📄 earth_science

Seismic evidence of regional stratification of Earth’s outermost core

Seismic evidence and 3-D modeling reveal that the Earth's outermost core exhibits regional stratification specifically beneath Large Low Shear Velocity Provinces, driven by reduced core-mantle boundary heat flux that promotes light element accumulation and links lowermost mantle heterogeneity to geomagnetic field dynamics.

Original authors: Yangtianli Zhou, Brandon Schmandt, Fenglin Niu, Sheng Yu

Published 2026-07-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yangtianli Zhou, Brandon Schmandt, Fenglin Niu, Sheng Yu

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A Secret Layer at the Bottom of the World

Imagine the Earth is like a giant, multi-layered cake. The very center is a solid iron ball (the inner core), surrounded by a swirling, liquid iron soup (the outer core). Above that is the mantle, which is like a thick, slow-moving layer of hot rock.

For a long time, scientists have wondered about the very top of that liquid iron soup, right where it touches the rock mantle. They suspected there might be a "stratified" layer there—a thin, stagnant layer where the liquid isn't churning and mixing like the rest of the core. Think of it like a layer of oil floating on top of water; it doesn't mix easily.

However, proving this was like trying to see a specific layer of fog inside a dark room from the outside. The evidence was messy, and scientists couldn't agree on whether this layer existed everywhere, only in certain spots, or not at all.

The New Discovery: It's Not Everywhere, It's "Regional"

This new study, led by researchers at Rice University and others, says: "We found the layer, but it's not a global blanket. It's more like a patchwork quilt."

They discovered that this stagnant, stratified layer exists primarily in two specific regions of the Earth's core: right underneath the massive "mountains" of rock in the lower mantle located beneath Africa and the Southwest Pacific.

How They Found It: The Seismic "Echo" Game

To see inside the Earth, scientists use earthquakes as flashlights. When an earthquake happens, it sends out seismic waves (vibrations) that travel through the Earth and bounce back to sensors on the surface.

The researchers focused on a specific type of seismic wave called SmKS.

The Analogy: Imagine you are in a large, empty cathedral (the Earth). You clap your hands (the earthquake). The sound bounces off the walls, the floor, and the ceiling.
The SmKS wave is a special echo that travels through the mantle, dips down into the liquid core, bounces off the very bottom of the core, travels back up through the core, and then exits back into the mantle to reach the sensors.
By measuring exactly how long these echoes take to return, scientists can figure out how fast the waves are traveling through the core. If the waves slow down or speed up, it tells us about the density and temperature of the material they passed through.

The Problem with Old Maps

Previous studies tried to map the whole core using simple, flat maps (1-D models). They assumed the Earth was perfectly symmetrical, like a ball bearing.

The Flaw: The Earth isn't a perfect ball bearing. The mantle above the core is lumpy and uneven.
The Result: When scientists ignored these lumps, their measurements were "noisy" and confusing. It was like trying to measure the speed of a car while driving over a bumpy road and blaming the car for the speed changes, rather than the road.

The New Method: A 3-D Simulation

The researchers in this paper did something different. They used a super-computer to create a 3-D simulation of the Earth.

They built a digital model that included the actual "lumps" and "bumps" in the mantle (using a model called SP12RTS).
They simulated how the seismic waves would behave in this realistic, bumpy environment.
This allowed them to filter out the "noise" caused by the rocky mantle and see the true signal coming from the core.

The "Light Element" Mystery

Once they cleaned up the data, they found something interesting:

Under the "lumpy" rock piles (called LLSVPs) in Africa and the Pacific, the seismic waves were moving slightly faster than expected.
Why? Usually, if a layer is stagnant and hot, waves slow down. But here, the waves sped up.
The Explanation: The researchers propose that because the rock piles above are so hot, they act like a lid, trapping heat and preventing the liquid iron below from churning vigorously.
Because the liquid isn't churning, light elements (like oxygen or silicon mixed into the iron) get stuck there.
The Analogy: Imagine a pot of soup. If you stir it, the ingredients mix evenly. If you stop stirring, the lighter ingredients (like foam or oil) float to the top and accumulate.
This accumulation of light elements makes the liquid slightly denser and changes its speed, which is what the seismic waves detected.

The Conclusion: The Mantle Controls the Core

The most important takeaway is that the mantle (the rock layer) is controlling the core (the iron layer).

The huge, ancient piles of rock at the bottom of the mantle are acting like a thermostat or a lid. They are changing how heat flows out of the core in specific regions. This causes the liquid iron to stop churning in those specific spots, allowing a layer of light elements to build up.

In short: The Earth's core isn't a uniform, churning pot of soup. It has specific "quiet zones" right under the giant rock piles of Africa and the Pacific, where the liquid is stagnant and layered, all because of what's happening in the rock layer above it.

Technical Summary: Seismic Evidence of Regional Stratification of Earth's Outermost Core

Problem Statement
The core-mantle boundary (CMB) represents a critical interface for Earth's thermal and magnetic dynamics. While the lowermost mantle is known to exhibit significant thermochemical heterogeneity, particularly in the form of Large Low Shear Velocity Provinces (LLSVPs), observational evidence for coupled heterogeneity in the overlying outermost core remains scarce. The existence and nature of a stratified layer at the top of the outer core have been debated for decades. Previous seismic studies relying on SmKS phases have suggested a global reduction in seismic velocity consistent with stratification; however, such a velocity decrease contradicts theoretical predictions that light-element enrichment (a likely cause of stratification) should increase velocity in iron alloys. Furthermore, conflicting studies report no evidence for stratification, and the small density contrasts required for stable stratification make detection difficult. Hypotheses regarding the spatial distribution of this stratification vary, proposing global layers, equatorial confinement, or regional variations driven by mantle structure, yet consistent seismic evidence for these specific patterns has been lacking.

Methodology
To address these uncertainties, the authors conducted a global survey of SmKS seismic phases ( $m=2, 3$ ), which graze the underside of the CMB. The study employed an expanded spatial sampling strategy, incorporating data from dense seismic arrays in China, Central and South America, and Antarctica to cover regions previously under-sampled.

The methodology involved three primary technical components:

Data Compilation and Clustering: A dataset of 1,961 high-quality single traces from 15 earthquakes was compiled. To mitigate biases from uneven source and station distributions, these traces were clustered into 65 groups using a density-based clustering algorithm (DBSCAN) based on azimuth and great-circle distance.
Travel Time Measurement: The study compared two measurement approaches. First, a traditional ray-theoretical method was used to validate consistency with prior studies, treating the S3KS waveform as a Hilbert-transformed SKKS template in a 1-D Earth model (PREM). Second, and more critically, the authors utilized 3-D full-wavefield simulations using the AxiSEM3D code. This approach replaced ray-tracing with synthetic waveforms generated using a 1-D core model (PREM) and a 3-D mantle velocity model (SP12RTS). This 3-D method accounts for source radiation patterns, phase crosstalk, and lateral velocity variations in the lower mantle, significantly improving measurement accuracy.
Model Testing: Three competing hypotheses for outermost core structure were tested against the observed travel time residuals: (a) Global stratification (a uniform low-velocity layer), (b) Equatorial stratification (velocity anomalies decaying toward the poles), and (c) Regional stratification beneath LLSVPs (velocity anomalies restricted to regions under the African and Pacific LLSVPs). The performance of these models was quantified using the Residual Sum of Squares (RSS) and assessed via 1,000 bootstrap sampling iterations to determine statistical significance against the null hypothesis (PREM).

Key Contributions and Results
The study provides the first seismic evidence supporting regionalized stratification of the outermost core, specifically beneath LLSVPs.

Refutation of Global and Equatorial Models: The 3-D simulation-based method revealed that nearly neutral residuals are observed when 3-D wave propagation and mantle heterogeneity are modeled, suggesting that a thick, strong global stratification is not required to fit the data. Statistical testing rejected both the global and equatorial stratification models, which showed low probabilities (0.11 and 0.07, respectively) of outperforming the PREM model.
Support for Sub-LLSVP Stratification: The model featuring stratification restricted to regions beneath LLSVPs yielded the lowest RSS. This model demonstrated a 0.76 probability of outperforming PREM and a 0.98 probability of outperforming the global stratification model.
Velocity Characteristics: Contrary to previous interpretations of velocity reductions, the preferred model indicates a subtle increase in compressional velocity (approximately 0.1%) in the outermost core beneath LLSVPs compared to complementary regions. This velocity increase is consistent with the accumulation of light elements in a thermally stratified, less vigorously convecting layer.
Physical Mechanism: The results align with geodynamic expectations where reduced CMB heat flux beneath long-lived thermochemical piles (LLSVPs) leads to a hotter, more stagnant outermost core. This environment promotes the accumulation of light elements, which, in this specific context, results in a slight velocity increase rather than the decrease often assumed in previous global models.

Significance
The paper claims that these findings resolve the long-standing discrepancy between seismic observations and theoretical predictions regarding outermost core structure. By demonstrating that stratification is likely regional rather than global, and that it is spatially coupled to lower mantle heterogeneity (LLSVPs), the study supports the hypothesis that lowermost mantle structures play a significant role in shaping the geomagnetic field and the physical state of the outermost core. The results suggest that the modern regional stratification is a slowly evolving component of Earth's lateral heterogeneity, regulated by mantle-controlled CMB heat flux and the concentration of light elements in the outer core. This work underscores the necessity of using 3-D wavefield modeling to disentangle mantle effects from core structure, providing a more accurate framework for interpreting the Earth's deep interior dynamics.