Spatial-temporal distribution and climatic statistical diagnosis and prediction of haze weather in North China during 2003-2014

Authors: Youfang Chen*, Peking university, Lixu Zhang, Peking university
Topics: Environment, China, Climatology and Meteorology
Keywords: North China, PM2.5 pollution, MODIS, reanalysis of meteorological elements data, climate change.
Session Type: Paper
Day: 4/14/2018
Start / End Time: 2:00 PM / 3:40 PM
Room: Southdown, Sheraton, 4th Floor
Presentation File: No File Uploaded

In North China, haze pollution has aroused wide concern of scientific community and the society. Relationship between them in North China and its surrounding areas is of great significance to better understanding the pollution characteristic. The mixed linear regression, cluster analysis, and other statistical analysis methods have been performed to study the spatiotemporal characteristics of haze pollution and its relationship with meteorological elements during 2003-2014 based on remote sensing AOD, reanalysis of meteorological data and ground-based fine particle matters observation data. The results showed that satellite remote sensing AOD had higher correlation coefficient (r=0.03) with the ground observation PM2.5 concentration. In North China and adjoining areas, PM2.5 exhibited obvious spatial distribution; the Taihang Mountains divided the area into southeastern mountains with significantly higher pollutant intensity and northwest with lower pollutant intensity. As for temporal variation, air pollution intensity was experienced a rapid increase since 2006. PM2.5 pollution showed significant positive correlation with relative humidity and longitudinal wind and north wind is apparently conducive to the accumulation of pollution compared with the south wind. Besides, it was the most favorable conditions for pollution at the height of 1000 to 2000 meters of the boundary layer. Moreover, a discrimination model was set up to predict PM2.5 pollution intensity and weather elements that affect PM2.5 were analyzed their long-term variation trend proceeding 1979 to 2014, five predictors selected were PM2.5 concentration the day before, relative humidity, boundary layer height, zonal wind and meridional wind. Prediction accuracy of most regions were more than 85%.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login