Inversion algorithm of black carbon mixing state based on machine learning

<p>The radiative properties of black carbon (BC) are significantly influenced by its mixing state. The single-particle soot photometer (SP2) is a widely recognized instrument for quantifying BC mixing state. However, the derivation of BC mixing state from SP2 is quite challenging. Since the SP...

Full description

Saved in:
Bibliographic Details
Main Authors: Z. Tian, J. Wang, C. Liu, J. Xing, Z. Zhang, Y. Jin, S. Shen, B. Wang, W. Nie, X. Huang, A. Ding
Format: Article
Language:English
Published: Copernicus Publications 2025-03-01
Series:Atmospheric Measurement Techniques
Online Access:https://amt.copernicus.org/articles/18/1149/2025/amt-18-1149-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849705919569461248
author Z. Tian
Z. Tian
J. Wang
J. Wang
J. Wang
J. Wang
C. Liu
C. Liu
J. Xing
J. Wang
J. Wang
Z. Zhang
Z. Zhang
Y. Jin
Y. Jin
S. Shen
S. Shen
B. Wang
B. Wang
W. Nie
W. Nie
X. Huang
X. Huang
A. Ding
A. Ding
author_facet Z. Tian
Z. Tian
J. Wang
J. Wang
J. Wang
J. Wang
C. Liu
C. Liu
J. Xing
J. Wang
J. Wang
Z. Zhang
Z. Zhang
Y. Jin
Y. Jin
S. Shen
S. Shen
B. Wang
B. Wang
W. Nie
W. Nie
X. Huang
X. Huang
A. Ding
A. Ding
author_sort Z. Tian
collection DOAJ
description <p>The radiative properties of black carbon (BC) are significantly influenced by its mixing state. The single-particle soot photometer (SP2) is a widely recognized instrument for quantifying BC mixing state. However, the derivation of BC mixing state from SP2 is quite challenging. Since the SP2 records individual particle signals, it requires complex data processing to convert raw signals into particle size and mixing states. Furthermore, the rapid accumulation of substantial data volumes impedes real-time analysis of BC mixing states. This study employs the Light Gradient-Boosting Machine (LightGBM), an advanced tree-based ensemble learning algorithm, to establish an inversion model that directly correlates SP2 signals with the mixing state of BC-containing particles. Our model achieves high accuracy for both particle size inversion and optical cross-section inversion of BC-containing particles, with a coefficient of determination <span class="inline-formula"><i>R</i><sup>2</sup></span> higher than 0.98. We further employ the SHapley Additive exPlanation (SHAP) method to analyze the importance of input features from SP2 signals in the inversion model of the entire particle diameter (<span class="inline-formula"><i>D</i><sub>p</sub></span>) and explore their underlying physical significance. Compared to the widely used leading-edge-only (LEO) fitting method, the machine learning (ML) method utilizes a larger coverage of signals encompassing the peak of scattering signal rather than the leading-edge data. This allows for more accurate capture of the diverse characteristics of particles. Moreover, the ML method uses signals with a high signal-to-noise ratio, providing better noise resistance. Our model is capable of accurately and efficiently acquiring the single-particle information and statistical results of the BC mixing state, which provides essential data for BC aging mechanism investigation and the assessment of further BC radiative effects.</p>
format Article
id doaj-art-9dbefcd7e7a4428983eea2e516bb71fa
institution DOAJ
issn 1867-1381
1867-8548
language English
publishDate 2025-03-01
publisher Copernicus Publications
record_format Article
series Atmospheric Measurement Techniques
spelling doaj-art-9dbefcd7e7a4428983eea2e516bb71fa2025-08-20T03:16:21ZengCopernicus PublicationsAtmospheric Measurement Techniques1867-13811867-85482025-03-01181149116210.5194/amt-18-1149-2025Inversion algorithm of black carbon mixing state based on machine learningZ. Tian0Z. Tian1J. Wang2J. Wang3J. Wang4J. Wang5C. Liu6C. Liu7J. Xing8J. Wang9J. Wang10Z. Zhang11Z. Zhang12Y. Jin13Y. Jin14S. Shen15S. Shen16B. Wang17B. Wang18W. Nie19W. Nie20X. Huang21X. Huang22A. Ding23A. Ding24State Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaJoint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, ChinaNational Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, Nanjing, 210023, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaDepartment of Civil and Environmental Engineering, University of Tennessee, Knoxville, TN 37996, USAJoint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, ChinaNational Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, Nanjing, 210023, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaState Key Laboratory of Climate System Prediction and Risk Management/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/China Meteorological Administration Aerosol-Cloud and Precipitation Key Laboratory, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaSchool of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing, 210044, ChinaJoint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, ChinaNational Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, Nanjing, 210023, ChinaJoint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, ChinaNational Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, Nanjing, 210023, ChinaJoint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, ChinaNational Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, Nanjing, 210023, China<p>The radiative properties of black carbon (BC) are significantly influenced by its mixing state. The single-particle soot photometer (SP2) is a widely recognized instrument for quantifying BC mixing state. However, the derivation of BC mixing state from SP2 is quite challenging. Since the SP2 records individual particle signals, it requires complex data processing to convert raw signals into particle size and mixing states. Furthermore, the rapid accumulation of substantial data volumes impedes real-time analysis of BC mixing states. This study employs the Light Gradient-Boosting Machine (LightGBM), an advanced tree-based ensemble learning algorithm, to establish an inversion model that directly correlates SP2 signals with the mixing state of BC-containing particles. Our model achieves high accuracy for both particle size inversion and optical cross-section inversion of BC-containing particles, with a coefficient of determination <span class="inline-formula"><i>R</i><sup>2</sup></span> higher than 0.98. We further employ the SHapley Additive exPlanation (SHAP) method to analyze the importance of input features from SP2 signals in the inversion model of the entire particle diameter (<span class="inline-formula"><i>D</i><sub>p</sub></span>) and explore their underlying physical significance. Compared to the widely used leading-edge-only (LEO) fitting method, the machine learning (ML) method utilizes a larger coverage of signals encompassing the peak of scattering signal rather than the leading-edge data. This allows for more accurate capture of the diverse characteristics of particles. Moreover, the ML method uses signals with a high signal-to-noise ratio, providing better noise resistance. Our model is capable of accurately and efficiently acquiring the single-particle information and statistical results of the BC mixing state, which provides essential data for BC aging mechanism investigation and the assessment of further BC radiative effects.</p>https://amt.copernicus.org/articles/18/1149/2025/amt-18-1149-2025.pdf
spellingShingle Z. Tian
Z. Tian
J. Wang
J. Wang
J. Wang
J. Wang
C. Liu
C. Liu
J. Xing
J. Wang
J. Wang
Z. Zhang
Z. Zhang
Y. Jin
Y. Jin
S. Shen
S. Shen
B. Wang
B. Wang
W. Nie
W. Nie
X. Huang
X. Huang
A. Ding
A. Ding
Inversion algorithm of black carbon mixing state based on machine learning
Atmospheric Measurement Techniques
title Inversion algorithm of black carbon mixing state based on machine learning
title_full Inversion algorithm of black carbon mixing state based on machine learning
title_fullStr Inversion algorithm of black carbon mixing state based on machine learning
title_full_unstemmed Inversion algorithm of black carbon mixing state based on machine learning
title_short Inversion algorithm of black carbon mixing state based on machine learning
title_sort inversion algorithm of black carbon mixing state based on machine learning
url https://amt.copernicus.org/articles/18/1149/2025/amt-18-1149-2025.pdf
work_keys_str_mv AT ztian inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT ztian inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT cliu inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT cliu inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jxing inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT jwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT zzhang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT zzhang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT yjin inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT yjin inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT sshen inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT sshen inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT bwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT bwang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT wnie inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT wnie inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT xhuang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT xhuang inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT ading inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning
AT ading inversionalgorithmofblackcarbonmixingstatebasedonmachinelearning