Regional climate impacts for relevant warming scenarios, drawing on the best available science
Organisations the world over are seeking more scientifically robust and flexible ways to consider regional climate change consequences and understand the associated uncertainties. Decision‑makers need credible, high resolution probabilistic assessments of climate hazards under relevant warming scenarios to shape investment and adaptation planning and policy. Legacy offerings are coarse, deterministic, or tied to few GCMs and scenarios, and provide limited insights on risks at the tails of distribution which drive costs.
The assessment of climate risk is rapidly maturing from a practice rooted in historical observation to a forward-looking discipline centred on probabilistic analysis. Climate change is no longer viewed as a distant, abstract environmental issue but as a present and material financial risk, compelling organisations to quantify its potential impact on assets, operations, and balance sheets. This will increasingly be required to meet regulatory obligations such as the EU Corporate Sustainability Reporting Directive and mandatory reporting requirements emerging in other jurisdictions. Energy market modellers and infrastructure planning is evolving to better capture the impacts of the changing climate on system design and costs.
We derive high-resolution regional climate impacts under any warming scenario. We provide climate impact data for the warming scenarios and at the resolution relevant to our clients and collaborators.
We develop platforms for serving the data that assist our collaborators and clients to use the data effectively, managing data volumes, version control, and enabling comparison to warming and impacts under other influential scenarios such as IPCC AR6, IEA, or NGFS.
We work with energy market modellers to provide regional climate impacts most relevant to the analysis of system reliability and performance under a changing climate.
We use the best available science. We are working with the global science community to advance the science and fill important gaps relevant to compound risks, and spatial and temporally correlated climate impacts that affect infrastructure and supply chain resilience and adaptation costs.
Methodology
The Climate Resource regional climate impact tool couples:
- The reduced complexity climate model MAGICC,
- Outputs from the World Climate Research Program's Coupled Model Intercomparion Project (CMIP) (building on our work on ESM input data and ESM model evaluation) and
- An implementation of methods underpinning MESMER-X. MESMER-X is a statistical emulation method that translates global climate model outputs into local-scale climate projections. This allows us to provide detailed regional projections efficiently while maintaining scientific rigor.
We explore uncertainty in climate hazards and avoid spurious accuracy. We can develop these for any emissions scenarios for which MAGICC can be used to assess the warming implications. This allows users to evaluate the transition and physical impacts of the scenarios they develop.
What is Climate Emulation?
Climate emulation is a statistical technique that learns the relationship between global temperature changes and local climate impacts from detailed Earth System Model(ESM) simulations. Once trained, the emulator can rapidly generate local projections for any global temperature pathway.
Think of it like this: instead of running expensive climate simulations every time, we've trained a statistical model that captures the essential patterns from those simulations.
Bias-Corrected Quantities
The climate indicators represent physical quantities (e.g., temperature in °C, precipitation in mm/year) derived from bias-corrected ESM output. Raw ESM output has been statistically adjusted to match observed historical climate patterns.
For variables that do show changes (like temperature anomaly), the baseline period is explicitly noted.
Source Data
Our projections are trained on bias-corrected data from leading Earth System Models (ESMs). Depending on the client needs, we can generate regional climate impacts from a large number of ESMs, or as few as five:
- GFDL-ESM4 (NOAA Geophysical Fluid Dynamics Laboratory)
- IPSL-CM6A-LR (Institut Pierre-Simon Laplace, France)
- MPI-ESM1-2-HR (Max Planck Institute)
- MRI-ESM2-0 (Meteorological Research Institute, Japan)
- UKESM1-0-LL (UK Met Office)
Using multiple models provides structural diversity and more robust uncertainty estimates. Each model represents the climate system slightly differently, so combining them gives us a better picture of the range of possible outcomes.
Understanding Uncertainty Bands
We derive the median (50th percentile) - the middle of the projected range. We also provide uncertainty bands that represent the 10th to 90th percentile range across 250 Monte Carlo realisations (50 per model). This captures:
- Model uncertainty: Different Earth System Models produce slightly different results
- Natural variability: Climate varies year-to-year even without human influence.
Forthcoming....spatially and temporally correlated climate impacts
We are extending this methodology, developing a machine learning model that generates super-ensembles and custom scenarios, at high spatial and temporal resolution, covering AR6 projection ranges, custom scenarios (IEA, NGFS), and user-specified scenarios. These methods will better quantify changes in spatially and temporally correlated climate impacts.
Licensing and access
We develop these projections for clients, based on the climate impacts most relevant to their modelling and analysis.
Get in touch at contact@climate-resource.com if you are interested in discussing this further.
