We are Building the World’s First Digital Twin to Enhance Our Climate Resilience

By Nicole Hu, Co-Founder & CTO

When Hurricane Ida slammed into the U.S. Gulf Coast in September 2021, the damage went far beyond flooded homes and buildings, impacting critical infrastructure. Energy companies in Louisiana scrambled to open new supply operations and restart oil pipelines; damaged roads and transport facilities slowed aerial surveys necessary to get power back online; and crews and equipment suppliers couldn’t get to their jobs. The worsening effects of climate change are wreaking havoc on infrastructure, supply chains and communities worldwide.

Louisiana’s devastation is just the tip of a melting iceberg.

Resilience to climate threats — a framework where organizations, communities and private and public sector actors understand, forecast and mitigate climate risk — is among our most critical global priorities. It demands an entirely new approach to understanding and acting on climate risk, one that mitigates climate threats and their ripple effects on businesses and communities, rather than scrambles to react after the damage has been done.

Instead of reacting to events, we need to model the future. That’s why One Concern is building the world’s first “digital twin” of cities, communities, and infrastructure to create a more resilient world for everyone.

What’s a “Digital Twin”?

Artificial intelligence and machine learning, paired with trillions of data points curated by expert data and disaster scientists, can create virtual models of our communities to model, understand and get ahead of climate risks. Think of it as a “Sim City” or interactive simulation for our planet — with digital twins, we have the power to granularly map cities building by building, block by block, and road by road, capturing every power base station, water pipeline, bridge and port.

Armed with better data and AI, we can run thousands of climate and extreme event simulations before a natural disaster to identify vulnerable communities and critical infrastructure in harm’s way. Our machine learning systems also “fill the gaps” within the data to account for the ongoing effects of climate change, allowing us to improve our predictive analytics and see into the future with greater confidence.

With digital twins we’re able to measure resilience using data like never before.

How We Build Our “Twin” Models

Our digital twin models are based on three key pillars:

  • We start with hazards like earthquakes, floods and windstorms. In the United States alone, we have 630 million earthquake data points from seismic hazard maps; 1.8 billion flood data points; and 57 million wind hazard data points.
  • We then map those hazards to networks, including substations, power lines, seaports, buildings, bridges, highways and so on.
  • Finally, we analyze vulnerability factors such as damage ratio, downtime and recovery curves to predict both direct and indirect impact and recovery period.

We then overlay these elements with our AI/ML analytics engine, which continues to digest new data from our data sources. We pull in weather and climate data, data on water pipes, power lines and other infrastructure data, as well as demographic data and structural and seismic data. We curate data from both public and private sources, such as advanced data sets from CoreLogic. We also infer or synthesize data and use on-ground data to validate the accuracy of our data and models.

The data and analytics are used as inputs for specific resilience metrics, such as the One Concern Resilience Score (1CRSTM), which reflects the One Concern Exceedance Probability (1CEPTM) and One Concern Downtime Static (1CDSTM) for a given property. These metrics and statistics serve a variety of valuation and risk assessment applications. We expect that the 1CRSTM will become the de facto global standard for measuring resilience.

Working with the World’s Leading Researchers

As much as we leverage technology, digital twins need more than machine learning and data to be effective. We also rely on a network of researchers within and beyond One Concern.

A strictly academic approach produces narrowly focused calculation components that may not work in the context of commercial application. A strictly commercial approach doesn’t benefit from the innovations continually developed in the academic research community. One Concern’s methodologies and calculation process workflows, however, reflect the best of both worlds.

One Concern’s Technology team keeps apprised of the latest research via deep relationships with academic field experts who are a part of a Technology Working Group that provides feedback and critique on the models. We are developing pipelines to continually sense changes in the built and natural environment by ingesting satellite imagery and IoT data (e.g. river gauges, weather radar, etc.) and applying AI on top of it.

Ignorance of Risk is No Longer an Option

With natural disasters consistently setting records, it’s no longer credible to claim we can’t have foreseen an extreme event or its devastating impacts. And current, conventional climate models are either not accessible to leaders or incapable of effectively informing against future threats.

Whether it’s safeguarding electric power and water sources or understanding how a disaster might affect the flow of vehicle traffic or supply chains, governments and private enterprises must identify the previously “unknown unknown” variables of a disaster event. With climate resilience now a strategic priority for most companies and countries, certain scenarios must be effectively simulated by leaders at organizations, institutions and communities, particularly given the exponentially changing climate threats we face. A digital twin makes this possible.

Resilience to climate threats is critical to a carbon neutral, safe and secure future. Without modeling the future with a digital twin, we will face a changing climate without knowing or understanding how to safeguard our most precious asset: people.

Nicole Hu is Co-Founder and CTO for One Concern.