The production of actionable climate science relies on effective communication of regional climate information and its associated uncertainties across sectors. To be of value beyond academic circles, climate data must be sufficiently credible (i.e., physically grounded), understandable (communicated in the vocabulary of the decision-makers), and useful for the particular decisions that need to be made. Comprehensive assessment of both dynamical and statistical climate models adds substantial value to their outputs, particularly when the evaluation criteria are the product of a two-way dialogue between scientists and end-users. Substantial progress has now been made on developing comprehensive frameworks for climate data assessment that incorporate process-oriented, feature-specific, and use-inspired metrics. These efforts have been particularly advanced over the continental U.S. (CONUS) under both the Department of Energy Hyperion and FACETS projects. This project continues these efforts so as to further (1) advance our understanding of processes at the climate-water-energy-land-decision interface, and (2) fundamentally improve our ability to perform credible climate modeling of particular regions. Here a multi-pronged approach has been developed to achieve these goals, by:
First, engaging climate scientists and stakeholders within working groups to build common understanding and strategy for achieving these goals. This project will further refine the engagement process from earlier work and understand how climate datasets are being used for specific decision applications.
Second, developing a storyline context to frame our assessment activities and provide a means to effectively interface our scientific pursuits with stakeholder interests. Formally, a storyline is a physically self-consistent unfolding of a past event, or of a plausible future event or pathway. These storylines are chosen to be representative of major climatic events that have impacted, or would impact, the policy or decision context.
Third, developing new metrics and leveraging existing metrics to evaluate and understand modeled processes, and subsequently inform credibility and uncertainty in light of a non-stationary climate system. We will perform deep dives into weather extremes, snowpack, low-flows and flooding regimes in rivers, water quality, and wind extremes. A key outcome will be the identification of process biases and errors that are most responsible for uncertainties in available products.
Fourth, using differential credibility analysis (DCA) to understand what aspects of predictions and projections of climate change are credible given our understanding of the processes and errors in these models. This process involves a broad assessment of model performance, and will target the validity of climate datasets for decision making.
Fifth, developing a deeper understanding of multi-sector interactions (those at the climate-water-energy-land-decision interface), the interplay between global and regional climate forcings and implications for key aspects of energy supply using climate simulations conducted at a range of spatial scales for scenarios of changing land use, irrigation and energy mix.