T3 Webinar Question and Answer Transcript

A Community Responds: A Multi-Agency Emergency Response to the I-35 Minneapolis Bridge Collapse (June 3, 2009)

Back to Webinar Files

Q.  I drive from south Minneapolis north quite a bit to a lake place. After a couple of months, my wife said to me, "What if the bridge collapsed and nobody noticed?" It got to the point that by the fall of 2007, the three or four million of us that live there seemed to get around without regard to the fact that the biggest bridge in town had gone down. Did that surprise anybody at Mn/DOT?
A. Jim Kranig: I think we were a little bit surprised that we were able to be as successful at building the additional capacity as we were. I don't know of anybody who really did anything on this scale before and certainly in the time constraints that we had. But once we were able to do all those projects, we added a lot of capacity. I think what really happened was people had longer trips, but it wasn't necessarily a really badly congested trip. Definitely people's travel added five, ten miles to a lot of people, but it was relatively good. We were pleasantly surprised, not totally surprised. I think it worked out probably better than most of us thought it would.
Q. In the lessons learned description, you described the EOC and how well it functioned. It just kind of crossed my mind, what if the EOC had been damaged? Did you or do you have a cutover plan that might have allowed another center to manage the event or the scene?
A. Heather M. Hunt: We don't have a mutual aid agreement with another jurisdiction to do EOC. However, there's redundancies throughout the metro area. Just about every agency has an EOC. We have a backup EOC at one of our public works facilities that's a little further away from the site. So our first thing would have been to relocate the EOC to the backup facility at our public works garage is actually where that's located.
Q. You said that fire had set up an incident command on the bridge and police had set up a different command post. Was there any time that they decided they were going to merge command posts and create a unified command or did that happen at any time during this incident?
A. Heather M. Hunt: As far as I know, it did not happen. They kept the separate command posts and then once police had worked through the determination that it was no longer a crime scene and it was going to be just a recovery operation, the police command post I think somewhat became of lesser-- or the fire command post continued the recovery operation. Police worked quite a bit with the Navy SEALs who sent in a dive team. They did a lot of search until the last missing persons were recovered.
Q.  Did EMS have a separate command post?
A. Heather M. Hunt: They did. They have an emergency response center located at the major hospital in Minneapolis. They ran mostly from that. They also had an impromptu command post that was set up in a mobile command vehicle at the scene.
Q. From the lessons learned, have they decided that there will be an agreement where there'll be a unified command if something happens in the future?
A. Heather M. Hunt: I think that's certainly being talked about. I don't know if any specific conclusions have been reached from that. But I know that they've been doing a lot of analysis on the incident and certainly know the value of a true unified command. It just didn't occur in the situation.
Q. A question a bout what you just mentioned, the mobile command centers. It sounds like you had but one, if any, available for this incident. Wouldn't it be a good idea to establish for cities like yours and all places, to have mobile command centers? I say that because at least I have some knowledge of airports after 911. Several of them did purchase mobile command centers and it has all the latest and interoperative communications facilities. It seems like that would be helpful.
A. Heather M. Hunt: We do have a number of mobile command post vehicles. They were just not co-located in this particular incident. Fire has one. Police has one. Hennepin County sheriff, we are in Hennepin County, have one. They were all used, but not co-located.
Q. I'm going to follow up on the incident command structure. Did the FBI ever join one of the other commands or did we have another separate command structure?ITS?
A. Heather M. Hunt: I am not sure of the answer to that. I believe that they were co-located in the police incident command area, but I don't know if they ran their own command or if they were integrated with police.
Q. First, how close did this incident come to any of your tabletop exercises or incident planning? The second question related would be what was the closest sort of incident that you had to plan for a tabletop that might have helped in the preparedness? Thanks.
A. Heather M. Hunt: I'm not familiar with all of the exercises that have been conducted. I believe that most of them have been around the metropolitan airport and plane crashes and so forth on the airport property, which is also in the City of Minneapolis. So I don't think that we ever expected a bridge to fall down. This was one of the true things that you just don't think that that would ever happen. I know there was never a bridge collapse exercise. Interestingly enough, though, when we implemented our new computer aided dispatch system just in March of that year, we did create a response plan for a bridge collapse over the Mississippi River and programmed that into our response plans in our CAD system. Here again though, I don't think that we envisioned a bridge collapse to the level that we experienced.
Q.  You mentioned training that was critical to the development of the interagency agreement or coordination plan. Could you expand on what training you were speaking of?
A. Heather M. Hunt: All of our EOC members, the officials that are designated to respond to the EOC, have had advanced incident management training and had participated in joint exercises together prior to this date. Also, all of our 911 staff has had basic NIMS and basic training in incident command structure. So everyone was familiar with it. It wasn't a textbook incident, but we felt that because people had that training, they were better prepared to deal with an incident of this magnitude
Q. In light of the initial responders when they got there and they just saw the magnitude of what happened, and a lot of them were jumping into the water from the pictures, how many injuries did you have from first responders? How many of them were injured during this rescue effort?
A. Heather M. Hunt: I don't have an exact answer for that, but there were no injuries that required hospitalization to any of the first responders.
Q. You mentioned there were 505 calls to the 911 center, but only 51 from the scene. I guess I'm curious where the other 454 phone calls came from.
A. Heather M. Hunt: They were people looking for people. They were people asking for information. They were media calls, many of them. Of course, there was business as usual. But our business as usual did decrease significantly during probably the first two days following the incident. We also had a call from a woman that was--I tried to find it, but I couldn't--who was irate because she was stuck in a traffic jam waiting to go over the 35W bridge and we'd better do something about that, because people are getting angry. Our operator very patiently and professionally explained the situation to her.
Q.  What is "all start" versus "fast start"?
A. Paul D. Linnee: This is applicable for sure in Motorola trunked systems. It may or may not be in others. I'm not intimate with them.

That is an attribute that is applied to the electronic definition of a talk group. Remember, we talked about priorities? Those are an attribute. All start versus fast start is also an attribute. So if you are dealing with the planning and the development of a given talk group and you are given the question, "Should I make this talk group all start or fast start?" here's what that means. If you make the talk group all start and you have a trunked radio system that is simulcast, multiple sites, your voice is going to emanate from X number of towers concurrently. If you say yes to the question all start, what you are saying is nothing I say will get broadcast by any of these sites until all of these sites have a radio channel available to carry my communication. That is all start. And, in an environment like this where some of the lower channeled sites in the outer part of the metro area, all of the channels may be busy, you will get a delayed access to the system, because you're not going to be allowed to talk until there's a channel available at every one of 30-some tower sites in our system. That's all start. Fast start is: "I don't care if there's only a channel available at one site, I've got to talk right now". I'm not saying you should use one or the other, but I'm saying that if you do the all start and you encounter a situation like this, you may have delayed system access. You may be like God and you may be able to talk to everybody at once, but you may have to wait for a while to do it. That is the plus and the minus. Now, as a result of this incident, a number of the talk groups had been awarded all start attributes at system design. As a result of this incident, that was taken away from many of the talk groups, partly in anticipation of the Republican National Convention which we had back it seems like a decade ago, but six months ago or whatever, last August. Things worked very well as a result of that. But this is one of the things about tweaking the system. Go ahead.
Q.  Were diversion routes set up before the bridge collapse?
A. Jim Kranig: No, we did not really have any pre-planned diversion routes, although a number of times when there was maintenance and such done on parts of 35W, we used trunk Highway 280 as another route. So the fact that we have the freeway mangement system over our entire freeway system, we can respond pretty quickly and we have data on all of it. So we really have a good idea of what we're seeing on normal days and all that. So we were able to respond pretty quickly to that. But we didn't go through and say, "Okay, if we lose this bridge, what would we have to do? If we lose this other bridge, what would we have to do?"

Paul D. Linnee: I'd like to add something to that. For the audience, picture the Golden Gate Bridge in San Francisco. Picture yourself being in San Francisco and you want to go north up into Marin County. If the Golden Gate Bridge goes down, you've got a big problem, because there are no bridges that get you there until you go all the way over to the Bay Bridge into Oakland and then drive up around San Francisco Bay. About 180 degrees opposite of that is the Mississippi River in Minneapolis. In downtown Minneapolis, I don't know the exact number, but I think it's like a dozen bridges cross the Mississippi River within about maybe a two mile run of the river. Now two of them happen to be interstate highways. The Interstate 94 bridge crosses the river maybe a mile downriver from the 35 bridge. Then there's a bunch of fairly high capacity city street bridges. So in this case, taking out that 35 bridge was not the critical thing you might think it might have been. Would you agree, Jim?

Jim Kranig: I think that overstates it a little, though, because all of those other bridges are pretty much at capacity on a normal day. So even though we have a lot of them, they're pretty much filled up. So that's why the City of Minneapolis had to do an awful lot of work to allow those streets that aren't normally carrying that much capacity to open up. We had to do a lot of work to basically be able to move the traffic around to them. But there are other bridges there. So I guess I kind of agree and kind of don't.
Q.  We've got communication system loading, specifically radio system loading. At what point did the radio system reach saturation for data and voice or do you know?
A. Paul D. Linnee: The radio system itself is not the backbone for the mobile data system in use in public safety in the metropolitan area. So there was no impact on data, per se. In terms of reaching capacity, I think we covered that in the earlier presentation, where we only had one instance where there was a priority three or higher access that was delayed more than 10 seconds.
Q.  Please describe the capability and resource gaps identified by the medical surge from the event.
A. Heather M. Hunt: The answer to that was that there were no capability or resource gaps. The gaps that were identified were mostly with communication, documentation, and patient disposition. At the time, under normal operations, EMS units use cellular phones to contact hospitals directly with incoming patient information. During a large scale incident, standard operating procedure detailed the use of radio and coordination of patient disposition through a centralized dispatch center. That was the EMS emergency dispatch center. However, EMS field personnel fell back on normal daily operating methodology and used the cell phone. This did hamper coordination and dispersion of victims and delayed patient dispositions and family reunification. The region has since made several changes to accommodate this issue, but it has not yet been resolved completely. I hope that answers the question.

Paul D. Linnee: I think one of the big issues going forward is going to be the bugaboo of cell phones. Cell phones are the bane and the boon to this business. To the extent that significant or important stuff takes place on cell phones or some of the push to talk variety cell phones, that stuff is off the main communication system for the public safety agency. It is not being recorded. It is not being documented. And it may not be heard by people who need to hear. So to a certain degree, you've got to go through the depot to catch the train, as opposed to hopping on it down the track. That may not seem convenient, but there is a reason for that.
Q.  Was personnel accountability a challenge in this event?
A. Paul D. Linnee:Heather referred to, I think, voluntary dispatching, voluntary self-assignment. There were some overreaction, no doubt about it. I referred in my presentation to the remote area monitoring. What happened was in that metropolitan area there's maybe 100 fire organizations, of which only a handful are full time; the rest are volunteer. Something akin to a statewide all call had been issued by the State Division of Emergency Management. What that meant for most of these volunteer fire departments out in the suburbs and the exurbs was that they were to respond to their stations and stage at their stations with the expectation being--literally we thought there were going to be hundreds of fatalities--that pretty soon the fire department 40 miles away would come in and cover for the one 25 miles away which would come in and cover for one 15 miles away and that one 15 miles away would be down at the bridge. Well, you get all these guys at the fire station. What's going on? They're watching CNN. They all take their radios and turn them on. That's what caused the remote system monitoring problems. Some of them said, "They need us. They need our boat. Let's go." They arrived and had to be managed.
Q.  Okay. I'd like to know what actually caused the bridge collapse. Can it be prevented? What monitoring and pre-warning detector systems are available? Are there any other types of bridge safety issues due to material and construction types used for the bridge in any inter-institutional issues on planning, coordination, and maintenance, which may cause the problem to go unnoticed? He is from the state Department of Transportation in California.
A. Jim Kranig: I think I'm not real comfortable talking too much about the details of the bridge. I think the best thing to do so that people don't get misled is probably refer them to the NTSB accident report. That can be found on NTSB.gov and then look for the publications on 35W bridge investigation. I think that would be the best thing. I will say I believe that the determination was the gusset plates were under-designed. Then, I suppose like in everything, there's other factors that contributed to it. But that was, I believe what they said the main cause. But really, the best thing if somebody really wants to get into it is go look at the report. I've printed it out here and it's, as you would expect, extremely thorough. In terms of the smart bridge technology that was included in the new bridge that went into place is to just assist in the research of how bridges react. I don't know that it so much is for a warning device as it is as research. But obviously, anything that gives you information about problems with the bridge would, by its very nature, give you the potential for warning. I know they have strain gauges in various locations and all of that information is being fed back to both Mn/DOT as well as the university researchers that are in structural engineering. I guess I'd have to take a look at that. There were quite a few questions in there and I'm not sure that I've answered all of them. Was there something about inter-jurisdictional?

Ray Fischer: They refer to it as inter-institutional issues on planning and coordination.

Jim Kranig: I think that's probably something that depends more on how you're organized and might vary substantially from state to state. For example, I know Mn/DOT is fairly centralized where we handle that work on our own bridges and that. Whereas, I believe in some states much of that is farmed out to counties and that. So I think you'd have to look at each individual case to make that determination.
Q.  Heather, you mentioned at least three potential HAZMAT dimensions of the event, had any of them transpired--the boxcars, the facility, and whether or not it was a terrorist event--that could have resulted in this further being compounded with a HAZMAT dimension. Also recognizing that it takes some time to determine that none of those events, in fact, took place. In the meantime, you had people being transported to your hospitals who, had they been exposed to a HAZMAT event, could very well have contaminated your trauma center. Is your radio system linked to your hospitals? In other words, are your hospitals on a talk group? Do you have the ability to let these hospitals know in advance of the potential of a HAZMAT incident that could involve a multiple casualty event? Was that utilized?
A. Heather M. Hunt: The hospitals at the Hennepin County Medical Center, which is the main hospital in Minneapolis, is on the system, as is the entire EMS response resource. So yes, there is communication capability through the system. I'm not sure that was a huge concern early on, but in the event that it was, yes, they would be able to communicate.
Q.  Do you have a protocol in place that when you have an event that could result in multiple casualties where there could be a HAZMAT connection, do you, as a matter of protocol, notify your hospitals when the event occurs?
A. Heather M. Hunt: I'd have to say not specifically. That would be up to the EMS agency. They do have the capability to do that. The hospital has its own contingency plans, which they did put in place in this incident.

Paul D. Linnee: Let me add to that. It so happens that in Minneapolis, the Minneapolis 911 CAD system is also the CAD system that serves Hennepin County Medical Center, such that if either entity creates an event, that event is mirrored in the other agency's response field. If they were to come together on a modification to what they refer to as nature codes for the structure collapse, for example, and add to that nature code an element that says maybe nature code H or structure collapse-H would be a structure collapse with a HAZMAT component or potential. Then behind that would have been all of the planning that would have said, "When one uses nature code HAZMAT H, not only do you go pick up bodies, but you wear contamination suits and so on and so forth." I think what Heather is saying is that's not in place, per se.
Q.  Are many of your talk groups encrypted?
A. Heather M. Hunt: In the Minneapolis subsystem we have only a handful that are encrypted; most are not.

Paul D. Linnee: In the system at large, encryption is a two-stage thing. The actual subscriber radio needs to have an option making it encryption capable. Then, once it's encryption capable, a given talk group may or may not be encrypted. If a given talk group is encrypted and you have an encryption capable radio and you are permitted to access that talk group, then you will be encrypted.
Q.  Any discussion on co-locating the city TMC, EOC, 911 centers in the Mn/DOT RTMC?
A. Heather M. Hunt: Not specifically, but we're continuously, of course, always looking at ways to have more efficient operations. I did mention that the City of Minneapolis is building a new EOC facility that will include some camera monitoring capabilities.

Jim Kranig: In terms of co-locating, I think that you have to, obviously, be careful not to get things too large. The other thing is I think what we've found that is very good for groups that work together on a daily basis to be actually co-located. As wonderful as the technology is to pull you together, it doesn't work to have everybody sitting at home just using the technology. You can overdo the consolidation as well. I think from our standpoint, we've found it's extremely beneficial to pull the state patrol traffic operations, freeway operations, and maintenance people all together, because we literally do things with each other all the time. Whereas, you need to set up those relationships and such to be able to, in the time of some major event, be able to reach out and work well with others. But I think you might be overstepping what would be really that useful if you tried to pull all kinds of groups together that maybe don't that frequently have to work with each other.
Q.  The topography in the area is flat, correct? Largely flat. Has that helped with the propagation? That's why the system works well in your area?
A. Paul D. Linnee: Yes, it is. If we would be Los Angeles County with a mountain range running through the middle, that would be a big impediment. However, unlike the Mississippi River that most people have in their imagination running through Tennessee and Arkansas, miles wide flat and flowing, in downtown Minneapolis, the river's actually in 100 foot gorge. Minneapolis, the reason it exists, the only place there ever has been a waterfall on the Mississippi River. So there's 100 foot gorge there. If you're down at the water level, you're 100 feet below the level of the freeway. So the radio system again, I pointed back to when I showed the map and where the bridge was in the middle of town. I don't care how deep a gorge is. If you've got 52 channels from about eight radio sites all homed in there, you're going to get good signal down in the gorge. Now if that gorge would have been 20 miles south of Minneapolis down in Dakota County, not so good.
Q.  How is the construction of the additional lanes on local roadways, the $7 million retrofits, handled without disrupting the additional volumes diverted from 35?
A. Jim Kranig: I'm not sure if I understand the question correctly. What we did was we constructed things at night and on weekends. In the case of the essentially three miles in each direction adding lanes on I-94, which is a very high volume roadway, we essentially closed it down in one direction at 10:00 on Friday evening and the other direction I believe it was closed down at midnight, with the guarantee that it would be open at 5:00 a.m. on Monday morning. So we just closed it completely off and it was a coordinated construction effort to behold for a traffic engineer and road designer. But it was the first time we had rain in about two and a half to three months. So it made it as hard as possible. But all the groups worked extremely well and they pulled it off. I was standing in the rain down there on Sunday evening thinking this was never going to be done and I got up extra early, because I thought we were going to have pandemonium the next morning. I checked the cameras at home and saw the lane lines painted and cars driving on it. Those were the kind of things we did. We really tried to stay completely out of the normal peak hours of everything.
Q.  You discussed interoperability of the voice systems. Does your system also address data communications? That possibly has already been answered.
A. Paul D. Linnee: As it relates to several agencies sharing the CAD system, yes. As it relates to complete interoperability with a mobile data terminal in a Minneapolis police car accessing the Hennepin County sheriff's CAD system, no, not yet, maybe not ever.
Q.  It's not really related to the incident, but earlier, Paul, you were talking about the funding source for your communication structure, $.04 per month. Has that been adjusted over time? What is your plan for ramping up for next generation 911?
A. Paul D. Linnee: The statewide 911 surcharge has grown significantly. It's up $.65 a month now statewide. But still, only $.04 of it is being used to defray radio system costs. The vast majority of the money spent on acquiring subscriber radios and system infrastructure at the local level between 1995 and 2001 was local general fund money, some of which was bonded, but it was local general fund money. The City of Minneapolis had to come up with some $20 odd million. Fortunately enough, the chairman of the public safety committee on the Minneapolis city council was, guess what, a member of the Metro Radio Board. She carried the vote on a seven to six margin to get the city to pay for the system. Since September 11, 2001, the question has not been: Where can you get money? The question is: How can you spend it all? We've gotten a lot of money from DHS and various other things for-- we even "bribe" people to get onto the system. We exempted all radio purchases from sales tax to get people to come onto the system if they bought radios that played into the system.
Q.  A question on the radio system. Is there a user fee for those agencies using the system? What's the life cycle on the existing system? What do you anticipate for replacement and the costs for that?
A. Paul D. Linnee: Some of these questions the answers are going to be a couple months old. The last I heard there is no user fee. There is discussion of implementing one, but there is no subscriber user fee, per se, yet. I believe that there is some, what do they call those, sunk funds or building up of some reserve funding being done to replace it. I know in the City of Minneapolis, the city bought it. The city owns it. The public works radio department maintains it. They rent the radios out to the fire department, the police department, the street department, the sewer department and so on and so forth. And the money that the police department pays to rent those radios is in the vicinity of $20, $30, $40 a month per radio. Some of that money is being put in the drawer hopefully labeled "Use me in 20 years to replace stuff." I don't know that 20 is the right number, but something along those lines.

back to top