C. G. Fletcher, P. J. Kushner, J. Cohen - Atmospheric Dynamics Group
C. G. Fletcher, P. J. Kushner, J. Cohen - Atmospheric Dynamics Group
C. G. Fletcher, P. J. Kushner, J. Cohen - Atmospheric Dynamics Group
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Can the Stratosphere Control the<br />
Extratropical Circulation Response to<br />
Surface Forcing?<br />
Chris <strong>Fletcher</strong> and Paul <strong>Kushner</strong><br />
<strong>Atmospheric</strong> Physics <strong>Group</strong><br />
University of Toronto<br />
Canada<br />
Judah <strong>Cohen</strong><br />
AER Inc., Lexington, MA, USA<br />
Thanks: Steven Hardiman (U Toronto), Michael Sigmond (UVic) and<br />
CFCAS funding<br />
AGU Chapman Conference, Santorini. September 27, 2007.
In this talk I will:<br />
• Explain why we are still interested in Siberian snow cover<br />
• Show results from a large ensemble of transient simulations<br />
using a low top model forced with anomalous Siberian snow<br />
extent<br />
• Demonstrate the large variability in tropospheric responses to<br />
snow forcing<br />
• Try to convince you that, to better predict how the troposphere<br />
will respond to snow forcing, you should look at the initial state<br />
of the stratosphere rather than the troposphere<br />
• Conclude that it’s probably not just about snow
Why are we interested in Siberian snow?<br />
• Snow forcing in autumn/fall is<br />
proposed as a source of<br />
seasonal predictability in the NH<br />
extratropics during winter<br />
• Dynamical mechanism<br />
involves wave-mean flow<br />
interaction in the stratosphere<br />
<strong>Cohen</strong> et al. [2007]
Results from NCEP Reanalysis<br />
• Correlations ~0.5 ⇒ suggests it’s unreliable<br />
• Stratospheric circulation anomalies can arise without<br />
a clear tropospheric precursor<br />
<strong>Cohen</strong> et al. [2007]<br />
<strong>Cohen</strong> et al. [2007]
Results from NCEP Reanalysis<br />
• Correlations ~0.5 ⇒ suggests it’s unreliable<br />
• Stratospheric circulation anomalies can arise without<br />
a clear tropospheric precursor<br />
<strong>Cohen</strong> et al. [2007]<br />
<strong>Cohen</strong> et al. [2007]
Results from ECHAM3 low top AGCM<br />
<strong>Cohen</strong> et al. [2007] Gong et al [2003]
Motivation<br />
• Reanalysis data are suggestive but not conclusive; r ~ 0.5<br />
• Previous modeling results are from a small ensemble, an old AGCM<br />
and with no information about response variability<br />
Research Questions:<br />
1. Can snow really act as a<br />
precursor to strat-trop<br />
interaction?<br />
2. If so, are there conditions<br />
where interaction is more<br />
likely to occur?<br />
<strong>Cohen</strong> et al. [2007]
GFDL AM2 low-top AGCM<br />
Full atmospheric GCM run in standard configuration for<br />
tropospheric seasonal/climate prediction:<br />
– Finite-volume dynamical core: 2 o x 2.5 o (lat x lon)<br />
– 24 vertical levels with lid at 1hPa; 4 above 100hPa<br />
– Rayleigh drag in top level sponge layer<br />
– Decorrelation timescales of NAM and geopotential height<br />
compare well with reanalyses
Experimental Design<br />
i. Set of 100 independent Oct 1st initial conditions from long preindustrial<br />
control run:<br />
– <strong>Atmospheric</strong> composition = 1870 levels<br />
– Climatological SST / sea ice
Experimental Design<br />
i. Set of 100 independent Oct 1st initial conditions from long preindustrial<br />
control run:<br />
– <strong>Atmospheric</strong> composition = 1870 levels<br />
– Climatological SST / sea ice<br />
ii.<br />
From each initial condition we fix snow mass at Oct 1st levels then<br />
run two new simulations Oct 1st - Dec 31st:<br />
(1) HIGH SNOW = Fixed Oct 1st snow + 40cm snow over Siberia<br />
(2) LOW SNOW = Fixed Oct 1st snow<br />
Snow Mass for Ensemble Member k<br />
HIGH SNOW k<br />
Initial<br />
Condition k<br />
LOW SNOW k<br />
40cm<br />
Siberia<br />
Time:<br />
Oct 1st . . . . . . . . . . . . . . . . . . . . . . .Dec 31st
Experimental Design<br />
i. Set of 100 independent Oct 1st initial conditions from long preindustrial<br />
control run:<br />
– <strong>Atmospheric</strong> composition = 1870 levels<br />
– Climatological SST / sea ice<br />
ii.<br />
From each initial condition we fix snow mass at Oct 1st levels then<br />
run two new simulations Oct 1st - Dec 31st:<br />
(1) HIGH SNOW = Fixed Oct 1st snow + 40cm snow over Siberia<br />
(2) LOW SNOW = Fixed Oct 1st snow<br />
Snow Mass for Ensemble Member k<br />
HIGH SNOW k<br />
Initial<br />
Condition k<br />
LOW SNOW k<br />
40cm<br />
Siberia<br />
Time:<br />
Oct 1st . . . . . . . . . . . . . . . . . . . . . . .Dec 31st<br />
iii.<br />
Response is defined as minus
Surface Albedo Response<br />
Snow Perturbation Region<br />
(typical January extent)<br />
Albedo response ~0.4-0.5
Early Surface Response to Snow Forcing<br />
Peak Cooling ~ 12 K in<br />
2 weeks<br />
SLP maximum ~ 6 hPa
Polar Cap Height Response<br />
70<br />
-25<br />
15
Polar Cap Height Response<br />
70<br />
-25<br />
15<br />
Response highly<br />
variable in the<br />
troposphere:
Can we Better Predict the Response?<br />
[Z] Initial Condition:<br />
5 days before perturbation<br />
48<br />
-12<br />
Strongest and most<br />
significant ‘predictor’ is<br />
located in the lower<br />
stratosphere<br />
Following Reichler et al. [2005]
Can we Better Predict the Response?<br />
[Z] Initial Condition:<br />
5 days before perturbation<br />
Initially<br />
WARM<br />
115<br />
Initially<br />
COLD<br />
50<br />
Strongest and most<br />
significant ‘predictor’ is<br />
located in the lower<br />
stratosphere<br />
25<br />
Following Reichler et al. [2005]
Strat-Trop Interaction Diagnostic<br />
Snow forcing begins Oct 1st, but strat-trop interaction is associated<br />
with WAF pulses whose timing is difficult to predict:<br />
- Find strongest WAF pulses then look at lagged SLP response<br />
- Does strat. initial condition influence interaction?<br />
d1-50: Find day of max upward<br />
∆WAF at 50 hPa<br />
(e.g. Polvani and Waugh [2004])<br />
Wait 10 days, then record<br />
30-day mean ∆SLP
30-day Mean ∆SLP Following WAF Pulses<br />
Init. WARM: 18/51 Init. COLD: 20/49<br />
(e.g. Polvani and Waugh [2004])
Annular Mode Response in SLP<br />
Init WARM<br />
Init COLD<br />
(e.g. Polvani and Waugh [2004])
Annular Mode Response in SLP<br />
Init WARM<br />
Init COLD<br />
p(NAM < -1.0):<br />
WARM = 0.18<br />
COLD = 0.10<br />
Subtle difference in NAM probabilities, but focused in areas of low seasonal<br />
predictability (e.g. East Coast US, Northern Europe).
Conclusions<br />
1. Can snow really act as a precursor to strat-trop interaction?<br />
- Response is highly variable but snow forcing does induce WAF pulses,<br />
causing strat. and trop. warming responses<br />
2. Are there conditions where interaction is more likely to occur?<br />
- Initial condition in polar strat. provides useful predictor of trop. response to<br />
snow forcing (better than trop. predictor)<br />
- Similar freq of interaction in “Init WARM” and “Init COLD” cases<br />
- “Init WARM”: interaction events produce more systematic downward prop<br />
and negative NAM response<br />
3. Is this really about the snow?<br />
- No. Results should also apply to SST/sea ice forcing
Conclusions<br />
1. Can snow really act as a precursor to strat-trop interaction?<br />
- Response is highly variable but snow forcing does induce WAF pulses,<br />
causing strat. and trop. warming responses<br />
2. Are there conditions where interaction is more likely to occur?<br />
- Initial condition in polar strat. provides useful predictor of trop. response<br />
to snow forcing (better than trop. predictor)<br />
- Similar freq of interaction in “Init WARM” and “Init COLD” cases<br />
- “Init WARM”: interaction events produce more systematic downward prop<br />
and negative NAM response<br />
3. Is this really about the snow?<br />
- No. Results should also apply to SST/sea ice forcing<br />
4. Results suggest a dynamical mechanism<br />
5. Limitations: strat. resolution/drag scheme, persistence?<br />
6. Ongoing: do we observe the same behavior in a strat-resolving<br />
model? Initial findings: confusing…<br />
<strong>Fletcher</strong>, C. G. et al. (2007) Geophys. Res. Let., In Press.<br />
Preprint: http://www.atmosp.physics.utoronto.ca/people/cgf/
The end.
Northern Annular Mode in SLP<br />
AM2<br />
Reanalysis<br />
Source: NCEP/CPC
Zonal mean climatologies<br />
ERA-40<br />
AM2-trop.<br />
AM2-strat.
Change language