Large earthquakes follow highly unequal ones

By analyzing seismic data and numerical simulations, this study demonstrates that large earthquakes tend to follow periods of high inequality in event distribution, suggesting that monitoring inequality indices like the Gini and Kolkata indices can serve as a tool for estimating proximity to criticality and predicting seismic hazards.

The Gist

The "Unfairness" Predictor: How Earthquakes Signal Trouble

Imagine you are watching a group of people playing a game of musical chairs. Most of the time, everyone moves around a little bit, and when the music stops, only one or two people are scrambling. It’s relatively "fair"—the movement is spread out among everyone.

But suddenly, the energy in the room changes. Instead of everyone moving a little, one person starts sprinting wildly, while everyone else stands perfectly still. The "inequality" of movement has skyrocketed. In the world of physics, that sudden spike in unfairness is often a warning that something massive is about to happen—like a giant crash.

This paper, written by researchers Sudip Sarkar and Soumyajyoti Biswas, suggests that we can use this exact concept of "inequality" to predict when a massive earthquake is looming.

---

1. The Concept: From Wealth to Earthquakes

Usually, when we hear the word "inequality," we think of money—the Gini index is a famous way to measure how much wealth is held by a tiny group of billionaires versus everyone else.

The researchers decided to apply this "wealth" math to energy.

• Normal times: Earthquakes are like a steady paycheck. You get lots of small "payouts" of energy, and they are somewhat consistent.
• Pre-disaster times: The energy distribution becomes "unfair." A few tiny events might release a lot of energy, or the pattern of energy release becomes wildly uneven.

The paper argues that as tectonic plates approach a "breaking point" (what scientists call criticality), the way they release energy becomes increasingly unequal.

2. The "Sandpile" and the "Train" (The Models)

Since we can't experiment on the actual Earth (it's too big and too deep!), the scientists used two digital "playgrounds" to test their theory:

• The Sandpile Model: Imagine slowly pouring sand onto a pile. Most of the time, a few grains slide down. But occasionally, a tiny movement triggers a massive avalanche. The researchers found that right before a "mega-avalanche," the tiny movements became very "unequal" in size.
• The Train Model: Imagine a long train being pulled over a bumpy, rocky track. The train moves in "jerks" (stick-slip motion). The researchers found that before a massive, system-wide jerk occurs, the smaller movements of the train become highly irregular and unequal.

3. The Real-World Test: Checking the Earth

The researchers didn't just stop at computer models; they looked at real earthquake data from North America, Japan, Southeast Asia, and Indonesia.

They used a "sliding window" method. Imagine looking at a history book through a magnifying glass that only shows 100 pages at a time. As you slide that glass through time, you calculate the "inequality score" (the Gini and Kolkata indices) for those 100 pages.

What they found:
Whenever a massive earthquake (a "big event") was about to happen, the "inequality score" of the preceding smaller earthquakes had spiked. The energy wasn't being released smoothly; it was becoming "clumpy" and "unfair."

4. Why does this matter? (The "So What?")

Predicting earthquakes is one of the hardest challenges in science. Currently, scientists look at things like the "b-value" (how many small vs. large quakes occur).

This paper offers a new tool for the toolkit. Instead of just counting how many quakes happen, we can monitor how unequal the energy release is.

The Takeaway:
If we see the "inequality" of seismic energy rising—meaning the energy is being distributed in a wildly uneven, "unfair" way—it acts like a cosmic alarm bell. It tells us that the tectonic plates are getting restless and are approaching a critical breaking point. It’s not a guarantee of when the big one will hit, but it’s a way to measure how close we are to the edge.

Technical Summary

Technical Summary: Large Earthquakes Follow Highly Unequal Ones

Authors: Sudip Sarkar and Soumyajyoti Biswas
Field: Statistical Physics / Seismology

---

1. Problem Statement

The fundamental challenge in seismology is the difficulty of forecasting large-magnitude earthquakes due to the inaccessibility of tectonic plate boundaries and the complex, non-linear nature of stress accumulation. While the Gutenberg-Richter (GR) law (describing magnitude distribution) and the Omori-Utsu law (describing aftershock decay) provide statistical regularities, they do not inherently provide a predictive signal for imminent catastrophic events.

The researchers address whether the inequality of energy release in a system approaching a critical point can serve as a precursor to large-scale "system-spanning" events (major earthquakes). They hypothesize that as a Self-Organized Critical (SOC) system nears a critical state, the distribution of its responses (energy releases) becomes increasingly unequal.

2. Methodology

The study employs a dual approach, combining numerical simulations of physical models with the analysis of real-world seismic catalogs.

A. Numerical Models:

• 2D Sandpile-like Model: A cellular automata model where stress increases linearly at each site. When a threshold ($S_c$) is reached, a cluster of sites undergoes a stress drop. This model is designed to reproduce GR and Omori-Utsu laws.
• Train Model: A simplification of the Burridge-Knopoff model, representing tectonic plates as blocks connected by elastic springs moving over a frictional surface. This model is equivalent to the quenched Edwards-Wilkinson model and exhibits SOC dynamics.

B. Inequality Indices:
To quantify the "unevenness" of energy release within a sliding temporal window ($W=100$ events), the authors use two indices borrowed from economics:

• Gini Index ($g$): Measures the area between the Lorenz curve (cumulative energy distribution) and the line of perfect equality.
• Kolkata Index ($k$): A generalization of the Pareto principle, identifying the fraction of events that account for a specific fraction of total energy.

C. Real-World Data Analysis:
The authors analyzed USGS earthquake catalogs (1975–2025) from four tectonically active regions: Southern Japan, South-East Asia, North America, and Indonesia. They converted magnitudes to energy ($E$) and applied a "dynamic reset" approach, where the window size expands until a large event ($\ge M7.5$) occurs, at which point the window resets.

3. Key Contributions

• Identification of a Universal Precursor: The paper demonstrates that elevated inequality indices ($g$ and $k$) are a universal feature of systems approaching criticality, regardless of whether they are simulated models or real tectonic plates.
• Bridging Inequality and $b$-value: The study provides a mathematical link between the increase in inequality indices and the observed decrease in the GR $b$-value (the exponent of the magnitude distribution) prior to large earthquakes.
• Quantitative Proximity to Criticality: The authors show that the intersection of observed $g$ and $n$ (where $n$ relates to the power-law exponent) values provides a quantitative estimate of how close a tectonic system is to a critical state.

4. Results

• Correlation with Magnitude: In both the sandpile and train models, larger avalanches (energy releases) consistently accumulate at higher $g$ and $k$ values.
• Empirical Validation: Analysis of real-world data shows a clear trend: large earthquakes are preceded by sequences of highly unequal energy releases. In all four studied regions, the largest 20 events occurred when the Gini and Kolkata indices were at or near their maximum values.
• Convergence of Indices: The study found that as a system approaches a critical point, the Gini and Kolkata indices tend to converge (approaching $\sim 0.87$), a behavior observed in both simulations and seismic data.
• Dynamic Reset Findings: The dynamic window method confirmed that high inequality is a reliable indicator of imminent large-scale energy release across different geographic tectonic settings.

5. Significance

This research shifts the focus of earthquake hazard assessment from simple frequency-magnitude counts to the statistical distribution of energy inequality.

The findings suggest that continuous monitoring of Gini and Kolkata indices in seismic time series could serve as a novel tool for earthquake forecasting. Because these indices are independent of the absolute size of events and focus on the functional form of the distribution, they offer a robust method to detect the "proximity to criticality," potentially providing a window of warning for imminent high-magnitude seismic events.