Assessing Skill in a Low-Resolution CESM Ensemble

Tuesday, May 13, 2014 - 07:00
Add to Calendar

Climate models are valuable tools for understanding how Earth's climate system is changing, yet they are inherently uncertain. Two key sources of uncertainty are related to initial conditions and internal model variability, which can increase the spread of climate projections. Here we present results from a 50-member climate change ensemble experiment, utilizing a low-resolution configuration of the fully-coupled Community Earth System Model (CESM), comprised of transient hind casts and projections (1850-2100) using the RCP8.5 forcing scenario. The transient simulations are initialized from different initial model states, sampled from a ~10,000 year fully-coupled unforced equilibrium simulation. We find this initial conditions uncertainty, which reflects unforced internal variability of the coupled ocean-atmosphere system, has a significant effect on projections of key climate change metrics, and the projected ranges increase with decreasing spatial scale. The model demonstrates considerable skill in simulating key regional climate processes. Further, the ensemble robustly captures the trends and variance structures of observed time series of some key climate change metrics across multiple spatial scales, such as temperature, precipitation, and sea-level changes due to thermal expansion of the ocean. Given the tradeoffs between model resolution and computational cost, our results indicate that ensemble frameworks such as presented here provide a useful resource for uncertainty quantification, integrated assessment, and analyzing climate change impacts.

Presentation file(s):