污染条件下成都市PM_(2.5)干沉降速率的模拟分析

SIMULATION OF PM_(2.5) DRY DEPOSITION VELOCITY IN CHENGDU UNDER CONTAMINATING CONDITION

利用成都市三瓦窑、沙河铺国控环监站2015年9月—2016年8月逐时PM_(2.5)监测数据,结合同期双流国际机场公布的地面气象要素(风场、温度、湿度和压强)以及温江站风速探空资料,首先统计分析了成都市风场、温度和湿度的基本特征,然后计算了污染条件下PM_(2.5)干沉降速率,并建立了适用于不同季节的GIFM模型和多元回归预测模型。结果表明:基本气象要素场的配置以及特殊地形导致了成都市PM_(2.5)干沉降环境恶劣,同时四季差异较大;污染条件下PM_(2.5)干沉降速率约为0.02~0.1 cm/s,表现为冬季<秋季<年<春季<夏季,四季的主要影响因子也不同,秋冬为湿度和压强,春季为温度,夏季为风速,且湿沉降强度过大时,会出现PM_(2.5)干沉降速率的"虚高"现象;GIFM模型和多元回归模型均能很好地预测污染条件下的PM_(2.5)干沉降速率,其预测能力均是夏季最好,冬季最差,春秋次之,通过对比分析表明GIFM模型的预测能力在各季节均优于多元回归模型。

Hourly PM2.5observation data（ Sept. 2015—Aug. 2016） from two environment monitoring stations of Sanwayao and Shahepu in Chengdu were used in this article,by combination of concurrent surface meteorological data（ e. g. Wind field,temperature,humidity and atmosphere pressure） from Shuangliu International Airport and sounding data of wind speed from Wenjiang station. The basic characteristics of wind field,temperature and humidity in Chengdu were first analyzed. Dry deposition velocity under polluted air condition was calculated,and then grey forecasting model,together with multi-regression analysis model for different seasons was also built. The results showed that： The configuration of basic meteorological factors and special terrain caused severe environment for dry deposition of PM2.5in Chengdu,with large differences among seasons.Winter had the lowest dry deposition rate,followed by Autumn,annual average,Spring and Summer. Dry deposition velocity of PM2.5ranged from 0. 02 cm/s to 0. 1 cm/s under contaminated condition. Main influential factors among seasons were different,where it was influenced by humidity and pressure in Autumn and Winter,temperature in Spring and wind speed in Summer. If the intensity of wet deposition was too high,the velocity would be artificially high. Both models made correct prediction of PM2.5dry deposition velocity under contaminated condition,with the best predictability in Summer,followed by spring,fall and winter. The contrast analysis showed that seasonal predictability of grey forecasting model was superior to that of multi-regression model.