Preprints
https://doi.org/10.5194/egusphere-2024-2496
https://doi.org/10.5194/egusphere-2024-2496
18 Oct 2024
 | 18 Oct 2024

Inversion Algorithm of Black Carbon Mixing State Based on Machine Learning

Zeyuan Tian, Jiandong Wang, Jiaping Wang, Chao Liu, Jinbo Wang, Zhouyang Zhang, Yuzhi Jin, Sunan Shen, Bin Wang, Wei Nie, Xin Huang, and Aijun Ding

Abstract. Black carbon (BC) radiative impact is significantly influenced by its mixing state. 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. Besides, the rapid accumulation of substantial data volumes impedes real-time analysis of BC mixing states. This study employs a light gradient boosting machine (LightGBM) to establish an inversion model which 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 R2 higher than 0.98. Further, we employed 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 (Dp) and explored 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 further BC radiative effects assessment.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Zeyuan Tian, Jiandong Wang, Jiaping Wang, Chao Liu, Jinbo Wang, Zhouyang Zhang, Yuzhi Jin, Sunan Shen, Bin Wang, Wei Nie, Xin Huang, and Aijun Ding

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2024-2496', Anonymous Referee #2, 03 Dec 2024
    • AC2: 'Reply on RC1', Jiandong Wang, 15 Dec 2024
  • CC1: 'Comment on egusphere-2024-2496', Xiaolong Fan, 06 Dec 2024
    • AC3: 'Reply on CC1', Jiandong Wang, 15 Dec 2024
  • RC2: 'Comment on egusphere-2024-2496', Anonymous Referee #1, 13 Dec 2024
    • AC1: 'Reply on RC2', Jiandong Wang, 15 Dec 2024
Zeyuan Tian, Jiandong Wang, Jiaping Wang, Chao Liu, Jinbo Wang, Zhouyang Zhang, Yuzhi Jin, Sunan Shen, Bin Wang, Wei Nie, Xin Huang, and Aijun Ding
Zeyuan Tian, Jiandong Wang, Jiaping Wang, Chao Liu, Jinbo Wang, Zhouyang Zhang, Yuzhi Jin, Sunan Shen, Bin Wang, Wei Nie, Xin Huang, and Aijun Ding

Viewed

Total article views: 294 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
207 75 12 294 26 1 1
  • HTML: 207
  • PDF: 75
  • XML: 12
  • Total: 294
  • Supplement: 26
  • BibTeX: 1
  • EndNote: 1
Views and downloads (calculated since 18 Oct 2024)
Cumulative views and downloads (calculated since 18 Oct 2024)

Viewed (geographical distribution)

Total article views: 304 (including HTML, PDF, and XML) Thereof 304 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 18 Dec 2024
Download
Short summary
The radiative effect of black carbon (BC) is substantially modulated by its mixing state, which is challenging to physically derive from the Single-particle soot photometer. This study establishes a machine learning-based inversion model, which can accurately and efficiently acquire the BC mixing state. Compared to the widely used Leading-Edge-Only method, our model utilizes a broader scattering signal coverage to more accurately capture diverse particle characteristics.