You describe yourself as a ‘disasterologist’. What exactly do you do and how did your role evolve into what it is now?

My actual job title is Director of Maritime Operations for an international humanitarian organization called Global Support and Development, but what I mean by disasterologist is that I study disasters. I think that’s done by trying to find answers to two questions: why do bad things happen, and what should we do about them? That’s a joint exercise in academic and operational problem solving, focused on the impacts of hazards and the disruption to the systems and networks upon which communities rely.

I think my career has been like a rock, skipping from right-place-right-time to right-place-right-time. I’ve been very fortunate to work with and learn from some incredible people in the trenches of some extremely complex problems. I think, for as long as I’ve been in this space, the roles have all been about finding answers to those two foundational questions – I’ve just been lucky enough to find a bunch of unique opportunities and approaches to attacking the problem.

You spent six years working in various positions for the US Coast Guard (USCG). What did you learn during that time that has changed your outlook on disaster management?

There are two phrases I picked up from my time at the National Search and Rescue School. The first is that search and rescue (SAR) is mostly search, which is a terrific sentiment relating to the work that’s required to be successful, operationally speaking. The second phrase is that we sometimes have to make decisions of 100 per cent consequence based on 50 per cent of the information.

During some SAR cases, we wouldn’t receive all of the information we’d want prior to starting a search. Sometimes, you hear ‘mayday’ on the radio and that’s the best you’ve got. No position, no number of people in distress, no description of the vessel or the problem. You just have the radio towers that picked up the call and that’s it. Within uncertainty, SAR operations require you to make the best possible decisions based on what you know.

It’s all about maximizing the information you can get a hold of – so in this situation above, we would identify which towers did or did not pick up that mayday call, and put out a broadcast to find out if any other sailors or aircrew heard the broadcast. If any aviation assets are in the area, we would ask them to overfly the area to see if they can spot a vessel in distress. Information gathering is key.

The cost of searches adds up. How much is it to launch a SAR operation?

Weirdly, I was once given the task of finding this out! We were the subject of some hoax calls, and eventually caught the perpetrator and took him to court in an effort to recover some of the costs we were incurring. If you add up the cost of two aviation assets, a boat, and the personnel coordinating the rescue, plus fuel, kit, maintenance, etc, it actually came to around US$250,000. The big issue, though, is not the financial burden – it’s the fact that we are spending our time looking for someone who isn’t there, potentially not being able to save the life of someone who does need us.

Is there a ‘perfect solution’ in SAR asset management?

I think people who have spent time in the SAR world are really good at identifying what ‘good enough’ looks like. We talk about ‘analysis paralysis’, which is where you spend so much time trying to develop the perfect solution that you lose your opportunity to enact any of the solutions that would have been good enough. In SAR, you’re very aware that the clock is ticking. For some, that stress weighs heavy, but if you have a good relationship with that stress, a sense of urgency can become the norm. Good enough can be good enough.

In SAR, you’re very aware that the clock is ticking

Speaking of stress, that’s another thing I think I picked up from my time in the USCG. People get into fields like SAR, emergency service, humanitarian aid, and disaster management because they want to help people. Sometimes, it’s really hard to turn your brain off, especially when you’ve spent some intimate time with someone’s worst moments. I learned how to disconnect when it was time to disconnect and I think that’s served me well as I’ve gotten into situations with a longer half-life than SAR operations. If you’re not able to turn your attention away from how rotten the world can be, disaster management can be brutal and unforgiving.

Coordination of assets is so key to effective SAR operations. What key attributes and technology are changing SAR asset management operations for the better?

Data and context. Data and context. Data and context. There are tons of new and innovative ways to collect and communicate data … but innovative doesn’t always mean useful. We see this a lot with folks who dip their toes into uncrewed aerial systems (UAS) during disasters. Sure, they can go collect imagery … but sometimes people don’t understand what problem they’re trying to solve. In the responses to Hurricanes Harvey and Dorian, we had UAS groups

I think there are tons of gains to be made in the coordination of non-traditional assets

show up saying they’ve collected some X-hundred gigabytes of imagery data that nobody asked for. Sweet. What do you want me to do with it? What’s the context of this data set? What question is it helping us answer?

I think there are tons of gains to be made in the coordination of non-traditional assets: the good Samaritans, the spontaneous responders, the drone hobbyists. The people who show up without any training or experience, but they show up. Tons of apps, mostly communication apps, get used here. How quickly can we give these sorts of resources the background, situational awareness, and instruction they need to be safe and effective?

On this topic, how can you harness the will of people to help effectively?

This is a tricky one. For instance, in Hurricane Harvey in 2017 [in Texas and Louisiana], I was working coordinating SAR assets. In the aftermath of the disaster, there were what we called ‘the Cajun Navy’ – groups of local people with boats or other vehicles that were helping retrieve people from flooded areas. But – and this is a big but – no one knew they were doing it. So we would get a call saying a group needed evacuating from a house, and when the helicopter reached the domicile, there was no one there because they’d already been picked up by the local teams. So we come back to this problem of time being wasted searching for someone that wasn’t there – communication and information gathering. It needs to be coordinated.

So, who is coordinating assets in this situation? Is there one overarching leader that needs to be working with everyone, whether military or civilian, to ensure that assets are being used effectively?

We’ve got the Incident Command System, which is part of the National Incident Management System. There are 16 emergency support functions within the system, of which, emergency SAR is function nine. One person (me) is placed in charge of all things SAR in any given scenario – wildfire, flood, etc. So, anyone with SAR assets comes under my jurisdiction and is managed by me.

Adaptability is key, and being able to identify areas in the system where there is some flexibility is important

While this system has a great purpose, it is not without its flaws – it can be bureaucratic and cumbersome. Adaptability is key, and being able to identify areas in the system where there is some flexibility is important. Of course, this does mean that sometimes people without the right experience see situations where the protocol may not be followed to the letter, and then think it is acceptable to do so, which can cause issues later on. But if you’ve got sufficient expertise, then knowing where you can make your own mind up and do the best in a given situation without following the rules to the letter, then that’s the reality of an ongoing SAR scenario.

Wildfires are becoming more common, more severe, and further reaching. Is the world prepared for what is going to happen? These kinds of disasters are, unlike some others, predictable to some extent. So what should areas prone to wildfires be doing to prepare?

This is a really interesting question. The cheeky answer is that bad things keep happening, so clearly we’re not prepared. Also, there’s probably an academic argument against the predictability of wildfires as a hazard. Sure, they’re more frequent and we ought to expect them, but it’s not like we have a schedule and list of addresses.

There’s a lot of really good, and interesting, work being done with respect to values-based emergency management (VBEM). Perhaps oversimplified, VBEM says that we accept risk in order to be near the things we value. I knowingly accept some risk to live in a coastal area in the mid-Atlantic USA, fully aware of my exposure to hurricanes and floods. It’s worth it to me. I sail, I surf, I paddleboard, and my family’s proximity and access to the water is a major reason why we live where we live.

VBEM outlines why people may choose to live where they live and the foundational point is based on the informed acceptance of risk. But risk isn’t fixed – it’s dynamic and based on more variables than I could list. As hazard profiles change and hazard exposure increases, risk increases. Developers have figured this out. Insurance companies have figured this out. Do people who live in wildfire-prone areas totally understand the dynamic changes to physical, financial, and social impacts of wildfire risk? Debatable. Better preparedness starts with better risk communication and better risk communication starts with better data.

How can emergency services be better equipped and prepared to respond to incidents like wildfires?

Risk communication is where we have the most room to improve, across all hazards. People make decisions based on their understanding of the information that’s available when it’s time to make a decision. Within that, there’s a ton of opportunity to make a bad decision.

Data that feeds risk information needs to be accurate, timely, and considerate of appropriate context. It needs to reach everyone it needs to reach. The people who need the risk information need to believe it, which means people need to trust the information source. If action is necessary, people need to understand their options, if the options are even realistic.

Speaking of information sources, and their uses, what do you see as the most useful way in which artificial intelligence (AI) can be used in SAR scenarios?

I think that this links back to the use of UAS, particularly by civilians. The most useful application of AI is in change detection – ie, analyzing images from before and after a disaster like a hurricane. This can give rescuers valuable information about where in an area has been particularly affected, and what assets could be deployed to be most effective in helping people there. So drones can be great at capturing the ‘after’ information, but if we have nothing to compare it with, then its use really is very limited, not telling us anything we didn’t already know.

And then we’ve got the issue of drones being in crowded airspace following a disaster. A civilian using a drone to be ‘helpful’ may be doing the opposite – we can’t fly a helicopter in airspace where there’s an unknown drone being operated, so we can’t be useful there.

Proactive communication is 10 times more effective than reactive communication

Proactive communication is 10 times more effective than reactive communication, so we need to be engaging with organizations before a disaster occurs to make sure that everyone knows how they can be most effective should the worst happen.