Uncertainty
The following sections describe the sources of uncertainty in climate projections, how they appear in CMIP6 climate projections for California, how to examine them using Analytics Engine tools, and how to interpret and communicate them responsibly in planning and decision-making contexts. The analyses presented here use near-surface air temperature (tas) and precipitation (pr) from the CMIP6 archive subsetted to California under the SSP3-7.0 scenario and accessed through the Cal-Adapt: Analytics Engine.
For background on climate projections and models more broadly, and for guidance on how to apply climate data in decision-making, see General Climate Guidance.
What are sources of uncertainty in climate projections?
Climate projections come from ensembles of models rather than a single simulation. When multiple models are run under the same emissions scenario, they produce a range of outcomes. That range is not a flaw, but a quantified expression of genuine scientific uncertainty about how the climate system works.
The IPCC distinguishes three principal sources of uncertainty in climate projections1:
- Model uncertainty (also called inter-model spread or structural uncertainty): differences in how modeling groups represent physical processes such as cloud feedbacks, ocean heat uptake, and land-surface exchanges.
- Scenario uncertainty: future greenhouse gas emissions depend on societal choices that have not yet been made.
- Internal variability: the climate system’s own unforced fluctuations, independent of any external forcing.
The time horizon of the simulations dictates the dominant source of uncertainty. Internal variability is most important in sub-decade scales; model uncertainty dominates over the 10–40 year planning horizon most relevant for infrastructure and resource decisions; and scenario uncertainty becomes the largest source of spread at multi-decadal to centennial scales.
Despite the uncertainties inherent in modeling future climate, they do not invalidate climate data; rather, they represent the complexity and range of potential future outcomes, and ultimately illustrate the critical importance of policy decisions made today and how they will determine the magnitude of future climate changes.
Climate modelers actively work to reduce model errors through improved model physics, increased computational power, and ensemble modeling techniques that aggregate results across multiple models. The next section describes how to examine these uncertainties in the Analytics Engine.
How can users examine uncertainties in climate data?
The goal of the climate projections in the Analytics Engine is not to provide a singular prediction of the future but rather to allow users to examine different plausible future outcomes. The data in Analytics Engine allows users to explore a range of potential future climate projections using different datasets, models, and emission scenarios. Users should be aware that there are uncertainties both in the amount of future greenhouse gas emissions as well as how the climate system will respond to these emissions. These uncertainties are a necessary part of planning for climate hazards and risks. Further, climate modeling processes are continually improving to provide better representations of the climate system.
Although all three sources of uncertainties are important to understand, the relative value of each type of uncertainty depends on the type of application, context, and time-period of decision-making. Therefore, disentangling the different sources of uncertainty and examining how each of them evolve over time, can help energy sector users better characterize and work with the specific uncertainties that are most useful for their context. For example, internal variability might be more important for planning processes that focus on the near-term (e.g. the Integrated Energy Policy Reports, IEPR assessments), whereas scenario uncertainty might be of greater value to consider for end-of-century planning (e.g. long-term distribution planning).
In the Analytics Engine, users can use the internal variability notebook and the model uncertainty notebook to perform some of these uncertainty analyses. Specific recommendations related to appropriate consideration of uncertainty are also included in the Model Selection and Temporal & Spatial Scales pages.