The Climate Calculation - Business success is increasingly defined by resilience to a changing world

Even the most detailed satellite images of California’s wildfires would have failed to reveal that businesses 60 miles away would have their power cut off by the authorities. Nor would they have indicated that regulators would respond to the crisis by drafting legislation to limit how much water data centers can consume.

“Just throwing more data at the problem without understanding how to extract what matters to you is unlikely to be productive,” explains Belinda Bates, head of development for climate resilience at Zurich Resilience Solutions (ZRS). “Our analysts often start by taking a broad view of climate risk across a company’s value chain, but quickly bring in their expertise to assess how operations or specific sites contribute to different risks, and to identify the steps they need to take to mitigate them.”

After severe flooding in 2016 disrupted production at Audi’s car manufacturing plant at Neckarsulm, Germany, engineers and scientists from ZRS were brought in to find a solution. After analysing the site and the surrounding region, they helped Audi to erect water-filled barriers around the facility that have prevented surface flooding ever since.

As weather-related risk becomes more volatile, resilience will move further up boardroom agendas. “We’ve gone through the most socially and economically costly year in human existence with regards to extreme events, Johan Rockström, director of the Potsdam Institute for Climate Impact Research, told an audience at the recent World Economic Forum in Davos. “Businesses have a great opportunity, but also a high degree of responsibility to map risks of extreme events across their entire value chain,” he added.

However, as businesses in California can attest, this is not as simple as it may seem. Start with physical risk. “Climate risks accumulate through a combination of interacting physical processes,” explains Emily Shuckburgh, academic director of the Institute of Computing for Climate Science at Cambridge University and director of Cambridge Zero, the university's climate change initiative. Businesses in California, for example, which are recovering from some of the deadliest wildfires in modern history, must now prepare for the increased risk of flooding and mudslides as rainwater struggles to permeate the scorched earth.

Next, consider the challenges of collecting accurate data across value chains. When analysts at the Bank of Japan assessed global exposure to environmental disruption, they found that economic data in countries such as Colombia was provided as a national average because the government lacked details of the precise locations of assets of Colombian banks. Other problems, such as the limited availability of data on subsidiaries or raw materials, can also hamper analysis.

The AI advantage

Together, physical and data challenges can make it difficult to take appropriate action. Fortunately, technology provides solutions. “Our models utilise neural networks not only to look at how natural hazards are changing, but how that changes the risk to a business,” says Iwan Stalder, head of accumulation management at Zurich Insurance Group. “We use neural networks to connect climate and catastrophe models,” he adds. “This helps us identify relationships and dependencies in large datasets that a human cannot do, because of the scale and complexity of the data, to better understand the business impact of a changing climate.”

Combining neural networks with other AI such as image recognition and machine learning will improve the accuracy of climate analysis in three distinct ways, says Professor Shuckburgh. “First, data-driven machine learning approaches are being incorporated into climate models to reduce the uncertainty in projections,” she says. “Second, AI is being used to correct for systematic biases and ‘downscale’ the projections – of the risk of extreme weather, for example – to particular locations such as a specific city. Third, AI approaches are being used to track meteorological conditions to specific impacts, and to map systemic risks that are connected through supply-chain networks.”

Future proofing

This level of climate analysis is already shaping the future. In 2024, Madrid City Council began to install permeable pavements, rain gardens and drainage ditches to mitigate the impact of floods that were increasingly threatening the city. The creation of shaded walkways and adoption of reflective materials will follow in 2025 as ZRS continues to help the city adapt.

“We've also worked with private companies to plan capex projects – from purchases of land to project planning and facility design – as well as integrating extreme weather and climate risk into supply-chain planning,” says Ms Bates. When sustainable technology supplier Johnson Matthey was looking to accelerate its decarbonization strategy, for example, ZRS developed quantifiable measurements for environmental risks across the company’s value chain, helping to determine where investment was most needed.

As more companies integrate climate resilience into their strategic planning, adaptation becomes a critical advantage. If an automotive company is looking to offset climate risk across its supply chain, for example, suppliers that can provide accurate data on their flood or fire resilience will be more likely to win contracts. Climate resilience will become more important as the severity and frequency of extreme weather and natural hazards increases. In this new economic reality, it is not the businesses that move fast and break things that will thrive, but the ones that can adapt to a volatile world.