Arctic Dipole Anomaly (DA) Drove the Record Lows in the Arctic Sea Ice Extent Jia Wang ([email protected]) NOAA Great Lakes Environmental Research Lab, Ann Arbor, Michigan Jinlun Zhang (1) , Eiji Watanabe (2), Kohei Mizobata (3), John Walsh (2), and, Xuezhi Bai (4), and Moto Ikeda (5) 1) APL, Univ. of Washington, WA USA 2) IARC, UAF, AK USA 3) Tokyo Univ. of Marine Science and Tech., Japan 4) University of Michigan, Cooperative Institute of Limnology and Ecosystems Research (CILER) 5) Hokkaido University, Japan Arctic Modeling Group: http://www.frontier.iarc.uaf.edu/~jwang/main.html
29
Embed
Arctic Dipole Anomaly (DA) Drove the Record Lows in the Arctic Sea Ice Extent
Arctic Dipole Anomaly (DA) Drove the Record Lows in the Arctic Sea Ice Extent. Jia Wang ( [email protected]) NOAA Great Lakes Environmental Research Lab, Ann Arbor, Michigan Jinlun Zhang (1) , Eiji Watanabe (2), Kohei Mizobata (3), John Walsh (2), and, Xuezhi Bai (4), and Moto Ikeda (5) - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Arctic Dipole Anomaly (DA) Drove the Record Lows in the Arctic
No correlation between the sea ice export in Fram Strait and AO (Vinje 2001; Hilmer and Jung 1999); Sea ice flux vs. SLP difference across Fram Strait (Kwok and Rothrock 1999);
Questions:
1) Is the AO/NAO the only dominant mode driving the Arctic ice-ocean system?
2) Are both Arctic sea ice circulation and sea ice export (sea ice thinning) only related to the AO (in a sense of cyclonic or anti-cyclonic anomaly)?
3) During neutral/negative AO phase (after year 2000), why sea ice set record lows one after one in the western Arctic, leading to a record minimum in September 2007!?
History of DA:1) Skeie (2000): BO; 2) Holland (2003); 3) Goose et al. 2003; 4) Semenov and Bengttson (2003)0) Wang et al. (1995)—Internal report of CCGCR of McGill
BO: EOF2, Skeie (2000)Ice export regresses to SLP in CCM2, Holland (2003)
EOF2 of Wang et al. (1995)Unpublished/internal report
One regime withmax. ice exportin coupled climatemodel
2. Data and Methods
• Atmosphere: NCEP/NCAR Reanalysis, 1948-2007
• SIC: Conventional (1901-Sep. 1978; Walsh and Chapman 1990), and SMMR/SSMI (NASA, Oct. 1978-2008; Parkinson 1989), 1x1 degree grid, Arctic Ocean and subpolar regions, 1901-2007
• Sea ice drift: IABP Dataset,1979-2002
Data and Methods (cont.)
• EOF analysis• Climatology/anomaly• Composite analysis and T/F-test• Correlation analysis/regression, & Monte Carlo simulation• Case study • Modeling:
1) Japan CCSR/NIES/FRCGC global GCM: 1900-20102) Regional Coupled Ice-Ocean Model (CIOM) in the pan Arctic and North Atlantic Ocean (Wang et al. 2002, 2005), and Bering/Beaufort/Chukchi seas (Wang et al. 2008)
3. Dipole-Anomaly (DA, 2nd EOF mode of SLP 70 North)
What is DA? Second mode of SLP of the Arctic Ocean (from 70-90N) Wu, Wang, Walsh (2006, J. Climate); Watanabe, Wang, Hasumi, Sumi (2006, GRL);Maslanik et al. (2007, GRL)
+DA
-DA
+AO
TDS
HighLowLow
-AO
Regression maps of AO and DA using GCM (K1) and NCEP
59%
19%
K1
K1
63%
14%
NCEP
NCEP
Regressed winter mean SLP anomalies to each EOF mode (NCEP) [hPa]
EOF 1st mode <AO> - Annular structure - In the positive AO phase high : Arctic region low : Mid-latitude
EOF 2nd mode <DA> - Dipole structure - In the positive DA phase high : Greenland Sea low : Laptev Sea
Each mode is independentby test of North et al. (1982)
GCM’s AO/DA composite anomalies of sea ice thickness and velocity
(AO +) – (AO -) (DA +) – (DA -)
Difference of sea ice thickness (cm) and velocity (cm/s)between the positive and negative phases
Circulation shows cyclonic anomaly. Thickness difference is not significant.
Circulation shows meridional anomaly.Thickness difference is significant.
IABP sea ice velocity regressed
to EOF1 (AO, upper) and EOF2
(DA, lower)
Sea ice volume fluxEOF1 (AO)
EOF2 (dipole)
Fram Strait ice volume transport vs. DA (r=.33) and AO (r=.04)
5. Summer DA and sea ice minima:1995, 1999, 2002, 2005, 2007, and
2008 (Wang et al. 2009, GRL, in press)
+DA
b)b)
a)
Record lows:
1995,
1999,
2002,
2005,
2007,
2008 (not record low,
but 2nd lowest ever!)
+AO
-AO
DA+
-DA
62% 13%
+DA
-DA+AO
-AO
EOF1 EOF2
EOF1 EOF2
DJF DJF
JJA JJA
a) b)
50% 16 %
d)c)
e) Winter
f) Summer
2007: NCEP/NCAR SLP and surface wind anomaly
+DA+DA
+DA
+DA Summer 2007 falls in state: -AO/+DA
SLP anomaly was a DA-dominated two-center structure, and the wind anomaly was meridional, blowing from the western to the eastern Arctic This DA-induced wind anomaly was responsible for the 2007 summer minimum
DA predicts record lows: 1995, 2002, 2007, and 2008 (+DA persists from W/S-S); 1999 and 2005 (-DA in W/S, but +DA in summer). So, summer DA is the key!
Since 1995, AO was near neutral and negative, while the DA was active.
Scatter plot with summer DA as x-axis and winter-spring DA as y-axis
Record low years: 1995, 2002, 2007 and 2008) in first quadrant with +DA persistent from winter-spring to summer. In 1999 and 2005, strong summer +DA contributed to the ice minimum.
Heat flux increased along Bering Strait (Mizobata et al., submitted)
Heat Flux was calculated using in situ observation and Satellite SST
0
1
2
3
4
5
6
7
-1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
Series1
Summer DA Index
Heat F
lux via B
ering Strait in T
W
2004 20072006
2005
20002003 2001
2002
Fig. 3bThe +DA strengthened inflow of the warm Pacific water since 2000s
Relation between Bering strait heat flux and summer DA index
+DA +DA
a) b)
c)
Wind Anomaly in m/s
Fig. 4
SLP and wind anomalies in Aug 2007 (Left) and 2008 (right)
Arctic Sea Ice National Snow and Ice Data Center
http://nsidc.org/arcticseaicenews/index.html
- 2nd smallest ice pack on record
- Large reduction in multi-year ice
-Most of ice pack is thin first year ice susceptible to melting
Comparison between the PIOMAS-simulated and SSM/I-observed ice extends for the period 1978-2008
Simulated compares well against the obs. The correlation is 0.93 in Sep. and 0.92 in Jan-Sep mean. The model reproduces summer ice minima in 1995, 2002, 2005, 2007 and 2008 as well, not for 1999
6. Conclusions• DA is the second dominant mode in the central Arctic
(local). Its dynamic impact is more important than the AO, while the local thermodynamic effect is also important, reflecting the feedback of the local ice anomaly to the atmosphere (Wu et al. 2006; Watanabe et al. 2006)
• DA-related wind anomaly is meridional: +DAWind anomaly blows from Pacific Arctic to Atlantic Arctic, enhancing TDS, sucking more Pacific Water inflow, driving away and melting more sea ice in Pacific Arctic (Woodgate et al. 2005; Shimada et al. 2006)-DAWind anomaly blows from Atlantic Arctic to Pacific Arctic, weakening TDS, blocking Pacific Water inflow, detaining more ice in the Pacific Arctic
Conclusions (cont.)• Summer sea ice minima in the 2000s, in particular 2007
summer (Zhang), was due to +DA (Wu et al., Watanabe et al. 2006; Wang et al. 2009), while AO was in its negative phase (Overland and Wang 2005; Maslanik et al. 2007)!
• DA has two impacts on sea ice of Pacific Arctic:
-- Direct (short-term, seasonal): driving sea ice and enhancing TDS, +SAT/SST, local ice/ocean albedo feedback (Wang et al. 2005)
-- Indirect (long-term, interannual): sucking in more warm Pacific Water inflow, +SAT/SST, melting more sea ice, enhancing local ice/ocean albedo feedback