Options Winter 2021: Jim Hall, Chair of the IIASA Science Advisory Committee, writes about how new data sets, systems analysis, and machine learning can help pinpoint hotspots of vulnerability and target measures to enhance resilience to future threats from climate change.
When natural disaster strikes infrastructure that delivers energy and water services or enables transport and digital communication, the consequences can be devastating far beyond the immediate destruction. So how can we reduce the risks of infrastructure failure cascading through economies and societies? Systems analysis and machine learning are revealing new ways to minimize the systemic risks.
The provision of new infrastructure can lock in patterns of development for years to come. Energy and transport systems have locked-in carbon-intensive behaviors and economic activities. Infrastructure decisions also lock in exposure to climate-related extremes. An analysis of all of the world’s land transport networks that my team did with the World Bank, found that 23% of roads and railways are exposed to climate-related hazards.
The consequences of climate-related shocks to infrastructure networks extend far beyond the direct costs of asset repair and reconstruction. Without transport connectivity, for instance, people can’t travel for work or access services like healthcare, and trade in essential commodities is disrupted. Billions of dollars of trade are at risk every year from climate-related disruptions to roads, railways, airports, and ports.
Until recently, the vulnerability of infrastructure networks to climatic extremes was mostly only studied at local and national scales. Some studies, including at IIASA, have looked at particular categories of global infrastructure exposure, like the risk of droughts to hydropower production, and cooling water shortages to thermoelectric power plants. Now, thanks to newly available big datasets and machine learning techniques, we are rapidly piecing together a complete picture of interdependent infrastructure networks covering energy, transport, water, and digital communications. This network analysis can be intersected with globally available datasets on climatic extremes and projections of how these threats may become more severe in the future. High-resolution spatial demographic and economic datasets are helping to quantify the consequences of failure, while new satellites are allowing us to observe the impacts of climatic extremes as they occur.
By analyzing the exposure of infrastructure networks to climatic extremes, their vulnerability to damage and disruption, the duration of disruption, and the pace of recovery, we are able to construct a picture of the resilience of infrastructure systems worldwide. This can help us to pinpoint hotspots of vulnerability and target measures to enhance resilience to future threats. That could include nature-based solutions like mangrove restoration, as well as physically strengthening assets to cope with more extreme events. Diversification and increasing stocks of commodities can make supply chains less fragile. Disaster risk finance can help governments to quickly access the resources they need to repair and recover. Knowledge from systems analysis is helping to test and implement these solutions.
Further info:
Koks, E.E., Rozenberg, J., Zorn, C., Tariverdi, M., Vousdoukas, M., Fraser, S.A., Hall, J.W., Hallegatte, S. (2019). A global multi-hazard risk analysis of road and railway infrastructure assets. Nature Communications, 10(1): 2677.