包头市细颗粒物遥感监测混合线性模型  

Mixed Linear Model of Remote Sensing Monitoring for Fine Particulate Matter (PM_(2.5)) in Baotou City

作　　者：同丽嘎[1] 李雪铭[2] 张靖[2,3] TONG Liga;LI Xueming;ZHANG Jing(College of Resources and Environment,Baotou Teachers’ College,Baotou,Innger Mongolia 014030,China;School of Urban and Environmental Sciences,Liaoning Normal University,Dalian,Liaoning 116029,China;College of environment and resources,Dalian Minzu University,Dalian,Liaoning 116600,China)

机构地区：[1]包头师范学院资源与环境学院,内蒙古包头014030 [2]辽宁师范大学城市与环境学院,辽宁大连116029 [3]大连民族大学环境与资源学院,辽宁大连116600

出　　处：《遥感信息》2019年第2期36-41,共6页Remote Sensing Information

基　　金：国家自然科学基金(41761036);国家自然科学基金青年科学基金(31500384)

摘　　要：以PM_(2.5)为主的颗粒物污染是影响我国城市空气质量的首要因素之一。针对城市尺度,高时间分辨率的MODIS气溶胶产品空间分辨率不足,且气溶胶光学厚度(AOD)与PM_(2.5)回归精度受到季节因素的影响而精度不足的问题,以Landsat-8OLI为数据源,选取季节因子为影响AOD变化的随机效应,以风速、地表温度、AOD等为固定效应,提出AOD与实测PM_(2.5)关系的混合线性模型。结果表明,与线性模型相比,该模型有较好的回归精度;包头市PM_(2.5)浓度时空差别明显,PM_(2.5)浓度冬季浓度水平最高,春季最低;从空间分布来看昆都仑区污染较高,青山区北部次之,东河区和九原区最低。PM 2.5 pollution is one of the most important factors affecting urban air quality in China.Because the spatial resolution of MODIS aerosol products with high temporal resolution is insufficient,and the shortage of monitoring and low accuracy of regression of “aerosol optical depth (AOD)-PM 2.5” relationship,Landsat-8 OLI imagery is selected,and the seasonal factors are chosen as random effects that affect the AOD changes though temporal scale.The wind speed,land surface temperature,AOD as the fixed effect,the mixed linear model of the relationship between AOD and PM 2.5 is established.The results show that:the model has good regression accuracy compared with the linear model;it is obviously that the spatial and temporal variation of PM 2.5 concentration in Baotou city.The PM 2.5 concentration is the highest in winter and lowest in spring.From the spatial distribution,higher pollution is in Kundulun district,followed by the northern part of Qingshan district,and the lowest pollution is in Donghe district and Jiuyuan district.

关 键 词：混合线性模型 PM2.5 深蓝算法 AOD 包头市 

分 类 号：K903[历史地理—人文地理学]