Precursors of the El Ni\~no phenomenon: A climate network analysis

Letter

Precursors of the El Niño phenomenon: A climate network analysis

Rupali Sonone and Neelima Gupte

Phys. Rev. E 103, L040301 – Published 16 April 2021

Abstract

The identification of precursors of climatic phenomena has enormous practical importance. Recent work constructs a climate network based on surface air temperature data to analyze the El Niño phenomena. We utilize microtransitions which occur before the discontinuous percolation transition in the network as well as other network quantities to identify a set of reliable precursors of El Niño episodes. These precursors identify 10 out of 13 El Niño episodes occurring in the period of 1979–2018 with an average lead time of approximately 6.4 months. We also find indicators of tipping events in the data.

Received 18 October 2019
Accepted 23 March 2021

DOI:https://doi.org/10.1103/PhysRevE.103.L040301

Physics Subject Headings (PhySH)

Degree distributions Discontinuous phase transition Network phase transitions Order parameters Percolation theory

Statistical Physics & ThermodynamicsNonlinear DynamicsNetworks

Authors & Affiliations

Rupali Sonone^1,* and Neelima Gupte^2,†

¹Department of Physics, Indian Institute of Technology Madras, Chennai 600036, India
²Department of Physics, Indian Institute of Technology Madras, Chennai 600036, India and Complex Systems and Dynamics Group, Indian Institute of Technology Madras, Chennai 600036, India

^*vaidehisonone@gmail.com
^†gupte@physics.iitm.ac.in

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 103, Iss. 4 — April 2021

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
The largest cluster $s_{1}$ (blue dotted line) and susceptibility $χ$ (green line) is plotted as a function of the link strength $C$ for the network. Plots (a) December 2013, (b) December 1990, and (c) December 1993, which are years which precede a very strong El Niño episode, strong El Niño episode, and a moderate El Niño episode, respectively. Plots (d)–(f) are for years when the El Niño episode peaks: (d) very strong El Niño episode, December 2015; (e) strong El Niño episode, December 1992; (f) moderate El Niño episode, December 1994.
Reuse & Permissions
Figure 2
The maximum jump in the susceptibility $χ$ for each year ( $1979 - 2018$ ). El Niño years are shown in red with a gradation according to their strength. There have been 13 such El Niño episodes out of which 1982, 1997, and 2015 are very strong El Niño years; 1987 and 1991 are strong El Niño years; 1986, 1994, 2002, and 2009 are moderate El Niño years; and 2004, 2006, 2014, and 2018 are weak El Niño years. A false alarm (in the year 2000) and the missed predictions (in the years 2001, 2003, 2008, and 2017) are solely based on susceptibility. Note that after taking into account other precursors (as summarized in Table 1) we get one false alarm and one missed prediction (in the year 2003).
Reuse & Permissions
Figure 3
(a) The jump $Δ χ$ as a function of the link strength $C$ . Inset: The susceptibility $χ$ as a function of the link strength $C$ for the network 1 year before a very strong El Niño episode, December 2014. (b) The scaling relation for the relative $Δ χ$ positions $(C_{c} - C_{i + 1} / C_{c} - C_{i})$ as a function of the index $i$ .
Reuse & Permissions
Figure 4
The histogram of the teleconnections in kilometers for the network. (a) One year before a very strong El Niño episode, December 1996. Inset: One year before strong La Niña episode, December 2009. (b) During a very strong El Niño episode, December 1997. Inset: During a strong La Niña episode, December 2010.
Reuse & Permissions
Figure 5
(i) The geographic distribution of degrees for the nodes of the network for $C = 0.5$ and above. (a) One year before a very strong El Niño episode, December 1996. (b) During a very strong El Niño episode, December 1997. (ii) The teleconnection plots of the networks at $C = 0.5$ and above. (c) One year before a very strong El Niño episode, December 1996. The same quantity in the red rectangular region in (c) is shown during a very strong El Niño episode, December 1997 in (d) indicating the teleconnections between the nodes with degree $\geq 70$ .
Reuse & Permissions

Physical Review E

covering statistical, nonlinear, biological, and soft matter physics

Precursors of the El Niño phenomenon: A climate network analysis

Rupali Sonone and Neelima Gupte

Phys. Rev. E 103, L040301 – Published 16 April 2021

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

References (Subscription Required)

Issue

Access Options

Physical Review E

covering statistical, nonlinear, biological, and soft matter physics

Precursors of the El Niño phenomenon: A climate network analysis

Rupali Sonone and Neelima Gupte

Phys. Rev. E 103, L040301 – Published 16 April 2021

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

References (Subscription Required)

Issue

Access Options

Authorization Required

Other Options

Download & Share

Images

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Log In

Search

Article Lookup

Paste a citation or DOI

Enter a citation