Preparing Chemical Facilities for NATECH Events

When natural disasters strike industrial facilities, the results can be catastrophic — and most companies aren't ready. In this episode of Process Safety with Trish and Traci, Trish Kerin and Traci Purdum explore NATECH events, where natural hazards collide with industrial risk. Drawing on real incidents including Arkema in Crosby, Texas, BioLab in Lake Charles and the Fukushima disaster, they examine why facilities consistently underestimate natural hazard risk, how to build truly complete ride-out and recovery plans, and why traditional PHAs fall short for NATECH scenarios. Kerin's bottom line: assume the event will happen, and prepare accordingly.

Transcript

Welcome to Process Safety with Trish and Traci, the podcast that aims to share insights from past incidents to help avoid future events. Please subscribe to this free, award-winning podcast on your favorite platform so you can continue learning with Trish and me in this series. I'm Traci Purdum, editor in chief of Chemical Processing, and with me, as always, is the one and only Trish Kerin, director of Lead Like Kerin. Hey, Trish.

Trish: Hey, Traci, how are you doing today?

Traci: I'm doing great, and I'm looking forward to our conversation today. We're tackling NATECH events — those moments when nature and industrial hazards collide. But before we get into the details, I wanted to hear from you about a weather event that's really put the risks of our industry into perspective for you. Putting you on the spot.

Trish: Oh, you are putting me on the spot with that one. I'll go with something from many years ago when I was a safety manager and a city in Australia suffered flooding well beyond anything we'd ever seen before. Weirs and dams failed simultaneously due to the extra water load. The city of Brisbane saw enormous flooding. The Brisbane River runs right through the middle of the city and through a whole lot of suburbs as well, and there was extensive damage. What really sticks in my mind personally was that rain event because at the time I was a safety manager in an LPG business. Our facilities had suffered quite substantially — we had pumps underwater, cylinders floating away. But I remember the TV footage: floating down the middle of the Brisbane River, heading toward the central business district, was a massive LPG bullet venting LPG as it went happily sailing down this flooded river, venting LPG as it went. It's one of those moments where you think, we didn't expect that to happen. It was a substantial flood, and that image has a personal attachment for me. The leadership team and I sat there watching it, making sure that wasn't our tank and that our tanks weren't floating anywhere. Then, just to get back into business, we had to inspect every customer facility where we were providing gas in cylinders. Every cylinder had to be inspected before it could be refilled and before customers could get their gas back on. These NATECH events can impact so many different parts of society and our industry as well.

Why Are Natural Disasters Underestimated?

Traci: And I think you mentioned it was unexpected — not just seeing that tank go down the river, but the unexpected ramifications. While I was prepping for this episode, I came across stats from a recent FM Global survey taken after hurricane season. It found that 62% of companies impacted by hurricanes admitted they were not completely prepared. What is it about natural hazard risk that causes even experienced facilities to underestimate it?

Trish: I think part of it is we just don't think it's going to happen to us. And that doesn't make a lot of sense when you consider hurricane season specifically. There are hurricanes that form in the Gulf every hurricane season — that's why it's called hurricane season. What's more rare is when they make landfall where chemical facilities are located. Obviously Lake Charles comes to mind. When Lake Charles was hit as opposed to a landfall in a purely residential or rural area, the damage profile is very different. In Australia, where I live, part of our country is subjected to cyclone season every year. Most of that area isn't heavily populated where cyclones typically form, but we have had instances where cities have been hit directly. In fact, back in 1974, the city of Darwin — a capital city of one of our states — was essentially decimated by a cyclone on Christmas Day, which falls within our November-to-March cyclone season. So we tend to think it'll happen somewhere else. We've got our emergency plans, we've got everything sorted. But the challenge with NATECH-type events is that we cannot prevent them. There is nothing we can do to prevent floodwaters or a hurricane or a cyclone or an earthquake or a wildfire. All we can do is mitigate. And one of the issues is that some of the help we look for — community firefighters, police directing traffic — those responders are busy serving the broader community. They don't have time to serve us anymore at that point. So one of the under-preparedness issues is that we don't go into these events ready to save ourselves if we need to.

Real-World Disaster Examples

Traci: You brought up a lot of good examples and I want to discuss a couple of specific, recent NATECH events and the facilities impacted. The first is the Arkema facility in Crosby, Texas, where organic peroxides ignited after refrigeration systems failed when a hurricane disabled power. What does that incident teach us?

Trish: That incident really teaches us a lot about how we prepare. They did start to manage the situation — they moved all the peroxides to refrigerated containers at higher elevations to reduce flood exposure. But the storm kept stalling. That particular hurricane hung just off the edge of Houston and created an enormous rain dump that just would not leave. It flooded far beyond what anyone had anticipated. They had moved materials to higher ground, but in the end, even that higher ground wasn't high enough. One of the challenges we face with NATECH is that we can't just dismiss certain scenarios as unlikely. We might say, well, we could get a 3-foot flood, but we're not likely to get a 6- or 8-foot flood. They did get massive floods. We sometimes, in a bit of wishful thinking, tell ourselves it won't be that bad — yes, we might flood, but not that much; we'll be OK.

Traci: I think it's the danger of the so-called 100-year storm. You hear about the 100-year storm and think, well, we just had one, so we won't have another one the following year. And there's a trick in that — that's not really what the 100-year storm designation means.

Trish: Right, that's not how statistics actually work. You could have three 100-year storms in three consecutive days — that is entirely possible. We're also now starting to see the emergence of the 500-year storm. And you could still have a couple of 500-year storms in the same week because our weather patterns and our climate are changing. We are seeing more and more extreme weather events at magnitudes we don't have recorded history for. But just because we've had a couple of 100-year storms in a row doesn't mean we're not going to have another one — and it doesn't mean we are going to have another one either. That's the problem with statistics. They don't tell you when something is going to happen. It's all about the odds, and even at very low odds, there's still a chance it can happen.

Traci: You mentioned Lake Charles a little while ago, and the 2020 event — Hurricane Laura. One facility, BioLab, had a chlorine release and a major fire after backup generators and suppression systems were compromised. Let's talk about that.

Trish: Yes. They had backup systems in place, but from memory, rainwater leaked in and shorted out the supply to some of those systems. A simple failure led to the systems they thought they had being inadequate. We need to remember it's not only about having the equipment — we need to make sure it currently works. When we talk about safety-critical elements, a key attribute we focus on is survivability: Is the control actually going to survive the initiating event? If it's a mitigation control and it does not survive the initiating event, it is not a control, because it will not work when you need it. We need to make sure these systems are understood, implemented and tested. And we need to ask how we are protecting those systems — including the external structure where something is stored — to ensure the reliability and survivability of that control when it matters.

Redundancies and Better Planning

Traci: Talking about redundancies — are there smart ways to plan for that a little better?

Trish: There are. When it comes to floodwater, elevation is key. Don't locate emergency systems at the low point or even a midpoint in your facility. Put them up on a plinth so they are protected from floodwater. Make sure the cabinets and outer buildings around critical equipment are suitably maintained. You can't let the roof rot a little, let a crack open up, let wildlife get in and create a gap. You're protecting the equipment inside and still doing all your testing, but if you're no longer protecting the structure keeping that equipment functional, you've lost your safeguard. You've got to make sure your maintenance system addresses the adequacy of the entire process — not just what you think is the critical part. Because if you lose integrity around the structure keeping something safe, you don't have anything keeping you safe.

Traci: You mentioned wildlife, and I did a story a while back interviewing people who lived through Hurricane Rita at the PPG Industries facility in Lake Charles. The plant credited the three P's — preparation, perseverance and powering up again — for getting through it. But even they acknowledged gaps: not enough food, no bedding for ride-out crews, unexpected wildlife inside the plant. What does a truly complete ride-out preparedness plan look like, and what do people most often overlook?

Trish: For events like hurricanes or major winter storms, we know these things are coming. We can see hurricanes forming and gaining intensity at least a couple of days out. We know through weather forecasting when a major winter storm is on the way. So in those types of incidents we actually have the opportunity to plan properly. That means asking: How long do we need to ride this out? How many people do we need to keep on site? Then we can determine what's needed — packaged water, not tap water since you may lose that; enough food; beds, bedding and clothing changes to maintain hygiene; and even chemical toilets if sanitation systems go down. If you've got three or four people riding out a storm to keep a plant safe, you've got to make sure they don't get sick too. Think about it as: if this number of people were completely isolated for X number of days, what do we need to cater for? Bring all of those resources in. The bigger challenge is earthquakes, because we generally don't get warning. When it hits, it hits. You can't keep water or food indefinitely, so you need a system — just like preventive maintenance — that periodically rotates those supplies so they remain suitable if you ever need to rely on them.

Not Just Hurricanes — Other Extreme Events

Traci: That's a great segue — NATECH conversations often center on hurricanes, but there are other extreme events. We're coming up on the 15-year anniversary of Fukushima, and I also want to touch on winter weather and other non-hurricane events and how we can better prepare.

Trish: Yes, we are coming up on the anniversary of the Tohoku earthquake, which then triggered the tsunami. Many people will remember the Fukushima Daiichi nuclear power plant was struck by that tsunami. Interestingly, they did have a seawall designed for tsunamis — but it wasn't tall enough for the wave they received, because they didn't think a wave that large was possible. The Tohoku earthquake was a magnitude 9.0 on the Richter scale — it was such a massive event it actually shifted the Earth's axis. There were many other impacts beyond Fukushima as well. For example, a refinery in Chiba prefecture in Japan had 17 LPG spheres and lost every single one in the resulting fire. All sorts of events occurred not just at Fukushima but across the region as a result of both the earthquake and the subsequent tsunami. Depending on where you are in the world, earthquakes are more or less likely. The West Coast all the way up through Alaska is earthquake-prone, so a significant event on the San Andreas Fault could create substantial disruption, damage and loss of life. When it comes to winterization, that's a lot easier to plan for. We know it's going to get cold, and we have a reasonable idea of how cold. It's about making sure we've taken all appropriate precautions — draining water from systems we don't need water in to prevent cracked pipes, putting heating blankets on sections of the plant that need to stay at certain temperatures to continue operating. We have well-established winterization techniques; the key is maintaining the discipline to implement them.

Help May Not Come — Prepare for That

Traci: I want to circle back to something you mentioned earlier. NATECH events often affect an entire region simultaneously, so contractors and emergency responders will all be dealing with their own impacts at the same time. How should facilities plan for a NATECH response when broader community support is also compromised? And how does crisis communication fit into that picture?

Trish: What we need to be doing is ensuring we have our own backup systems, recognizing that it's entirely possible the state-supplied communication systems will go down. We know this will happen — they will fail at some point during an emergency, either through line failure or simply the overwhelming of network capacity. So how are you going to initiate and manage communication within your own facility? It might mean having fixed-line communication between control rooms, for example, rather than relying on radio networks that could be overwhelmed or telephone networks that could go down. It's about understanding your own communication needs and having them covered. It also means having clear, developed response plans. If you've got a fire and the local fire brigade cannot reach you, how do you deal with that? What minimum firefighting capability must you have on site? Because you can't guarantee the fire brigade will get to you — and the goal may not be to extinguish, it may simply be to control the burn. We need to be realistic: in extreme events, the response and outcome may look quite different from normal. We're not necessarily looking at recovery — we're looking at just coping until we can eventually get to recovery. That requires some hard conversations about what we can accept. Your insurers will obviously also want to be part of those conversations, since they'll be helping foot the bill. FM Global does a lot of work in this space, helping companies prepare for different natural event scenarios. Regardless of who your insurer is, they have significant experience dealing with natural hazard events — that's core to their business, whereas an individual company may not have that experience. Don't be afraid to talk to your insurers and ask for their advice. Chances are they'll have good ideas you haven't thought of.

Traci: I actually visited FM Global headquarters, and they don't just rely on hearsay — they practice it. I got to feed 2-by-4s into a machine that launched them into windows and shingles to test structural integrity. They fling bowling balls at windows. They set boats on fire in large warehouses to verify that sprinkler systems maintain their integrity under high heat and flames. So yes, I agree — rely on them to better understand the risks.

Trish: Absolutely.

Traci: All right — so we've ridden out the storm and made it through. Now we're into one of the most dangerous phases: coming back online. What inspection and verification protocols should be in place before operations resume?

Trish: That is where we really need to conduct detailed pre-startup safety reviews. We need to check the integrity of our equipment and the functionality of our safety-critical controls, because they may have been damaged during the event. Those integrity inspections need to cover not only primary containment but structural elements as well — for example, are tank foundations still suitable for handling the load when we refill with product, or has there been weakening that could cause collapse? We need structural integrity reviews as well as equipment integrity reviews: pipelines, tanks, vessels, pumps, emergency systems, sprinklers, detection systems. And for safety-critical controls and instrumented systems, we need to conduct function testing — we need to prove those systems work as intended before startup. Making sure you have fully documented pre-startup safety review checklists for various parts of the plant is absolutely critical. Do not take shortcuts in the effort to get back up faster. The worst thing you can do is rush and miss a critical fault that causes another incident or takes you offline again. Take your time, do every check, assure yourself positively that your system is safe to operate — then you can start it up again.

Risk Assessment

Traci: Looking ahead and getting it on paper — do you think existing PHAs are adequate for evaluating NATECH risk, or do we need new tools designed specifically for this threat category?

Trish: There have actually been some NATECH risk assessment tools developed. The European Union has certainly led the way, and one of the core reasons is the way traditional PHAs handle likelihood. When we go through a standard PHA, we look at consequence and likelihood. The problem with applying likelihood to NATECH events is that we actually just have to assume the event will happen — and then make sure we have mitigation in place to minimize the consequence. We can't stop natural events; they will happen. Our traditional PHAs are not structured that way. They look at risk and conclude, yes, the consequence may be relatively high, but the likelihood is very, very low, so we don't need to do anything more. That approach leads you to miss implementing key controls. That's why dedicated NATECH risk assessments have been developed. Researchers Elisabeth Krausmann and Ernesto Salzano have done significant work in this area, and there are books and technical papers on NATECH risk assessment methodology well worth incorporating into your planning.

Traci: I'll look those up and link them in the transcript so folks can find them easily. Now, putting the advice hat on you — what one piece of advice would you give to someone just starting to build or refresh their preparedness program today?

Trish: Regardless of how unlikely it seems, it can happen.

Traci: Anything you want to add that we didn't touch on?

Trish: I think we covered most of it, but I'd re-emphasize: go out and look at your response systems. If you lost the ability to have the fire brigade show up. If you lost communication. If your workers were stranded on site — could you cope? Do you have enough in place? That's really what you need to focus on and understand, and make sure you can adequately plan for it.

Traci: Well, Trish, you've pointed out many times that we can't stop these events, but we can respond to them better — and you've helped our audience do exactly that. Unfortunate events happen all over the world, and we will be here to discuss and learn from them. Subscribe to this free podcast so you can stay on top of best practices. You can also visit us at ChemicalProcessing.com for more tools and resources aimed at helping you run efficient and safe facilities. On behalf of Trish, I'm Traci, and this is Process Safety with Trish and Traci. Thanks again, Trish.

Trish: Stay safe.