摘要
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。
Most areas of China are facing serious air pollution problems,spatio-temporal modeling of heavy pollution areas has important significance.This paper established a space-time predictive model of PM2.5concentration in Jing-Jin-Ji area based on Bayesian method,which not only took into account the temporal and spatial variability of PM2.5concentration,but also used the meteorological data as covariates to make prediction.Experimental result showed that this method had good prediction accuracy,and its index of agreement on the test site reached higher than 0.9,which indicated that it could be used for spatiotemporal modeling and prediction of PM2.5concentration at the regional level.
出处
《测绘科学》
CSCD
北大核心
2016年第2期44-48,共5页
Science of Surveying and Mapping
基金
国家科技支撑计划项目(2012BAC20B06)
关键词
贝叶斯
时空预测
PM2.5浓度
Bayesian
spatio-temporal prediction
PM2.5 concentration
