There exist increasingly serious and dangerous photochemical smog events in some cities caused by increasing motor vehicles. The prediction and early warning system, however, has not started to work. Therefore, we had better monitor the photochemical smog using remote sensing data and, at the same time, build the early warning system. Studies reveal a distinct positive correlation between aerosol optical depth (AOD) and air pollution monitored by MODIS data. When AOD<0.3, air is clean; when 0.3<AOD<1.0, there exists weak air pollution; and when AOD>1.0, air pollution becomes serious. Based on analyzing the correlation between AOD and air pollution and using data of air pollution monitors and Air pollution Index (API), this paper has divided the early warning system of Urban Photochemical Smog Pollution into five grades: nil, faint, weak, relatively strong and strong, Verification with the example of Guangzhuo has laid the foundation for the application of the Advanced Warning Mechanism System to the Urban Photochemical Smog Pollution.