Sea-ice in decadal and long-term simulations with the Max Planck Institute Earth System Model

Abstract

A new type of simulations, so-called decadal predictions, was part of the Coupled Model Intercomparison Project Phase 5 (CMIP5) protocol. In contrast to long-term simulations that are run by global Earth system models (ESMs) to project future climate under various forcing scenarios, the decadal short-term (~10-30 years) simulations aim at predicting some aspects of slow natural variability by initializing the simulations from an observed climate state. In this work, the representation of sea-ice in the CMIP5 initialized decadal predictions from the Max Planck Institute Earth System Model (MPI-ESM) is compared to the one from the MPI-ESM uninitialized free-running hindcast simulations from this model and to observations. The aim of the work is to start exploring the science question whether the initial conditions in the hindcast lead to more accurate retrospective predictions of sea-ice compared to the uninitialized simulations. Each ensemble member of the MPI-ESM decadal simulations is initialized with a standalone experiment from the underlying ocean model (MPI-OM) at the end of the previous year. In this simulation temperature and salinity anomalies are nudged towards three-dimensional ocean fields from the National Center of Environmental Predictions (NCEP) / National Oceanic and Atmospheric Administration (NOAA) reanalysis. For the evaluation, the observational datasets of the Advanced Microwave Scanning Radiometer-Earth observing System (AMSR-E), the National Snow and Ice Data Center (NSIDC) and the Hadley Centre Sea Ice and Sea Surface Temperature (HadISST) were used.