This article comes from WeChat public account: Jizhi Club (ID: swarma_org) < span class = "text-remarks">, author: Fan Jingfang

Introduction

Since 1900, at least 30 El Nino incidents have occurred.

Recently, Meng Jun and Dr. Fan Jingfang from the Potsdam Institute for Climate Impact Research, Professor Chen Xiaosong and collaborators from Beijing Normal University have proposed a method based on entropy theory to predict the intensity of El Niño events.

For the just arrived 2020, based on the System Sample Entropy method proposed below, the authors of the paper predict that El Niño will have a high probability of happening again in the second half of this year, and will develop into a medium or even high intensity El Niño. event. This article is the author’s interpretation of this PNAS paper.

Thesis title:

Complexity-based approach for Ei Niño magnitude forecasting before the spring predictability barrier

Address of the paper: https://www.pnas.org/content/117/1/177.short

I. What is El Niño?

El Niño is a cyclical climate phenomenon. It refers to the phenomenon that the seawater in the eastern equatorial Pacific warms abnormally every few years, and it occurs on average every 2-7 years. The El Niño-Southern Oscillation phenomenon (El Niño-Southern Oscillation, ENSO) is a climatic phenomenon of the surface temperature and sea breeze of the eastern Pacific near the equator .

El Niño is the warming phase in the El Niño-Southern Oscillation. At present, there are many opinions on the mechanism of this phenomenon in the scientific community. It is generally believed that the occurrence of the El Niño phenomenon is related to the change of the atmospheric circulation—Water circulation—at the Pacific equator. The low-temperature stage in the El Niño-Southern Oscillation phenomenon is called the La Niña phenomenon.

In the middle stages of El Niño and La Niña, the Western Pacific Low and the Eastern Pacific High (or Walter Circulation) will cause sea breeze from east to west The warm seawater and air on the surface of the East Pacific Ocean are blown from east to west, and the low-temperature seawater at the lower level is swept upward, resulting in lower seawater temperature on the surface of the East Pacific Ocean. Before El Niño occurred, the westerly wind on the surface of the equatorial Pacific weakened or turned, which caused the sea temperature in the eastern Pacific to rise, which eventually led to El Niño’soccur. During the El Niño, the normal patterns of tropical precipitation and atmospheric circulation were disrupted and triggered extreme climate events worldwide.

Origin: The term “El Niño” is derived from Spanish and originally meant “holy infant”. At the beginning of the 19th century, in Spanish-speaking countries such as Ecuador and Peru in South America, fishermen discovered that every few years, a warm current southward along the coast appeared from October to March of the next year, causing the surface layer The temperature of seawater has risen significantly. The cold current of Peru was prevalent in the eastern coast of the Pacific Ocean in South America. The fish moving with the cold current made the Peruvian fishery one of the four largest fishing grounds in the world. However, as soon as this warm current appeared, fishes who liked cold water would die in large numbers. The fishermen were devastated. Because this phenomenon is most serious around Christmas, fishermen who have suffered natural disasters and have no choice but to call it the Son of God-Holy Baby.

Definition: The frequency of the El Niño phenomenon is irregular. This phenomenon usually occurs every 2 to 7 years and can last up to 18 months. According to the definition of the US Atmospheric and Oceanic Administration (NOAA) , if the Ocean El Nino Index (Oceanic Niño Index2) exceeds 0.5oC and lasts at least five months, then the event is called the El Niño event (El Niño Episodes) Conversely, if the Ocean El Nino Index is below -0.5oC and lasts for at least five months, the event will be referred to as the La Nina Event (La Nina Episodes) < / span>.

Figure 1: The normal Pacific pattern: The warm pool in the west drives deep atmospheric convection. From Wikipediap>

Figure 2: El Niño: Warm sea water and atmospheric convection move eastward. From Wikipedia

Second, El Niño’s influence

1. Climate Impact

El Niño affects the global climate and disrupts normal weather patterns, which may cause severe rain, flood, extreme heat or cold weather in some places, which may lead to animal disease outbreaks (including zoonotic diseases and foodborne diseases) , as well as plant diseases and insect pests and forest fires. In previous El Niño incidents, people living in fisheries in some areas were severely affected.

Observations of the El Niño events since 1950 have shown that the occurrence and effects of El Niño are characterized by diversity and uncertainty, which has increased the difficulty of related predictions. During most El Niño events, the more clear impacts included below-average rainfall in Indonesia and northern South America, while above-average rainfall occurred in southeastern South America, eastern equatorial Africa, and southern United States.

Impact on China:

For China, El Nino is prone to warm winters, heavy rains and floods in the south, high temperatures and droughts in the north, and cold summers in the northeast. Extreme weather is more likely to cause danger than pure temperature changes.

(1) Typhoons have decreased. The number of tropical storms (Typhoon) and the number of landings in China’s coastal areas were less than in normal years.

(2) The summer monsoon is relatively weak, and the monsoon rain belt is south. It is located in central China or south of the Yangtze River. The northern region is prone to drought and high temperature in summer, and the south is prone to low temperature and flood. The severe floods in China in the past century, such as the floods in the middle and lower reaches of the Yangtze River in 1931, 1954, and 1998, occurred in the following year when the El Nino phenomenon occurred.

(3) In the winter following the El Niño phenomenon, warm winters are prone to occur in northern China.

2. Socio-ecological impact on humans and nature

Economy: El Niño broadly affects commodity prices and the macroeconomics of different countries. It can limit the supply of rain-driven agricultural products; reduce agricultural production, construction and service activities; cause food prices to rise and inflate; and can trigger social unrest in poor, commodity-dependent countries that rely heavily on imported food. In addition, most countries have experienced short-term inflationary pressures following the El Niño shock, while global energy and non-fuel commodity prices have risen.

Health: Extreme weather conditions associated with El Niño are related to changes in the incidence of epidemics. For example, the El Niño cycle is associated with an increased risk of certain diseases transmitted by mosquitoes, such as malaria, dengue and rift valley fever. The vicious cycle of malaria in India, Venezuela, Brazil and Colombia is now linked to El Niño. After a heavy rain and flood, another mosquito-borne disease has erupted in temperate areas of southeastern Australia, namely Australian encephalitis (Murray Valley, Encephalitis, MVE ) , this is related to the La Niña incident. During the 1997-98 El Niño event, extreme rainfall occurred in northeastern Kenya and southern Somalia, followed by a severe outbreak of Rift Valley Fever.

Three, El Niño’s Prediction

The occurrence of the El Niño event has predictive significance for climate disasters in various regions of the world, so its prediction has become one of the most important elements in climate prediction. Early and accurate prediction of the occurrence and intensity of El Niño is of great significance in preventing or reducing the economic, agricultural, social and other losses it causes worldwide, and it will help the government and relevant institutions to formulate policies to deal with the impact of El Niño.

Barriers to spring prediction

At present, the traditional El Niño prediction method can only give a more accurate forecast within a range of 6 months in advance. This phenomenon is called a spring prediction obstacle. (I.e., accurate forecasts cannot be given in the spring or earlier of the year in which El Niño occurred) . Figure 3 shows the prediction capabilities of 18 dynamics and 8 statistical climate models. We see that almost all models lost their ability to accurately predict El Niño during the spring.

Figure 3: Schematic diagram of the prediction capabilities of traditional climate models. From NOAA

Long-term forecast of El Niño events

To overcome the above-mentioned obstacles to spring forecasting, a team from the Potsdam Institute for Climate Impact has proposed a new framework for predicting the beginning of the El Niño event-the climate network approach. Different from general complex networks, nodes are considered in the climate network as sites in the spatial grid of the global climate dataset; the network is defined according to the statistical similarity between the corresponding time series pairs of climate data between the two sites Side. The climate network method enables people to understand the dynamic evolution of the climate system at different time and space scales from a new perspective.

The current climate network-based method can predict the beginning of the El Niño event one year in advance [3, 4, 5], and its prediction accuracy is over 89%. Much higher than traditional climate models. And this series of methods successfully carried out a successful early warning of the events of 2014-2016 and 2018-2019.

Prediction of El Niño event intensity

Although we can overcome the obstacles of the spring forecast to successfully predict the beginning of the El Niño event, there is no way to predict its strength for a long time.degree. The strength of prediction is important because stronger El Niño usually leads to more extreme events (such as floods, droughts or severe storms) A serious impact on the economy, society and ecosystems.

Recently, Meng Jun and Dr. Fan Jingfang from the Potsdam Institute for Climate Impact Research, Professor Chen Xiaosong and collaborators from Beijing Normal University have proposed a method based on entropy theory to predict the intensity of El Niño events [6]. The research paper has been published online in the world famous journal American Academy of Sciences PNAS (https://www.pnas.org/content/117/1/177.short) , for the first time, overcomes the “spring forecast obstacles” that have long plagued El Niño predictions, and will advance the occurrence of the El Niño phenomenon, especially the intensity forecast by one year.

The author of this article proposes a new method based on information entropy theory—System Sample Entropy—to calculate the El Niño region. (Nino 3.4) < / span> Complexity of near-surface air or sea surface temperature (including temperature disorder over time and synchronization or coherence of temperature changes at different locations) . Using this method, the authors found that there is a very strong and stable linear relationship between the complexity of temperature changes in the Nino 3.4 region and the intensity of the El Niño phenomenon. (January 1 to December 31) The greater the complexity of temperature changes in the Nino 3.4 region, the greater the intensity of the El Niño event that will occur in the next year.

Based on this finding, the authors propose a set of magnitudes based on the complexity of temperature changes in the Nino 3.4 area per year (quantified by System Sample Entropy in this area) / span> to predict the occurrence and intensity of El Niño in the coming year. The method currently successfully predicts the year of occurrence of 9 of the 10 El Niño events from 1984 to 2019, and 21 of the 24 years without El Niño, especially the prediction of El Niño intensity.The average error is only 0.23 degrees Celsius.

For the just arrived 2020, based on the System Sample Entropy method proposed in the article, the authors predict that El Niño will have a high probability of recurring in the second half of this year, and develop into a moderate or even high intensity El Niño event. , Its predicted intensity is 1.48 + -0.25 degrees Celsius (Figure 4) .

Fourth, Climate Change and El Nino

The latest report of the Intergovernmental Panel on Climate Change shows that stronger El Niño may occur more frequently as the climate warms. If so, climate events related to El Niño will also become more severe, but the impact of climate change may be more than just the severity of the El Niño phenomenon and related consequences. Some studies suggest that global warming may increase the total area affected by El Niño worldwide. Therefore, when El Niño occurs, its impact on the weather may be more severe than it is today.

Figure 4: System entropy theory predicts the beginning and intensity of the El Niño event one year in advance. Source [6]

References

[1]. https://baike.baidu.com/item/El Niño

[2]. Oceanic Niño Index (ONI) [3 month running mean of ERSST.v4 SST anomalies in the Niño 3.4 region (5oN-5oS, 120o-170oW)]

[3]. J. Ludescher et al., Improved El Nin˜ o forecasting by cooperativity detection. Proc. Natl. Acad. Sci. USA 110, 11742–11745 (2013 ).

[4]. J. Meng, J. Fan, Y. Ashkenazy, S. Havlin, Percolation framework to describe El Nin˜ o conditions. Chaos 27, 035807 (2017) .

[5]. J. Meng, J. Fan, Y. Ashkenazy, A. Bunde, S. Havlin, Forecasting the magnitude and onset of El Nin˜ o based on climate network. New J. Phys. 20, 043036 (2018).

[6]. J. Meng, J. Fan, et al, Complexity-based approach for El Nino magnitude forecasting before the spring predictability barrier. Proc. Natl. Acad. Sci . USA 117, 177-183 (2020).

This article comes from WeChat public account: Jizhi Club (ID: swarma_org) author: Fan Jingfang