Spatial and temporal distribution of stratospheric turbulence from global high-resolution radiosonde data

Using global high-resolution radiosonde data from 2014 to 2025, this study maps the spatial and temporal distribution of stratospheric turbulent diffusivity, revealing significant regional enhancements driven by mountain waves and convection, a potential aerosol injection zone near the tropical tropopause, and a notable increasing trend over the decade that could improve predictions of stratospheric plume dispersion.

The Gist

Imagine the Earth's atmosphere as a giant, multi-layered ocean. We are used to thinking about the waves and currents in the surface layer (the troposphere) where we live and fly. But there is a vast, invisible ocean above us called the stratosphere.

For a long time, scientists thought this upper layer was calm and still. However, this new study reveals that the stratosphere is actually full of invisible, chaotic "eddies" and "whirlpools" known as turbulence. These aren't the bumps you feel on a plane ride; they are microscopic swirls that act like a giant mixer, blending air, heat, and chemicals together.

Here is a simple breakdown of what the researchers found, using some everyday analogies.

1. The Invisible Mixer

Think of the atmosphere as a giant pot of soup. If you want to mix in a spice (like a pollutant, a rocket exhaust, or a geoengineering aerosol), you need a spoon to stir it. That "spoon" is turbulence.

• The Problem: It's very hard to see these tiny swirls. They are too small for satellites to spot and too rare for planes to catch.
• The Solution: The researchers used 370 weather balloons launched all over the world over the last decade. These balloons are like high-tech thermometers and wind socks that take thousands of measurements as they float up. By looking at how the temperature and wind change in tiny steps (every 5 to 10 meters), the scientists could "see" the invisible mixing happening.

2. What Makes the Stratosphere "Stir"?

The study found that the stratosphere is mostly stable (like a calm lake), but it gets stirred up in two main ways:

• The "Wind Shear" Effect (The Dominant Force): Imagine two layers of a deck of cards sliding past each other at different speeds. The friction between them creates a ruffle. In the sky, when wind layers blow at different speeds, they create friction that turns into turbulence. The study found that 80% of the turbulence happens this way, even when the air is very stable.
• The "Mountain Wave" Effect: When wind blows over a mountain, it creates ripples in the air above it, like water flowing over a rock in a stream. These ripples can travel all the way up to the stratosphere and break, creating turbulence.
• The "Convective" Effect: Sometimes, massive thunderstorms act like a rocket booster, shooting air and energy up into the stratosphere, causing a splash of turbulence.

3. Where is the Turbulence Happening?

The researchers mapped out where the atmosphere is most "stirred up." They found hotspots in:

• Turkey, India, Malaysia, and Japan: These areas are like the "whirlpool zones" of the stratosphere.
• Mountainous regions: Places like the Rocky Mountains and the Andes act as launchpads for turbulence.
• The "Tropical Injection Zone": There is a specific sweet spot just above the equator (around 17 km high) where the air is particularly good at mixing.

4. Why Does This Matter? (The Geoengineering Connection)

This is the most exciting part for the future. Scientists are discussing Stratospheric Aerosol Injection (SAI)—a form of geoengineering where we might intentionally spray tiny particles into the sky to reflect sunlight and cool the Earth.

• The Challenge: If you spray these particles from a plane, you don't want them to clump together or stay in one spot. You want them to spread out quickly and evenly, like sugar dissolving in hot tea.
• The Discovery: The study found that the "sweet spot" for injection is right above the tropical tropopause (around 17 km). This is the "fast lane" for mixing. If we inject particles there, the natural turbulence will help spread them out much faster than if we injected them elsewhere.

5. The Atmosphere is Changing

The study also looked at the last 10 years (2015–2025) and found a surprising trend: The stratosphere is getting more turbulent.

• The Analogy: Imagine the atmosphere is getting "stiffer" (more stable), like a thicker gel. You might think this would stop mixing, but the opposite is happening. Because the air is stiffer, the "eddies" that do manage to form have to be stronger and deeper to break through.
• The Result: The mixing power of the atmosphere has increased by about 3.5 times over the last decade. This suggests that as our climate changes, the way pollutants and aerosols spread in the sky is also changing.

Summary

This paper is like a new map for the invisible currents of the upper atmosphere. It tells us:

1. Where the air is most turbulent (Turkey, India, and the tropics).
2. How it happens (mostly wind friction, sometimes mountains).
3. Why it matters (it controls how fast our planet mixes chemicals and how we might one day cool the Earth).
4. That it's changing (the atmosphere is mixing more vigorously than before).

By understanding these invisible currents, we can better predict how the climate works and make smarter decisions about how we might one day manage the Earth's temperature.

Technical Summary

1. Problem Statement

Stratospheric turbulence is a critical but difficult-to-observe atmospheric process that governs the mixing of momentum, trace gases, and aerosols. While turbulence in the boundary layer is well-studied, stratospheric turbulence remains largely overlooked due to its relative weakness and observational challenges.

• Knowledge Gap: There is a lack of reliable, global observational constraints on stratospheric turbulent diffusivity ($K$). This scarcity hinders the improvement of numerical weather prediction, climate modeling, and the assessment of geoengineering strategies like Stratospheric Aerosol Injection (SAI).
• Specific Need: Understanding the spatial and temporal distribution of $K$ is essential for predicting the dispersion of aircraft plumes and the efficacy of SAI, where turbulent diffusion controls the initial spreading and concentration of injected aerosols. Previous studies relied heavily on sparse aircraft campaigns or focused on turbulence occurrence rather than diffusivity magnitude.

2. Methodology

The study utilizes a unique, large-scale dataset and a specific theoretical framework to estimate turbulent diffusivity globally.

• Data Source:

  • Dataset: Global high-resolution operational radiosonde data (HVRRD) from 370 stations.
  • Time Period: October 2014 to December 2025 (11 years and 3 months).
  • Resolution: 1s and 2s sampling intervals, corresponding to vertical resolutions of ~5m and ~10m.
  • Variables: In-situ measurements of temperature, pressure, humidity, and horizontal wind up to ~30 km altitude.
  • Preprocessing: Data were interpolated to uniform vertical grids and smoothed with a 60m moving-average filter to remove instrumental noise and balloon pendulum motion.

• Turbulence Estimation Method:

  • Framework: The study employs the Minimum Richardson Number ($Ri_{min}$) method proposed by Ko and Chun (2019).
  • Advantage: Unlike the Thorpe method (which only detects overturning/static instability), the $Ri_{min}$ method captures turbulence in statically stable but shear-unstable conditions ($0 < Ri < 0.25$), which is dominant in the stratosphere.
  • Turbulence Classification:
    • PosRi: Statically stable layers with strong shear ($0 < Ri < 0.25$).
    • NegRi: Statically unstable layers with overturning ($Ri < 0$).
  • Diffusivity Calculation: Turbulent diffusivity ($K$) is estimated using a first-order Smagorinsky closure:
    $$K = (0.2 L)^2 |Def|^{0.5} (0.25 - Ri_{min})$$
    Where $L$ is the turbulence layer thickness, $Def$ is the total deformation (approximated by Vertical Wind Shear, VWS, due to lack of horizontal derivative data), and $Ri_{min}$ is the minimum Richardson number within the layer.

3. Key Contributions

• First Global Climatology: Provides the first comprehensive global observational map of stratospheric turbulent diffusivity ($K$) derived from operational radiosondes.
• Regime Differentiation: Systematically separates turbulence into PosRi (shear-driven) and NegRi (overturning) regimes, quantifying their distinct contributions to global diffusivity.
• SAI Relevance: Identifies specific altitude and latitude regions favorable for Stratospheric Aerosol Injection based on turbulent mixing characteristics.
• Decadal Trends: Analyzes the temporal evolution of stratospheric turbulence over the 2015–2025 period, linking changes to atmospheric stabilization.

4. Key Results

A. Global Occurrence and Characteristics

• Magnitude: Log-transformed diffusivity ($\log_{10} K$) ranges from -0.5 to 2. The mean $\log_{10} K$ is 0.554 ($\approx 3.58 \, m^2 s^{-1}$).
• Dominant Regime: 80% of turbulence cases occur in the PosRi regime (statically stable, shear-driven). Only 20% are NegRi (overturning).
• Intensity: Although less frequent, NegRi cases exhibit stronger diffusivity and larger layer thickness ($L$) compared to PosRi cases.
• Drivers: PosRi turbulence is characterized by shallow layers ($L < 200m$) driven by shear, while NegRi involves deeper layers ($L > 350m$) associated with mature overturning.

B. Spatial Distribution and Sources

• Hotspots: Enhanced $K$ values are observed over Turkey, India, Malaysia, Japan, and major mountain ranges (Rockies, Andes, Appalachians).
• Mechanisms:
  • Mountain Waves: Dominant in Turkey, India, and the Rockies, evidenced by high station elevation and zero-wind ($U_0$) ratios.
  • Convection: Dominant in Malaysia and low latitudes, evidenced by high ratios of deep convective profiles (CP210K).
  • Shear: High VWS contributes to high $K$ in mountainous regions, while low $Ri$ (instability) drives high $K$ in convective regions.

C. Latitude-Altitude Structure

• Upper Stratosphere ($>27$ km): A pronounced global maximum in $K$ exists, driven by strong Vertical Wind Shear (VWS) associated with the polar vortex and ozone-related thermal structures.
• Tropical Tropopause Layer (TTL): A distinct local maximum in $K$ is found just above the tropical tropopause (Equator to 15°N, at ~17 km altitude).
  • Implication: This region represents a "sweet spot" for Stratospheric Aerosol Injection (SAI). It offers high turbulent diffusivity for rapid aerosol dispersion while being accessible to transport aircraft (unlike the >27 km region).
• Seasonality: High-latitude turbulence peaks in the winter hemisphere due to the strengthening of the polar vortex and associated shear. Tropical patterns remain relatively consistent year-round.

D. Temporal Trends (2015–2025)

• Increasing Diffusivity: A significant positive trend in global mean $\log_{10} K$ was observed, increasing by $3.5 \times 10^{-3} \, m^2 s^{-1}$ per year.
• Drivers of Trend: The increase is primarily driven by the thickening of turbulence layers ($L$), not by changes in wind shear.
• Regime Shift: The fraction of PosRi cases has increased significantly, while NegRi cases have decreased.
• Interpretation: This suggests the stratosphere is becoming more statically stable (consistent with lower stratospheric cooling), yet the turbulence events that do occur are becoming vertically deeper and more intense to overcome this stability.

5. Significance and Implications

• Geoengineering (SAI): The study provides critical observational constraints for SAI models. The identification of the ~17 km tropical maximum suggests that injecting aerosols in this specific region could maximize initial dispersion efficiency.
• Modeling: The results offer observationally based reference values for turbulent diffusivity, aiding in the parameterization of turbulence in global climate models and Lagrangian plume models.
• Climate Change: The observed trend toward a more stable yet intermittently more intense turbulence environment offers new insights into how stratospheric mixing may evolve under ongoing climate change.
• Aviation: The findings help constrain the dispersion of aircraft exhaust plumes, improving predictions of contrail formation and aviation-induced climate effects.

Limitations: The study acknowledges that radiosonde data reflects turbulence after generation, limiting the diagnosis of initiation mechanisms. Additionally, the spatial distribution of stations is inhomogeneous, and the 11-year record, while significant, is short for establishing robust long-term climate trends. Future work should integrate these observations with reanalysis and model data.