A33M-0387
An Evaluation of the Seasonal Arctic Sea Ice Predictions from CFSv2

Wednesday, 16 December 2015
Poster Hall (Moscone South)
Qiong Yang1, Muyin Wang1 and James E Overland2, (1)University of Washington Seattle Campus, Seattle, WA, United States, (2)NOAA Seattle, Seattle, WA, United States
Abstract:
The rapid reductions in Arctic sea ice have been observed in the past several decades,

especially at the end of the summer melt season in September. It is necessary to have a

reliable seasonal forecast of Arctic sea ice. In this study, we examined the Arctic sea ice

predictions produced by NCEP Climate Forecast System version 2 (CFSv2) in the real-

time operational mode. Forecasts were initialized monthly for two-year period (March

2014 to September 2015). Forecasts of sea ice extent (SIE) and concentration (SIC) were

evaluated against the sea ice analysis (HadISST_ice) from the Hadley Center. We found

that the Arctic September SIE forecasts from CFS were overestimated with the biases in

SIC mainly originated from the Beaufort Sea, Laptev Sea and Fram Strait. For 2014, we

found that the forecast initialized from March with the lead-time of 6 months gave the

best September SIE forecast while the forecast initialized from July with the lead-time of

2 months had the worst September SIE forecast. In order to understand the forecast biases

in September sea ice, the atmospheric forecasted forcings including incoming

solar/Infrared radiation, upward solar/infrared radiation from surface, latent and sensible

heat flux, 2-meter air temperature, cloud fraction, sea level pressure and 10-meter wind

from CFSv2 were evaluated using the European Center for Medium-Range Weather

Forecasts Interim Re-Analysis (ERA-Interim) .