运用贝叶斯方法的PM_(2.5)浓度时空建模与预测Spatio-temporal modeling and prediction of PM_(2.5) concentration based on Bayesian method
朱亚杰,李琦,侯俊雄,范竣翔,冯逍
摘要(Abstract)：
针对当前我国大部分地区正面临严重的空气污染问题,对重污染区域进行时空建模具有重要的意义。该文基于贝叶斯时空模型建立了京津冀区域的PM2.5浓度时空预测模型,该模型充分考虑了PM2.5浓度的时间变异特性与空间分布特性,并引入了气象数据作为协变量对没有监测站的位置进行预测。实验结果表明,该方法具有很好的预测精度,其在测试站点上的拟合优度达到了0.9以上,能够应用于区域级PM2.5浓度的时空分布建模与预测。
关键词(KeyWords)： 贝叶斯;时空预测;PM2.5浓度
基金项目(Foundation): 国家科技支撑计划项目(2012BAC20B06)
作者(Author): 朱亚杰,李琦,侯俊雄,范竣翔,冯逍
DOI: 10.16251/j.cnki.1009-2307.2016.02.009
参考文献(References)：
- [1]CRESSIE N.The origins of kriging[J].Math Geol.,1990,22(3):239-252.
- [2]CARUSO C,QUARTA F.Interpolation methods comparison[J].Computers&Mathematics with Applications,1998,35(12):109-126.
- [3]LI J,HEAP A D.A review of comparative studies of spatial interpolation methods in environmental sciences:performance and impact factors[J].Ecological Informatics,2011,6(3):228-241.
- [4]CRESSIE N,WIKLE C K.Statistics for spatio-temporal data[Z].John Wiley&Sons,2011.
- [5]LI L.An application of a shape function based spatiotemporal interpolation method to ozone and populationbased environmental exposure in the contiguous U.S.[J].J.Env.Inform.,2008,12(2):120-128.
- [6]LI L,REVESZ P.Interpolation methods for spatiotemporal geographic data[J].Computers,Environment and Urban Systems,2004,28(3):201-227.
- [7]TIAN J,LI L,PILTNER R,et al.Spatiotemporal interpolation methods for air pollution exposure[C]//Ninth Symposium on Abstraction,Reformulation,and Approximation.SARA,2011:75-81.
- [8]LE N D,ZIDEK J V.Interpolation with uncertain spatial covariances:a Bayesian alternative to Kriging[J].Journal of Multivariate Analysis,1992,43(2):351-374.
- [9]GELFAND A E.Hierarchical modeling for spatial data problems[J].Spatial Statistics,2012(1):30-39.
- [10]ALLENBY G M,ROSSI P E,MCCULLOCH R E.Hierarchical bayes models:apractitioners guide[J].SSRN Electronic Journal,2005.
- [11]HANDCOCK M S,STEIN M L.A Bayesian analysis of Kriging[J].Technometrics,2012,35(4):403-410.
- [12]CHRISTAKOS G,SERRE M L.BME analysis of spatiotemporal particulate matter distributions in North Carolina[J].Atmospheric Environment,2000,34(20):3393-3406.
- [13]BAKAR K S.Bayesian analysis of daily maximum ozone levels[D].Southampton:University of Southampton,2012.
- [14]SARTO S D,RANALLI G M,SHUVO BAKAR K,et al.Bayesian spatio-temporal modeling of urban air pollution dynamics[C]//The METMA VII and GRASPA14 Conference.Torino(IT),September 10-12,2014.
- [15]BAKAR K S,SAHU S K.spTimer:spatio-temporal Bayesian modeling using r[J].Journal of Statistical Software,2013:1-26.
- [16]LIU R F,SOETJIPTO R.Analysis of three Bayesian network inference algorithms:variable elimination,likelihood weighting,and gibbs sampling[J].J.Artif.Intell.Res.,2004,1(1).
- [17]WILLMOTT C J,ACKLESON S G,DAVIS R E,et al.Statistics for the evaluation and comparison of models[J].Journal of Geophysical Research-Oceans,1985,90(C5):8995-9005.