CE01 - Milieux et biodiversité : Terre fluide et solide

Non linear Atmospheric Response to Sea surface temperature for a physically consistent ENSO (El Niño Southern Oscillation) paradigm – ARiSE

non linear Atmospheric Response to Sea surface temperature for a physically consistent ENSO (El Niño Southern Oscillation) paradigm

ENSO is the dominant climate variability phenomenon on earth. It develops as the result of air-sea interactions in the tropical Pacific, and has major societal impacts globally. Understanding and modelling ENSO correctly is thus a pre-requisite for seasonal climate prediction. ENSO’s warm (El Niño) and cold (La Niña) phases are not symmetrical, with much larger amplitude (extreme) El Niño than La Niña events.

Toward a better understanding and modelling of ENSO asymmetries

Over the last decade, many studies have identified nonlinear oceanic and atmospheric processes that appear to play a key role in ENSO asymmetries. There is however currently no consensus on the relative importance of those various oceanic and atmospheric processes in determining the overall ENSO asymmetry. The first objective of the ARiSE project is thus to quantify the respective roles of those oceanic and atmospheric nonlinear processes in setting the asymmetrical ENSO properties, in terms of amplitude, spatial pattern, duration and predictability. <br />ENSO conceptual models provide a powerful tool to understand ENSO properties in observations or more complex coupled general circulation models. Those idealized models are however often linear, and studies attempting to incorporate nonlinearities in those models are still relatively scarce. The second objective of that project is to build simple models of the nonlinear atmospheric response during ENSO, and nonlinear ENSO conceptual models. <br />Finally, climate models used for seasonal predictions or in the IPCC climate change projections currently strongly underestimate ENSO asymmetrical features. This limits their ability to provide reliable seasonal forecasts, or to predict the effect of climate change on ENSO. The third objective of ARiSE is thus to understand which misrepresented atmospheric and/or oceanic physical processes are responsible for this improper ENSO representation. This should provide useful clues as how to improve ENSO in climate models.

In work package 1 (WP1), we applied a humidity budget to atmospheric re-analyses (i.e. products that are strongly constrained by observations) to develop an idealized equation of the atmospheric convection (i.e. precipitation) non-linear response to ocean temperature anomalies. This equation was then implemented in a model of atmospheric dynamics (DREAM), that can either be run in linear or nonlinear mode. This will allow to separately evaluate the influences of nonlinearities associated with atmospheric dynamics / convection on the surface wind response to ocean temperature, a key element for ENSO development.
WP2 aims at developing idealized models of the atmospheric response during ENSO events and of the ENSO phenomenon itself. The most classical ENSO simple model (the recharged oscillator) will be modified to represent the main oceanic and atmospheric nonlinearities, and understand their effect on ENSO asymmetry.
WP3 focuses on the most complex climate models: Coupled General Circulation Models (CGCM). One of our objectives is to describe the physical mechanisms (and more specifically the nonlinearities) that contribute to the development of extreme El Niño events in the CNRM CGCM. Our second objective will be to quantify ENSO asymmetrical features and important nonlinear processes in CGCM simulations from the CMIP6 database and in various observational databases. This should allow to determine which nonlinear physical processes account most for the underestimated ENSO asymmetry in CGCMs.

At the end of year 1 (*), the main results obtained so far follow. (* The project really started about 6 months after the initial targeted date due to difficulties to recruit good candidates and was then slowed down by the sanitary crisis).
We developed a very simple methodology to account for the effect of ENSO-induced changes in atmospheric vertical stability on the influence of the Sea Surface Temperature (SST) on tropical rainfall (Izumo et al. 2020). We also developed a simplified atmospheric moisture budget, that allows to predict very accurately the rainfall response to tropical SST anomalies. This equation was then applied to ENSO, and demonstrated that the nonlinear transport of moisture anomalies by circulation anomalies plays a key role in the amplitude asymmetry between El Niño and La Niña, and in the genesis of extreme El Niño rainfall anomalies (Srinivas et al. 2021 submitted). We furthermore demonstrated that nonlinearities in the atmospheric dynamical response to rainfall anomalies tended to partially cancel the above convective nonlinearity (M. Béniche, master report 2020).
We have also highlighted an ENSO predictability asymmetry, depending on the oceanic initial conditions. Anomalously low western pacific heat content evolve more predictably toward a La Niña than anomalously high western Pacific heat contents, which can either evolve toward a neutral state or El Niño. The physical mechanism that explains this predictability asymmetry is linked to the nonlinear relation between intraseasonal “Westerly Wind Bursts” and ENSO, with more numerous and more random WWBs during El Niño events onset (Planton et al. 2021 in revision). Finally, we have also assessed ENSO’s realism (in particular its asymmetry) in the CMIP6 CGCM simulations database (Planton et al. 2021), confirming that ENSO asymmetrical features are severely underestimated in this generation of models, as they were in previous ones.

At the end of the first year of the ARiSE project, our current perspectives are:
1) Coupling our simplified moisture conservation equation with the DREAM atmospheric dynamics model provides a nonlinear transfer function between the ENSO SST anomalies and the surface wind response, that will allow to study the detailed role of the various atmospheric nonlinearities (convection, dynamics, wind to wind stress bulk relation) on the overall Bjerknes feedback nonlinearity.
2) Accounting for observed nonlinearities in the various terms of the recharge oscillator ENSO conceptual model will allow to characterize the potential impact of those nonlinearities on the representation of ENSO asymmetries.
3) The CNRM model strongly underestimates ENSO asymmetries. We hence decided to build an experimental strategy for estimating the role of various oceanic and atmospheric nonlinearities on ENSO asymmetry based on forced simulations from its oceanic and atmospheric components. In parallel, we will investigate statistical links between the representation of these various nonlinear physical processes and the ENSO asymmetrical features in the CMIP6 database.

Selected publications and book chapters

1. Planton, Y. Y., Guilyardi, E., Wittenberg, A. T., Lee, J., Gleckler, P. J., Bayr, T., S. McGregor, M. J. McPhaden , S. Power, R. Roehrig, J. Vialard and A. Voldoire (2021). Evaluating climate models with the CLIVAR 2020 ENSO metrics package. Bulletin of the American Meteorological Society, 102(2), E193-E217.
2. Izumo T., J. Vialard, M. Lengaigne, I. Suresh, 2020: Relevance of relative sea surface temperature for tropical rainfall interannual variability, Geophysical Research Letters, doi: 10.1029/2019GL086182
5. Planton Y., J. Vialard, E. Guilyardi, M. Lengaigne, 2021: Asymmetric influence of oceanic heat content on ENSO prediction, Climate Dynamics, in revision.
6. G. Srinivas, J. Vialard, M. Lengaigne, T. Izumo & E. Guilyardi, 2021: Mechanisms of the ENSO asymmetrical precipitation response, J. Climate, submitted
7. Kug J.-S., J. Vialard, J.-Y. Yu, Y.-G. Ham, M. Lengaigne, 2020: Remote forcing: Influence of climate variability outside the tropical Pacific, El Niño Southern Oscillation in a Changing Climate, 247-265, doi:10.1002/9781119548164.ch11.
8. Guilyardi E., A. Wittenberg, M. Lengaigne, A. Capotondi, S. Thual, 2020: ENSO modelling: history, progress and challenges, El Niño Southern Oscillation in a Changing Climate, 199-226, doi:10.1002/9781119548164.ch9.
9. Cai W., G. Wang, L. Wu, M. Collins, A. Timmermann, S. Power, M. Lengaigne, 2020: ENSO to greenhouse forcing, El Niño Southern Oscillation in a Changing Climate, 289-307, doi:10.1002/9781119548164.ch13.
10. Sprintall, J., S. Cravatte, B. Dewitte, Y. Du and A. Sen Gupta, 2020: Oceanic Teleconnections, chapter 15 in «El Nino in a Changing Climate« AGU Book, ISBN: 978-1-119-54816-4, 528 pages.

The El Niño Southern Oscillation (ENSO) is the leading mode of interannual climate variability on earth. It consists of irregular, alternating phases of warm (El Niño) or cold (La Niña) Sea Surface Temperature (SST) anomalies in the tropical Pacific Ocean, but has global climate impacts through atmospheric teleconnections. It is predictable 2-3 seasons in advance, so ENSO is also a major source of global seasonal climate predictability. State-of-the-art ENSO forecasts however have not performed well for “extreme” El Niño events such as those in 1982-83, 1997-98 or 2015-2016. With climate projections suggesting that such extreme events may become more frequent, it is vital to improve our understanding of the processes involved in these ENSO events.

ENSO grows through the Bjerknes feedback, a positive feedback loop between the ocean and the atmosphere. In this feedback loop, an SST anomaly induces deep atmospheric convection. The resulting changes in surface wind drive an ocean response that strengthens the initial SST anomaly, allowing ENSO events to grow. While the oceanic component of this feedback loop is well understood, there are far fewer studies of its atmospheric component. Yet, the atmospheric component of the Bjerknes feedback exhibits strong non-linearities which have been recently suggested to play a key role in extreme El Niño events. Deep convection indeed only occurs for SSTs above ~27.5°C, implying that the atmosphere will be more responsive to SST fluctuations in the warm western Pacific than in the colder eastern Pacific. The conceptual models used to understand ENSO do not account for this non-linearity, or for other potentially-important sources of atmospheric non-linearities such as heat flux feedbacks. Additionally, the capacity of Coupled General Circulation models (CGCMs) used for ENSO research and forecasting to reproduce the non-linear atmospheric response to ENSO SST anomalies has not been systematically evaluated so far. Previous work suggests that this non-linearity likely plays a key role in linking the models’ systematic biases to their documented underestimation of the Bjerknes feedback.

Developing tools to quantify, understand and model the non-linear wind response to ENSO SST anomalies is thus a vital prerequisite to a) understanding extreme El Niño mechanisms and b) being able to diagnose the source of ENSO biases in CGCMs. ARiSE proposes to use observational analyses and a hierarchy of atmospheric and coupled models (from conceptual to general circulation models) to better describe the non-linear atmospheric response to ENSO and its impact on ENSO properties, and in particular extreme El Niño events.

In Work Package (WP) 1, we will use observations and two types of atmospheric GCMs to produce a seasonally-dependent transfer function between SST and wind stress variations, and explore its non-linearity. In WP2, we will develop the simplest possible atmospheric model that encapsulates essential dynamical and thermodynamical non-linearities for ENSO, and investigate their impact on ENSO, in an intermediate and a conceptual model. In WP3, we will use the results from the previous WPs to improve our understanding of ENSO in CGCMS: a) we will diagnose the specific mechanisms of extreme El Niños in a CGCM; b) we will evaluate links between atmospheric non-linearities and ENSO biases in the CMIP database. This will provide tools for improving ENSO representation in state-of-the-art CGCMs used in forecast mode.

The current project gathers a unique blend of oceanographers and atmospheric scientists with good expertise in ENSO, including developers of some models used in the project. This project also benefits from the support of several renowned international collaborators on their own funds. ARiSE is a unique opportunity to make a step change in our understanding of tropical air-sea interactions, the role of the atmosphere in extreme El Niños with potentially large societal impacts.

Project coordination

Jérôme Vialard (Laboratoire d'océanographie et du climat : expérimentations et approches numériques)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

CECI Climat, Environnement, Couplages et Incertitudes
CNRS - CNRM Centre National de la Recherche Scientifique - Centre national de recherches météorologiques
LOCEAN Laboratoire d'océanographie et du climat : expérimentations et approches numériques
CNRS-LEGOS Centre National de la Recherche Scientifique - Laboratoire d'études en géophysique et océanographie spatiales

Help of the ANR 457,947 euros
Beginning and duration of the scientific project: February 2019 - 48 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter