The primary objective of this study is to develop day and night road traffic noise maps for Guangzhou using Geographical Information Systems (GIS) and Global Positioning Systems (GPS). First, as speeddensity relation is used to estimate the traffic volume from GPS data collected from floating cars. Meanwhile, the attributes of the roads and buildings are automatically exported from GIS. Second, a single vehicle noise emission model is combined with a noise propagation model to formulate a regional traffic noise calculation model that accounts for the traffic noise attenuation in an urban area. The algorithm is optimized by intelligently dividing the computational grids, filtering the traffic noise sources automatically and performing a quick index of the estimation objects. Finally, the day and night road traffic noise levels in Guangzhou are estimated to create two traffic noise maps. The accuracy of the developed algorithm is validated by conducting a traffic noise monitoring experiment in several districts of Guangzhou with different road types. The results show that the average error between the estimated and measured values is below 2.0 dB.