Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

A chemistry-informed deep learning network for mitigating the stratospheric OH data gap.

Wenjie Yin^1,2, Chen Zhou¹, Wuhu Feng^2,3

¹School of Earth and Space Science Technology, Wuhan University, Wuhan, China.

Science Advances

|July 8, 2026

Summary

Related Concept Videos

Inductive Effects on Chemical Shift: Overview

Inductive Effects on Chemical Shift: Overview

The protons in unsubstituted alkanes are strongly shielded with chemical shifts below 1.8 ppm. Methine, methylene, and methyl protons appear at approximately 1.7, 1.2 and 0.7 ppm, while the proton signal from methane appears at 0.23 ppm. An electronegative substituent, such as chlorine, withdraws the electron density from the protons, increasing their chemical shift. Progressive substitution of the hydrogens in methane by chlorine shifts the proton signals increasingly downfield, to 3.05 ppm in...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Cosmic ray-driven electron-induced reaction theory does not quantify spatiotemporal variations in lower-stratospheric ozone and temperature.

Proceedings of the National Academy of Sciences of the United States of America·2026

Same author

Simulated Annealing-Guided Geometric Descent-Optimized Frequency-Domain Compression-Based Acquisition Algorithm.

Sensors (Basel, Switzerland)·2026

Same author

The influence of short-lived halogens on atmospheric chemistry and climate.

Nature·2025

Same author

Aerosol, clouds and particles: general discussion.

Faraday discussions·2025

Same author

Transport and chemistry: general discussion.

Faraday discussions·2025

Same author

Ongoing large ozone depletion in the polar lower stratospheres: the role of increased water vapour.

Faraday discussions·2025

A new deep learning model, DRCAT, reconstructs global hydroxyl (OH) radical data, filling critical observational gaps since 2009. This chemistry-informed network provides reliable stratospheric OH profiles and shows generalizability to extreme events.

Area of Science:

Atmospheric Chemistry
Stratospheric Science
Machine Learning Applications

Background:

Hydroxyl (OH) radical is crucial for stratospheric ozone depletion and chemical family interactions.
Global OH observations are sparse, with a significant data gap since 2009.
Existing methods to fill OH data gaps have limitations.

Purpose of the Study:

To develop a novel method for reconstructing continuous global stratospheric OH profiles.
To address the critical observational void in satellite-based OH measurements.
To provide a reliable dataset for understanding stratospheric chemistry.

Main Methods:

Developed DRCAT, a chemistry-informed deep learning network.
Utilized coincident satellite observations of chemically related species for training.

Related Experiment Videos

Trained the network on two years of stratospheric chemistry data.

Main Results:

DRCAT successfully reconstructed a continuous global stratospheric OH dataset from 2004 to present.
The model outperformed traditional chemical transport models and steady-state approximations.
DRCAT accurately predicted OH enhancements following the 2022 Hunga volcanic eruption, demonstrating robustness.

Conclusions:

DRCAT offers a robust solution for filling observational gaps in stratospheric OH data.
The deep learning framework is scalable for reconstructing other short-lived atmospheric species.
This approach enhances our ability to monitor and understand stratospheric chemical processes.