Volcanic Climate Impacts and ENSO Interaction

1
Volcanic Climate Impacts and ENSO Interaction
Georgiy Stenchikov Department of Environmental
Sciences, Rutgers University, New Brunswick,
NJ Thomas Delworth and Andrew Wittenberg NOAA
Geophysical Fluid Dynamics Laboratory, Princeton,
NJ Motivations Radiative forcing from volcanic
aerosols causes strong forced climate
variations that interfere with the major modes of
climate variability. Volcanic natural
experiments provide an opportunity to learn more
about important aspects of the climate system
behavior including the long discussed statistical
connection between explosive volcanism and El
Nino. However, nonlinearity of this interaction
makes it difficult to make sense from
observations and model simulations. Therefore
we conducted ensembles of specifically designed
numerical experiments with comprehensive coupled
ocean-atmosphere GCMs to address the following
questions How does ENSO modulate climate
response to volcanic forcing? What is volcanic
impact on ENSO cycle? How does Volcano/ENSO
interaction depend on the strength of ENSO events
and magnitude of volcanic forcing? We mostly
focused not on the ENSO statistics but on the
detailed structure of Volcano/ENSO interaction.
We hope this analysis will help to better
understand the observed statistical connection
between explosive volcanism and ENSO.
2
We have used GFDL Coupled Climate Model
CM2.1 with 2ºx2.5ºL24 Atmosphere and 1ºx1ºL50
Ocean and Volcanic Aerosols implemented by
Stenchikov et al. (2006)
Radiative Impact of Explosive Volcanism in
1980-1995 Dominated other Forcings Producing
Significant Transient Cooling Pinatubo
- 0.75 W/m2 average for 10 years El Chichon
- 0.50 W/m2 average for 10 years
3
Volcanic Impacts could be seen in the all
components of Climate System Time-depth plot of
global mean temperature anomaly (K) caused by
volcanic and solar forcings calculated as the
ensemble mean of the GFDL IPCC AR4 experiments
minus the control integration (Delworth,
Ramaswamy, and Stenchikov, 2005)
1900
1940
1860
1980
4
Observed (CRU, Phil Jones) and Simulated Surface
Air Temperature Anomalies (K) from the GFDL IPCC
AR4 Runs
5
Nino3.4 Index from observations and from ensemble
of GFDL CM2.1 IPCC AR4 Runs
Observed Simulated
PINATUBO
EL CHICHON
in GFDL CM2.1 IPCC AR4 runs
6
El Nino, La Nina, and Neutral Initial Conditions
for 20-year Ensemble Runs
7
ENSO EFFECT ON RESPONSE Global Surface Air
Temperature Response to Pinatubo Forcing in the
Runs with the El Nino and La Nina Initial
Conditions
8
Observed Surface Air Temperature Response
The top panel shows the three observed ENSO
indices used in the iterative method for
separating volcano and ENSO signals. Removal of
the estimated ENSO signals yields the residual
lower tropospheric temperatures shown at the
bottom panel Santer et al., 2001
9
PINATUBO
10
Volcanic Impact on ENSO of Different Amplitude
PINATUBO
11
Mechanism of ENSO response to volcanic forcing
12
Effect of Strong Volcanic Forcing 3xPinatubo and
5xPinatubo
13

• Conclusions
• We used the GFDL coupled ocean-atmosphere CM2.1
  modeling system to conduct ensembles of runs with
  El Nino, La Nina, and Neutral initial conditions
  varying strength of ENSO events and magnitude of
  volcanic forcing. Our findings are as follows
• El Nino tends to delay maximum cooling caused by
  volcanic aerosols, while in the La Nina cases
  cooling develops earlier in time. Global
  temperature in the coupled system relaxes for
  about 7 years in agreement with observations.
• In the runs with El Nino initial conditions
  volcanic cooling decreases amplitude of El Nino
  but causes warming of equatorial SST in the year
  following an El Nino event. In the runs with the
  neutral initial conditions volcanic impact tends
  to produce El Nino-like response in the second
  year after volcanic eruption. However, this
  effect is fairly weak.
• The warming effect in the year after El Nino is
  robust. It gets stronger in the runs with weaker
  El Nino and with increase of volcanic forcing.
  The 1xPinatubo and 3xPinatubo cooling impact in
  the first year after eruption could only decrease
  El Nino amplitude, while 5xPinatubo forcing
  reduces El Nino almost completely.
• Volcanic cooling affects the Bjerknes Feedback.
  Associated SST warming is caused by reduction
  of the strength of the upwelling because of
  weakening of the trade winds according to the
  Ocean Dynamical Thermostat Mechanism of Clement
  et al. (1996).
• Decreasing of El Nino amplitude caused by
  volcanic cooling also could weaken the relaxing
  Kelvin wave that makes relaxation process less
  intensive and could cause a warming effect in
  the Eastern Equatorial Pacific in the year
  following El Nino.