The single most important place to start adding value in emergency preparedness is. . .

Since complex is often as simple as it gets in emergency planning and coordination, the place to start by way of adding much more value are those initiatives already underway to manage the complex interconnectivities and vulnerabilities of lifeline infrastructures.

This means capitalizing on existing opportunities beyond the official emergency management structures and plans at the local, regional, state and federal levels there. The aim is to leverage existing initiatives that have already “seen the light.” The priority in focusing on those who actively acknowledge the centrality of interconnectivities is made all the more visible because these are still early days in thinking through emergency management in terms of infrastructural connectivities.

When it comes to our recent US Pacific Northwest research, ongoing professional efforts focusing around inter-infrastructural connectivities were the Cascadia Rising exercises in Oregon and Washington State, the Cascadia Lifelines Dependencies Collaborative (“CSZ Lifelines Group”) in Oregon, the Regional Disaster Preparedness Organization (RDPO) in Washington State, various city and county groups in both states, as well as state personnel with emergency support functions, whose duties and responsibilities explicitly entail lifeline interconnections.

A priority is assembling and undertaking major table-top exercises and improvisation drills with these groups around unfolding Magnitude 9 earthquake scenarios centered around shifting interconnectivities of water, electricity, telecoms and roads in western Oregon and Washington State. The core competency called for in these table-tops is in the area of interconnectivities. These people are targeted because they already work outside their infrastructural or sectoral siloes.

The advantage of starting with ongoing or already-existing major initiatives is that they involve professionals who know much more by way of what needs to be done in preparing for large-scale emergencies. This means that when asked, “Have you read this report on seismic vulnerability here?,” and they answer “No, we haven’t,” no one should assume these professionals aren’t as knowledgeable as they must be. (I.e., the professionals may explain their “no” by referring to work already done in actual emergency operations to address shifting and shifted interconnectivities and vulnerabilities, with the staff and resources they have.)

Source: The above is a slightly edited extract from E. Roe and P.R. Schulman (2025). The Centrality of Restoration Resilience Across Interconnected Critical Infrastructures for Emergency Management: A Framework and Key Implications. Oregon Research Institute: Springfield, OR (accessed online at https://www.ori.org/wp-content/uploads/2025/08/FinalReport_10Aug2025.pdf). Research design, references and other particulars can be found there.

Infrastructures are more vulnerable when their latent interconnectivities are not managed ahead before an emergency

An interviewee in our recent Pacific Northwest research gave the example of a major opportunity that was not missed when altering the local hazard mitigation plan to replace an existing culvert by a new bridge:

So in the case of the county, they’re well positioned because they already had this plan, like we see this culvert is undersized, we continually see water over the road and we want to replace it with a bridge and that was the plan. If they had to wait and put that bridge in their general capital improvement process, it could have taken them another decade to replace it, right? But because they had a plan and that’s the direction they were already headed, now that the culvert washed away, it’s going to accelerate in the direction of the change that they were already headed, which is towards the bridge, which is great. . .They thought ahead of it and now they’re taking advantage of it in terms of trying to get additional funding”

As another interviewee put it, the hazard mitigation plan becomes a way to think more strategically about the federal funding component in critical infrastructure development at the local level.

Put this way, forward planning–in our terms, managing ahead–has a major role in anticipating and taking advantage of already-existing funding and construction opportunities. When the focus is on mitigating major emergencies, missed opportunities to correct for known vulnerabilities are mitigable errors to be avoided. As such, the hazard mitigation plan also becomes a mechanism to think through how the bridge would alter road transportation in ways another culvert would not.

The public good of “stabilization” in immediate emergency response and initial service restoration

A federal emergency manager whose national and international experience included working in one of the Pacific Northwest states whose emergency management we recently researched described his experience in a disaster:

But as, you know, as they start bringing systems up and we run into this, everybody gets pretty excited, right? They’re like, “Oh, water plant’s up and running. It’s great. You know, we’re out of the bottled-water business.” Or, you know, people have water to their taps now. And then,. . it’s off again. And so we plan those things, we really never plan a kind of a hard cut off of any of it. We continue until we’re sure everything is stable. And that kind of the world I live in is the world of stabilization. . .It might not be a long term solution, but they work. And that’ll then allow all of those other systems that are tied to it—so water, wastewater or interconnected power with all of that. It allows you time when you once get it stable, time [for] people that really know what they’re doing to go and figure out if there are any other problems with the systems, figure out where the gaps are going to be. . .

[For] example, anytime they’re putting the grid back together, power will come up in an area. And at some point, they’re gonna have to take that back offline in order to do something somewhere else. And so, it’s not necessarily stable at that point. So if we are providing food and water, we may still have to provide food and water because there’s going to be a time where it might be a couple of days that the power’s gonna go back down.
So, you know, there’s that interconnectivity. It seems like that is the world we live in. It all, everything we do, every piece of it impacts something else. . .

Here stabilization is explicitly tied to shifts in the interconnectivities between and among critical infrastructures as the disaster unfolds. But how can this challenge of stabilization during immediate response and initial service restoration be better addressed in advance before the emergency?

Part of the answer centers on a much more nuanced understanding of vulnerabilities in infrastructures.

Discussions of infrastructure vulnerabilities that we are familiar with frequently focus on physical components, like corrosion in gas pipelines. Sometimes, physical properties considered vulnerabilities are focused on precisely because they follow from the system’s strengths: “So we do have a fairly robust [internal communications] system. The vulnerability is that we are responsible for that system. No one else is going to go fix it, so we have to rely on our internal resources to then have access to the sites to fix any problems that arise,” a electricity transmission interviewee told us.

The vulnerabilities of special interest in our research, though, start off as significant weaknesses before disasters. These begin when the interconnected infrastructures fail to anticipate the need for special capacities—in lateral telecommunications and communications, for instance—with respect to the demands arising from their shifting or shifted interconnectivities. Once the emergency is underway, sequential dependencies up and down chains of command are of course important (e.g., declarations of emergency to release funds under the Incident Command System). But they do not and cannot offer sufficient options in micro-coordination for requisite variety to match the shifting demands and capabilities imposed by the shocks, surprises and contingencies of a major catastrophe. Here other forms of interconnectivity–reciprocal, mediated and pooled–are required. In important cases, the resources for them are also known beforehand.

It is a mitigable error not to anticipate (predict and prepare for) the vulnerabilities that come with shifts to these other interconnectivity configurations. Coping with and responding jointly to interconnectivity shifts by infrastructure operators and emergency managers for initial service restoration without (1) prior contacts, (2) repositories and pools of back-up material, equipment and facilities, and (3) the availability of robust communication channels is an known error to avoid. It is also known that all of this subsequent micro-coordination takes place against the constraining limits that the infrastructures involved necessarily focus also on their own priorities, sequences of tasks, and duties. Yes, “First get the plant up and running!,” but they may be unable to do so without helping to restore initial service provision of other infrastructures they depend upon and/or depended upon them.

Which brings us back to stabilization. This prior, jointly-undertaken preparation extends beyond the priorities or commercial interests of a single infrastructure. Managing ahead for shifting infrastructure interconnectivity is a public good in interconnected service restoration of critical infrastructures during a disaster and its unfolding. This status must be recognized as such by policymakers, emergency management officials, and senior executives in lifeline infrastructures.

Source: The above consolidates slightly edited extracts from E. Roe and P.R. Schulman (2025). The Centrality of Restoration Resilience Across Interconnected Critical Infrastructures for Emergency Management: A Framework and Key Implications. Oregon Research Institute: Springfield, OR (accessed online at https://www.ori.org/wp-content/uploads/2025/08/FinalReport_10Aug2025.pdf). Research design, references and other particulars can be found there.

The importance of career orientations in emergency planning and response

“I can’t say enough good things about planning and how important it is,” a state emergency manager in the US Pacific Northwest told us recently, “but you realize the gaps in plans when you’re dealing with such catastrophic events that we’ve dealt with in the past 18 months to two years…There’s a lot that needs to be decided on the fly because it hasn’t been planned for or it’s not going to work, the plan didn’t consider all the factors because every emergency is different”.

It is crucial to be very clear about what is being said in the above. It isn’t just the commonplace that no plan can anticipate all contingencies that arise in a disaster. It’s also that even longstanding emergency managers have not encountered some of the emergencies now underway. There are two parts to this encounter: the actual presence and type of major disaster and the career orientation of those involved.

With respect to responding to recent winter storms, a district emergency planner conceded, “We get them but we don’t get them often enough to be good at them”. “We haven’t had a really major emergency that cuts through all the lifeline [infrastructures] at the same time,” a city public works engineer with long experience told us. A water treatment plant operator, with years’ experience, hadn’t had “any [real] emergencies except for the last year with an ice storm that took out our power”.

At issue is also the career orientation and training of those doing the planning and responding. “If your specialty is putting out fires, then you’re not [equally as] good at long-range planning. . . It’s not their strength,” in the view of one interviewee . A county emergency planner and coordinator expanded: “We can’t rely on the civil defense era style of planning in which you had to have every possible scenario documented and then you end up with a thousand page plan no one reads through”. “Emergency management is an old game,” this informant added; “We still see a lot of the old guard in this. . .working against innovation even if it’s not intentional on their part”.

On the upside, change the career orientation can mean changing the presiding orientation to planning and response. “You can always tell who has had a first-responder experience” and, in that interviewee’s opinion, “the good fortune to be a first responder” was helpful. From our research perspective, those who have experience across the full cycle of infrastructure operations—from normal through disrupted and failed and into recovery and establishment of a new normal (if there is to be one)—are better positioned to assess the different interinfrastructural configurations, shifts and challenges to systemwide control variables (like electricity frequency or water-flow pressures) core to effective micro-coordination across the infrastructures during initial emergency response and service restoration as a disaster unfolds.

It is necessary to reiterate that, as shared improvisation is a core competency in immediate response and backbone service restoration, effective emergency coordination, as redefined here, requires not only the participation of emergency responders and others in the Incident Command System, but also, just as important, participation from those in the control rooms and the wraparound support staff in the backbone infrastructures—electricity, water, telecoms and roads. So too for the planning side and in changing the mix of professionals involved.

We were told by an operations planner at a major electricity transmission company that they now had a real-time engineer team working with dispatchers, undertaking real time analysis and study of outages likely to happen in the next day and even in consultation with outside utilities dependent on that supply. “Are our bottles lining up, so we’re seeing the same things?,” insisted the informant by way of avoiding unpleasant surprises.

The first line of response against cascading failure across interconnected infrastructures are those infrastructures, not the state and federal Incident Command System

Inter-infrastructural cascades may be more granular with respect to duration and open to management than assumed in formal modeling and some planning. Certainly, interviewees in our recent research on a Magnitude 9 earthquake in the Pacific Northwest described major emergencies as more punctuated than as single headlong rush of disasters.

Rapid infrastructure cascades can, of course, happen and were observed in New Orleans during and after Hurricane Katrina and, later, in northern Japan after the earthquake and tsunami of March 2011. Yet cascade models by and large assume an unmanaged, near-immediate escalation in failure probabilities and consequences across interconnected systems. Really-existing infrastructures, like water, electricity and telecommunications, are however managed. Individual infrastructures do not generally fail instantaneously (brownouts may precede blackouts, levees may seep long before failing), and the transition from normal operations to failure across systems can be a punctuated duration.

This means there is a granularity in both space and time between infrastructures in which reliability management can make a difference in the probability of failure of an individual infrastructure and failure across infrastructures. To put the point from the opposite direction, some of the recorded near misses and close calls demonstrate that operators have the time, albeit sometimes just in time, to prevent (significantly more) knock-on effects from initial disruptions or outright failures.

This means that any assumption that infrastructures are not managed in failure is highly misleading. In many cases neither infrastructure control rooms nor their reliability contributions disappear during or after a large system failure. This also means disaster response actually begins in the infrastructures.

It begins before the formal activation of the emergency management infrastructure with its Incident Command System (ICS), such as incident management teams (IMTs) and emergency operations centers (EOCs). “When the M9 hits,” said a city water distribution manager, “my group, we’re going to be the first in. . .We’re the first responders for the water system. I may even have to call someone who lives nearby and tell them to drive up to our major water tank and close the shut-off valve.”

Sources: The above is a slightly edited consolidated extract from:

E. Roe and P.R. Schulman (2025). The Centrality of Restoration Resilience Across Interconnected Critical Infrastructures for Emergency Management: A Framework and Key Implications. Oregon Research Institute: Springfield, OR (accessed online at https://www.ori.org/wp-content/uploads/2025/08/FinalReport_10Aug2025.pdf).
E. Roe and P.R. Schulman (2016). Reliability and Risk: The Challenge of Managing Interconnected Infrastructures. Stanford Business Books: Stanford CA.

Too many organizations and positions to coordinate?

I remember coming across a figure of hundreds of different organizations responsible for some part of the San Francisco Bay Delta–municipal, state, federal, private, more–and wondering how was it ever possible to coordinate all these different entities for the purposes managing the delta smelt collapse then underway. Only later in our research on a Magnitude 9 earthquake in the US Pacific Northwest did I realize I had jumped the gun and misstated the coordination problem.

When it comes to the projected M9 events, we learned Oregon and Washington State had a huge diversity of different organizational and network formats to address the events. Here is a sample of a few different formats interviewees mentioned in no order:

the Clark Regional Emergency Services Agency as a special district (and other district types, like special purpose districts); a Regional Disaster Preparedness Organization with members from multiple jurisdictions and with special committees and subcommittees; the Pacific Northwest Economic Region, a longstanding multi-state statutory entity that takes a regional economic perspective on inter-infrastructural emergencies; an intergovernmental and water coordinator funded by a consortium but located in a utility; a multi-country water suppliers forum; a citywide disaster policy council; federal defense staff embedded in a state and other emergency management units when liaising over a catastrophic event; a multi-agency coordinating group at the state level, whose members change with type of incident; an infrastructure working group and a critical infrastructure branch at the Oregon state level; the Western Region Mutual Assistance Group dealing with electricity; a disaster resilience action group across one city’s agencies; a watershed-based water providers group; a senior policy group in a large electricity transmission provider; special state task forces, e.g., to “restore power and roadways at the same time” as one state emergency put it; a water utility with a dedicated account manager in the electric utility; and a Statewide Interoperability Coordinator as well as a State Resilience Officer for the state of Oregon, among others.

Given different contingencies and local contexts, such diversity in organizational formats is what we would expect when requisite variety is key to immediate response and initial service restoration. Success again is defined as the match between then-existing task demands and then-existing response capabilities. Indeed, we would expect to find many more, real-time functioning formats, not just by location but also by type of sub-event. Ensuring local or regional requisite variety in options and understanding they necessarily include on-the-fly improvisations is not helpfully understood as first and foremost “a coordination problem.”

This also means counseling against any dismissal of such variety as “unnecessary organizational duplication” or in need of “organizational consolidation and integration.” The worst thing to do is to impose central command and control on the real-time need for requisite variety, case by case.

Another feature of having diverse organizational and network formats is they enable participants to wear different hats for different occasions–another kind of requisite variety. The department head who is also a member of the city’s emergency management working group is able to speak with his or her citywide hat on in ways not open to them as a department head only. The statewide emergency manager having responsibilities for firefighting is also the liaison on such issues in the state’s emergency coordination center; other statewide managers shift to being boots on the ground with respect to their function.

More formally and to revert to formal terminology, the single resource—in this case, the professional—provides multiple services as demands change, which in turn is crucial for meeting real-time matches in requisite variety. There are downsides, of course, to wearing multiple hats when it comes to emergency management. One interviewee noted a possible reluctance of city officials to have a new full-time emergency manager position because there were occasions when city officials considered that function to be part of their job. Another interviewee, however, thought it was a good thing that a wide range of departmental staff had emergency management duties, even if it took up a small percentage of their time.

Source: The above two sections are a slightly edited extract from E. Roe and P.R. Schulman (2025). The Centrality of Restoration Resilience Across Interconnected Critical Infrastructures for Emergency Management: A Framework and Key Implications. Oregon Research Institute: Springfield, OR (accessed online at https://www.ori.org/wp-content/uploads/2025/08/FinalReport_10Aug2025.pdf). Research design, references and other particulars can be found there.

Answers to Three Big Questions for Major Emergency Management: Why plan if it’s going to be that bad? What does success even look like then? How much pre-disaster mitigation is really needed?

1. If the disaster is going to be that devastating, why even plan?

Here is what one emergency manager said of what many of our interviewees treated as an inevitable Magnitude 9 earthquake on the Cascadia Subduction Zone off the shores of the US Pacific Northwest:

A lot of my contemporaries and planners on the emergency side have all said that the devastation is going to be so great that the ability to get resources in to help out survivors is going to be so limited that this thing is going to be so protracted that there is not going to be enough preparation in the homes of people—the individual communities are not prepared enough to last long enough and there’s going to be a lot of subsequent deaths after day-7 of the event. There’s not going to be enough response ability to come in because infrastructure is so broken. There’s no roads, no bridges, no airports, no shipping ports, no communication, no electricity, no freshwater, all that stuff. It’s going to be one of those biblical proportions kinds of disaster.
(A federal emergency manager working in the two states.)

This is not hyperbole to interviewees. An M9 Cascadia earthquake (when combined with magnitude 8.0 aftershocks and massive tsunami impacts) will be unimaginably catastrophic precisely in light of prior experience and training with lesser emergencies such as wildfires, winter storms and flooding.

Given these many uncertainties, isn’t anyone’s guess as good as the next? “Why are we even planning? It’s going to be that bad,” one interviewee reported county officials asking. It can feel like an avalanche you’re just not going to get ahead of, reflected a state senior emergency management official.

Given so much uncertainty that comes with the M9 earthquake, it is no surprise emergency managers rely on their pre-disaster plans and processes as a starting point. The importance of activating the federal an state Incident Command System (ICS) was repeatedly mentioned. This means that instead of concluding you can’t really be prepared for something as unimaginable as the unfolding M9 events, what concerns most of the emergency managers we interviewed is the lack of further preparedness in respect to what can still be prevented regardless.

Some fires are clearly preventable, e.g., through prior vegetation management. So too if pre-disaster efforts such as mitigations, two-week ready supply programs of essential materials including food and water for households, and other preparedness planning can help reduce the pressure for immediate response. Or reduce the recovery period after events through the reduction of damage that would have otherwise been incurred in the absence of such measures, these too should not be neglected.

The insight here is that, for those interviewed, the real-time unpredictability and unexpected contingencies ahead in the M9 events carry their own information about interconnected infrastructure systems in failure and that information can be useful for managing or coping ahead. This is especially true for those real-time professionals whose core competencies revolve around systemwide failures: They are likely to know beforehand something about how the system in failure will affect other interconnected infrastructures.

Interviewees stressed the need to focus on infrastructure components and facilities they know will fail in the M9 events, whether or not the knowledge has been formalized into an agency’s risk register and risk assessments. That there are no guarantees pre-disaster efforts will actually mitigate is beside the point for experienced emergency managers who have witnessed or been directly involved in disasters elsewhere.

They have seen how better pre-disaster efforts would have made a difference there. That is their job. One core competency of emergency managers is to identify pre-disaster opportunities—including new options and strategies for increased requisite variety to improve real-time disaster response, and not just in their own infrastructures. Seismically strengthening a water infrastructure, as one interviewee confirmed, would better inform emergency planning and projects for the road and wastewater infrastructures adjacent to the water lines.

2. What does success even look like in a disaster?

“What does success look like?” a senior state emergency manager asked, and answered from his experience: “Success in every disaster is that you didn’t have to get improvisational immediately. You can rely on prior relationships and set up a framework for improvisation and creativity.” Why? Because contingencies that can’t be planned for require improvisation. “I was lucky,” reported a statewide emergency manager involved in an ice storm that happened during the Covid-19 lockdown as “my neighborhood was only out of power for six hours. . .”

Contingency—happenstance, accident, chance, coincidence—will matter even more for the Magnitude 9 earthquake in the US Pacific Northwest: “What side of the river will I be on?” replied an infrastructure operator to our question of what would be a performance standard in the M9 events. Being at work, holding water in the main reservoirs, and getting some of the wells up and running would be a success, the interviewee added. Opening a road from point A to point B, that’s a success, to paraphrase a state emergency manager for highways. “Doing the best with what we have” was the frequent response. “Being here and do as best as we can, would be considered a win,” put one infrastructure operator. But how can we know that the “best” was done? The best with respect to what?

“Doing their best” has very specific meaning by way of being a performance standard for immediate response and initial service restoration: It is when responders find or create a match between task environment demands and response capabilities. More formally: Effective performance can be understood as the contingent correspondence of task environment demands and the response capabilities (resources, skills, options, strategies) to meet those demands in real time. The term, “contingent,” conveys the sense that the conjunction of capabilities and demands can be fortuitous and is by no means assured through pre-disaster mitigations, formal preparedness plans, and other agency arrangements, like mutual aid agreements.

Seeking requisite variety in matching unpredictable/uncontrollable task demands with highly contingent resource capabilities is, we argue, a strategy and performance standard appropriate for both immediate emergency response and initial service restoration. This is because skills in assembling options under highly volatile conditions remain central to enabling joint improvisations across the interconnected critical infrastructures. When assembled, these are accomplishments in real time, and measures of success (or not) later on should reflect that.

All easier said than done, of course. The fact that there can be no guarantee the shared infrastructural improvisations—these impromptu but major interconnections between and among infrastructures—will be effective means there is a premium placed on people already skilled in improvisation. This is why newer infrastructure employees and emergency staff may well not (yet) have those skills. It also accounts for those much-remarked-upon people, outside the emergency management infrastructure, who step forward during a disaster and do the needful because they already have life skills in working with what is at hand.

An emergency planner and coordinator remarked, “I think what makes a good emergency manager is you feel uncomfortable being off-balance. . .That’s one of the reasons I was drawn to the field. When nobody has the answer that’s when I feel most capable in my job”. Activation of the Incident Command System (ICS) helps provide some structure to trying to meet the challenge. What achieving requisite variety provides is time-, duration- and site-specific organization that can add up to some semblance of stability. As one state coordinator put it: “My responsibility on the response side leading into recovery is stabilization, [which means] this is not going to get any worse. We’ve restored a foundational level of service, whether that is permanent or temporary. . .”

3. How much pre-disaster mitigation is needed?

It is not possible to answer the question, How much mitigation of infrastructures is needed before a disaster? “When is ‘resilient-enough’ enough?,” asked an interviewee. Maybe we don’t have to fix every road before the earthquake, another put it. While understandable sentiments, no amount of money or political-will beforehand would be enough to dislodge the central and strategic role of improvisation in the unfolding events of a major disaster like an M9 earthquake in the US Pacific Northwest.

These inter-infrastructural improvisations on the spot end “saving the asset” (the term of one of our interviewees), which in our terminology was preventing failure in critical services interconnected through a shared substation or tower. Other examples of what we are calling joint improvisations during emergencies were also mentioned in our interviews, the point being in the words of one county emergency planner and emergency coordinator: “There’s a lot of improv that has to happen here”.

The crucial point of joint improvisions involving emergency responders and infrastructure operators is that they come in unpredictable forms contingent on then-specific demands and then-existing capabilities. One state coordinator involved in communications management during emergencies told us about convening an online group of competing companies and infrastructure providers:

During a winter storm we had a utility or provider say we’ve got fiber cuts in this area, we don’t have the fiber to replace it in that area, our resources are in this other area—that allowed us to look at the group and say now is the time for some teamwork: Can anyone else solve that problem and be a good team member? And we’ve seen a lot of that sort of problem-solving manifest among the agencies with very little input from us. Another example might be a cellular carrier who is a competitor of another carrier going “Hey, we’re going to fill our generator, can we top off your fuel tank while we’re up there?. . .But I don’t think [those kinds of cooperation] would occur if we didn’t coordinate it and get everybody on the same call and provide a platform for them to kind of air those sorts of things.

To characterize these one-off improvisations, like topping off a fuel tank, as incidental or side work or what mates just do for each other, is to miss entirely the point that they are essential for professionals undertaking effective emergency response. When it comes to immediate response under M9 conditions, there is no workaround for improvisation. At the point of its undertaking, the joint improvisation is the manifest interconnectivity between the infrastructures involved under conditions of otherwise failure.

Source: The above are slightly edited extracts from E. Roe and P.R. Schulman (2025). The Centrality of Restoration Resilience Across Interconnected Critical Infrastructures for Emergency Management: A Framework and Key Implications. Oregon Research Institute: Springfield, OR (accessed online at https://www.ori.org/wp-content/uploads/2025/08/FinalReport_10Aug2025.pdf). Research design, references and other particulars can be found there.

What does success look like in a disaster?

Contingency—happenstance, accident, chance, coincidence—will mater even more for the Magnitude 9 earthquake in the US Pacific Northwest: “What side of the river will I be on?” replied an infrastructure operator to our question of what would be a performance standard in the M9 events. Being at work, holding water in the main reservoirs, and getting some of the wells up and running would be a success, the interviewee added. Opening a road from point A to point B, that’s a success, to paraphrase a state emergency manager for highways. “Doing the best with what we have” was the frequent response. “Being here and do as best as we can, would be considered a win,” put one infrastructure operator. But how can we know that the “best” was done? The best with respect to what?

All easier said than done, of course. The fact that there can be no guarantee the shared infrastructural improvisations—these impromptu but major interconnections between and among infrastructures—will be effective means there is a premium placed on people already skilled in improvisation. This is why newer infrastructure employees and emergency staff may well not (yet) have those skills. It accounts for the much-remarked-upon people, outside the emergency management infrastructure, whose life skills in working with what is at hand step forward and do the needful in a disaster.

An emergency planner and coordinator remarked, “I think what makes a good emergency manager is you feel uncomfortable being off-balance. . .That’s one of the reasons I was drawn to the field. When nobody has the answer that’s when I feel most capable in my job”. Activation of the Incident Command System helps provide some structure to trying to meet the challenge. What achieving requisite variety provides is time-, duration- and site-specific organization that can add up to some semblance of stability. As one state coordinator put it: “My responsibility on the response side leading into recovery is stabilization, [which means] this is not going to get any worse. We’ve restored a foundational level of service, whether that is permanent or temporary. . .”

Which “gaps in communication” during emergency management?

Much has been written in the literature on the importance of communication in emergency management, and our interviewees, like others, report communication gaps in their after-action reports.

The value-added in our Pacific Northwest research on a Magnitude 9 earthquake there is to underscore how gaps in communications arise because the interinfrastructural connectivity shift spatially and temporally as disaster events unfold. In these cases, the forms and flows of communications follow from the interconnectivities and shifts when those configurations change. Here it isn’t only that communications between and among emergency responders and infrastructure operations establish a follow-on interconnectivity; as their prior interconnectivities change and shift, so do communications change in form and content. This has an important implication for the diversity of communication means, as we shall see.

Start with an interinfrastructural example up to failure and immediate emergency response:

· Vessels come into a port and shipments are off-loaded there onto truck and rail for onward transport (sequential interconnectivity with serial dependencies);

· If there is a major service disruption, the port may take a more active role in coordinating which vessels have priority, how shipments are off-loaded and stored temporarily, and the modes of transporting onward (mediated interconnectivity by the port as a focal infrastructure);

· If the disaster is more extensive, the vessels may have to coordinate from ship pilot to ship pilot, without the assistance of port authorities or others (reciprocal interdependence); and

· The Incident Command System set up immediately after the disaster may make coordinating the waterways for emergency uses as one of its first priorities (pooled interconnectivity centered around the focal ICS or a vessel traffic control unit).

Although a critical infrastructure in normal operations differs dramatically when in failure, sequential or serial interconnectivity at the infrastructure systemwide level doesn’t disappear, nor would it the M9 events. Interconnectivity configurations, however, would most assuredly change (including those for reciprocal, mediated and pooled).

For example, restoration of the electric transmission grid would start sequentially with the 500kV lines, then the 230kV, then the 115kV. An emergency manager for an electric utility said that M9 outages would initially be treated like any other outage, following a multi-step process to bringing back the system. We were also told by a statewide emergency manager that response to a major ice storm consisted of first closing the affected state roads, followed by power companies coming in to move the downed lines, and then crews coming in to remove the fallen trees.

In other words, the shifts in configurations and their elements have a major impact on the pace and composition of emergency management activities. This includes interinfrastructural communications. That is why diversity (more formally, requisite variety) in mechanisms for communications under dynamic conditions—land lines, cellphones, satellite phones, CB radio, Starlink and more–is so important at these times.

Official cellphone capabilities dropped in one area as a result of an ice storm event, while personal internet hotspots remained available for key personnel during the same event. As interconnectivities shifted, so did the mechanisms for joint communication. The need for such requisite variety explains why interoperability among communications systems, as important as it is, cannot substitute for that needed diversity in disaster communication.

How much pre-disaster mitigation is needed?

Improvisational behavior is mentioned repeatedly in the emergency management literature. Our research findings add value by stressing the importance here of an extremely important subsample of inter-infrastructural improvisations. An impromptu berm is built around a substation or a fire break bulldozed around a communications tower, both of which are critical. Why? Because of the collocation of critical components that supported, in the case of the high-valued tower, state police, forest service and transportation. These inter-infrastructural improvisations on the spot end “saving the asset” (the term of one of our interviewees), which in our terminology was preventing failure in critical services interconnected through a shared substation or tower. Other examples of what we are calling joint improvisations during emergencies were also mentioned in our interviews, the point being in the words of one county emergency planner and emergency coordinator: “There’s a lot of improv that has to happen here”.

During a winter storm we had a utility or provider say we’ve got fiber cuts in this area, we don’t have the fiber to replace it in that area, our resources are in this other area—that allowed us to look at the group and say now is the time for some teamwork: Can anyone else solve that problem and be a good team member? And we’ve seen a lot of that sort of problem-solving manifest among the agencies with very little input from us. Another example might be a cellular carrier who is a competitor of another carrier going “Hey, we’re going to fill our generator, can we top off your fuel tank while we’re up there?. . .But I don’t think [those kinds of cooperation] would occur if we didn’t coordinate it and get everybody on the same call and provide a platform for them to kind of air those sorts of things.

To characterize these one-off improvisations, like topping off a fuel tank, as incidental or side work or what mates just do for each other, is to miss entirely the point that they are essential for professionals undertaking effective emergency response. When it comes to immediate response under M9 conditions, there is no workaround for improvisation. At the point of its undertaking, the joint improvisation is the manifest interconnectivity between the infrastructures involved under conditions of otherwise failure.