Summary
The paper provides a new dataset designed to make it quicker and easier to model climate scenarios where SRM is used, compared to those where it is not. This could allow more scenarios to be created and shared to help governance decisions, such as what happens if one country decides to deploy SRM alone, or if SRM deployment suddenly stops.
Abstract
In this paper we present GeoMIP-pattern, the first global geoengineering pattern scaling dataset. This dataset is useful to generate custom solar radiation modification scenarios and to emulate the GeoMIP model output with low data volume. Temperature, precipitation, and relative humidity patterns are derived from ScenarioMIP SSP5-8.5 and GeoMIP G6sulfur model output data using a two-step approach: the first scaling patterns are obtained by estimating a linear regression between the field of interest and the SSP5-8.5 annual global mean surface temperature, allowing for the calculation of no stratospheric aerosol injection scenarios for a global spatial grid. The second patterns are produced by regressing the field of interest on the annual global temperature difference between the G6sulfur and SSP5-8.5 experiments, which in combination with the first pattern emulates the stratospheric aerosol injection scenario. The dataset contains validation statistics, including unit root hypothesis tests novel to the field of pattern scaling, which demonstrate that the pattern slope adequately captures local changes, with most regions showing no remaining trend or nonstationarities in the residuals.
