Machine learning based bias correction for modis aerosol optical depth in Beijing

Mengjuan Wang; Meng Fan; Zhibao Wang; Liangfu Chen; Lu Bai; Yuanlin Chen; Mei Wang

doi:10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

Machine learning based bias correction for modis aerosol optical depth in Beijing

Mengjuan Wang, Meng Fan, Zhibao Wang, Liangfu Chen, Lu Bai, Yuanlin Chen, Mei Wang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Downloads (Pure)

Abstract

Aerosol refers to suspensions of small solid and liquid particles in the atmosphere. Although the content of aerosol in the atmosphere is small, it plays a crucial role in atmospheric and the climatic processes, making it essential to monitor. In areas with poor aerosol characteristics, satellite-based aerosol optical depth (AOD) values often differ from ground-based AOD values measured by instruments like AERONET. The use of 3km DT, 10km DT and 10km DTB algorithms in Beijing area has led to significant overestimation of AOD values, highlighting the need for improvement. This paper proposes the use of machine learning techniques, specifically support vector regression (SVR) and artificial neural network (ANN), to correct the deviation of AOD data. Our approach leverages ground-based monitoring data, meteorological reanalysis data and satellite products to train the models. Our results show that the ANN model outperforms the SVR model achieving R2, RMSE and Slope values of 0.88, 0.12 and 0.97, respectively, when applied to nearly two decades of data from 2001 to 2019. This study significantly improves the accuracy of MODIS AOD values, reducing overestimation and bringing them closer to ground-based AOD values measured by AERONET. Our findings have important applications in climate research and environmental monitoring.

Original language	English
Title of host publication	Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39
Editors	O. Altan, F. Sunar, D. Klein
Publisher	Copernicus Gesellschaft mbH
Pages	395-402
Number of pages	8
DOIs	https://doi.org/10.5194/isprs-archives-XLVIII-M-1-2023-395-2023
Publication status	Published - 21 Apr 2023
Externally published	Yes
Event	39th International Symposium on Remote Sensing of Environment 2023 - Antalya, Turkey Duration: 24 Apr 2023 → 28 Apr 2023

Publication series

Name	International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Volume	XLVIII-M-1-2023
ISSN (Print)	1682-1750
ISSN (Electronic)	2194-9034

Conference

Conference	39th International Symposium on Remote Sensing of Environment 2023
Abbreviated title	ISRSE-39
Country/Territory	Turkey
City	Antalya
Period	24/04/2023 → 28/04/2023

Keywords

Aerosol optical depth
artificial neural network
bias correction
machine learning
support vector regression

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.5194/isprs-archives-XLVIII-M-1-2023-395-2023Licence: CC BY

Machine learning based bias correction for modis aerosol optical depth in Beijing
Copyright 2023 The Authors. This is an open access article published under a Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited.
Final published version, 1.53 MBLicence: CC BY

Cite this

Wang, M., Fan, M., Wang, Z., Chen, L., Bai, L., Chen, Y., & Wang, M. (2023). Machine learning based bias correction for modis aerosol optical depth in Beijing. In O. Altan, F. Sunar, & D. Klein (Eds.), Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39 (pp. 395-402). (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences; Vol. XLVIII-M-1-2023). Copernicus Gesellschaft mbH. https://doi.org/10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

Wang, Mengjuan ; Fan, Meng ; Wang, Zhibao et al. / Machine learning based bias correction for modis aerosol optical depth in Beijing. Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39. editor / O. Altan ; F. Sunar ; D. Klein. Copernicus Gesellschaft mbH, 2023. pp. 395-402 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences).

@inproceedings{ad75848a51c94eeaa611b9c9d0f88826,

title = "Machine learning based bias correction for modis aerosol optical depth in Beijing",

abstract = "Aerosol refers to suspensions of small solid and liquid particles in the atmosphere. Although the content of aerosol in the atmosphere is small, it plays a crucial role in atmospheric and the climatic processes, making it essential to monitor. In areas with poor aerosol characteristics, satellite-based aerosol optical depth (AOD) values often differ from ground-based AOD values measured by instruments like AERONET. The use of 3km DT, 10km DT and 10km DTB algorithms in Beijing area has led to significant overestimation of AOD values, highlighting the need for improvement. This paper proposes the use of machine learning techniques, specifically support vector regression (SVR) and artificial neural network (ANN), to correct the deviation of AOD data. Our approach leverages ground-based monitoring data, meteorological reanalysis data and satellite products to train the models. Our results show that the ANN model outperforms the SVR model achieving R2, RMSE and Slope values of 0.88, 0.12 and 0.97, respectively, when applied to nearly two decades of data from 2001 to 2019. This study significantly improves the accuracy of MODIS AOD values, reducing overestimation and bringing them closer to ground-based AOD values measured by AERONET. Our findings have important applications in climate research and environmental monitoring.",

keywords = "Aerosol optical depth, artificial neural network, bias correction, machine learning, support vector regression",

author = "Mengjuan Wang and Meng Fan and Zhibao Wang and Liangfu Chen and Lu Bai and Yuanlin Chen and Mei Wang",

year = "2023",

month = apr,

day = "21",

doi = "10.5194/isprs-archives-XLVIII-M-1-2023-395-2023",

language = "English",

series = "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences",

publisher = "Copernicus Gesellschaft mbH",

pages = "395--402",

editor = "O. Altan and F. Sunar and D. Klein",

booktitle = "Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39",

address = "Germany",

note = "39th International Symposium on Remote Sensing of Environment 2023, ISRSE-39 ; Conference date: 24-04-2023 Through 28-04-2023",

}

Wang, M, Fan, M, Wang, Z, Chen, L, Bai, L, Chen, Y & Wang, M 2023, Machine learning based bias correction for modis aerosol optical depth in Beijing. in O Altan, F Sunar & D Klein (eds), Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLVIII-M-1-2023, Copernicus Gesellschaft mbH, pp. 395-402, 39th International Symposium on Remote Sensing of Environment 2023, Antalya, Turkey, 24/04/2023. https://doi.org/10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

Machine learning based bias correction for modis aerosol optical depth in Beijing. / Wang, Mengjuan; Fan, Meng; Wang, Zhibao et al.
Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39. ed. / O. Altan; F. Sunar; D. Klein. Copernicus Gesellschaft mbH, 2023. p. 395-402 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences; Vol. XLVIII-M-1-2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Machine learning based bias correction for modis aerosol optical depth in Beijing

AU - Wang, Mengjuan

AU - Fan, Meng

AU - Wang, Zhibao

AU - Chen, Liangfu

AU - Bai, Lu

AU - Chen, Yuanlin

AU - Wang, Mei

PY - 2023/4/21

Y1 - 2023/4/21

N2 - Aerosol refers to suspensions of small solid and liquid particles in the atmosphere. Although the content of aerosol in the atmosphere is small, it plays a crucial role in atmospheric and the climatic processes, making it essential to monitor. In areas with poor aerosol characteristics, satellite-based aerosol optical depth (AOD) values often differ from ground-based AOD values measured by instruments like AERONET. The use of 3km DT, 10km DT and 10km DTB algorithms in Beijing area has led to significant overestimation of AOD values, highlighting the need for improvement. This paper proposes the use of machine learning techniques, specifically support vector regression (SVR) and artificial neural network (ANN), to correct the deviation of AOD data. Our approach leverages ground-based monitoring data, meteorological reanalysis data and satellite products to train the models. Our results show that the ANN model outperforms the SVR model achieving R2, RMSE and Slope values of 0.88, 0.12 and 0.97, respectively, when applied to nearly two decades of data from 2001 to 2019. This study significantly improves the accuracy of MODIS AOD values, reducing overestimation and bringing them closer to ground-based AOD values measured by AERONET. Our findings have important applications in climate research and environmental monitoring.

AB - Aerosol refers to suspensions of small solid and liquid particles in the atmosphere. Although the content of aerosol in the atmosphere is small, it plays a crucial role in atmospheric and the climatic processes, making it essential to monitor. In areas with poor aerosol characteristics, satellite-based aerosol optical depth (AOD) values often differ from ground-based AOD values measured by instruments like AERONET. The use of 3km DT, 10km DT and 10km DTB algorithms in Beijing area has led to significant overestimation of AOD values, highlighting the need for improvement. This paper proposes the use of machine learning techniques, specifically support vector regression (SVR) and artificial neural network (ANN), to correct the deviation of AOD data. Our approach leverages ground-based monitoring data, meteorological reanalysis data and satellite products to train the models. Our results show that the ANN model outperforms the SVR model achieving R2, RMSE and Slope values of 0.88, 0.12 and 0.97, respectively, when applied to nearly two decades of data from 2001 to 2019. This study significantly improves the accuracy of MODIS AOD values, reducing overestimation and bringing them closer to ground-based AOD values measured by AERONET. Our findings have important applications in climate research and environmental monitoring.

KW - Aerosol optical depth

KW - artificial neural network

KW - bias correction

KW - machine learning

KW - support vector regression

U2 - 10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

DO - 10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

M3 - Conference contribution

T3 - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences

SP - 395

EP - 402

BT - Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39

A2 - Altan, O.

A2 - Sunar, F.

A2 - Klein, D.

PB - Copernicus Gesellschaft mbH

T2 - 39th International Symposium on Remote Sensing of Environment 2023

Y2 - 24 April 2023 through 28 April 2023

ER -

Wang M, Fan M, Wang Z, Chen L, Bai L, Chen Y et al. Machine learning based bias correction for modis aerosol optical depth in Beijing. In Altan O, Sunar F, Klein D, editors, Proceedings of the 39th International Symposium on Remote Sensing of Environment, ISRSE-39. Copernicus Gesellschaft mbH. 2023. p. 395-402. (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences). doi: 10.5194/isprs-archives-XLVIII-M-1-2023-395-2023

Machine learning based bias correction for modis aerosol optical depth in Beijing

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this