Climate Model Simulations of Extreme Cold-Air Outbreaks (CAOs)

1
Climate Model Simulations of Extreme Cold-Air
Outbreaks (CAOs)
Steve Vavrus Center for Climatic
Research University of Wisconsin-Madison John
Walsh International Arctic Research
Center University of Alaska-Fairbanks Bill
Chapman Diane Portis Department of Atmospheric
Sciences University of Illinois
2
Importance of Cold-Air Outbreaks

• CAO Incursion of extremely cold polar air mass
  into the middle
• -lower latitudes of Eurasia and North America
• CAOs cause many deaths each winter (30/year in
  U. S., over
• 1,000 fatalities during 2003 in Bangladesh and
  India)
• Result in billions of dollars of economic
  losses, especially in
• agricultural sector (e.g., Florida citrus
  industry)
• Cold surges strongly affect climatological heat
  balance in subtropics
• and may influence onset of El Niños (via
  westerly wind bursts)

3
Cold-Air Outbreaks are complex phenomena . . .

• Occurrence of CAOs is highly variable and
  associated with low-
• frequency circulation modes (Arctic
  Oscillation, PNA pattern)
• CAOs may be triggered by many factors, both
  local and remote
• Climate models may help explain CAO behavior,
  but they have been
• underutilized for this purpose
• Trend of CAOs in past decades-century has not
  followed mean
• warming trend in either U. S. or Europe

4
Observed Variability of Extreme Events
United States
Cold Waves
United States
Heat Waves
Kunkel et al., 2002
5
Observed Variability of Extreme Events
United States
Cold Waves
Walsh et al., 2001 (J. Climate)
United States
Heat Waves
Kunkel et al., 2002
6
Observed Variability of Extreme Events
United States
Cold Waves
Walsh et al., 2001 (J. Climate)
United States
Heat Waves
Kunkel et al., 2002
7
Objectives

• Evaluate ability of climate models to represent
  CAOs
• (frequency, magnitude, synoptics, event types)
• Use models and NCEP/NCAR Reanalysis to determine
  the
• relative importance of thermodynamic and
  dynamic processes
• Analyze precursor conditions and forcing factors
  (e.g., SSTs,
• snow cover, clouds, circulation) to explain
  origin of CAOs
• Estimate how and why CAOs may change in the
  future, due to
• enhanced greenhouse forcing

8
Observed and Simulated Extreme Cold in GCMs
9
Our Definition of a Cold Air Outbreak (CAO)
Two or more consecutive days in which the daily
mean temperature is at least two standard
deviations below the DJF mean temperature
10
Simulated and Observed CAO Characteristics in
CCM3 AMIP Runs (driven by observed SSTs from
1979-1998)
11
Simulated and Observed CAO Characteristics in
CCM3 AMIP Runs (driven by observed SSTs from
1979-1998)
12
Skewness -0.37
13
CAO
WAO
14
Very Cold Days Actual (Expected)
T lt Tmean- 2s 4.2 (2.3)
CAO
WAO
Very Warm Days Actual (Expected)
T gt Tmean 2s 1.0 (2.3)
15
Wintertime Temperature Skewness (Observed)
16
Wintertime Temperature Skewness (AMIP runs, CSM1,
and Observed)
17
CCM3 Wintertime Temperature Skewness in Pacific
Northwest
Modeled Skewness -1.22 Observed Skewness
-1.19
18
CSM1 Simulation of a CAO (February 24 - March 8)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
Response of CCM3 AMIP Simulation to ENSO
26
Response of CCM3 AMIP Simulation to ENSO
T42
rCAO-Nino3.4 -0.18
X
T85
rCAO-Nino3.4 0.04
OBSERVED
X
rCAO-Nino3.4 -0.39
27
Simulated Future CAOs in CSM1 (Transient CO2
Forcing)
28
Simulated Future Wintertime Circulation Changes
CSM1
L
L
L
L
H
29
Simulated Future Wintertime Circulation Changes
CSM1
CCSM3
300 hPa Height Anomalies (m)
L
L
Sea Level Pressure Anomalies (hPa)
L
L
H
30
Conclusions and Future Work

• Climate models seem capable of reproducing the
  first-order
• characteristics of cold air outbreaks
• Capability of models to represent other major
  aspects of CAOs
• is under investigation
• Relative roles of various forcing terms need to
  be established
• (e.g., SST variability, snow cover, circulation
  modes, etc.)
• Future behavior of CAOs may be a complicated
  function of
• mean temperature, circulation changes, and
  remote forcings

31
Our Definition of a Cold Air Outbreak (CAO)
Two or more consecutive days in which the daily
mean temperature is at least two standard
deviations below the DJF mean temperature (stand
ard deviation defined here as the average daily
standard deviation of interannual air
temperature variations during winter) Standard
Deviation S (sDec.1 sDec.2 sDec.3 . . .
sFeb. 28)/90