09
^{CLIMATE CHANGE ADAPTATION}
Adapting to a fast-changing climate

Why utilities must act now to manage the unpredictable

^{CLIMATE CHANGE ADAPTATION}

Adapting to a fast-changing climate – why utilities must act now to manage the unpredictable

In January 2025, Storm Éowyn tore across the UK and Ireland as the region’s most powerful windstorm in a decade. In a matter of hours, it cut power to more than a million homes, knocked out telecoms networks, and triggered knock-on disruption to transport and water services. It wasn’t an isolated case and showed a real risk facing utilities everywhere: a single climate event cascading through an entire interconnected system in real time.

The impacts of climate change are putting power networks globally under huge stress. Heatwaves and prolonged droughts make generation less efficient, limit how much power transmission and distribution lines can carry and make equipment degrade faster. Wildfires pose direct threats to overhead lines, substations and network integrity.

Flooding, storms, extreme rainfall, and the soil erosion they bring, damage equipment and restrict access for maintenance and recovery operations, so outages last longer. And cold weather affects equipment performance, puts more mechanical stress on lines and drives sharp peaks in demand.

The effects of this exposure on customers and the bottom line are already substantial. In the US, JD Power found that 45% of utility customers experienced a power outage in the first half of 2025. Almost half (48%) of those outages were caused by extreme weather.

The picture looks to worsen globally over time. The Oak Ridge National Laboratory found that major outages cost $121 billion in 2024. Capgemini’s “Climate adaptation: Harnessing tech-driven resilience to create sustainable value” report found that without action, a typical large listed utility could lose up to $4.6bn/year in the 2050s.

What’s more, climate risk is now volatile, financially material, and something investors and regulators increasingly expect to see priced and managed. This means utilities must commit capital now against risks that don’t behave in predictable ways. It’s a substantial challenge, but the choice is clear: adapt now or pay later.

A single hazard can cause system-wide failures

The challenge for utilities isn’t only that extreme weather events are becoming more frequent and intense. It’s that those hazards are compounding and cascading across a single interconnected system.

The grid is increasingly being stressed by five overlapping hazard families: heatwaves and droughts, wildfires, storms and tropical cyclones, floods and heavy rainfall, and cold waves with wet snow. Two or more of these hazards often strike at one time, compounding the stress on the system. For example, heatwaves increase conductor sag, transformer stress, and cooling demand at the same time as drought creates combustible conditions and reduces field accessibility.

This means that what appear as separate risks on paper become a single compound event in practice: higher load, weaker assets, greater fire ignition risk, and slower restoration. As a result, utilities can no longer treat thermal risk and wildfire risk as separate workstreams.

One hazard can also affect multiple parts of the system, causing the disruption to cascade and potentially become a system-wide failure. Storm Éowyn was an example of this: a single storm brought down both power and telecoms, then rippled into transport and water services – underscoring the interdependence of modern infrastructure.

This is why climate risk is now an uncertainty problem for utilities. The traditional single-hazard, static resilience planning models cannot capture these kinds of compound, cascading failures. Instead, utilities must move beyond hazard-specific mitigation towards integrated adaptation.

Larger network operators are generally furthest ahead on this journey, with the most mature ones treating resilience as a systematic, risk-based capability. They’re also concentrating resources on the most critical infrastructure and partnering with specialists to get better at anticipating, absorbing and recovering from increasingly severe events.

The cost of adaptation versus the cost of inaction

Climate adaptation has become a strategic priority across the sector, not only among those leading the field. But measures to improve resilience, such as asset hardening, automation, and network redesign, require substantial upfront investment. By contrast, the cost of failing to invest – outages, infrastructure damage, restoration costs, regulatory penalties, and wider societal disruption – is immediate and significant.

This asymmetry helps to explain why many utilities still don’t consistently assess and prioritize large-scale resilience investments, even as the potential cost of not doing so rises. And unlike in other sectors, there’s little opportunity to transfer climate-related infrastructure risk to insurers.

Which pot the money comes from can present another problem. Traditional regulatory models reward CapEx, where physical assets generate predictable, rate-based returns. Yet many of the most effective solutions, such as predictive analytics, real-time monitoring and real-time optimization, are OpEx-driven and harder to justify. This can create an investment bias towards capital-intensive options over more flexible, lower-cost alternatives.

The fundamental challenge is that resilience has moved far beyond a network-design problem.

Today, utilities must decide how much capital to commit now, to guard against risks that don’t behave predictably.

Assessing climate risk: a prerequisite to effective resilience planning

Before utilities can decide where to invest, which assets to strengthen, or how to prepare for extreme events, they first need a clear view of their vulnerabilities and priorities.

A structured climate risk assessment provides that view by identifying potential threats and organizing them by priority. High-priority risks are both highly likely to happen and to have potentially severe effects on operations or customers. Medium-priority risks may need targeted mitigation or monitoring, while lower-priority risks may be managed through existing controls.

Identifying where to intervene most urgently is only the first step. The challenge is turning those findings into practical adaptation and resilience measures that strengthen the network before, during and after extreme events.

Leading utilities are increasingly adopting resilience frameworks that combine infrastructure hardening, operational preparedness, real-time monitoring, and rapid-response capabilities within a single integrated architecture.

Figure 1: A sample climate risk assessment, with risks organized by severity and likelihood

As Figure 2 shows, in these frameworks, resilience can be broken into four interconnected stages: anticipate, alert and monitor, analyze and respond, and intervene and recover.

Figure 2: The four stages of resilience

Anticipate

The priority is anticipating future risks and reducing exposure before events occur. This includes using climate projections, hazard forecasting, network modeling, and asset vulnerability assessments to understand future operating conditions. It also encompasses long-term adaptation actions such as strengthening existing infrastructure, replacing old cables, undergrounding selected assets, improving vegetation management practices, and embedding climate risk considerations into asset investment strategies.

Alert and monitor

Resilience increasingly depends on situational awareness. Utilities are expanding the use of sensors, weather-monitoring systems, connected devices, and network monitoring platforms to warn them about abnormal operating conditions early. By connecting advanced SCADA, ADMS, and outage management systems to intelligent sensors and actuators at the edge, operators can also detect developing risks, monitor asset conditions in real time, and mobilize resources before failures happen.

Analyze and respond

Once an event has begun, utilities need to rapidly assess impacts and make informed operational decisions. Technologies such as drones, LiDAR, edge computing, intelligent substations, and AI-enabled analytics can give them a clearer view across the network and help them identify faults faster. Advanced network control capabilities, predictive analytics, and self-healing functions allow operators to shorten outages and speed up repairs in increasingly complex operating environments.

Intervene and recover

The final stage focuses on maintaining a continuous service and recovering faster. This includes using temporary generation, reconfiguring network assets, mobilizing an emergency workforce, managing spare parts, and coordinating with telecommunications, water, transport, and emergency services. The goal isn’t simply to restore power quickly, but to minimize the wider societal impacts that can emerge when critical infrastructure systems fail simultaneously.

Underlying all four stages is a set of capabilities that increasingly determine resilience. Network digitization, advanced data platforms, AI-enabled decision support, integrated operational technology (OT) and information technology (IT) systems, and workforce transformation all provide the foundation for more adaptive and responsive networks.

Importantly, resilience doesn’t come from any single technology or intervention. Each utility will need to define an adaptation pathway that reflects its specific geography, climate exposures, asset base, regulatory environment, and investment priorities. What matters is having a structured framework that links climate risk assessment directly to resilience action.

Looking ahead: how AI could turn a periodic sequence into a continuous, connected loop

As we’ve shown, technology underpins each of these stages. But as climate risks become more complex and less predictable, the volume of data these technologies generate is outpacing what operators can process manually. Ultimately, this will drive utilities to adopt AI-enabled, model-driven operations, in which the four separate steps act as a continuous, connected loop.

In practice, that means combining OT, IT, weather intelligence, and asset data into one live view of the network’s exposure to climate risks. This means operators can monitor conditions and respond to emerging risks in real time, rather than analyzing each system separately after an event.

It also means giving intelligence more room to act: supporting, and in some cases carrying out, interventions within clear governance and safety boundaries, instead of only informing a human’s decision. Once intelligence can act, resilience will evolve from a series of discrete projects into a continuously adaptive capability embedded throughout operations. A loop of monitoring, identifying risks, modeling scenarios, responding, prioritizing resources, and learning – rather than fixed planning and investment cycles.

It may be a way off yet, but this AI-enabled approach will help to close the gap between the unpredictable nature of climate risk and how utilities have traditionally managed it.

Final word

The message for utilities is clear: understanding climate risks isn’t enough. As those risks become more systemic and interconnected, resilience must be managed as a continuous cycle of assessment, prioritization, preparation, response, and recovery – not a series of one-off investments.

The utilities that connect climate scenarios, risk intelligence, operational readiness, and investment planning will be best placed to protect their networks, their customers, and their capital in an increasingly volatile climate. Whether the next event is another Storm Éowyn or something we haven’t seen yet, these utilities will do more than just keep the lights on.

Authors

Thomas Hernandez

Executive Vice President, Global Head of Energy & Utilities, Capgemini

Mercy Paul

Director, Lead Energy & Utilities Industry Platform, Capgemini

Philippe Vié

Senior Advisor, Energy and Utilities, Capgemini

Manuel Castro

Vice President, Energy & Utilities Lead, Capgemini