Investigation of relationships between meteorological conditions and high PM10 pollution in a megacity in the western Yangtze River Delta, China Academic Article uri icon

abstract

  • © 2017, Springer Science+Business Media Dordrecht. Nanjing, a megacity in the western Yangtze River Delta of China, has been suffering high PM10 pollution during recent decades. The meteorological conditions that affect the formation of high PM10 pollution in Nanjing remain unknown. In this study, 12-year daily PM10 concentrations and ground-level meteorological conditions from 2001 to 2012 were analyzed. The relationships between PM10 concentrations and meteorological parameters including wind direction, wind speed, temperature, relative humidity, and precipitation were investigated with multiple methods, including occurrence frequency, back trajectory, multiple linear regression (MLR), and artificial neural network (ANN) analysis. The results show that PM10 pollution in Nanjing was severe during 2001–2012. The 12-year mean is 109 μg/m3, which is 56% higher than the grade II annual PM10 standard of China. Declining trends are found for both the annual averages and the episode averages during the most polluted events, with a rate of 3.5 and 4.0 μg/m3 per year, respectively. High PM10 pollution events are more likely to occur on no-precipitation days when wind is from east with speed of 1.0 to 3.0 m/s, temperature is 10 to 20 °C, relative humidity is below 80%, and sea-level pressure is 1015–1025 hPa. The 72- and 24-h back trajectory analysis shows that over half of the air masses on the high PM10 pollution days are from east-south with an average transport distance of around 400 km in 72 h and 200 km in 24 h, respectively. Including the previous day concentration in the MLR models substantially improves model performance for predicting daily PM10, with an increase of the Pearson correlation coefficient from 0.25 to 0.61. The ANN model generally performs better than the MLR models and can be considered as a better method for future PM10 forecasting. However, both models have difficulties accurately predicting very severe pollution events, and more factors should be considered in future studies.

author list (cited authors)

  • Zu, Y., Huang, L., Hu, J., Zhao, Z., Liu, H., Zhang, H., Ying, Q. i., & Chen, M.

citation count

  • 11

publication date

  • April 2017