Get Complete Project Material File(s) Now! »
A brief history of El Niño phenomenon
At the end of 19th century, Carranza (1891) described for the first time the characteristics of a strong climatic event that took place along the Peruvian coast in 1891. He depicted an abnormal southward current apparently originating from the Guayaquil Gulf, and that was opposite to the southerly current. This abnormal current disrupted the oceanic and atmospheric climatological conditions, resulting in warmer than usual ocean conditions and heavy precipitations along the Peruvian coast and, as a consequence, flooding in the Northern edge of Peru in April and May of 1891. This event was so strong that precipitation anomalies reached cities such as Lima [77ºW, 12ºS] and Ica [75.7ºW, 14ºS] (Rodriguez, 2001) – cities that usually have no precipitation all year around. The southward extension of this unusual event can be appreciated from Figure 1.1bis that displays the Sea Surface Temperature (SST) interannual anomaly in March (Figure 1.1bis-a) and April (Figure 1.1bis-b) of 1891 over the South Eastern Pacific region and highlights that the SST anomaly increased by ~2°C from March to April on average along the coast of Peru. This event was mostly described based on its coastal expression and it was not clear at that time, how such event was connected to the equatorial variability. The first document that ever mentions “El Niño” along the coast of Peru was published two years later in the bulletin of the Sociedad Geográfica de Lima, written by Carrillo (1892), in which he reported the various evidences of a “strange” southward current observed in the ocean off Peru just after Christmas: La corriente del Niño (El Niño current).
It was the El Niño that occurred in 1925, the largest one registered at the time and documented by an ornithologist (Murphy, 1926) that first enticed the concern of the international scientific community on the El Niño phenomenon. Whereas the meteorologist H. P. Berlage was the first to suggest that El Niño could be linked to the large scale circulation in the tropical Pacific (Cushman, 2004), the first theory was proposed by Sir Bjerknes, following earlier works by Sir Walker who was sent to India to forecast the monsoon in order to improve the cotton production- an important raw material for Europe at that time. Sir Walker observed that the fluctuations of pressure at Tahiti and Darwin were opposite and they were varying a lot from one year to the other.
He used these data (the difference between both) to forecast the Indian monsoon that is now known to be associated with the El Niño Southern Oscillation (ENSO). Sir Bjerknes was the first to suggest that, the unusual phenomenon which was occurring along the coast of Peru (the episodic warming of sea water at Christmas time) could be related to the large scale fluctuations identified by Sir Walker. He spurred the concept that the zonal contrast in SST across the equatorial Pacific is tightly linked to the so-called Walker circulation driven by the difference in sea level pressure between Tahiti and Darwin.
El Niño theories and forecasts
What was not explained by this theory is the oscillatory nature of ENSO: warm El Niño events can be followed by a cold phase (i.e. La Niña) and comes back after some time, the later defining its “periodicity”. The 80s and 90s were the decades of intense development in the theory of ENSO, which was allowed in part thanks to the development of the observing system in the tropical Pacific, both in situ (TOGA-TAO program launched in 1985) and from space (the first altimeter (GEOSAT) was launched in 1985). In particular, altimetry allowed for the quasi-synoptic observation of planetary equatorial waves, which permitted testing theories based on the reflections of equatorial waves onto the Pacific meridional boundaries. In these theories (for instance, the so-called delayed action oscillator, see #1.3.2 for more details), the delayed effect of the reflected wave is the process by which an El Niño event is suppressed, providing a concept for its oscillating nature.
Even if, at the time, the observation of long equatorial waves permitted to understand its role on the oscillatory process of El Niño, the basin scale adjustment associated with the free propagating waves could not provide a consistent view of the timescales of ENSO evolution and periodicity. The recharge-discharge theory emerged in the late 90s in part in order to resolve such issue. The recharge-discharge oscillator considers a fast adjustment that takes place at the scale of the tropical Pacific basin. Along the equator, the trade winds induced a zonal transport but also a meridional transport due to the effect of the Coriolis force (the so-called Sverdrup transport). The latter replenishes the equatorial band in warm waters when the thermocline has risen (recharge process) or evacuates the warm waters from the equatorial band (discharge), which provides the slow negative feedback for the system to oscillate (see details in Section 1.3).
The theory is based on the observation that there is a delayed adjustment between the SST and the heat content over the equatorial Pacific (the so-called warm water volume). This theory has provided the most general framework for studying ENSO for the last two decades. At the time when I started this thesis, the recharge-discharge oscillator was still the main paradigm for interpreting the interannual variability in the equatorial Pacific. The recharge-discharge oscillator provides a solid theoretical basement for studying various aspects of ENSO, in particular its changes in properties (frequency, amplitude, predictability) over long-timescales. For instance, from the 80s, ENSO experienced an increase in its amplitude, which has been attributed to the warmer mean state after the climate shift of the late 70s. An increase in ENSO activity may be related to a decrease in the stability of the oscillator, which can be diagnosed from observations or Reanalysis products based on this formalism (An and Jin, 2001). Hence, along with improvements in numerical modeling and data assimilation, at the end of the XXe, we came to the idea that El Niño could be predicted with at least 2 or 3 seasons in advance.
A new type of El Niño
However, the observations since the beginning of the 21st century have wiped out this expectation. In particular, it has been observed that since the 2000s, the predictive value of the recharge-discharge oscillator has been limited since the lag between the NINO3 SST [150°W-90°W; 5°S-5°N] and the WWV index has decreased significantly (McPhaden, 2012). First, this has been suggested to be due to changes in the location of the center of ENSO wind stress anomalies. In particular, it can be shown that the further to the west the wind stress forcing, the smaller the lag between NINO3 SST and the WWV mode (Clarke, 2010; Fedorov, 2010). Also, dissipation processes could play a significant part in explaining the lag between the NINO3 SST and WWV mode (Thual et al., 2013). However, it was not clear if such change in ENSO characteristics observed after 2000 could be fully interpreted in the light of the recharge-discharge theory. In fact, a new type of El Niño, known as the Central Pacific El Niño (CP El Niño, Kug et al., 2009) or Modoki El Niño (Ashock et al., 2007), characterized by weaker SST anomalies located in the central Pacific was described, which, since then, has put the community in front of a new challenge (Capotondi et al., 2015).
This thesis is a contribution to the efforts of the ENSO community for understanding such change. It focuses on the investigation of the equatorial variability in the Pacific over the last century, when only El Niño of one type has taken place. Those new El Niño events have a center of action in the central Pacific conversely to the Eastern Pacific or Cold Tongue El Niño that has the most severe impacts for Peru.
As a summary of where the thesis stands within the history of El Niño research, the Figure 1.2bis provides a schematic of the chronology of ENSO research over the Twentieth century.
Motivations and objectives of the thesis
It is now well known that ENSO involves a variety of variability timescales: from intraseasonal to interannual. The intraseasonal atmospheric variability in the tropical Pacific has a particular interest because it has been observed prior to the development of recent El Niño events and possibly involved in the rectification of the ENSO (Kessler and Kleeman, 2000). This intraseasonal atmospheric variability is in particular composed by the Madden and Julian oscillation (Madden and Julian, 1994) and the Westerly Wind Bursts (stochastic in nature) that are associated to periods of variability ranging from 10 to 60 days. Surprisingly, the IntraSeasonal Kelvin wave (ISKw) has a much wider spectrum of variability ranging from (50 days)-1 to the semi-annual frequency with a particular energy peak in the (90 -120 days)-1 frequency band (Dewitte et al., 2008a). A recent study (Gushchina and Dewitte, 2012) also indicates that the relationship between ENSO and the intraseaonal atmospheric variability could be distinct among the two types of El Niño.
It appears important to document the characteristics of the IEKws and its relationship with the CP El Niño development and decay, considering that the ISKw is influential on the spread of ENSO forecasts (Wang et al., 2011) and that prediction systems exhibit a distinct forecast skill depending on the El Niño type (Hu et al., 2012; Xue et al., 2013).
Table of contents :
Chapter 1: Overview
1.1 Introduction (Version française)
1.1.1 Bref historique du phénomène El Niño
1.1.2 Emergence des théories et prévisions du phénomène El Niño
1.1.3 Un nouveau type d’événement El Niño
1.1.4 Motivations et objectifs de cette thèse
1.2 Introduction (English version)
1.2.1 A brief history of El Niño phenomenon
1.2.2 El Niño theories and forecasts
1.2.3 A new type of El Niño
1.2.4 Motivations and objectives of the thesis
1.3 ENSO theories
1.3.1 Bjerknes theory
1.3.2 The delayed oscillator
1.3.3 Discharge and recharge oscillator theory
1.4 The Central Pacific El Niño (CP El Niño)
1.5 ENSO and the Kelvin wave
1.6 Processes impacting the Kelvin wave in the equatorial Pacific
Chapter 2: The 2002/03 El Niño: Equatorial waves sequence and their impact on Sea Surface Temperature
2.1 Overview
2.2 Article published in Journal Geophysical Research- Oceans
2.3 Thermodynamics associated to the El Niño phenomenon over 1990 – 2011
2.3.1 Introduction
2.3.2 Data and methodologies
2.2.3 Results
Chapter 3: The Central Pacific El Niño Intraseasonal Kelvin wave
3.1 Overview
3.2 Article published in Journal of Geophysical Research – Oceans
Chapter 4: On the change in thermocline seasonal variability along the equatorial Pacific from before and after 2000
4.1 Overview
4.2 Draft Article to be submitted to Journal of Geophysical Research – Oceans
Chapter 5: Conclusions and Perspectives
