Estimation of the Arctic Aerosol Optical Depth Based on the Synergistic Integration of Multi-Source Data
-
摘要: 北极地区是全球气候变化的敏感区，其突出的增暖放大效应（AA）与气溶胶辐射强迫密切相关。气溶胶光学厚度（AOD）作为表征大气气溶胶消光特性的关键指标，对解析气溶胶在环境和气候系统中的作用具有关键意义。卫星遥感技术已成为获取全球或区域尺度AOD数据的重要手段，但受观测原理限制及北极复杂地表环境影响，现有卫星产品存在明显数据缺失。贝叶斯最大熵法（BME）是AOD数据融合的常用方法，但传统BME采用最小二乘法构建协方差时，难以有效处理高维参数空间的复杂性与非平稳性。本文基于中分辨率成像光谱仪（MODIS）与多角度成像光谱辐射计（MISR）的AOD产品，通过引入具备全局搜索能力的粒子群优化算法（PSO）改进协方差建模过程，构建PSO-BME融合算法以提升数据融合的稳定性与精度。研究结果表明：PSO-BME能够有效融合MODIS和MISR的AOD数据并实现缺失数据填补。融合AOD在双源覆盖区RMSE降至0.055，EE达78%，MAE为0.04，相关系数0.7，且在无观测区仍保持可接受精度；年均空间覆盖率从MODIS的15.45%和MISR的1.45%提升至32.7%。时空分布分析显示，融合产品空间连续性明显改善，能更真实反映AOD整体变化特征；时空演变规律揭示北极气溶胶分布受本地气象条件与中低纬污染跨境传输的双重影响。Abstract: The Arctic is a climate-sensitive region where Arctic Amplification is influenced by aerosol radiative forcing. Aerosol Optical Depth (AOD) as key parameter characterizing the extinction properties of atmospheric aerosols, plays a critical role in understanding the influence of aerosols on environmental and climate systems. However, single-satellite AOD products exhibit large uncertainties and data gaps in the Arctic due to sensor limitations and complex surface conditions. The Bayesian Maximum Entropy (BME) method is commonly used for AOD data fusion, yet the traditional BME approach, which employs least squares to model covariance, struggles to effectively handle the complexity and non-stationarity of high-dimensional parameter spaces. Based on AOD products from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Multi-angle Imaging Spectro Radiometer (MISR), this study introduces a Particle Swarm Optimization (PSO) algorithm with global search capability to improve the covariance modeling process, resulting in a PSO-BME fusion algorithm that enhances the stability and accuracy of data integration. The results demonstrate that the PSO-BME method effectively integrates MODIS and MISR AOD data and successfully fills data gaps. In regions covered by both sources, the fused AOD achieves an RMSE of 0.055, an EE of 78%, an MAE of 0.04, and a correlation coefficient of 0.7, while maintaining acceptable accuracy in unobserved areas. The annual spatial coverage increased from 15.45% (MODIS) and 1.45% (MISR) to 32.7%. Spatiotemporal distribution analysis shows that the fusion product significantly improves spatial continuity and more accurately reflects overall AOD variations. Furthermore, the spatiotemporal evolution patterns reveal that aerosol distribution in the Arctic is influenced by both local meteorological conditions and cross-border transport of pollutants from mid- and low-latitudes.
-
Key words:
- Arctic /
- Aerosol Optical Depth /
- Bayesian Maximum Entropy /
- Data Fusion
-
图 1 北极地区21个AERONET测站分布（灰色：观测300–
1000 天；深蓝色：>1000 天），并叠加2022年8月1日Terra/MODIS（浅蓝）、Aqua/MODIS（浅绿）与MISR（棕）的日尺度观测覆盖。Fig. 1 Distribution of 21 AERONET stations in the Arctic. Sites are colored by data availability: gray (300–
1000 days) and dark blue (>1000 days). Daily coverage from Terra/MODIS (light blue), Aqua/MODIS (light green), and MISR (brown) on 1 August 2022 is also shown.
图 2 基于PSO-BME方法的多源卫星AOD产品融合框架的示意图
Fig. 2 Conceptual framework illustrating the fusion of multi-source satellite AOD products using the PSO-BME approach.
图 3 2019年6月MODIS与MISR AOD去除时空趋势前后对比图。a原MODIS AOD直方图；b去时空趋势后MODIS AOD直方图；c原MISR AOD直方图；d去时空趋势后MISR AOD的直方图。
Fig. 3 Comparison before and after removing spatiotemporal trends of 2019.6. a) Original MODIS AOD histogram; b) MODIS AOD histogram after spatiotemporal trend removal; c) Original MISR AOD histogram; d) MISR AOD histogram after spatiotemporal trend removal.
图 7 MODIS、MISR AOD产品和融合AOD产品2010—2020年季节覆盖率对比
Fig. 7 Comparison of seasonal coverage between MODIS, MISR, and Fusion AOD products over the period 2010—2020.
图 4 2010—2020年北极地区MODIS与MISR AOD产品与AERONET观测值的散点对比（分季节），黑色实线为y=x参考线，两侧对称包络线界定EE范围。a)总体MODIS；b)春季MODIS；c)夏季MODIS; d)秋季MODIS; e)总体MISR; f)春季MISR; g)夏季MISR; h)秋季MISR。
Fig. 4 Seasonal scatterplot comparisons between MODIS and MISR AOD products and AERONET observations over the Arctic from 2010 to 2020. The solid black line denotes the 1:1 reference line, while the symmetric envelope lines indicate the Expected Error range. a) All MODIS; b) Spring MODIS; c) Summer MODIS; d) Autumn MODIS; e) All MISR; f) Spring MISR; g) Summer MISR; h) Autumn MISR
图 5 以AERONET AOD数据为基准，验证MODIS、MISR和融合后的AOD数据。a)原始MODIS数据; b)原始MISR数据; c)整个北极的融合AOD; d) MODIS、MISR同时存在的融合AOD; e)MODIS存在MISR缺失的融合AOD; f) MISR存在MODIS缺失的融合AOD; g)MODIS、MISR同时缺失的融合AOD。
Fig. 5 Validation of MODIS, MISR, and fusion AOD products against AERONET AOD benchmarks; a) Original MODIS; b) Original MISR; c) Entire Arctic region; d) MODIS & MISR concurrent coverage area; e) MODIS available, MISR missing region; f) MISR available, MODIS missing region; g) Both MODIS & MISR missing area.
图 6 不同融合输入参数下 PSO-BME 与 LS-BME 在北极区域的 AOD 融合精度对比，实线表示PSO-BME结果，虚线表示LS-BME结果. a) RMSE; b) R; c) RMB; d) MAE; e) N; f) Within EE
Fig. 6 Comparison of AOD fusion accuracy and stability between PSO-BME and LS-BME over the Arctic under different fusion input parameter configurations. Solid lines represent the PSO-BME results, and dashed lines represent the LS-BME results a) RMSE; b) R; c) RMB; d) MAE; e) N; f) Within EE
图 8 2010-2020年AOD数据周平均覆盖率随时间的演变特征
Fig. 8 The evolution characteristics of the weekly average coverage rate of AOD data from 2010 to 2020 over time.
图 9 MODIS和MISR AOD以及融合AOD产品2010—2020年季节均值的空间分布图
Fig. 9 Spatial patterns of seasonal mean AOD derived from MODIS, MISR, and Fusion products over the period 2010—2020.
图 10 MODIS和MISR AOD以及融合AOD产品2010—2020年各季节AOD线性趋势的空间模式，黑色圆点代表通过90%置信水平的显著性检验区域。
Fig. 10 Spatial distribution of seasonal linear trends in AOD from MODIS, MISR, and fusion products over the period 2010—2020. Black dots denote areas where the trends are statistically significant at the 90% confidence level.
表 1 原始及融合AOD的精度分析结果
Tab. 1 The accuracy analysis results of the original and fusion AOD.
AOD Data RMSE MAE RMB R N EE MODIS 0.073 0.05 1.76 0.69 1333 69.3% MISR 0.042 0.02 1.17 0.70 173 95.4% 融合AOD（整个北极） 0.074 0.05 1.80 0.63 2313 67.3% 融合AOD（MODIS、MISR同时存在） 0.055 0.04 1.69 0.70 165 78.0% 融合AOD（MODIS存在MISR缺失） 0.072 0.05 1.70 0.72 1168 70.4% 融合AOD（MISR存在MODIS缺失） 0.030 0.02 1.17 0.91 8 100.0% 融合AOD（MODIS、MISR同时缺失） 0.076 0.05 1.84 0.57 972 64.7% 
下载: 导出CSV
表 2 不同卫星数据覆盖组合下 PSO-BME 融合结果的精度验证统计
Tab. 2 Statistical accuracy assessment of PSO-BME fusion results under different satellite data coverage combinations.
不同实测比例区域 区域 RMSE MAE RMB R N EE% MODIS实测比例为0～20%，
MISR实测比例为0MODIS 0.047 0.04 1.80 0.58 59 84.7 整个北极 0.062 0.05 1.89 0.56 192 65.6 MODIS存在MISR缺失 0.047 0.04 1.80 0.58 59 84.7 MODIS、MISR同时缺失 0.068 0.05 1.92 0.53 133 63.1 MODIS实测比例为0～20%，
MISR实测比例为0～10%MODIS 0.083 0.06 1.96 0.56 375 63.2 MISR 0.040 0.02 1.27 0.69 89 97.8 整个北极 0.073 0.05 1.83 0.57 898 67.5 MODIS、MISR同时存在 0.087 0.08 2.02 0.46 82 43.5 MODIS存在MISR缺失 0.080 0.05 1.76 0.61 293 69.9 MODIS缺失MISR存在 0.032 0.02 1.20 0.96 7 100 MODIS、MISR同时缺失 0.066 0.05 1.74 0.59 532 69.9 MODIS实测比例为20～40%，
MISR实测比例为0～10%MODIS 0.072 0.05 1.63 0.75 710 70.6 MISR 0.050 0.02 1.10 0.78 56 92.9 整个北极 0.079 0.05 1.71 0.70 1027 66.7 MODIS、MISR同时存在 0.054 0.03 1.50 0.86 55 83.6 MODIS存在MISR缺失 0.073 0.05 1.65 0.74 653 69.5 MODIS、MISR同时缺失 0.094 0.06 1.87 0.54 316 58.2 MODIS实测比例为40～60%，
MISR实测比例为0～10%MODIS 0.069 0.05 1.67 0.74 899 71.0 MISR 0.044 0.02 1.08 0.75 84 94.0 整个北极 0.076 0.05 1.72 0.70 1224 67.7 MODIS、MISR同时存在 0.052 0.04 1.52 0.84 83 84.3 MODIS存在MISR缺失 0.071 0.05 1.68 0.74 815 69.7 MODIS、MISR同时缺失 0.085 0.06 1.88 0.54 324 58.6
下载: 导出CSV
-
[1] 余光明, 徐建中, 任贾文. 青藏高原大气气溶胶研究进展[J]. 冰川冻土, 2012, 34(3): 609−617.Yu Guangming, Xu Jianzhong, Ren Jiawen. The progress of aerosol research in the Tibetan Plateau[J]. Journal of Glaciology and Geocryology, 2012, 34(3): 609−617. [2] Rantanen M, Karpechko A Y, Lipponen A, et al. The Arctic has warmed nearly four times faster than the globe since 1979[J]. Communications Earth & Environment, 2022, 3(1): 168. doi: 10.1038/s43247-022-00498-3 [3] Perovich D K, Polashenski C. Albedo evolution of seasonal Arctic sea ice[J]. Geophysical Research Letters, 2012, 39(8): L08501. doi: 10.1029/2012gl051432 [4] Boisvert L N, Petty A A, Stroeve J C. The impact of the extreme winter 2015/16 arctic cyclone on the Barents–Kara Seas[J]. Monthly Weather Review, 2016, 144(11): 4279−4287. doi: 10.1175/MWR-D-16-0234.1 [5] Nummelin A, Li C, Hezel P J. Connecting ocean heat transport changes from the midlatitudes to the Arctic Ocean[J]. Geophysical Research Letters, 2017, 44(4): 1899−1908. doi: 10.1002/2016GL071333 [6] Middlemas E A, Kay J E, Medeiros B M, et al. Quantifying the influence of cloud radiative feedbacks on arctic surface warming using cloud locking in an earth system model[J]. Geophysical Research Letters, 2020, 47(15): e2020GL089207. doi: 10.1029/2020GL089207 [7] Pithan F, Mauritsen T. Arctic amplification dominated by temperature feedbacks in contemporary climate models[J]. Nature Geoscience, 2014, 7(3): 181−184. doi: 10.1038/ngeo2071 [8] Park J Y, Kug J S, Bader J, et al. Amplified Arctic warming by phytoplankton under greenhouse warming[J]. Proceedings of the National Academy of Sciences of the United States of America, 2015, 112(19): 5921−5926. [9] Campen H I, Arévalo-Martínez D L, Artioli Y, et al. The role of a changing Arctic Ocean and climate for the biogeochemical cycling of dimethyl sulphide and carbon monoxide[J]. Ambio, 2022, 51(2): 411−422. doi: 10.1007/s13280-021-01612-z [10] Shindell D, Faluvegi G. Climate response to regional radiative forcing during the twentieth century[J]. Nature Geoscience, 2009, 2(4): 294−300. doi: 10.1038/ngeo473 [11] Bond T C, Doherty S J, Fahey D W, et al. Bounding the role of black carbon in the climate system: a scientific assessment[J]. Journal of Geophysical Research: Atmospheres, 2013, 118(11): 5380−5552. doi: 10.1002/jgrd.50171 [12] Twomey S. The influence of pollution on the shortwave albedo of clouds[J]. Journal of the Atmospheric Sciences, 1977, 34(7): 1149−1152. doi: 10.1175/1520-0469(1977)034<1149:tiopot>2.0.co;2 [13] Kaufman Y J, Tanré D, Boucher O. A satellite view of aerosols in the climate system[J]. Nature, 2002, 419(6903): 215−223. doi: 10.1038/nature01091 [14] Popp T, De Leeuw G, Bingen C, et al. Development, production and evaluation of aerosol climate data records from European satellite observations (Aerosol_cci)[J]. Remote Sensing, 2016, 8(5): 421. doi: 10.3390/rs8050421 [15] 唐健雄. 基于GF-5 DPC数据的PM2.5估算方法研究[D]. 廊坊: 北华航天工业学院, 2023.Tang Jianxiong. Research on PM2.5 estimation method based on GF-5 DPC data[D]. Langfang: North China Institute of Aerospace Engineering, 2023. (查阅网上资料, 未找到对应的英文翻译, 请确认) [16] 杨军, 咸迪, 唐世浩. 风云系列气象卫星最新进展及应用[J]. 卫星应用, 2018(11): 8−14.Yang Jun, Xian Di, Tang Shihao. Latest developments and applications of the Fengyun series meteorological satellites[J]. Satellite Application, 2018(11): 8−14. (查阅网上资料, 未找到对应的英文翻译, 请确认) [17] 顾俊杰, 戴一枫, 金林林. 卫星遥感气溶胶光学厚度的相关研究综述[J]. 科技资讯, 2017, 15(19): 219−220, 222.Gu Junjie, Dai Yifeng, Jin Linlin. A review on the research on satellite remote sensing aerosol optical thickness[J]. Science & Technology Information, 2017, 15(19): 219−220, 222. (查阅网上资料, 未找到对应的英文翻译, 请确认) [18] 晏阳天, 白学志. 北极气溶胶时空变化特征及其对向下短波辐射的影响[J/OL]. 海洋科学, 1−19. https://link.cnki.net/urlid/37.1151.P.20250513.1627.002, 2025-08-23.Yan Yangtian, Bai Xuezhi. Temporal and spatial variation characteristics of Arctic aerosol and its influence on downward short-wave radiation[J/OL]. Marine Sciences, 1−19. https://link.cnki.net/urlid/37.1151.P.20250513.1627.002, 2025-08-23. [19] Swain B, Vountas M, Deroubaix A, et al. Retrieval of aerosol optical depth over the Arctic cryosphere during spring and summer using satellite observations[J]. Atmospheric Measurement Techniques, 2024, 17(1): 359−375. doi: 10.5194/amt-17-359-2024 [20] 杨艳丽, 常亮. MODIS北极气溶胶光学厚度的精度验证与分布特征研究[J]. 极地研究, 2022, 34(1): 62−71. doi: 10.13679/j.jdyj.20210028Yang Yanli, Chang Liang. Accuracy validation and distribution characteristics of Arctic MODIS aerosol optical depth[J]. Chinese Journal of Polar Research, 2022, 34(1): 62−71. doi: 10.13679/j.jdyj.20210028 [21] Tomasi C, Kokhanovsky A A, Lupi A, et al. Aerosol remote sensing in polar regions[J]. Earth-Science Reviews, 2015, 140: 108−157. doi: 10.1016/j.earscirev.2014.11.001 [22] Schmale J, Sharma S, Decesari S, et al. Pan-Arctic seasonal cycles and long-term trends of aerosol properties from 10 observatories[J]. Atmospheric Chemistry and Physics, 2022, 22(5): 3067−3096. doi: 10.5194/acp-22-3067-2022 [23] Li Deren. Remote sensing image fusion: a practical guide[J]. Geo-Spatial Information Science, 2017, 20(1): 56. [24] 贾永红, 李德仁. 多源遥感影像像素级融合分类与决策级分类融合法的研究[J]. 武汉大学学报(信息科学版), 2001, 26(5): 430−434.Jia Yonghong, Li Deren. An approach of classification based on pixel level and decision level fusion of Multi-source images in remote sensing[J]. Geomatics and Information Science of Wuhan University, 2001, 26(5): 430−434. [25] Nguyen H, Cressie N, Braverman A. Spatial statistical data fusion for remote sensing applications[J]. Journal of the American Statistical Association, 2012, 107(499): 1004−1018. doi: 10.1080/01621459.2012.694717 [26] Yang Jing, Hu Maogui. Filling the missing data gaps of daily MODIS AOD using spatiotemporal interpolation[J]. Science of the Total Environment, 2018, 633: 677−683. doi: 10.1016/j.scitotenv.2018.03.202 [27] Christakos G. A Bayesian/maximum-entropy view to the spatial estimation problem[J]. Mathematical Geology, 1990, 22(7): 763−777. doi: 10.1007/BF00890661 [28] He Junyu, Christakos G, Wu Jiaping, et al. Spatiotemporal BME characterization and mapping of sea surface chlorophyll in Chesapeake Bay (USA) using auxiliary sea surface temperature data[J]. Science of the Total Environment, 2021, 794: 148670. doi: 10.1016/j.scitotenv.2021.148670 [29] Tang Qingxin, Bo Yanchen, Zhu Yuxin. Spatiotemporal fusion of multiple‐satellite aerosol optical depth (AOD) products using Bayesian maximum entropy method[J]. Journal of Geophysical Research: Atmospheres, 2016, 121(8): 4034−4048. doi: 10.1002/2015JD024571 [30] Yu H L, Wu Yuzhang, Cheung S Y. A data assimilation approach for groundwater parameter estimation under Bayesian maximum entropy framework[J]. Stochastic Environmental Research and Risk Assessment, 2020, 34(5): 709−721. doi: 10.1007/s00477-020-01795-z [31] Chen Li, Liang Shuang, Li Xiaoli, et al. A hybrid approach to estimating long-term and short-term exposure levels of ozone at the national scale in China using land use regression and Bayesian maximum entropy[J]. Science of the Total Environment, 2021, 752: 141780. doi: 10.1016/j.scitotenv.2020.141780 [32] Li Aihua, Bo Yanchen, Zhu Yuxin, et al. Blending multi-resolution satellite sea surface temperature (SST) products using Bayesian maximum entropy method[J]. Remote Sensing of Environment, 2013, 135: 52−63. doi: 10.1016/j.rse.2013.03.021 [33] Xia Xinghui, Zhao Bin, Zhang Tianhao, et al. Satellite-derived aerosol optical depth fusion combining active and passive remote sensing based on Bayesian maximum entropy[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 4100313. doi: 10.1109/tgrs.2021.3051799 [34] Zhu Hao, Cheng Tianhai, Li Xiaoyang, et al. Fusion of multisource satellite AOD products via Bayesian maximum entropy with explicit a priori knowledge[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 4103112. doi: 10.1109/tgrs.2023.3285943/mm1 [35] Aldabash M, Bektas Balcik F, Glantz P. Validation of MODIS C6.1 and MERRA-2 AOD using AERONET observations: a comparative study over turkey[J]. Atmosphere, 2020, 11(9): 905. doi: 10.3390/atmos11090905 [36] Tao Minghui, Wang Jun, Li Rong, et al. Characterization of aerosol type over East Asia by 4.4 km MISR product: first insight and general performance[J]. Journal of Geophysical Research: Atmospheres, 2020, 125(13): e2019JD031909. doi: 10.1029/2019JD031909 [37] Abdou W A, Diner D J, Martonchik J V, et al. Comparison of coincident Multiangle Imaging Spectroradiometer and Moderate Resolution Imaging Spectroradiometer aerosol optical depths over land and ocean scenes containing Aerosol Robotic Network sites[J]. Journal of Geophysical Research: Atmospheres, 2005, 110(D10): 2004JD004693. doi: 10.1029/2004JD004693 [38] Kahn R A, Gaitley B J, Garay M J, et al. Multiangle Imaging SpectroRadiometer global aerosol product assessment by comparison with the Aerosol Robotic Network[J]. Journal of Geophysical Research: Atmospheres, 2010, 115(D23): D23209. doi: 10.1029/2010jd014601 [39] Li Long, Shi Runhe, Zhang Lu, et al. The data fusion of aerosol optical thickness using universal kriging and stepwise regression in East China[C]//Proceedings of SPIE 9221, Remote Sensing and Modeling of Ecosystems for Sustainability XI. San Diego: SPIE, 2014: 922112. [40] Levy R C, Mattoo S, Munchak L A, et al. The Collection 6 MODIS aerosol products over land and ocean[J]. Atmospheric Measurement Techniques, 2013, 6(11): 2989−3034. doi: 10.5194/amt-6-2989-2013 [41] Chung M. Gaussian Kernel Smoothing[M]. University of Wisconsin-Madison, 2020. (查阅网上资料, 未找到本条文献信息, 请确认并补充出版地信息) [42] Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning[M]. New York: Springer, 2001. [43] Kennedy J, Eberhart R. Particle swarm optimization[C]//Proceedings of International Conference on Neural Networks. Perth: IEEE, 1995: 1942-1948. [44] Tyagi N, Bhargava D, Ahlawat A. Implementation of particle swarm optimization algorithm inspired by the social behaviour of birds[C]//Proceedings of the 2024 4th International Conference on Technological Advancements in Computational Sciences. Tashkent: IEEE, 2024: 750-754. [45] Karaboga D. An idea based on honey bee swarm for numerical optimization[R]. Technical Report - TR06, Kayseri: Erciyes University, 2005. [46] 翟盘茂, 任福民, 张强. 中国降水极值变化趋势检测[J]. 气象学报, 1999, 57(2): 81−89.Zhai Panmao, Ren Fumin, Zhang Qiang. Detection of trends in China's precipitation extremes[J]. Acta Meteorologica Sinica, 1999, 57(2): 81−89. [47] Remer L A, Tanré D, Kaufman Y J, et al. Validation of MODIS aerosol retrieval over ocean[J]. Geophysical Research Letters, 2002, 29(12): 1618. [48] Shaw G E. The arctic haze phenomenon[J]. Bulletin of the American Meteorological Society, 1995, 76(12): 2403−2414. [49] Damoah R, Spichtinger N, Forster C, et al. Around the world in 17 days – hemispheric-scale transport of forest fire smoke from Russia in May 2003[J]. Atmospheric Chemistry and Physics, 2004, 4(5): 1311−1321. doi: 10.5194/acpd-4-1449-2004 [50] Stohl A. Characteristics of atmospheric transport into the Arctic troposphere[J]. Journal of Geophysical Research: Atmospheres, 2006, 111(D11): D11306. doi: 10.1029/2005jd006888 -
计量
- 文章访问数: 22
- HTML全文浏览量: 14
- PDF下载量: 0
- 被引次数: 0
