Why El Niño Sometimes Goes Haywire: A Compass for the Climate ✛

Original: "Statistical Response of ENSO Complexity to Initial Condition and Model Parameter Perturbations"

arXiv:2401.03281 · 2024-01-06 · CC BY 4.0 · 2 min

Atmospheric and Oceanic Physics Applied Statistics

Scientists have found a way to know in advance which nudges most strongly affect the Pacific Ocean's moods.

The Pacific Ocean hosts a huge pendulum of warm water. It swings from the coast of Australia to South America and back, and each swing changes weather worldwide. Sometimes this pendulum acts like a broken clock: it slows down, then suddenly speeds up, causing droughts in some places and floods in others. Scientists have long tried to figure out what pushes the pendulum off course, but previously they only tracked a single path, which in a chaotic system quickly falls apart.

The new approach works like a compass that shows not the pendulum itself but the hidden currents around it. Instead of guessing one single scenario, scientists look at a whole cloud of possible ocean states and measure how much that cloud changes after a small nudge. For this, they use a special measure: relative entropy, which shows how different one probability cloud is from another. If the nudge barely changes the cloud, the system does not notice it; if the cloud spreads or shifts, the nudge hit a sensitive spot.

It turned out that the most sensitive spot depends on how far ahead we look. For short timescales, sea surface temperature matters most. For medium timescales, it is the currents that carry heat. And for long timescales, it is the depth of the layer where water cools sharply, because that is where the heat reserve hides.

The most surprising discovery: some influences do not change the average state of the ocean but make rare and dangerous events much more likely. For example, the friction of wind on water may not move the pendulum from its place, but it can make it slip into extreme positions more often. Previously, such hidden risks were impossible to spot because everyone watched only the average path. Now forecasters have a compass that points to these hidden currents, and in the future it will help warn about climate surprises in advance.

Whole graph →

The extreme El Niño of 1997–1998 was so strong that it raised global sea surface temperatures enough to make 1998 the warmest year on record at that time. This study shows that perturbations to initial conditions at the peak of such an event can influence the system's statistics even two years later.

P(p^\delta(x_t), p(x_t)) = \int p^\delta(x_t) \log \frac{p^\delta(x_t)}{p(x_t)} dx_t

A measure of statistical response: how much the probability distribution of the system state changes after perturbation

P(p^\delta(x_t), p(x_t)) = \frac{1}{2} \delta \cdot I(p(x_t)) \delta + O(|\delta|^3)

For small perturbations, the response is proportional to the square of the perturbation magnitude, and the direction of maximum response is given by the leading eigenvector of the Fisher information matrix

Concepts

ENSO El Niño-Southern Oscillation El Niño index spring predictability barrier Multivariate ENSO Index ERSST Jin-Timmermann model equatorial wave propagation Southern Oscillation recharge oscillator model La Niña NOAA CPC TAO/TRITON moorings statistical method natural variability Teleconnection Warm water volume relative ONI stochastic dynamical system statistical noise

I am the author

Why El Niño Sometimes Goes Haywire: A Compass for the Climate ✛