资源环境科技发展态势分析平台

Ice mass loss from the Amundsen Sea Embayment ice streams in West Antarctica is a major source of uncertainty in projections of future sea level rise. Physically based ice flow models rely on a number of parameters that represent unobservable quantities and processes, and accounting for uncertainty in these parameters can lead to a wide range of dynamic responses. Here we perform a Bayesian calibration of a perturbed parameter ensemble, in which we score each ensemble member on its ability to match the magnitude and broad spatial pattern of present-day observations of ice sheet surface elevation change. We apply an idealized melt rate forcing to extend the most likely simulations forward to 2200. We find that diverging grounding line response between ensemble members drives an exaggeration in the upper tail of the distribution of sea level rise by 2200, demonstrating that extreme future outcomes cannot be excluded.