Building resilient water and marine infrastructure through adaptive planning

Unlike many past disruptions, climate change moves in a broadly known direction. Scientists have modelled sea level rise, while regional projections show where there is rising drought risk. The pace and severity of these events, however, are harder to pin down. As planners, how we approach projects has changed. We can work with a range of plausible futures, instead of only historical averages.

The cost of managing failure is already climbing. Insurance losses from extreme weather keep rising, putting pressure on both insurers and asset owners. The economic toll of severe supply restrictions runs into the billions of dollars in today’s terms, and Cape Town came close to running out of water entirely in 2018.

Extreme rainfall has degraded raw water quality in catchments and constrained treatment capacity in major cities. Bushfires have damaged catchments for years after the event. Flooding disrupts infrastructure near rivers and coastlines, and prolonged heat stresses electrical and mechanical systems. Planning for a single hazard misses the point because these risks compound.

Three questions should sit at the top of any resilience plan:

1. What are the system’s operating limits?
2. If those limits go, what does failure look like for customers, the community and the wider economy?
3. What signals indicate the limits are about to be tested? 

For decades, infrastructure planning was based around solving a single issue, like a demand forecast or a fixed sea level allowance. When conditions were more stable, this was an effective way to plan. Now, the same asset needs to perform across a wide range of possible outcomes.

Adaptive planning prepares for multiple conditions and times each commitment to how those conditions unfold. It starts with mapping the full range of scenarios a system may face, including the extreme combinations that push assets past their current limits.

Early work identifies conceptual options across conventional and climate-independent solutions. Readiness activities then test viability through pre-feasibility studies and site investigations. Only once that groundwork is complete does the project move into detailed design, business cases and delivery. Each stage carries defined decision points where leaders can decide what to do next based on the evidence.

Formal frameworks such as the Water Security Plan for Greater Melbourne and connected towns put this thinking into practice. These plans recognise that resilience decisions are governance decisions, shaped by service expectations, risk tolerance and what can realistically be funded.

What makes these decisions difficult, however, is that service expectations rarely change. Communities expect reliable supply and safe infrastructure regardless of conditions, while ports expect operability, safety, access and quick recovery from any disruption. 

Risk tolerance and funding, on the other hand, are not fixed in the same way. While adaptive planning does not completely remove this tension, it brings it into view so decision-makers can work with it much more easily.

Asset owners and operators can start with actions that keep options open without over-committing:

1. Set aside land first. Space disappears fast, and a future project that lacks one loses years and money looking for the next-best option.
2. Pilots and trials come next, since new technologies are easier to back once they already have a history of working.
3. Demand management closes the supply gap, which stretches the lead time before stakeholders have
   to act.
4. Modular design allows assets to grow in stages instead of being rebuilt.
5. Monitoring triggers come last, which tie the next decision to something measurable rather than to whoever holds the budget that year.

Resilient planning starts from a clear picture of the overall water supply. Rainfall-dependent sources such as dams, run-of-river systems and groundwater are typically cheap to operate, but their reliability depends on the climate. As inflows decline or become more erratic, that reliability is threatened.

Climate-independent sources such as seawater desalination and potable reuse cost more to operate and need much more energy. Unlike rainfall-dependent sources, though, this water is available regardless of drought, reducing exposure to long-term drying trends.

Australia’s desalination plants illustrate this trade-off. Operators built Sydney’s desalination plant as a drought asset, but it now runs much more often, partly to support supply when poor raw water quality limits surface water treatment. 

Melbourne’s desalination plant used to sit on standby for years, but operators are now bringing it up to near-full output as dry conditions return. These are insurance assets where operators pay for them through the good years and call on them in the bad ones. The economic case rests on what they prevent.

The lesson from large desalination and water security programs is that the decisive choices are often made well before construction, from how the problem is framed, how uncertainty is carried, what options are kept alive and when leaders are prepared to commit.

The same logic applies to coastal and marine infrastructure. Climate scientists have now bounded near-term sea level rise projections reasonably well, particularly over the next thirty years. This gives room for designers to plan for future conditions without locking in worst-case assumptions today.

One example of this in action is a government-owned coastal park dealing with sunny day flooding. Local ordinances required new infrastructure to account for long-term sea level rise, which meant a protective wall around 2.4 metres (eight feet) tall around the existing facilities. A wall that high would have met the regulation, but it would have also closed off the views that draw people to the marina and the surrounding historic buildings.

Our design team chose an alternative — a wall engineered structurally for the future elevation, but built initially to roughly one metre (3.3 feet) to address near-term risk. Visualisations of the current and eventual elevations helped regulators and the community accept the approach. Permits and funding followed. When monitoring shows the triggers have arrived, park operators can simply raise the existing wall by doweling a reinforced cap into the top instead of replacing the whole structure.

Some disruption during extreme events is acceptable, but what matters is how quickly the system recovers. In some locations, combining engineered solutions with natural systems such as dunes or mangroves supports that recovery while decreasing long-term maintenance demands.

Keeping enough options open for long enough, without giving every option equal attention forever, requires disciplined monitoring, early readiness work and honest conversations about service expectations, risk tolerance and funding. While adaptive planning cannot remove hard choices altogether, it does make them visible early enough to manage. The key is to: 

• Map the full scenario range, including the combinations that push assets past their current limits.
• Start preparatory work early, since approvals and site reservation often take five to ten years and the obvious sites disappear first.
• Build monitoring trigger points into governance so decisions move on evidence instead of sentiment.

The work is easier when drawing on what others have already learned. Whatever the challenge, somewhere a peer has worked through a version of it. The solution rarely transfers cleanly, but the journey usually does. We’re happy to be a starting point for that conversation.