cate January 18, 2003 time 6:00 p.m. North Dakota Emergency Management 2003b [cited 11 April 2003 2003]. Available from http://www.state.nd.us/dem//Sitrep/2002/sr02-007.pdf.

Oak Ridge National Laboratories. Appendix C communications support network: system design criteria and evaluation guide for the chemical stockpile emergency prepardness program. 2005 [cited. Available from http://emc.ornl.gov/EMCWeb/EMC/PG/PG_HTML/App_C.html.

Palmer, B. 2005. Interview with Bernard Palmer meteorologist in charge, National Weather Service, Columbia, South Carolina. Columbia, 27 April.

Partnership for Public Awareness. 2004. The emergency alert system (EAS): an assessment. Partnership for Public Awareness 2003 [cited 16 August 2004 2004]. Available from http://www.partnershipforpublicwarning.org/ppw/docs/eas_assessment.pdf.

Runyon, B. 2004. Interview with National Weather Service data acquisition program manager Bill Runyon and John Zeiter science and operations officer National Weather Service New Braunfels, Texas. New Braunfels, Texas, 12 July 2003.

Spicer, W. 2005. Interview Wes Spicer Operations Manager for Clear Channel Broadcasting KKYR Texarkana, Arkansas. Tekarkana, AR, 17 October 2005.

Stensby, R. 2004. Interview with Rick Stensby General manager of Clear Channel Minot North Dakota. Minot, 25 August 2004.

Topping, J. 2005. Are you ready? A step-by-step emergecy preparness guidebook to prepare your station. Washington D.C.: National Assoication of Broadcasters.

Wagner, S. 2005. Most residents in survey rate rescue attempt as 'poor' part 1. The Forum 2002 [cited 12 January 2005]. Available from http://www.in-forum.com/specials/minot/part2.cfm?id=16521.
Welsh, E. 2005. Interview with Earl Welsh, chief engineer at Clear Channel, WBBQ. Aiken, 14 January 2005.