Induced surface fluxes: a new framework for attributing Arctic seaice volume balance biases to specific model errors
A new framework is presented for analysing theproximate causes of model Arctic sea ice biases, demon-strated with the CMIP5 model HadGEM2-ES (Hadley Cen-tre Global Environment Model version 2 – Earth System).In this framework the Arctic sea ice volume is treated as aconsequence of the integrated sur...
| Main Authors: | , , , , |
|---|---|
| Format: | Text |
| Language: | unknown |
| Published: |
Zenodo
2019
|
| Subjects: | |
| Online Access: | https://dx.doi.org/10.5281/zenodo.3565622 https://zenodo.org/record/3565622 |
| Summary: | A new framework is presented for analysing theproximate causes of model Arctic sea ice biases, demon-strated with the CMIP5 model HadGEM2-ES (Hadley Cen-tre Global Environment Model version 2 – Earth System).In this framework the Arctic sea ice volume is treated as aconsequence of the integrated surface energy balance, viathe volume balance. A simple model allows the local de-pendence of the surface flux on specific model variables tobe described as a function of time and space. When theseare combined with reference datasets, it is possible to esti-mate the surface flux bias induced by the model bias in eachvariable. The method allows the role of the surface albedoand ice thickness–growth feedbacks in sea ice volume bal-ance biases to be quantified along with the roles of modelbias in variables not directly related to the sea ice volume. Itshows biases in the HadGEM2-ES sea ice volume simulationto be due to a bias in spring surface melt onset date, partlycountered by a bias in winter downwelling longwave radia-tion. The framework is applicable in principle to any modeland has the potential to greatly improve understanding of thereasons for ensemble spread in the modelled sea ice state.A secondary finding is that observational uncertainty is thelargest cause of uncertainty in the induced surface flux biascalculation. |
|---|