In climate change terms, 2023 hit the kind of milestones that wouldn’t be rewarded in an annual performance review. It was the hottest year on record by a large margin, and global greenhouse gas (GHG) emissions once again climbed, with atmospheric CO₂ reaching a record 419 parts per million, according to the EU’s Copernicus Climate Change Service. At S&P Global Sustainable1, these sobering facts further motivate our work to improve the data and models we use to understand the business risks and opportunities posed by a changing climate.

In this blog, we describe three recent enhancements to our climate physical risk modeling methodology that together provide more granular analysis of physical risks: updated climate model data, a new pluvial (rain-driven) flood hazard, and updated wildfire and drought metrics.

The latest climate model data

Climanomics is a risk analytics platform that calculates the financial impact of climate risk on physical assets or real estate investments within a portfolio. The platform can conduct analysis across eight decades for four emissions scenarios and covers any location on the planet.

All hazard models included in the Climanomics platform are now built from the latest generation of climate model data, known as CMIP6. CMIP stands for the Coupled Model Intercomparison Project, the global climate science community’s framework for aligning climate models, experiments and data, and CMIP6 was developed in support of the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Climanomics uses downscaled CMIP6 datasets provided by the NASA Earth Exchange, which significantly improve the resolution of analysis for many hazards.

Compared to CMIP5, the prior generation of climate model data, CMIP6 offers three key advantages:

Data from more climate models. Our CMIP6-based platform now uses 32 models, as opposed to 21 models in the CMIP5-based version. The integration of a broader range of institutions and models, each with its own strengths and weaknesses, helps scientists distinguish which trends in the data are attributed to model biases.

Broader climate sensitivity. Climate sensitivity is a measure of the relationship between changes in global CO₂ concentration and change in global temperature. Most of the previous climate model data, including CMIP5 models, found that a doubling of atmospheric CO₂ would result in a rise in global temperature of 1.5 degrees C to 4.5 degrees C. However, recent data underpinning CMIP6 finds that a doubling of atmospheric GHG emissions would result in larger changes in global temperature, ranging from 1.8 degrees C to 5.6 degrees C. This modeling enhancement is important to ensure that future climate change impacts are not underestimated.

New climate scenarios. CMIP5 scenarios were based on physical changes to the atmosphere caused by anthropogenic impacts to GHG emissions and land use change. The CMIP6 scenarios are enhanced by the incorporation of additional socioeconomic factors alongside those physical impacts. Known as the Shared Socioeconomic Pathways (SSPs), the CMIP6 scenarios are based on five distinct narratives for future socioeconomic development and provide a consistent logic for the qualitative projections of land use, energy use, population, emissions and other factors, embedded within each scenario.

CMIP5 focused on a range of GHG emissions scenarios called Representative Concentration Pathways (RCPs). The RCPs represent the way emissions increase the amount of excess heat that becomes trapped on Earth. This energy is referred to as radiative forcing and is measured in watts per square meter. RCP 4.5, for example, is a scenario in which global emissions peak in 2040, resulting in a radiative forcing level of 4.5 watts of heat per square meter by 2100.

CMIP6 combines the SSPs with the RCPs to create scenarios that couple emission pathways with socioeconomic developments. CMIP6 prioritizes some scenarios, shown in dark blue in the diagram below, which are also used in the updated Climanomics platform. For example, SSP3-7.0 is a scenario with comparably high non-CO₂ and high aerosol emissions, in which CO₂ emissions roughly double from current levels by 2100. It is also known as the “regional rivalry” scenario, which has no additional climate policies and instead features a revival of nationalism and regional conflicts, with low investment in technological development, continued population growth, and strong inequality.

The light blue boxes show other scenarios considered important but at a lower priority in CMIP6. White boxes show the SSP/RCP combinations that are possible, while blank spaces reflect that some forcing levels were found to be incompatible with some SSP narratives.

Pluvial flooding

Pluvial, or rain-driven, flooding is now included in Climanomics, adding a hazard that acutely impacts urban areas. Due to their higher proportion of asphalt, concrete, and other impervious surfaces, cities and towns are more prone to flash flooding resulting from intense rainfall events, which are becoming more frequent and severe in many locations because of climate change.

S&P Global uses CMIP6 daily precipitation data and applies a statistical model to determine the intensity of extreme rainfall events. The pluvial hazard model projects the annual frequency of the historical baseline 100-year precipitation rate. That relates to the pluvial hazard metric of annual frequency of the 100-year flood depth: a hazard value of 0.1, for example, means that pluvial flooding of a depth that historically happened only every 100 years has become an event that happens every 10 years.

This allows municipalities, investors, property owners and other stakeholders in urban areas to understand projected change in exposure to extreme precipitation events under different climate scenarios.

Updated drought and wildfire models

In 2023, Climanomics drought and wildfire models were updated to provide more specificity and granularity. The updated drought metric transitioned away from using the Self-Calibrating Palmer Drought Index (scPDSI)¹ to the Standardized Precipitation Evapo-transpiration Index (SPEI).² One of the key advantages of using the SPEI is the ability to capture drought conditions on various time scales, which in turn allows for the identification of different types of drought, including agricultural, meteorological and hydrological drought. Low SPEI values indicate drought conditions.

The updated wildfire model transitions from the Z-Index³ (a measure of moisture anomalies used to assess wildfire conditions based solely on temperature and precipitation) to the Fire Weather Index⁴ (FWI). The FWI is a meteorologically-based index used worldwide to estimate fire danger. In addition to temperature and precipitation, it considers the effects of surface winds and relative humidity on fuel moisture and fire behavior. High FWI values indicate more favorable conditions for the ignition and spread of wildfires. The enhanced model also employs a land cover mask to differentiate areas that contain or are adjacent to burnable vegetation.

S&P Global Sustainable1 makes ongoing improvements to climate hazard models incorporated in Climanomics. Future improvements to the hazards discussed here will include multiple pluvial return periods and probability of wildfire events.