Developing Actionable Regional Climate Models and Data for Communities and Decision-makers Across Alaska

We present a novel, high-resolution land-atmosphere regional climate model (RCM) configuration with multi-decadal historical and future climate simulations for the greater Alaska and Yukon River region. To address diverse climate information needs and advance understanding of Arctic change, key model decisions were informed by Indigenous decision-makers that responded to a targeted survey. Advances to two state-of-the-science models were made. First, to enable climate change assessments at Arctic community scales, the Regional Arctic System Model (RASM) was modified from its 50 km atmosphere-land grid to 4 km. Next, to advance RASM land-modeling and leverage community models, the Community Terrestrial Systems Model (CTSM) was coupled to RASM with enhancements made for hydrological modeling. This configuration, forced by ERA5 reanalysis, has resulted in simulations of historical (water years 1991-2021) precipitation, temperature, streamflow, and snowpack (among other variables) with unprecedented skill and fidelity for a land-atmosphere coupled RCM. An ensemble of six future simulations for the mid-21st century, WY2035-2065, using two different approaches to address community desires for a range of futures to inform decision making was then developed. The two future approaches are the pseudo global warming (PGW) approach, a climate perturbation approach applied to historical weather, and direct Earth System model downscaling.

Our co-design effort has resulted in a unique fit-for-purpose model and dataset for a broad range of interdisciplinary climate change research studies and intended to serve Arctic Community needs, with a total of 330 simulation years available. Here we highlight the model development advances, characteristics of the resultant dataset, and examples of projected future regional hydrometeorological change across the future ensemble including contrasts across the PGW and direct downscaled futures.