Bar-Ilan News

Spokesman office Tel.: 03-531-8121 Fax: 03-738-4079
  • Israel Prize-winning physicist Prof. Shlomo Havlin, of the Department of Physics

    Israel Prize-winning physicist Prof. Shlomo Havlin, of the Department of Physics

    Early Warning: Physicists Forecast "El Niño" for 2020

    Date: 2019-12-31 Hour: 14:37

    The significant weather phenomenon "El Niño" could soon occur again in the Pacific region, according to researchers from Justus Liebig University Giessen (JLU), the Potsdam Institute for Climate Impact Research (PIK) and Bar-Ilan University.  While commonly used models do not currently show any signs of an upcoming storm, the researchers predict "El Niño" by the end of 2020.

    El Niño, which occurs when changing ocean conditions disrupt weather patterns, is in the forecast for 2020, according to German and Israeli researchers.  The prediction method they developed is based on an algorithm that relies on a network analysis of air temperatures in the Pacific region.  This algorithm correctly predicted the last two El Niño events (in 2014 and 2018) more than a year in advance. Now it is predicting one by the end of 2020.  Among other things, such long-term forecasts can assist farmers in preparing themselves and adjusting their sowing accordingly.

    A probability of around 80 percent

    "This novel climate network approach is very promising for improving El Niño prediction, said Prof. Shlomo Havlin, an Israel Prize-winning physicist from Bar-Ilan University who was involved in developing the algorithm.

    "Conventional methods are unable to make a reliable El Niño forecast more than six months in advance. With our method, we have roughly doubled the previous warning time," stresses JLU physicist Armin Bunde, who initiated the development of the algorithm together with his former PhD student Josef Ludescher.

    Hans Joachim Schellnhuber, Director Emeritus of PIK, explains: "This clever combination of measured data and mathematics gives us unique insights -- and we make these available to the people affected.” He points out that the prediction method does not offer one hundred percent certainty: "The probability of 'El Niño' in 2020 is around 80 percent. But that's pretty significant."

    Josef Ludescher, now at PIK, emphasizes: "We also predicted the absence of another 'El Niño' in 2019 at the end of last year. Only since July have official forecasts agreed with ours." The team is currently expanding the algorithm in order to be able to forecast the strength and length of the weather phenomenon in the future.

    Empty fishing nets, torrential rainfall, extended droughts

    With a conventional early warning period of at most half a year, people in the tropics and subtropics are poorly prepared for the often-devastating consequences of "El Niño" (Spanish for "the Godly Child").  El Niño can occur at irregular intervals around Christmas and result in empty fishing nets and torrential rainfall in Peru, as well as extended droughts in parts of South America, Indonesia, Australia and Africa. In addition, the Indian subcontinent may experience a change in monsoon patterns and California may experience more precipitation.

    For their investigations, the researchers used a network of atmospheric temperature data in the tropical Pacific consisting of 14 grid points in the equatorial "El Niño" core area, and 193 points in the Pacific outside this core area. The physicists had discovered that already in the year before the eruption of an "El Niño", the teleconnection effect between the air temperatures inside and outside of the core area becomes considerably stronger. In particular, they used this effect to optimize their prediction algorithm.

    The prediction method was first published in 2013 in an article in Proceedings of the National Academy of Sciences. Reliable data from the period between the beginning of 1950 and the end of 2011 were available to the researchers for the investigations. The period between 1950 and 1980 served as a learning phase for determining the alarm thresholds. With the help of this algorithm, the "El Niño" events could then be predicted and compared with actual events. In 80 percent of the cases, the alarm was correct and the "El Niño" event could be accurately predicted the year before.