The Atmospheric Response to Realistic Reduced Summer Arctic Sea Ice Anomalies - Arctic Sea Ice Decline: Observations, Projections, Mechanisms, and Implications - page 88

The Atmospheric Response to Realistic Reduced Summer
Arctic Sea Ice Anomalies

Uma S. Bhatt, 1 Michael A. Alexander, 2 Clara Deser, 3 John E. Walsh, 4 Jack S. Miller, 5
Michael S. Timlin, 6 James Scott, 2 and Robert A. Tomas 3

The impact of reduced Arctic summer sea ice on the atmosphere is investigated
by forcing an atmospheric general circulation model, the Community Climate
Model (CCM 3.6), with observed sea ice conditions during 1995, a low-ice year.
The 51 experiments, which spanned April to October of 1995, were initiated with
different states from a control simulation. The 55-year control was integrated using
a repeating climatological seasonal cycle of sea ice. The response was obtained from
the mean difference between the experiment and control simulations. The strongest
response was found during the month of August where the Arctic displays a weak
local thermal response, with warmer surface air temperatures and lower sea level
pressure (SLP). However, there is a significant remote response over the North
Pacific characterized by an equivalent barotropic (anomalies are collocated with
height and increase in magnitude) structure, with anomalous high SLP collocated
with a ridge in the upper troposphere. The ice anomalies force an increase (decrease)
in precipitation north of (along) the North Pacific storm track. A linear baroclinic
model forced with the transient eddy vorticity fluxes, transient eddy heat fluxes,
and diabatic heating separately demonstrated that transient eddy vorticity fluxes
are key to maintaining the anomalous high over the North Pacific. The model's
sensitivity to separately imposed ice anomalies in the Kara, Laptev-East Siberian,
or Beaufort seas includes SLP, geopotential height, and precipitation changes that
are similar to but weaker than the response to the full sea ice anomaly.

1 Geophysical Institute, Department of Atmospheric Sciences,
University of Alaska Fairbanks, Fairbanks, Alaska, USA.
2 NOAA Earth System Research Laboratory, Boulder, Colorado,
USA.
3 National Center for Atmospheric Research, Boulder, Colorado,
USA.
4 International Arctic Research Center, University of Alaska
Fairbanks, Fairbanks, Alaska, USA.
5 ARSC, University of Alaska Fairbanks, Fairbanks, Alaska, USA.
6 Department of Atmospheric Sciences, University of Illinois at
Urbana-Champaign, Urbana, Illinois, USA.

1. INTRODUCTION

Summer sea ice in the Arctic decreased at a rate of 4-6%
per decade [ Deser et al. , 2000] through the 1990s, and the
melt rate has accelerated to 10% per decade [ Stroeve et al. ,
2007; National Snow and Ice Data Center, press release,
1 October 2007, available at nsidc.org/news/press/2007_
seaiceminimum/20071001 _ pressrelease.html) in the 2000s.
In the 1990s the melting of Arctic ice was consistent with
the positive phase of the North Atlantic Oscillation (NAO),
which is characterized by enhanced storminess and warm
moist air penetration into the Arctic. The NAO has ap-
proached more neutral values since 2000, yet the ice melt has
accelerated. The observed influx of plugs of warm Atlantic
layer water into the Arctic provides one likely mechanism