中文摘要：

在利用CAPPS2.0进行空气质量自动化预报的前期工作中,为了验证模式的准确性,对辽宁省14个主要城市2003年1月的31 d、4月、7月和10月各12 d作出了模拟,分别得到各城市SO2、NO2和PM10的污染指数预报值。将各污染指数每天模拟的预报值与实际监测值进行对比,发现春、夏、秋季的预报效果非常好,而冬季预报值明显偏高。对预报值与监测值的相关性检验表明两者相关性非常显著,冬季预报值出现的是一个系统误差,可以通过修改模式进行订正。

英文摘要：

In the process of preparing to use CAPPS2.0 numerical model to do automatic prediction of the air quality,the accuracy of the model has been verified by simulating pollution scenarios over the time period of the whole January,twelve days in April,July and October for the 14 main cities of Liaoning province,and obtained the numerical prediction results of the pollution index of SO2 ,NO2 and PM10 for every city. Contrast of everyday indices between forecast and observation showes that the forecast results are rather fine in spring,summer and autum while clearly higher in winter. The relativity tests between forecast and observation indicate that their correlation is very notable. A systematic error always exists in winter forecast,it should be corrected by means of modifing the model.