Pacific Climate Change: Past, present and future

1
Pacific Climate Change Past, present and future

• Amy Clement
• Division of Meteorology and Physical Oceanography
• Rosenstiel School of Marine and Atmospheric
  Science
• University of Miami

CLIVAR Pacific Panel meeting Feb 15-17, Honolulu,
Hawaii
2
Constraining the equatorial zonal SST gradient

• (1) Any perturbation to the gradient will be
  amplified by coupled interactions (i.e. Bjerknes
  feedback, Djikstra and Neelin 1995)
• (2) Atmospheric convection is sensitive to this
  gradient? expect global impacts
• Talk outline
• Mechanisms
• Examples from past, present and future using
  observations (both paleo instrumental) and
  models

3
Mechanisms How would the gradient respond to an
idealized forcing?

• Increased E/W gradient in a warmer climate
• Cooling by upwelling the east (Clement et al.
  1996)
• Ocean ventilation from the windy subtropics
  (Seager and Murtugudde 1997, Liu 1999)

• Decreased E/W gradient in a warmer climate
• Latent heat flux damping in warm pool (Knutson
  and Manabe 1995)
• High cloud feedbacks (Meehl and Washington 1996)

4
Past Climate Changes
LGM
Greenland ice core temperatures from Alley (2000)
The Two-Mile Time Machine
20th century changes
Source http//www.giss.nasa.gov
Future scenarios
PCMDI
5
Last Glacial Maximum
Faunal reconstructions CLIMAP 1980- little change
in tropical SST (inconsistent with land data)
Hostetler and Mix (1999)- large cooling in the
EEP
Hostetler and Mix (1999)

• Geochemical studies
• Lea et al. (2000)- 3oC cooling in EEP slightly
  larger than western site (159oE)
• Koutavas et al. (2002)- 1.2oC cooling in EEP
  smaller cooling than western site

From Koutavas et al. 2002
6
Hadley center LGM Simulation (Hewitt et al.
2003, also Kitoh et al. 2001)
Rosenthal and Broccoli 2004
NCAR CSM LGM simulation (Shin et al. 2002, also
Liu et al. 2002)
7
Climate events of the last millenium
Mean dO18 (warm/wet vs. cool/dry) from Palmyra
corals (Cobb et al. 2003)
Results from Mann et al. (2004) ZC model with
solar and volcanic forcing
MWP
LIA
Relatively warm central Pacific Reduced zonal
gradient Wet conditions in N. America
Relatively cold central Pacific Increase zonal
gradient Medieval Drought in N. America
(Herweijer et al. 2006)
8
La Nina-like footprint for the Medieval Warm
Period From Herwiejer et al. (2006) submitted to
J. Climate
9
Kaplan
20th century SST trends
ERSST
HadiSST
10
Linear trend (K/50yr) 1900-1950
Linear trend (K/50yr) 1951-2000

• Data issues
• Different forcing (solar irradiance, e.g. Meehl
  2003)

• Weakening of Walker circulation in SLP and
  reanalyses (Zhang and Song 2006)

11
Future changes
SAT trend (K/ century) in CMIP2 1/year
experiments (Collins CMIP Modelling Groups
2005)
12
Collins et al. (cont.)
13

• Collins et al. (2005) conclusion
• The most likely scenario (p.59) in a
  model-skill-weighted histogram of CMIP models is
  for no trend towards either mean El Nino-like or
  La Nina-like conditions. However, there remains a
  small probability (p0.16) to a change to El
  Nino-like conditions of the order of one standard
  El Nino per century in the 1/year CO2 increase
  scenario.
• Results from IPCC AR4 simulations for ENSO
  behavior under increasing CO2 are inconclusive
  (Guilyardi 2005, Meehl et al. 2005, van
  Oldenborgh et al. 2005, Merryfield 2006)

14
Atmospheric circulation indices appear to be more
robust (Tanaka et al. 2005)
All models simulate a weaker Walker circulation
under SRES A1B scenario (GFDL too, B. Soden,
Personal communication)
15
Meridional gradient response also appears to be
robust (Liu et al. 2005)
16
Concluding points/Questions

• LGM results are inconclusive w.r.t. E/W gradient
  (both in models and observations)
• Climate events of the last millenium
• Little Ice Age- reduced E/W gradient
• Medieval Warm Period- increased E/W gradient
  with global impacts
• Related to solar forcing?
• 20th century
• E/W gradient changes are different for early and
  late 20th century- different forcing/response or
  data problems?
• Late 20th century reduced E/W gradient consistent
  with atmospheric circulation change
• 21st century
• Models do not predict a large (consistent)
  response in E/W gradient
• BUT, do predict a consistent response in
  atmospheric circulation change
• To what extent are ocean processes/coupled
  interactions in the Pacific important in the
  response to external climate forcing?

17
Mid-Holocene ENSO suppression
strong ENSO
GRADUAL or ABRUPT?
weak ENSO
18
Mid-Holocene ENSO suppression
19
Mid-Holocene ENSO suppression
20
(No Transcript)
21
(No Transcript)