基于数据场模型的南京市气溶胶光学厚度反演

doi:10.6046/zrzyyg.2023187

摘要
图/表
参考文献
相关文章
Metrics

全文: PDF(4470 KB)

HTML
输出: BibTeX | EndNote (RIS)

摘要

高亮度地表上空的气溶胶光学厚度(aerosol optical depth, AOD)反演对于城市大气环境监测和污染岛研究等具有重要意义。该文综合考虑像元的光谱特性和空间特性,提出了一种基于数据场模型(data field method,DFM)的AOD反演方法。该方法基于Landsat OLI数据,开展了南京市上空AOD的反演研究,并分析不同空间结构、数据场强度和下垫面特征对DFM方法的影响,最后与传统的深蓝算法和MOD04产品等进行了对比分析。结果表明: DFM反演结果与CE318太阳光度计观测数据的相关系数为0.936,均方根误差、平均绝对误差和相对平均偏差分别为0.151,0.120和1.139,平均相对误差和误差比分别为22.7%和72.7%,与地面实测数据具有较高的一致性。地表数据场强度高于12的区域,DFM算法具有较好的反演效果,且在夏春季反演效果优于秋冬季,在乡村、工业区等下垫面变化剧烈区域具有更好的反演效果。

	服务

	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	缪晨阳
	陈健
	马本浩

关键词 ：气溶胶光学厚度, 数据场, 6S模型, 结构函数

Abstract：

The inversion of aerosol optical depth (AOD) above the bright surface holds great significance for monitoring urban atmospheric environment and investigating urban pollution islands. This study proposed an AOD inversion method using the data field method (DFM) by incorporating both spectral and spatial characteristics of pixels. Using the Land OLI data, this study conducted AOD inversion over Nanjing and analyzed the impacts of different spatial structures, data field intensity, and underlying surface features on the DFM. This method was then compared with the traditional Deep Blue algorithm and MOD04 products through a detailed comparative analysis. The results indicate that the DFM inversion results had a correlation coefficient of 0.936 with observations obtained using aCE318 sun photometer, a root mean square error of 0.151, a mean absolute error of 0.120, an average relative error of 22.7%, a relative mean deviation of 1.139, and an error ratio of 72.7%, demonstrating high consistency with ground-based measurements. In areas with surface data field intensity exceeding 12, the DFM algorithm exhibited superior inversion performance and proved more effective in the spring and summer than in autumn and winter. Particularly, this algorithm achieved enhanced inversion results in rural and industrial areas characterized by rapid changes in their underlying surfaces.

Key words： aerosol optical depth data field method 6S model structure function method

收稿日期: 2023-06-28 出版日期: 2024-12-23

ZTFLH:

TP79

基金资助:国家高分遥感卫星数据与产品共享交换服务平台项目“长江大保护高分综合应用”(30-Y60B01-9003-22/23)

通讯作者: 陈健(1978-),男,博士,副教授,研究方向为大气环境遥感、定量遥感。Email: chjnjnu@163.com。

作者简介: 缪晨阳(1999-),男,硕士研究生,研究方向为大气遥感。Email: 1025299135@qq.com。

引用本文:

缪晨阳, 陈健, 马本浩. 基于数据场模型的南京市气溶胶光学厚度反演[J]. 自然资源遥感, 2024, 36(4): 295-303.
MIAO Chenyang, CHEN Jian, MA Benhao. Inversion of aerosol optical depth in Nanjing City based on data field method. Remote Sensing for Natural Resources, 2024, 36(4): 295-303.

链接本文:

https://www.gtzyyg.com/CN/10.6046/zrzyyg.2023187 或 https://www.gtzyyg.com/CN/Y2024/V36/I4/295

Tab.1 Landsat8 OLI数据列表

Fig.1 气溶胶光学厚度反演流程

Fig.2 气溶胶光学厚度空间分布影像

Tab.2 AOD反演结果对比分析表

Tab.3 AOD反演结果的不确定性统计表

理论	函数形式	属性相关性	空间权重
势函数	$\sum_{i=1}^{m} \sum_{j=1}^{n}\left\|\rho_{x}-\rho_{i, j}\right\| \cdot \sqrt{-\left(\frac{\left\\|x-x_{i, j}\right\\|}{\sigma}\right)^{2}}$	$\| ρ x - ρ i, j$ \|	$\sqrt{-\left(\frac{\left\\|x-x_{i, j}\right\\|}{\sigma}\right)^{2}}$
变异函数	$1 2 N (h) ∑ i = 1 N (h) [ρ (x i) - ρ (x i + h)] 2$	$[ρ (x i) - ρ (x i + h)] 2$	$1 2 N (h)$

Tab.4 势函数与变异函数对比^[11]

Fig.3 2017年7月秋季地表反射率、结构函数与能量场归一化场强分布直方图

Fig.4 影像地表空间特征

Fig.5 DFM,SFM,DB AOD反演结果与EE之比

Fig.6 300 m分辨率地表反射率数据场强度统计图

Fig.7 不同地表下垫面数据场强度统计图

[1]	Boiyo R, Kumar K R, Zhao T L. Spatial variations and trends in AOD climatology over East Africa during 2002-2016:A comparative study using three satellite data sets[J]. International Journal of Climatology, 2018, 38:e1221-e1240.
[2]	丁莹, 冯徽徽, 邹滨, 等. 长株潭城市群气溶胶时空分布与传输规律[J]. 中国环境科学, 2020, 40(5):1906-1914.
	Ding Y, Feng H H, Zou B, et al. Spatial-temporal distribution and transport characteristic of aerosol in Changsha-Zhuzhou-Xiangtan urban agglomeration[J]. China Environmental Science, 2020, 40(5):1906-1914.
[3]	Li W, Shao L. Transmission electron microscopy study of aerosol particles from the brown hazes in Northern China[J]. Journal of Geophysical Research:Atmospheres, 2009, 114(D9):D09302.
[4]	Kaufman Y J, Sendra C. Algorithm for automatic atmospheric corrections to visible and near-IR satellite imagery[J]. International Journal of Remote Sensing, 1988, 9(8):1357-1381.
[5]	Hsu N C, Jeong M J, Bettenhausen C, et al. Enhanced deep blue aerosol retrieral algorithm: The second generation[J]. Journal of Geo physical Resrarch:Atmospheres, 2013, 118(16):9296-9315.
[6]	赵金涛. 基于Landsat-8 OLI影像的气溶胶光学厚度反演研究[D]. 银川: 宁夏大学, 2021.
	Zhao J T. Research on the inversion of aerosol optical thickness based on Landsat-8 OLI image[D]. Yinchuan: Ningxia University, 2021.
[7]	朱琳, 孙林, 杨磊, 等. 结构函数法反演气溶胶光学厚度中像元的间隔设置[J]. 遥感学报, 2016, 20(4):528-539.
	Zhu L, Sun L, Yang L, et al. Pixel distance settings in aerosol optical depth retrieval through the structure function method[J]. Journal of Remote Sensing, 2016, 20(4):528-539.
[8]	Liu G R, Chen A J, Lin T H, et al. Applying SPOT data to estimate the aerosol optical depth and air quality[J]. Environmental Modelling and Software, 2002, 17(1):3-9.
[9]	Zhou G, Zhang Y, Ma Z, et al. A geostatistics-based method to determine the pixel distance in a structure function model for aerosol optical depth inversion[J]. Atmosphere, 2017, 8(1):6.
[10]	Xia Z L, Li H, Chen Y H, et al. Identify and delimitate urban hotspot areas using a network-based spatiotemporal field clustering method[J]. ISPRS International Journal of Geo-Information, 2019, 8(8):344.
[11]	冯霞. 基于变异函数和数据场的高分遥感影像空间结构特征描述[D]. 武汉: 武汉大学, 2014.
	Feng X. Spatial structural feature description for high resolution remote sensing images based on variogram and data field[D]. Wuhan: Wuhan University, 2014.
[12]	刘楠. 基于光谱空间数据场模型的遥感影像边缘提取[D]. 武汉: 武汉大学, 2005.
	Liu N. Edge detection of remote sensing image based on data field model in spectral space[D]. Wuhan: Wuhan University, 2005.
[13]	Wang S, Wang D, Li C, et al. Clustering by fast search and find of density peaks with data field[J]. Chinese Journal of Electronics, 2016, 25(3):397-402.
[14]	Tao S Y. A review of recent research on the East Asian summer monsoon in China[J]. Monsoon Meteorology, 1987:60-92.
[15]	《中国城市统计年鉴—2018》编委会与编辑部. 陈小龙总编,中国城市统计年鉴,中国统计出版社, 2018,4-5.
	Editorial Board and Editorial Department of China Urban Statistical Yearbook-2018. Chen Xiaolong as Editor-in-Chief,China Urban Statistical Yearbook published by China Statistics Press, 2018,4-5.
[16]	Gerace A, Montanaro M. Derivation and validation of the stray light correction algorithm for the thermal infrared sensor onboard Landsat8[J]. Remote Sensing of Environment, 2017, 191:246-257.
[17]	Smirnov A, Holben B N, Eck T F, et al. Cloud-screening and quality control algorithms for the AERONET database[J]. Remote Sensing of Environment, 2000, 73(3):337-349.
[18]	朱佳恒. 复杂地形与下垫面条件下的ERA5再分析地表温度降尺度研究[D]. 南京: 南京信息工程大学, 2022.
	Zhu J H. Reanalysis of ERA5 under complex terrain and underlying surface conditions; Downscaling of surface temperature[D]. Nanjing: Nanjing University of Information Science and Technology, 2022.
[19]	陶建斌, 舒宁, 沈照庆. 基于数据场聚类的遥感影像分类方法研究[J]. 国土资源遥感, 2008, 20(3):20-23,26.doi:10.6046/gtzyyg.2008.03.05.
	Tao J B, Shu N, Shen Z Q. A study of the method for clssification of remote sensing images based on data field cluster[J]. Remote Sensing for Land & Resources, 2008, 20(3):20-23,26.doi:10.6046/gtzyyg.2008.03.05.
[20]	徐涵秋. 利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J]. 遥感学报, 2005, 9(5):589-595.
	Xu H Q. A study on information extraction of water body with the modified normalized difference water index (MNDWI)[J]. Journal of Remote Sensing, 2005, 9(5):589-595.
[21]	李兆麟, 左洪超, 罗雯. 利用6S模型的自定义气溶胶类型反演北京地区气溶胶光学厚度[J]. 甘肃科技, 2019, 35(6):33-39.
	Li Z L, Zuo H C, Luo W. Aerosol optical depth retrieval with a custom aerosol type using 6S models over Beijing area[J]. Gansu Science and Technology, 2019, 35(6):33-39.
[22]	程晨, 陈健, 李鑫慧. 基于TM图像的南京市气溶胶光学厚度反演[J]. 国土资源遥感, 2013, 25(3):90-96.doi:10.6046/gtzyyg.2013.03.16.
	Cheng C, Chen J, Li X H. Retrieving aerosol optical depth over Nanjing City based on TM image[J]. Remote Sensing for Land and Resources, 2013, 25(3):90-96.doi:10.6046/gtzyyg.2013.03.16.
[23]	Layeb A. Novel feature selection algorithms based on crowding distance and Pearson correlation coefficient[J]. International Journal of Intelligent Systems and Applications, 2023, 15(2):37-42.
[24]	Hsu N C, Tsay S C, King M D, et al. Aerosol properties over bright-reflecting source regions[J]. IEEE Transactions on Geoscience and Remote Sensing, 2004, 42(3):557-569.
[25]	田信鹏, 孙林, 刘强, 等. 北京地区Landsat8 OLI高空间分辨率气溶胶光学厚度反演[J]. 遥感学报, 2018, 22(1):51-63.
	Tian X P, Sun L, Liu Q, et al. Retrieval of high-resolution aerosol optical depth using Landsat8 OLI data over Beijing[J]. Journal of Remote Sensing, 2018, 22(1):51-63.

[1]	孟清, 白红英, 赵婷, 郭少壮, 齐贵增. 秦岭山地对气溶胶的生态屏障效应[J]. 国土资源遥感, 2021, 33(1): 240-248.
[2]	申振宇, 高小红, 汤敏. 高海拔复杂地形区SPOT6图像大气校正方法对比及精度验证[J]. 国土资源遥感, 2020, 32(1): 81-89.
[3]	施益强, 邓秋琴, 吴君, 王坚. 厦门市MODIS气溶胶光学厚度与空气质量指数的回归分析[J]. 国土资源遥感, 2020, 32(1): 106-114.
[4]	李恺霖, 张春桂, 廖廓, 李丽纯, 王宏. 福建省空气清新度卫星遥感监测[J]. 国土资源遥感, 2019, 31(4): 151-158.
[5]	张明明, 程东兵, 齐建华, 胡建春, 罗晶. 基于卫星遥感的池州市气溶胶光学厚度时空分布[J]. 国土资源遥感, 2017, 29(4): 147-155.
[6]	岳梦雪, 秦昆, 张恩兵, 张晔, 曾诚. 基于数据场和密度聚类的高分辨率影像居民区提取[J]. 国土资源遥感, 2017, 29(3): 92-97.
[7]	孔金玲, 杨晶, 孙晓明, 杨姝, 柳富田, 杜东. 多光谱遥感影像大气校正与悬沙浓度反演——以曹妃甸近岸海域为例[J]. 国土资源遥感, 2016, 28(3): 130-137.
[8]	范东福, 杨书运, 吴必文, 范东勤, 王俊, 江波, 陈晓龙. 基于AOD数据的秸秆焚烧监测[J]. 国土资源遥感, 2015, 27(2): 133-138.
[9]	佃袁勇, 方圣辉, 徐永荣. 一种协同反演气溶胶与水汽含量的高光谱图像大气校正算法[J]. 国土资源遥感, 2015, 27(2): 22-28.
[10]	程晨, 陈健, 李鑫慧. 基于TM图像的南京市气溶胶光学厚度反演[J]. 国土资源遥感, 2013, 25(3): 90-96.
[11]	夏双, 阮仁宗, 张月, 颜梅春. 气溶胶光学厚度对蓝藻水华信息提取的影响[J]. 国土资源遥感, 2013, 25(1): 33-38.
[12]	马晓红, 余涛, 高海亮, 陈兴峰, 谢玉娟, 韩杰. 内蒙古辐射校正场特性评价与应用潜力分析[J]. 国土资源遥感, 2011, 23(4): 31-36.
[13]	孙源, 顾行发, 余涛, 高海亮. 环境星CCD数据大气校正研究[J]. 国土资源遥感, 2010, 22(4): 6-9.
[14]	张川, 刘少峰, 刘衍宏, 裴晓寅. 基于辐射传输模型的Hyperion光谱重建应用评价[J]. 国土资源遥感, 2010, 22(2): 30-34.
[15]	陶建斌, 舒宁, 沈照庆. 基于数据场聚类的遥感影像分类方法研究[J]. 国土资源遥感, 2008, 20(3): 20-23.

Viewed

Full text

Abstract

Cited

Shared

Discussed