Why improve quantification of uncertainty?

Providing information with associated probabilities to quantify the uncertainties involved is not something new. Previous UKCIP climate scenarios included a limited presentation of the associated uncertainties. However, as understanding of impacts and adaptation has grown there has been an increasing call for a more rigorous and quantified treatment of uncertainty. UKCP09 is, in part, a response to this.

In detail

In previous UK climate change scenarios (e.g. UKCIP98 and UKCIP02), information on uncertainties was generally restricted to presentation of single alternative climate scenarios for each of four SRES emission pathways, along with limited information on what other climate models were suggesting.

This was the result of the capabilities of the modelling and scenario development communities (supply side), who were naturally drawn to striving for more accurate projections of the future and hence invested the majority of their resources (financial and computing) into improving climate models. It also suited the expressed requirements of most users (demand side), who were requesting single future climate scenarios that could be used to explore the impacts of climate change, and use this knowledge to develop scoping studies and optimal solutions which were characteristic of the earlier impacts and adaptation assessment approaches.

In common with other fields of earth and environmental sciences, the recognition that we will always be faced with uncertainties in the climate model projections has changed the focus of both the demand-side and supply communities.

Users have recognised the limitations of searching for a single optimal strategy in the face of uncertainty, and have started exploring resilient and flexible strategies that require more information about the associated uncertainties. Simultaneously, recent advancements in climate modelling and computer capabilities have led to much greater investigation of model uncertainty, which allows climate change information which better quantifies the main uncertainties in projections of climate futures. UKCP09 is thus a response to these changing demands and a realisation of the supply capabilities. Users of UK climate scenarios have been quite vocal about their concerns regarding the lack of quantitative information on the uncertainties associated with future climate. This concern is related to the limited utility for decision and policy making where there is inadequate reference to the associated uncertainties.

Expression of this concern was particularly evident during the users’ consultations undertaken by UKCIP following the release of the UKCIP02 scenarios and leading up to UKCP09. In those consultations users expressed the need for improved consideration and quantification of associated uncertainties and for inclusion of the results of a broader spectrum of climate models (more than just the Met Office Hadley Centre models) to be taken into account in the future development of scenarios/projections. It has become clear, that this request is primarily related to the need for information to support more robust (or informed) decisions.

Evidence suggests that decisions taken and policies developed with more of an understanding and consideration of the associated uncertainties are potentially superior in that this knowledge can be used to:

• Inform the selection of strategies and options while not limiting those explored due to incomplete information or false confidence.

• Inform the choice of available strategies and options through assessment of robust options given the potential risks and range of climate futures.

• Inform the selection and timing of adaptation measures consistent with the evolution of climate (and other) risks and opportunities.