A rapid-application emissions-to-impacts tool for scenario assessment: Probabilistic Regional Impacts from Model patterns and Emissions (PRIME)

Climate policies evolve quickly, and new scenarios designed around these policies are used to illustrate how they impact global mean temperatures using simple climate models (or climate emulators). Simple climate models are extremely efficient, although some can only provide global estimates of climate metrics such as mean surface temperature, CO2 concentration and effective radiative forcing. Within the Intergovernmental Panel on Climate Change (IPCC) framework, understanding of the regional impacts of scenarios that include the most recent science is needed to allow targeted policy decisions to be made quickly. To address this, we present PRIME (Probabilistic Regional Impacts from Model patterns and Emissions), a new flexible probabilistic framework which aims to provide an efficient mechanism to run new scenarios without the significant overheads of larger, more complex Earth system models (ESMs). PRIME provides the capability to include features of the most recent ESM projections, science and scenarios to run ensemble simulations on multi-centennial timescales and include analyses of many key variables that are relevant and important for impact assessments. We use a simple climate model to provide the global temperature response to emissions scenarios. These estimated temperatures are used to scale monthly mean patterns from a large number of CMIP6 ESMs. These patterns provide the inputs to a "weather generator"algorithm and a land surface model. The PRIME system thus generates an end-to-end estimate of the land surface impacts from the emissions scenarios. We test PRIME using known scenarios in the form of the shared socioeconomic pathways (SSPs), to demonstrate that our model reproduces the ESM climate responses to these scenarios. We show results for a range of scenarios: the SSP5-8.5 high-emissions scenario was used to define the patterns, and SSP1-2.6, a mitigation scenario with low emissions, and SSP5-3.4-OS, an overshoot scenario, were used as verification data. PRIME correctly represents the climate response (and spread) for these known scenarios, which gives us confidence our simulation framework will be useful for rapidly providing probabilistic spatially resolved information for novel climate scenarios, thereby substantially reducing the time between new scenarios being released and the availability of regional impact information.