To construct self-defined aerosol models and enhance the accuracy of aerosol optical depth (AOD) retrieval, a novel satellite-based method for estimating the volume percentages of aerosol components based on a stacking model was proposed. First, a dataset was constructed utilizing MODIS(MODerate resolution imaging spectroradiometer ) data and ERA5 meteorological data. To address the challenge of limited training samples, Tab-DDPM (tabular denoising diffusion probabilistic model) was employed to generate synthetic samples. Subsequently, a stacking model integrating random forest, XGBoost, and CatBoost was established to accurately estimate the volume percentages of aerosol components. Based on these estimated proportions, self-defined aerosol models were constructed to facilitate improved AOD retrieval. Experimental results indicate that compared with single machine learning models, the estimation accuracy of aerosol components is significantly improved by the stacking model. Validation against ground-based AERONET(AErosol RObotic NETwork) observations in the Beijing and Xianghe regions demonstrates that AOD retrieved based on self-defined aerosol models exhibits high consistency, with correlation coefficients r of 0.80 and 0.83, root mean square errors (RMSE) of 0.23 and 0.20, and mean bias errors (MBE) of 0.09 and 0.03, respectively. Compared with the traditional continental aerosol model (which yields RMSEs of 0.29 and 0.28), systematic errors are significantly reduced by the proposed method. Furthermore, in comparison with the official MODIS AOD product, although the correlation coefficients are slightly lower (official product r≈0.90), superior error control capability is demonstrated by the proposed method in terms of RMSE (official product values of 0.29 and 0.22) and MBE. The results demonstrate that the model mismatch problem is effectively addressed, and the accuracy of AOD retrieval is significantly enhanced by the proposed method.
GU Wei-liang, WAN Yong, DAI Yong-shou, LIU Yu.
Stacking Model for Volume Percentages of Aerosol Components Retrieval and Its Application to Aerosol Optical Depth Retrieval[J].
Science Technology and Engineering, 2026, 26(23): 10080-10092 DOI:10.12404/j.issn.1671-1815.2506569
ZhaoHui, LiMenghua, KangXiong, et al. Spatiotemporal variations of aerosol optical depth in Ningxia and its driving factors[J]. Acta Scientiae Circumstantiae, 2026, 46(1): 01022.
WenShaoxin, LiuHaixin, GaoYepeng, et al. Temporal and spatial evolution of AOD in Beijing-Tianjin-Heibei Region and its meteorological interpretation[J]. Science Technology and Engineering, 2022, 22(30): 13569-13577.
[5]
陈丽. 多源气溶胶光学厚度时空代表性研究[D]. 南京: 南京信息工程大学, 2025.
[6]
ChenLi. Study on the spatiotemporal representativeness from multisource aerosol optical depth[D]. Nanjing: Nanjing University of Information Science and Technology, 2025.
GaoQi, GuoXincheng, MengNina, et al. Analysis on characteristics and potential sources of aerosols in Beijing-Tianjin-Hebei Region[J]. Science Technology and Engineering, 2021, 21(30): 13185-13195.
[9]
刘子岩. 无实测大气样本的静止卫星气溶胶智能反演[D]. 赣州: 江西理工大学, 2025.
[10]
LiuZiyan. Intelligent aerosol retrieval for geostationary satellites without measured atmospheric samples[D]. Ganzhou: Jiangxi University of Science and Technology, 2025.
MiXiaonan, BaiLinyan, ZhaoYongqiang, et al. Analysis of urban heat island effect in Taiyuan based on multi-source satellite data[J]. Science Technology and Engineering, 2021, 21(32): 13650-13658.
FanJingjing, LiJinsong, ChenJie, et al. Mountain fire monitoring and early warning method of transmission line based on Fengyun-3E Satellite[J]. Science Technology and Engineering, 2025, 25(8): 3247-3257.
LuoTianyang, XuXiaofeng, WuHaopeng, et al. Spatial-temporal variation characteristics of aerosols in Xuzhou using multi-sources observations[J]. Environmental Monitoring in China, 2024, 40(5): 206-223.
[17]
GriggsM. Satellite measurements of tropospheric aerosols[J]. Advances in Space Research, 1982, 2(5): 109-118.
[18]
GeogdzhayevV I, MishchenkoI M, RossowB W, et al. Global two-channel AVHRR retrievals of aerosol properties over the ocean for the period of NOAA-9 observations and preliminary retrievals using NOAA-7 and NOAA-11 data[J]. Journal of the Atmospheric Sciences, 2002, 59(3): 262-278.
[19]
AhnC, TorresO, JethvaH. Assessment of OMI near-UV aerosol optical depth over land[J]. Journal of Geophysical Research: Atmospheres, 2014, 119(5): 2457-2473.
[20]
KaufmanY J, TanréD, RemerL A, et al. Operational remote sensing of tropospheric aerosol over land from EOS moderate resolution imaging spectroradiometer[J]. Journal of Geophysical Research: Atmospheres, 1997, 102(D14): 17051-17067.
[21]
LevyR C, RemerL A, DubovikO. Global aerosol optical properties and application to moderate resolution imaging spectroradiometer aerosol retrieval over land[J]. Journal of Geophysical Research: Atmospheres, 2007, 112: D13210.
[22]
HsuC N, TsayS, KingD M, et al. Aerosol properties over bright-reflecting source regions[J]. IEEE Transaction on Geoscience and Remote Sensing, 2004, 42(3): 557-569.
[23]
HsuC N, TsayS C, KingM D, et al. Deep blue retrievals of Asian aerosol properties during ACE-Asia[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(11 ): 3180-3195.
[24]
WeiJ, LiZ, PengY, et al. MODIS Collection 6. 1 aerosol optical depth products over land and ocean: validation and comparison[J]. Atmospheric Environment, 2018, 187: 428-440.
[25]
LyapustinA, WangY, KorkinS, et al. MODIS collection 6 MAIAC algorithm[J]. Atmospheric Measurement Techniques, 2018, 11(10): 5741-5765.
[26]
CaiH, ZhongB, LiuH, et al. An improved deep learning network for AOD retrieving from remote sensing imagery focusing on sub-pixel cloud[J]. GIScience & Remote Sensing, 2023, 60(1): 2262836.
[27]
SuT, IstvanL, LiZ, et al. Refining aerosol optical depth retrievals over land by constructing the relationship of spectral surface reflectances through deep learning: application to Himawari-8[J]. Remote Sensing of Environment, 2020, 251: 112093.
[28]
WangW, WangN, ChenB. Retrieving hourly aerosol optical depth for geostationary satellite FY-4B/AGRI by surface-related dynamic spectral reflectance ratio method[J]. Advances in Space Research, 2025, 75(3): 2484-2505.
LiLi, LiChangjun, ZhaoHaijun, et al. Retrieval of aerosol optical depth using FY-4A data based on custom model[J]. Journal of Marine Meteorology, 2022, 42(4): 106-116.
[31]
MaoQ, HuangC, ChenQ, et al. Satellite-based identification of aerosol particle species using a 2D-space aerosol classification model[J]. Atmospheric Environment, 2019, 219: 117057.
[32]
ChenQ, HuangC, YuanY, et al. Spatiotemporal distribution of major aerosol types over China based on MODIS products between 2008 and 2017[J]. Atmosphere, 2020, 11(7): 703.
FanJiao, GuoBaofeng, HeHongchang. Retrieval of aerosol optical thickness with MODIS data over Hangzhou[J]. Acta Optica Sinica, 2015, 35(1): 9-17.
[37]
BibiH, AlamK, ChishtieF, et al. Intercomparison of MODIS, MISR, OMI, and CALIPSO aerosol optical depth retrievals f or four locations on the Indo-Gangetic plains and validation against AERONET data[J]. Atmospheric Environment, 2015, 111: 113-126.
[38]
WanY, ZhangX, FanC, et al. A joint method for wave and wind field parameter inversion combining SAR with wave spectrometer data[J]. Remote Sensing, 2022, 14(15): 3601.
[39]
KotelnikovA, BaranchukD, RubachevI, et al. Tabddpm: modelling tabular data with diffusion models[C]// Proceedings of the 40th International Conference on Machine Learning (ICML 2023). Honolulu: PMLR, 2023: 17564-17579.
[40]
LiuY, ZhangZ, SongH, et al. An improved stacking model for predicting myocardial infarction risk in imbalanced data[J]. Health Information Science and Systems, 2025, 13(1): 16.
[41]
JiangE, NieB, CaoZ, et al. Non-invasive blood pressure estimation using multi-domain pulse wave features and random forest regression[J]. Electronics, 2025, 14(7): 1409.
HeTing, WenJunming. FTmPSO algorithm with constraint-handling mechanism for constrained multi-modal optimization and performance verification using Wilcoxon rank-sumtest[J]. Information & Computer, 2025, 37(14): 175-177.