The hourly WRF forecast wind fields with a resolution of 1 km is used as the input of the CALMET diagnostic model to generate wind fields with different temporal and spatial resolutions, which drive CALPUFF to obtain concentration fields with a resolution of 50 m per minute. The impact of the temporal and spatial resolution of CALMET meteorological fields on the concentration fields and the calculation time of each scheme are analyzed. The results show that satisfactory wind field and concentration field can be obtained even with coarse temporal and spatial resolution at stable wind direction and high wind speed conditions. The temporal and spatial resolution has a significant impact on the wind and concentration fields when the wind direction changes and the wind speed is low. The difference between concentration fields driven by various meteorological schemes can be as high as 40%. During the transition of the wind field, the accuracy of the concentration field will worsen with finer meteorological grid if the modeling time step of CALMET is greater than 30 minutes. The longer the modeling time step is, the more significant the deviation of the concentration field is. Considering the calculation time and the accuracy of the concentration field simulation, CALMET meteorological scheme with a time step of 10 min and a grid resolution of 400 m is recommended in the emergency early warning of air pollution accidents.

采用WRF输出的逐小时1 km分辨率预报风场作为CALMET诊断模式的输入, 生成不同时空分辨率的CALMET诊断风场, 耦合 CALPUFF得到逐分钟50 m分辨率浓度场, 在此基础上分析CALMET气象场时空分辨率对浓度场的影响, 并统计不同气象时空分辨率方案的计算耗时。结果表明, 当风向稳定且风速较大时, 较粗的时空分辨率亦能得到满意的风场和浓度场; 当风向转变且风速较小时, 时空分辨率对诊断风场和浓度场影响显著, 不同气象方案的浓度场差异可以高达40%; 风场转变期间, 当CALMET的时间步长大于30 min时, 加密气象网格会降低浓度场的模拟精度, 时间步长越长, 浓度场偏离越显著。综合考虑计算耗时和浓度场模拟准确性, 推荐在大气污染事故应急预警中采用10 min时间步长和400 m网格距的CALMET气象方案。