成都市空气质量预报系统的应用及预报效果评估

Application of Chengdu Air Quality Forecast System and Evaluation of Forecast Effect

介绍了成都市空气质量预报系统的模式设置和排放源处理,并评估其对2017年2月成都市的气象要素和PM10、PM2.5、SO2、NO2小时浓度的24小时预报效果。结果表明,系统能较好的预报成都市主要气象要素的逐小时变化情况,平均气温、气压和相对湿度的相关系数均在0.72以上,但对累积降水的预报效果仍需优化;系统能合理的反映各污染物的时空分布,PM10和PM2.5的预报效果达优秀水平,但存在不同程度的高估;NO2的预报值与实况值的时间变化趋势一致性最高,空间分布对应较好,但模式预报对其存在一定程度的低估;对SO2有显著的高估,但其空间分布模拟效果最佳。优化源的空间分配,及时更新源排放清单,同时针对不同排放源提高小时尺度排放清单分辨率可能是未来提高成都市空气质量预报系统预报准确率的有效途径。

This paper introduced the setting of Chengdu air quality forecast system and the treatment of emission source, and evaluated the 24-h forecast performance for meteorological factors, PM 10 , PM 2.5 , SO 2 and NO 2 hourly concentration in Chengdu during February 2017.The results showed that the hourly variation of the main meteorological elements in Chengdu can be well forecasted by the system, the correlation coefficients of mean temperature, air pressure and humidity are all above 0.72, however, the forecast effect of precipitation should still be optimized. The spatial and temporal distribution of each pollutant could be reasonably reflected by the system, and the prediction effect of PM 10 and PM 2.5 is excellent, but there are different degrees of overestimation. The predicted value of NO 2 and the actual value have the highest consistency in temporal variation trend, as well as spatial distribution, but it is somewhat underestimated by model forecasting. The value of SO 2 is obviously overestimated, but its spatial distribution is best simulated. Optimizing the spatial allocation of sources, updating the source emission inventories in time, and improving the temporal resolution on hour scale of emission inventory for different emission sources may be an effective way to improve the forecast accuracy of Chengdu air quality forecast system in the future.