Related Experiment Video
Updated: Jan 14, 2026

Investigating the Relationship between Sea Surface Chlorophyll and Major Features of the South China Sea with Satellite Information
Published on: June 13, 2020
A novel approach to predict the arctic stratospheric ozone from stratospheric polar vortex dynamics using explainable
Anish Kumar1, Joyjit Mandal2, Sina Mehrdad3
1Institute for Meteorology, Leipzig University, Leipzig, Germany. anish.kumar@uni-leipzig.de.
Arctic ozone levels are decreasing, prompting the need for predictive algorithms. This study uses machine learning on stratospheric polar vortex dynamics to forecast ozone, showing promising accuracy for future climate modeling.
Area of Science:
- Atmospheric Chemistry
- Climate Science
- Machine Learning
Background:
- Arctic stratospheric ozone depletion has trended downwards since 2019.
- An Arctic ozone hole was reported in the Stratospheric Polar Vortex (SPV) in 2020, causing concern.
Purpose of the Study:
- To develop an efficient algorithm for predicting Arctic ozone levels.
- To utilize the morphological and dynamical properties of the SPV for ozone prediction.
Main Methods:
- An explainable machine learning approach using XGBoost.
- Feature engineering based on physics-based properties of the SPV.
Main Results:
- XGBoost achieved an R-squared score of 0.80 and a correlation of 0.91 with observed ozone levels.
- The algorithm accurately predicted daily and seasonal ozone variations, including the 2020 low, with a ~20 Dobson unit overestimation.
- The model demonstrated strong alignment with observations in certain years.
Conclusions:
- Dynamical parameters of the SPV are effective for predicting chemical ozone loss.
- The developed algorithm can serve as a tool for projecting future Arctic ozone variability using climate model data.
- This approach offers a chemistry-independent method for ozone variability prediction.
Related Concept Videos
Precipitation Processes
Polar Equations of Conics
Predicting Molecular Geometry
Predicting Reaction Outcomes
Curvilinear Motion: Polar Coordinates
The particle's location is described using a unit vector along the radial direction. Deriving the particle's position...
Potential Due to a Polarized Object