邓悦,刘纪平,刘洋,徐胜华

• 1:中国测绘科学研究院
• 2:中南大学地理信息系

摘要(Abstract)：

针对地理加权回归(GWR)模型无法克服小样本数据下异常值影响的问题,该文利用贝叶斯地理加权回归(BGWR)模型对北京地区2016年10月1日至12月29日长达90d的PM2.5监测数据进行了浓度模拟。该方法通过加入贝叶斯先验信息,选取不同的平滑函数,在局部空间样本稀少的情况下,有效降低了异常值和"弱数据"对回归结果的影响,更加真实地反映了PM2.5浓度空间分布。实验结果表明,基于不同平滑函数的3种BGWR模型校正决定系数分别达到了0.799、0.801和0.867。平均比GWR模型提升了28%,比OLS模型提升了32%,证实了BGWR模型在模拟PM2.5浓度分布时具有更好的适用性。

关键词(KeyWords)： 贝叶斯处理;地理加权回归;贝叶斯地理加权回归;北京;PM2.5浓度模拟

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划资助项目(2016YFC0803108)

作者(Author): 邓悦,刘纪平,刘洋,徐胜华

DOI: 10.16251/j.cnki.1009-2307.2018.10.006

参考文献(References)：

• [1]POPEⅢC A,THUN M J,NAMBOODIRI M M,et al.Particulate air pollution as a predictor of mortality in a prospective study of U.S.adults[J].American Journal of Respiratory and Critical Care Medicine,1995,151(3Pt 1):669-674.
• [2]汉瑞英,陈健,王彬.利用LUR模型模拟杭州市PM2.5质量浓度空间分布[J].环境科学学报,2016,36(9):3379-3385.(HAN Ruiying,CHEN Jian,WANG Bin.Application of LUR models for simulating the spatial distribution of PM2.5concentration in Hangzhou,China[J].Acta Scientiae Circumstantiae,2016,36(9):3379-3385.)
• [3]许刚,焦利民,肖丰涛,等.土地利用回归模型模拟京津冀PM2.5浓度空间分布[J].干旱区资源与环境,2016,30(10):116-120.(XU Gang,JIAO Limin,XIAO Fengtao,et al.Applying land use regression model to estimate spatial distribution of PM2.5concentration in Beijing-Tianjin-Hebei region[J].Journal of Arid Land Resources and Environment,2016,30(10):116-120.)
• [4]吴健生,谢舞丹,李嘉诚.土地利用回归模型在大气污染时空分异研究中的应用[J].环境科学,2016,37(2):413-419.(WU Jiansheng,XIE Wudan,LI Jiacheng.Application of land-use regression models in spatialtemporal differentiation of air pollution[J].Environmental Science,2016,37(2):413-419.)
• [5]覃文忠,王建梅,刘妙龙.地理加权回归分析空间数据的空间非平稳性[J].辽宁师范大学学报(自然科学版),2005,28(4):476-479.(QIN Wenzhong,WANG Jianmei,LIU Miaolong,et al.Spatial nonstationarity of geographically weighted regression analysis of spatial data[J].Journal of Liaoning Normal University(Natural Science Edition),2005,28(4):476-479.)
• [6]BRUNSDON C,FOTHERINGHAM A S,CHARLTON M E.Geographically weighted regression:a method for exploring spatial nonstationarity[J].Geographical Analysis,1996,28(4):281-298.
• [7]LESAGE J P.A spatial econometric examination of China’s economic growth[J].Geographic Information Sciences,1999,5(2):143-153.
• [8]LESAGE J P.A family of geographically weighted regression models[M]∥Advances in spatial econometrics:methodology,tools and applications.Berlin,Germany:Springer,2004:241-264.
• [9]SUBEDI N.Local modeling of tree crown area using Bayesian geographically weighted regression[D].Syracuse,New York:State University of New York,2012.
• [10]崔长彬,姜石良,张正河.河北县域经济影响因素的空间差异分析:基于贝叶斯地理加权回归方法[J].经济地理,2012,32(2):39-45.(CUI Changbin,JIANG Shiliang,ZHANG Zhenghe.The analysis of spatial variability of influencing factors to county economy in Hebei:based on BGWR[J].Economic Geography,2012,32(2):39-45.)
• [11]丁刚.基于BGWR模型的区域创新能力建设中政府效能测评[J].技术经济,2013,32(4):20-26,47.(DING Gang.Evaluation of provincial government efficiency in regional innovation capability construction in China based on BGWR model[J].Technology Economics,2013,32(4):20-26,47.)
• [12]王翠云,胡学平,相旭东.PM2.5影响因素的主成分回归分析与预测[J].安庆师范学院学报(自然科学版),2015,21(4):24-27.(WANG Cuiyun,HU Xueping,XIANG Xudong.Analyzing and forecasting the influence factor of PM2.5 based on principal component regression[J].Journal of Anqing Teachers College(Natural Science Edition),2015,21(4):24-27.)
• [13]于建华,虞统,魏强,等.北京地区PM10和PM2.5质量浓度的变化特征[J].环境科学研究,2004,17(1):45-47.(YU Jianhua,YU Tong,WEI Qiang,et al.Characteristics of mass concentration variations of PM10and PM2.5in Beijing area[J].Research of Environmental Sciences,2004,17(1):45-47.)
• [14]周丽,徐祥德,丁国安,等.北京地区气溶胶PM2.5粒子浓度的相关因子及其估算模型[J].气象学报,2003,61(6):761-768.(ZHOU Li,XU Xiangde,DING Guoan,et al.The correlation factors and pollution forecast model for PM2.5concentration in Beijing area[J].Acta Meteorologica Sinica,2003,61(6):761-768.)
• [15]王浩宇,王三平,郭志明,等.太原市冬季SO2、PM10和PM2.5浓度城郊变化特征研究[J].环境与可持续发展,2014,39(5):190-193.(WANG Haoyu,WANG Sanping,GUO Zhiming,et al.SO2、PM10and PM2.5concentrations variation characteristics study between urban and suburban areas in winter in Taiyuan[J].Environmental and Sustainable Development,2014,39(5):190-193.)