Atmospheric Sciences [A]

A11G
 MC:2008  Monday  0800h

ENSO and Global Change: Past, Present, and Future I


Presiding:  D Sun, University of Colorado/CIRES & NOAA/Earth System Research Laboratory; A Clement, Rosenstiel School of Marine and Atmospheric Science; J Yu, Department of Earth System Science, University of California, Irvine

A11G-01 INVITED

El Nino/Southern Oscillation: Mid-Holocene, Present and Future

* Battisti, D S battisti@washington.edu, University of Washington, Dept. of Atmospheric Sciences 351640, Seattle, WA 98195-1640, United States
Roberts, W doctor.binkle@gmail.com, Maplecroft, The Towers St Stephen's Road, Bath, BA1 5JZ, United Kingdom
Vimont, D J dvimont@wisc.edu, University of Wisconsin-Madison, Department of Atmospheric and Oceanic Sciences 1225 W. Dayton Street, Madison, WI 53706, United States

Proxy data of sea surface temperature (SST) and precipitation in the tropical Pacific suggests the variance in ENSO was greatly reduced in the mid-Holocene compared to the modern climate. Recent studies report that full-physics climate models (e.g., those used in the recent IPCC assessment) with realistic ENSOs in their modern day climate simulations also show reduced ENSO variance when the models are forced by mid- Holocene insolation and greenhouse gas concentration. In most of these studies the reduction in variance is thought to be due to changes to the climatological mean state and - although the climate models simulate similar mean state changes - authors have offered different hypotheses for which mean state changes are important for the changes in ENSO. None of the hypothesis put forward have been tested, however, presumably because of the difficultly of designing a set of experiments with the full physics models that would illuminate which of the mean state changes were important, and the extraordinary computational resources that would be required to perform the experiments. Here we introduce a new tool for evaluating the physics responsible for ENSO - in nature and in full- physics climate models (such as those used in IPCC). The tool is a linear version of the Zebiak/Cane model with updated parameters estimated from observations and theory and with the climatological mean fields determined from observations or from the full-physics climate models. We demonstrate the tool can be used to determine the spatial and temporal characteristics of ENSO in nature. We also demonstrate how to use the tool to unambiguously determine how and why ENSO changes in the full-physics climate models when the mean state changes due to external forcing, such as Milankovitch or anthropogenic climate change. Applying the tool to the output of the NCAR CCSM, for example, shows that ENSO variance is reduced in the mid- Holocene in this model because the tropical Pacific SST is reduced, weakening the Bjerknes Feedback (ocean mean state changes destabilize ENSO, but not enough to compensate for the stabilization due to reduced SST). In contrast, applying the tool to the identical experiments using the HADCM3 shows that ENSO is reduced in that model because of a weaker thermocline that stabilizes the ENSO mode. Finally, we show how the tool can be used to quantitatively estimate the influence of mean state biases in the present generation full-physics climate models on the ENSO biases simulated by these models. We also outline a method for side-stepping the mean state biases to obtain better projections of how anthropogenic climate change will impact the spatio-temporal structure of ENSO and its teleconnections.

A11G-02 INVITED

A Paleoclimatic Context for Recent ENSO Activity

* Cole, J jecole@email.arizona.edu, Department of Atmospheric Sciences, University of Arizona, Tucson, AZ 85721, United States
* Cole, J jecole@email.arizona.edu, Department of Geosciences, University of Arizona 1040 E. 4th St, Tucson, AZ 85721, United States
Tudhope, S sandy.tudhope@ed.ac.uk, School of GeoSciences Grant Institute, Edinburgh University West Mains Road, Edinburgh, EH9 #JW, United Kingdom
Ault, T tault@email.arizona.edu, Department of Geosciences, University of Arizona 1040 E. 4th St, Tucson, AZ 85721, United States
Barnett, H hbarnett@email.arizona.edu, Department of Geosciences, University of Arizona 1040 E. 4th St, Tucson, AZ 85721, United States
Thompson, D thompsod@email.arizona.edu, Department of Geosciences, University of Arizona 1040 E. 4th St, Tucson, AZ 85721, United States

Is recent activity in the ENSO system unusual? Does it reflect the influence of anthropogenic climate forcing? Or does it fall within the bounds of natural variability? To address these questions, paleoclimate data are becoming increasingly critical. We can use paleoclimate records to define the range of natural variability and elucidate the mechanisms by which the system responds to changing radiative forcing. An expanding database of annual paleoclimate records allows us to characterize decadal variability during recent centuries, and older records point to intervals where ENSO variance may have been at least as great as today's. Paleoclimate records suggest that modern ENSO variability is not without precedent. But paleodata have their own associated uncertainties that must be considered. A dataset of 6 replicate coral isotopic records allows a close look at 20th century variability that highlights both the strengths and pitfalls of such records. Between-core differences in mean and variance suggest that replication is critical. We will expand this perspective to look at Indo-Pacific variability over the past few centuries to explore the influence of decadal and longer-term variability. A coherent signal can be identified that is strongly coupled to ENSO. Finally we highlight examples in the past where ENSO variance appears to have been larger than present. Changes in radiative forcing are typically not associated with either decadal variability or periods of enhanced activity, suggesting that internal variability in ENSO is large enough to make detection of an anthropogenic signal difficult from SST reconstructions alone. Nonetheless, if we can use coral records to identify specific long- term responses that are expected from modeling experiments, we may have greater confidence in attribution.

A11G-03 INVITED

Evolution of the 2006-2008 ENSO cycle

* McPhaden, M J michael.j.mcphaden@noaa.gov, NOAA/PMEL, 7600 Sand Point Way NE, Seattle, WA 98116, United States

This presentation describes the development of the El Nino/Southern Oscillation (ENSO) cycle of warm (El Niño) and cold (La Niña) events during 2006-08. Emphasis is on the interplay between large scale, low frequency deterministic dynamics and episodic intraseasonal wind forcing in affecting the observed variability. Efforts to forecast the El Nino and La Nina are reviewed, with discussion of factors affecting their predictability. Perspectives on possible influences of global warming on the ENSO cycle, which exhibited unusual behavior in the first decade of the 21st century, will be also be presented.

A11G-04

Where was ENSO strongest?

* Cane, M A mcane@ldeo.columbia.edu, Lamont-Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, NY 10964, United States
Chen, D dchen@ldeo.columbia.edu, Lamont-Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, NY 10964, United States
Kaplan, A alexeyk@ldeo.columbia.edu, Lamont-Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, NY 10964, United States

Mark A. Cane, Dake Chen, Alexey Kaplan The description of this session begins: "Historical SST records suggest that for the past three decades, ENSO has been anomalously strong" and goes on to ask why. In this talk we dispute this interpretation of the historical record from within the historical record. In particular, we suggest that the most "anomalously strong" period in the historical ENSO record is the late nineteenth century. This claim requires a discussion of how we measure "ENSO strength". We also speculate on possible reasons for the strength of ENSO in this earlier period. Finally, we consult the models, and in reiteration of the collective conclusion of all speakers at this session, find that the riddles the models provide are inelegant and disobliging, lacking the cryptic wisdom of the classical oracles.

A11G-05

Understanding El Niño in Ocean-Atmosphere General Circulation Models: progress and challenges

* Guilyardi, E Eric.Guilyardi@locean-ipsl.upmc.fr, Walker Institute, University of Reading, Department of Meteorology PO box 243, Reading, RG6 6BB, United Kingdom
* Guilyardi, E Eric.Guilyardi@locean-ipsl.upmc.fr, LOCEAN/IPSL (CNRS), UPMC, case 100 4, place Jusieu, Paris, 75252 Cede, France
Wittenberg, A Andrew.Wittenberg@noaa.gov, GFDL, Forrestal Campus, US Route 1, Princeton, NJ 08542, United States
Fedorov, A alexey.fedorov@yale.edu, Yale University, Department of Geology and Geophysics PO Box 208109, New Haven, CT 06520, United States
Collins, M matthew.collins@metoffice.gov.uk, Met Office Hadley Centre, Fitzroy Rd, Exeter, EX1 3PB, United Kingdom
Wang, C Chunzai.Wang@noaa.gov, NOAA/AOML, Physical Oceanography Division NOAA Atlantic Oceanographic and Meteorological Laboratory 4301 Rickenbacker Causeway, Miami, FL 33149, United States
Capotondi, A Antonietta.Capotondi@noaa.gov, University of Colorado & NOAA, CIRES,Campus Box 216, Boulder, CO 80309-0216, United States
van Oldenborgh, G oldenborgh@knmi.nl, KNMI, Wilhelminalaan 10, de Bilt, 3732, Netherlands
Stockdale, T Tim.Stockdale@ecmwf.int, ECMWF, Shinfield Park, Reading, RG2 9AX, United Kingdom

Determining how El Niño and its impacts may change over the next ten to hundred years remains a difficult scientific challenge. Ocean-atmosphere Coupled General Circulation Models (CGCMs) are routinely used both to analyze El Niño mechanisms and teleconnections and to predict its evolution on a broad range of timescales, from seasonal to centennial. The ability to simulate El Niño as an emergent property of these models has largely improved over the last few years. Nevertheless, the diversity of model simulations of present-day El Niño indicate current limitations in our ability to model this climate phenomenon and anticipate changes in its characteristics. A review of the several factors that contribute to this diversity, as well as potential means to improve the simulation of El Niño, is presented.

http://ams.allenpress.com/perlserv/?request=get- abstract&doi=10.1175%2F2008BAMS2387.1

A11G-06

Understanding ENSO modulation in the GFDL CM2.1 coupled GCM

* Wittenberg, A T andrew.wittenberg@noaa.gov, NOAA/GFDL, Forrestal Campus, US Route 1, Princeton, NJ 08542, United States

The GFDL CM2.1 global coupled GCM exhibits strong interdecadal and intercentennial modulation of its ENSO behavior. To the extent that such modulation is realistic, it puts large error bars on ENSO metrics diagnosed from centennial and shorter records -- with important implications for model assessment and intercomparison, climate projections, and historical and paleo records. We shall discuss the causes and effects of this ENSO modulation, including its links to the background climatology, inter-event memory, seasonal phase locking, and decadal predictability.

http://www.gfdl.noaa.gov/~atw/research/conf/agu_fall_2008

A11G-07

A Simple Theory for ENSO Amplitudes In IPCC AR4 models

* Jin, F jff@hawaii.edu, Department of Meteorology University of Hawaii at Manoa, 2525 Correa Rd, Honolulu, HI 96822, United States
Kim, S seonk@hawaii.edu, Department of Meteorology University of Hawaii at Manoa, 2525 Correa Rd, Honolulu, HI 96822, United States

Because of the great global impacts of ENSO, how does ENSO activity respond to the global warming is one of the central questions of the climate change. In an attempt to address this question, we analyzed the IPCC AR4 simulations. The so-called controlled simulations of the current climate models exhibit great variations in simulating the ENSO activity (measured by NINO3 variance for instance) with huge range of a factor 2 too strong to a factor ½ too week, when compared with the observed ENSO variance. The great variations of ENSO amplitude in these different models are closely related to the differences in the coupled instability of ENSO, the later is measured by the so-called Bjerknes (BJ) instability index (Jin et al 2007). There is a clear and strongly nonlinear relation between the BJ index and ENSO variance. A simple theory is proposed to demonstrate that this nonlinear relation is an expected behavior near criticality where ENSO is either marginally damped or unstable. The diverse results from the IPCC models fit beautifully with the theoretical formula relating the ENSO amplitude to growth rate. The potential implications of this theory of ENSO amplitude to the estimation of future ENSO strength will be discussed.

A11G-08

Analytical Theory for the Quasi-Steady and Low-Frequency Equatorial Ocean Response to Wind Forcing

* Clarke, A J aclarke@ocean.fsu.edu, Department of Oceanography The Florida State University, 117 N. Woodward Avenue P.O. Box 3064320, Tallahassee, FL 32306-4320, United States

Analytical theory is used to examine the linear response of a stratified ocean to large-scale, low-frequency wind forcing. The following results, applied mainly to the equatorial Pacific, were obtained. (i) Provided that the wind stress curl vanishes at large distance from the equator, a general Sverdrup solution is valid in the quasi-steady (zero frequency ) limit. The flow is horizontally non-divergent and there is no disturbance east of the wind forcing region, i.e., there is no El Nino – like response. The meridionally averaged zonal flow towards the western boundary layer is zero so that there is no net mass flow into the boundary layer and the large-scale boundary condition is therefore satisfied. (ii) Consistent with observations and other previous work, for finite but small frequency, there are two modes of motion. One is a 'tilt' mode in which the equatorial sea level and thermocline are tilted by the in-phase zonal wind stress and the other is an equatorial warm water volume (WWV) mode in which the discharge of warm water (negative WWV anomaly) lags the wind stress forcing by a quarter of a period. (iii) The WWV mode does not exist when there is no wind stress curl. (iv) When the frequency omega approaches zero, the amplitude of the WWV mode approaches zero like the square root of omega, consistent with the absence of the unsteady WWV mode in the zero frequency limit. However, even at decadal frequencies observations show that there is an unsteady WWV mode, suggesting that dissipation of the large-scale flow may play an important dynamical role at low frequencies.