To speak of uncertainty implicitly

Climateurope2 has an ambition to standardise climate services and that includes uncertainty. However, uncertainty can mean different things in different contexts.

Last year, Climateurope2 organised a workshop on communicating climate uncertainty to find out how uncertainty was being communicated in different sectors of the climate information value chain: from national weather services, to risk assessment enterprises down to local decision makers who need to make their infrastructure and land use planning more resilient to our changing climate.

Perhaps the most unusual lesson we learned was not to talk about uncertainty at all. There are many ways in which uncertainty can be inferred. For instance, by showing the consequences of just two different climate scenarios, it is understood that there is uncertainty about what the future climate has in store.

Breaking down uncertainty

From a data provision perspective, uncertainty quantification and communication can be considered in terms of different aspects.

Examples of uncertainty aspects are:

• Location of uncertainty in the value chain from emissions information through climate models and impact models and down to adaptation decisions.
• Nature of the uncertainty: whether it is intrinsic to the system, due to incomplete knowledge or due to lack of calculation capacity.
• Level of uncertainty from statistical uncertainty through scenario uncertainty to recognised ignorance.

The aspect of the uncertainty will determine the type of communication. For instance, a statistical level of uncertainty that is typical of weather forecasts can be communicated with graphs, whereas a scenario level of uncertainty would be better communicated with a narrative or storyline approach.

Scenarios are used to understand climate change on generational timescales. The climate our children will experience in their lifetimes depends on policy choices that are being made now. At those timescales, uncertainty about how we, as humans, play our part in the climate system is one of the biggest unknowns.

Tailor uncertainty communication to the user

Knowing the intended use of the climate information allows us to deliver those aspects of uncertainty that are most relevant with a precision that makes sense. We also need to consider the user's level of understanding and the information that it is possible to obtain from climate scenarios.

Placing the user at the centre of the communication means that we can tailor the information to their concerns. For instance, if resilience to extreme events is the most important aspect of climate change to the user, the scenario uncertainty can be reduced by constraining the analysis to only those scenarios that manifest the extreme events of concern. However, any assumptions or analysis methods that constrain the uncertainty context must be communicated transparently alongside the findings.

We need standards for uncertainty communication

Communicating uncertainty that is relevant to the user in a clear way can be an important tool for building trust. When climate risks remain obvious despite the uncertainties, more confidence can be placed in the assessment and bolder choices can be made.

Given the many different aspects of uncertainty, a key challenge of uncertainty communication is to be transparent without confusing users, especially when different datasets are being combined. If a user has different uncertainty information from different channels it can become confusing and trust may be lost. This is an area where standards and protocols have the potential to bring much needed clarity.

So what’s next?

We’ll be focusing on outreach in the coming months to find out how different sectors are using climate information, what their unique needs are and where those needs overlap. With this we aim to develop strategies for the delivery of information about climate uncertainty.