the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Global aerosol typing classification using a new hybrid algorithm utilizing Aerosol Robotic Network data
Xiaoli Wei
Qian Cui
Leiming Ma
Feng Zhang
Wenwen Li
Peng Liu
Abstract. Aerosols have great uncertainty owing to the complex changes in their composition in different regions. The radiation properties of different aerosol types differ considerably and are vital in studying aerosol regional and/or global climate effects. Traditional aerosol-type identification algorithms, generally based on cluster or empirical analysis methods, are often inaccurate and time-consuming. Hence, we aimed to develop a new aerosol-type classification model using an innovative hybrid algorithm to improve the precision and efficiency of aerosol-type identification. An optical database was built using Mie scattering and a complex refractive index was used as a baseline to identify different aerosol types by applying a random forest algorithm to train the aerosol optical parameters obtained from the Aerosol Robotic Network sites. The consistency rates of the new model with the traditional Gaussian density cluster method were 90 %, 85 %, 84 %, 84 %, and 100 % for dust, mixed-coarse, mixed-fine, urban/industrial, and biomass burning aerosols, respectively. The corresponding precision of the new hybrid algorithm (F-score and accuracy scores) was 95 %, 89 %, 91 %, and 89 %. Lastly, a global map of aerosol types was generated using the new model to characterize aerosol types across the five continents. This study utilizing a novel approach for the classification of aerosol will help improve the accuracy of aerosol inversion and determine the sources of aerosol pollution.
- Preprint
(2996 KB) - Metadata XML
- BibTeX
- EndNote
Xiaoli Wei et al.
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-1754', Anonymous Referee #1, 12 Oct 2023
The manuscript " Global aerosol typing classification using a new hybrid algorithm utilizing Aerosol Robotic Network data" aims to develop a new aerosol-type classification model using an innovative hybrid algorithm to improve the precision and efficiency of aerosol-type identification. The study shows good consistency between the new method and traditional Gaussian density cluster method, with consistency rates of 90%, 85%, 84%, 84%, and 100% for dust, mixed-coarse, mixed-fine, urban/industrial, and biomass burning aerosols, respectively. Overall, the manuscript provides a well-structured and clear overview of the study design, methodology, and results. The authors communicate the significance of their findings in addressing the issue of classifying aerosol type accurately and efficiently in global scale, which has important implications for aerosol inversion and aerosol pollution study. However, there are a few areas where the manuscript could beÂ
improved.Â
1. The authors should elaborate on why they chose Mie scattering model to build aerosol optical database for classifying the aerosol type.
2. It would be helpful to provide more context on the limitations of their approach and future directions for research in this area.
3. The manuscript needs to have more information in the result about the improvements in calculation time efficiency of aerosol type classification with a specific scale.
4. The manuscript requires a clearer explanation of how the random forest model was implemented and any potential biases associated with the model.
5.More relevant literature review should be included, especially those from the last three years.
Overall, the manuscript presents a valuable contribution to the field of aerosol typing classification research, and with some revisions, it has the potential to be published in this journal.Citation: https://doi.org/10.5194/egusphere-2023-1754-RC1 -
RC2: 'Comment on egusphere-2023-1754', Anonymous Referee #2, 29 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1754/egusphere-2023-1754-RC2-supplement.pdf
Xiaoli Wei et al.
Xiaoli Wei et al.
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
196 | 93 | 11 | 300 | 6 | 7 |
- HTML: 196
- PDF: 93
- XML: 11
- Total: 300
- BibTeX: 6
- EndNote: 7
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1