Accurately linking scenario-driven climate hazards to economic impacts means bringing together rapidly developing areas of research in fields that are not normally linked, especially when it comes to joining physical science with the realms of economics, finance and ecology. Society and economic productivity are strongly affected by various climate hazards, and the work of quantifying the impacts at specific locations in the future based on scenarios of human-caused changes is at the forefront of an ever-increasing set of specialized data, models and knowledge.  For this reason, it is important to identify the strong, weak and missing linkages between climate hazards and economic variables (GDP, productivity, etc.), to better quantify impacts to inform public and private policies and plans. Doing so will also help policymakers and companies understand the scale of adaptation and resilience measures required. For example, some of our recent work estimates climate-change impacts on GDP.

Understanding and quantifying the likely economic and financial impacts of climate change requires confronting the challenge of linking discipline-specific methods, models, and data.  We consider building the collaboration pathways among different disciplines to be one of the frontiers of climate science.

Some of the critical assets to address this challenge consist of models of varying degrees of complexity (earth system models, macroeconomic models, integrated assessment models, ecosystem models, etc.), along with an extensive and growing set of diverse physical and economic data. These models and data have specialized characteristics not readily visible to researchers from other disciplines that can lead to mismatches in assumptions and methods. The cross-disciplinary challenge is to forge, at an appropriate level of detail, a common understanding of how the diverse models and data can be linked in a consistent way.

The difficulty of addressing this challenge is increased because, in the case of climate change, many parts of these disciplines are also operating at the frontiers of their respective knowledge. There are many important aspects not fully understood, such as cloud-climate feedback on the physical side and GDP-growth impacts on the economic side. The multiple effects of climate change on temperature, precipitation, storms, wildfires and other phenomena mean that we are dealing with spatially and temporally dependent climate and weather events, or compound events, both from a physical perspective and in terms of their downstream economic effects. Additionally, the interactions of individual agents, such as households, companies and governments add complexity because collectively their actions create macroeconomic outcomes.

One particularly important and daunting challenge is linking compound climate and weather events to economic impacts. There are two elements of this problem that make it difficult to tackle: first, the need for a better approach to modeling compound climate and weather events; and second, the need for more interdisciplinary collaboration to link this cutting-edge climate science with work being done in the fields of finance and economics. 

At the S&P Global Climate Center of Excellence (CCoE), one of our projects is to develop an improved set of scenarios related to compound weather and climate events that provide physical hazard risks where the co-occurrence of different variables is known to be physically consistent. These include co-occurrences of extreme temperature and humidity associated with heat waves or the freezing rain, heavy snow, cold waves, landslides and strong winds associated with severe winter storms — not only at a given location, but across broad spatial and temporal scales.  We know from episodic historical data, such as the US National Oceanic and Atmospheric Administration’s billion-dollar disasters series, that the spatial scale, duration and time between extreme weather and climate events are important considerations in assessing economic impacts and policy changes. CCoE researchers are working to address both heat waves and winter storms in a realistic manner consistent with their compound properties.

Compound weather events can have other negative effects beyond their direct impacts on communities or businesses, and modeling these effects requires addressing their spatial and temporal attributes. For instance, during extreme heat and cold waves, the electrical grid’s ability to transfer energy depends on the spatial extent and duration of the extreme temperatures. Climate scenarios often fail to provide this type of information. We are working to address these types of climate dependencies through joint research across S&P Global.

Another focus for our researchers is advancing our knowledge of the absolute value associated with any change in weather and climate variables. Important for many applications, this has traditionally been addressed by downscaling global climate models (GCMs) to address local variations and remove systematic biases in the models. Numerous approaches have been developed over the past several decades to address this issue, and recently, machine learning techniques show promise. Nonetheless, a critical challenge remains in ensuring the compound co-occurrence of these values is physically realistic. We have seen instances where the standard downscaling and bias-correction algorithms have been applied, yet the co-occurrence of variables like high temperature and humidity is not realistic.  

While much of our work is directed at physical climate science, we are also addressing cross-disciplinary methods and results. For example, we are currently comparing each of the underlying data sources that lead to estimates of country-specific climate impacts on GDP growth, as calculated both by CCoE using the latest CMIP6 GCM data, and by the Network for Greening the Financial System. This entails comparison of multiple data sources on historical temperatures and detailed comparison of the underlying equations in each step of the calculation of GDP impacts.

This work is ongoing, but we have already found differences that appear to be related to cross-disciplinary gaps in communication. Once completed, we will share these results with the broader research community to help close these gaps and foster better estimates of the complex interactions and feedback between the economy and climate change.

Finally, we understand and embrace the fact that collaboration across disciplines and institutions is necessary to make much needed advances in our understanding of these complex problems. We have several collaborations in place, among which is a multi-day research workshop jointly sponsored by S&P Global and the World Climate Research Programme. This workshop, scheduled for November 2024, will seek to address the linkage of physical, economic and financial models and data.