Applying the anomaly-based weather analysis on Beijing severe haze episodes

Smoggiest days like heavy rainfall, tornadic and hail storms are considered as weather extremes. The method of anomaly-based meteorological variable analysis has been recently used to analyze and predict weather extremes. Thus, in this study, the same method was also applied to analyze ten severe haze episodes with PM2.5 exceeding 300 micrograms per cubic metre (μg/m3) for >24 h occurred during 2015–2016 in Beijing or North China. The anomalous spatial features of tropospheric meteorological variables show that the haze intensity indicated by PM2.5 concentration is strengthened gradually following the five factors of (1) an anomalous temperature inversion in the lower troposphere, (2) negative geopotential anomaly near the surface, (3) positive geopotential anomaly at the upper troposphere, (4) southwesterly wind anomaly and (5) positive anomaly of specific humidity in the lower troposphere. The correlation coefficient between the geopotential anomaly factor at the upper troposphere and PM2.5 concentration at Beijing is 0.56 reaching the level of 95% confidence in December 2016. Forecast evaluation analyses reveal that the European Centre for Medium-Range Weather Forecasts (ECMWF) model is able to predict most anomalous temperature-pressure structures related to haze episodes and to indicate haze intensities for leading about one week.