What does this research mean for the field?

Implementing a stochastically perturbed parameter (SPP) scheme in sea ice models enhances ensemble spread by approximately 10%, improving the calibration of probabilistic sea ice forecasts and significantly increasing seasonal atmospheric forecast skill at high and mid-latitudes. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To assess how introducing a stochastically perturbed parameter scheme into the SI3 sea ice model impacts seasonal forecast calibration and atmospheric skill in the Northern Hemisphere.

September 2, 2026Weather and Climate DynamicsOpen Access

The impact of stochastic sea ice perturbations on seasonal forecasts

View Full Paper

Ask AI

Bookmark

Authors

KSKristian Strømmen MMMichael Mayer ASAndrea Storto

Discussion

Member takes

Overview

Climate modeling study reveals improved sea ice and atmospheric seasonal forecast skill across the Northern Hemisphere, indicating stochastic parameter perturbations enhance ensemble reliability.

Key Points

To assess how introducing a stochastically perturbed parameter scheme into the SI3 sea ice model impacts seasonal forecast calibration and atmospheric skill in the Northern Hemisphere.
Implemented a stochastically perturbed parameter (SPP) scheme into SI3, the sea ice component of the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecast System (IFS).
Evaluated probabilistic seasonal ensemble forecasts for Northern Hemisphere summer and winter against unperturbed control forecasts.
Inclusion of the SPP scheme expanded ensemble spread for sea ice concentration (SIC) and sea ice thickness (SIT) by approximately 10% relative to unperturbed forecasts, yielding better-calibrated probabilistic predictions.
Perturbations induced robust shifts in the mean state, including reduced mean SIC, winter ice redistribution from the central Arctic to the ice edge, and altered lower-tropospheric air temperature biases up to 850 hPa.
A significant improvement occurred in winter seasonal forecast skill for 500 hPa geopotential height (Z500) over the Euro-Atlantic domain, projecting partially onto the North Atlantic Oscillation.

Cite This Study

Strømmen et al. (2026) studied this question.

synapsesocial.com/papers/6a97e20ec562ede874ec6140 https://doi.org/10.5194/wcd-7-1593-2026

View Full Paper

Ask AI

Bookmark

September 2, 2026Weather and Climate DynamicsOpen Access

The impact of stochastic sea ice perturbations on seasonal forecasts

View Full Paper

Ask AI

Bookmark

Authors

KSKristian Strømmen MMMichael Mayer ASAndrea Storto

Discussion

Member takes

Overview

Climate modeling study reveals improved sea ice and atmospheric seasonal forecast skill across the Northern Hemisphere, indicating stochastic parameter perturbations enhance ensemble reliability.

Key Points

To assess how introducing a stochastically perturbed parameter scheme into the SI3 sea ice model impacts seasonal forecast calibration and atmospheric skill in the Northern Hemisphere.
Implemented a stochastically perturbed parameter (SPP) scheme into SI3, the sea ice component of the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecast System (IFS).
Evaluated probabilistic seasonal ensemble forecasts for Northern Hemisphere summer and winter against unperturbed control forecasts.
Inclusion of the SPP scheme expanded ensemble spread for sea ice concentration (SIC) and sea ice thickness (SIT) by approximately 10% relative to unperturbed forecasts, yielding better-calibrated probabilistic predictions.
Perturbations induced robust shifts in the mean state, including reduced mean SIC, winter ice redistribution from the central Arctic to the ice edge, and altered lower-tropospheric air temperature biases up to 850 hPa.
A significant improvement occurred in winter seasonal forecast skill for 500 hPa geopotential height (Z500) over the Euro-Atlantic domain, projecting partially onto the North Atlantic Oscillation.

Cite This Study

Strømmen et al. (2026) studied this question.

synapsesocial.com/papers/6a97e20ec562ede874ec6140 https://doi.org/10.5194/wcd-7-1593-2026

View Full Paper

Ask AI

Bookmark