Abstract. Aiming at the complicated environment of Vehicular Ad Hoc Networks(VANET),in this paper,a region-based adaptive clustering algorithm (ZACA) is proposed.Through classifying the road environment into segment and intersection, respectively calculating the connectivity according to the different road environments, and combine it with the regional location and the traditional synthesis weights of the Weighted Clustering Algorithm,the election of the cluster head under different road models and the maintenance of cluster based on the network area are realized.Simulation result shows this method can adapt to vehicular network environment effectively, and also can improve the stability of the cluster and reduce the clustering overhead.