Resilient by necessity: What happens when everything is down?

When power, communications and access collapse following a natural disaster, infrastructure delivery is stripped back to first principles. What remains reveals how systems really work under pressure

info icon Cannot load this file

0:00 / 0:00
Nexus Antoinette Krause-Frances Neeson-Luke Parker_1380 X 862 px_GreenBG Hero.jpg
Cannot load photo
There was an error loading this image

In brief

  • When extreme weather hits, infrastructure does not fail in isolation. Entire lifeline systems break down at once.
  • Recovery depends less on perfect plans and more on trust, local knowledge and practical decisions made under pressure.

The first thing that disappears after a major cyclone is communication.

Power lines fall. Mobile towers fail. Internet connections vanish. Roads are cut by floodwater and landslides. Communities that were connected hours earlier suddenly find themselves isolated.

When Cyclone Seroja struck Western Australia’s Mid-West in April 2021, the scale of disruption was immediate. More than 26,000 households lost power. Nearly 200 mobile phone towers went offline. Across a region spanning more than 130,000 square kilometres, towns that rarely experienced cyclones were left without the basic systems that modern infrastructure depends on.

For the engineers tasked with helping restore those lifelines, even sending a fee proposal became an exercise in improvisation. With no internet and only intermittent mobile reception, the proposal was delivered by text message.

Across the Tasman Sea, Cyclone Gabrielle delivered a similar blow to New Zealand’s North Island in 2023. The damage to critical infrastructure, including bridges and roads, was both widespread and devastating.

Mobile phone coverage in parts of Tairāwhiti (Gisborne) disappeared for days. Communities could only be contacted by knocking on doors. Shops remained open but payments had to be made in cash because electronic systems had failed.

In moments like these, infrastructure delivery is stripped back to first principles. The systems that normally govern planning, procurement and design suddenly fall away. What remains is something far more fundamental. Trust, local knowledge and the ability to make practical decisions when perfect information does not exist.

When everything goes dark

The first realisation in the hours after a cyclone or a natural disaster is not just that infrastructure is damaged. It is that everything is connected.

  • Lose power and communications go with it
  • Lose communications and coordination slows to a crawl
  • Lose roads and even the simplest repairs become logistically complex

What looks like a series of individual failures is, in reality, a single system breaking apart.

“You don’t realise how dependent everything is until it all stops at once. That’s when you see how the system really works.”

Luke Parker
Geotechnical Team Leader, GHD

This is what disaster recovery really looks like. Not a neat sequence of damage assessments and design responses, but what teams on the ground often describe as blackout operations.

And in those conditions, modern infrastructure delivery does something unexpected. It goes backwards.

Spreadsheets move from cloud-based systems to paper. Decisions are tracked in notebooks instead of dashboards. Coordination happens through conversations, not platforms.

When digital systems disappear, the work does not stop. It just becomes more direct, more manual and more dependent on people making decisions with limited information.

Disasters rarely take out a single asset. They disrupt entire lifeline networks at once. Power, water, transport and communications fail together and they recover together.

Which means the challenge is not just to fix what is broken. It is to reconnect what has been separated.

Trust becomes critical infrastructure

When systems go down, there is no time to build new relationships. You work with the ones you already have.

In Mingenew, one of the communities hit hardest by Cyclone Seroja, that made all the difference. Existing relationships meant the response could start immediately. Teams operated as an extension of the local Shire Council, not as external advisors.

That kind of alignment is hard to manufacture under normal conditions. In a disaster, it becomes essential.

In that gap, trust carries the load:

  • Decisions can be made without perfect information
  • It allows scope to evolve as conditions change
  • Teams can act before every detail is resolved

When everything else is down, trust becomes the system that keeps things moving.

“When you lose internet, power and phones, the only thing left is trust, initiative and sometimes a pen and a clipboard.”

Antoinette Krause
Mid West Region, Technical Director - Civil, GHD

When the rules no longer fit reality

Trust can get work started. But it cannot override the rules that shape what gets built.

In theory, disaster recovery is straightforward. Assess the damage, secure funding and restore infrastructure to how it was before. In practice, it is rarely that simple.

In Northampton, on Western Australia’s Mid-West coast, the coastline had changed so significantly after Cyclone Seroja that “like-for-like” restoration was no longer possible.

What followed was not a technical exercise, but a negotiation.

Design options had to be reworked, coastal studies commissioned and policy settings revisited. A working group of local government, state agencies and funding bodies met fortnightly for more than a year to align on what could actually be delivered.

Not because the engineering was complex. In many cases, the solutions were relatively straightforward. But because the system around the engineering struggled to adapt to a situation it was never designed for.

This is one of the defining challenges of disaster recovery. The systems that govern infrastructure delivery are built for stable conditions. They assume 1) time for planning, 2) clarity of scope and 3) a shared understanding of what success looks like.

Disasters remove all three. And they force decisions to be made before all the answers are known.

A disaster at scale

If Cyclone Seroja exposed how infrastructure systems break down, Cyclone Gabrielle showed what happens when that breakdown occurs across an entire region.

Gabrielle swept across New Zealand’s North Island with a scale and intensity that few New Zealanders had experienced before. It was estimated that 800,000 landslides were triggered by this event across the North Island. In Hawke’s Bay and Tairāwhiti, the impact was immediate and widespread. Roads were cut by slips and floodwaters, bridges were damaged or destroyed and entire communities were left isolated.

The numbers tell part of the story. Across Hawke’s Bay, the scale of the damage was immense. At least 16 bridges were totally destroyed, with 28 significantly damaged. More than 100 kilometres of roads had to be rebuilt. In Tairāwhiti, eight bridges were destroyed and 54 sustained major damage while two state highway bridges were also lost across the wider East Coast network.

Hillsides gave way under intense rainfall and rivers changed course. In some areas, forestry and large woody debris was carried downstream, compounding flood damage and blocking critical infrastructure. The scale of landslides alone reached levels rarely seen in a single event, turning entire transport corridors into a sequence of failures rather than isolated repair sites.

This was not a matter of fixing individual assets. It was about restoring connectivity across a network where multiple points had failed at once. A repaired culvert meant little if the road leading to it was still cut off. A rebuilt bridge had limited value if surrounding routes remained impassable.

The challenge was no longer linear. It became systemic and that is where traditional delivery models begin to struggle.

Because when failure spreads across the network, recovery is not measured by what is rebuilt first. It is measured by how quickly communities can reconnect.

In some locations, communities did not wait for formal solutions. Temporary diversion routes were built using whatever materials were available, restoring access within days rather than months. They did not meet design standards, and they were never intended to last. But they worked.

The bottom line

When everything is down, infrastructure is reduced to its essentials - not systems, platforms or standards, but people, decisions and connections.

Cyclone recovery shows that resilience depends less on perfect design and more on the ability to act when conditions are uncertain.

Because the first priority is not perfection. It is reconnection.

The question is what that means for how infrastructure should be designed in the first place.

Was this article useful to you?

Thank you for your feedback.

Kids playing outside

Bold moves.
Sustainable growth.

Explore how leaders are focusing resources, simplifying complexity and scaling impact.