C33A-07 - Model-observation comparison of sea ice floe size distribution in the Arctic
Presentation Information
TitleC33A-07 - Model-observation comparison of sea ice floe size distribution in the Arctic
AbstractFloe size distribution (FSD) represents varying sizes of sea ice floes and is a key variable for both the thermodynamics and dynamics of sea ice in the marginal ice zones (MIZ). In recent years a set of FSD processes have been included in sea ice models. However, existing observations are still sparse in space and time, making evaluation of FSD models performance difficult and hindering further developments. Here, we present a novel observational FSD data product we have developed from 1-m resolution MEDEA satellite and from 0.5-m resolution NASA Worldview Imagery products. We have created the dataset for the Chukchi Sea (70° N, 170° W) for 2006-2014 and Fram Strait (84.9° N, 0.5° E) for 2000-2014 for summer months May-July as demonstration cases to assess model performance. We compare the novel FSD observations with outputs from the four following FSD model configurations implemented in the stand-alone CICE sea ice model and used by the modelling community to assess the model performance. These are: (i) the diagnostic-prognostic Waves-in-Ice module and Power law Floe Size Distribution (WIPoFSD) model; (ii) the two fully prognostic configurations of the CPOM-FSD model; and (iii) the fully prognostic model FSDv2WAVE. Among the four, the prognostic FSDv2WAVE model simulates the most fragmented sea ice and has the most spatially variable FSD both in the Chukchi Sea and Fram Strait. Simulated FSD in the WIPoFSD is in a range of the observation; the two CPOM-FSD prognostic models overestimate the mean values of the FSD. By discarding floes with diameters smaller than 20 metres, the statistics of the simulated FSD improves, suggesting that the models are still missing physical processes for the small sea ice floes. This research contributes to improving MIZ modelling and development of model parameterisations for large scale climate models, thus leading to a more accurate projection of the future Arctic sea ice and climate. For generating novel FSD observation, YW and BH acknowledge financial support by NERC grants NE/S002545/1 and NE/R000654/1. AB is funded by NERC, reference NE/R016690/ and NE/R000654/1. YA acknowledges support from NERC grants NE/N018044/1, NE/R000085/1, NE/T000546/1 and NE/R015953/1. CH was supported by NASA grant 80NSSC20K0959 and by Schmidt Futures.
SectionCryosphere (C)
AuthorYanan
InstitutionUniversity of Huddersfield
In Person/Virtual StatusVirtual: I am presenting and participating virtually.