Connectivity is an important metric of network performance. In opportunistic sensor networks (OSNs), the frequent topology change caused by node mobility leads to the challenges of representation of connectivity. Building connectivity model for OSNs is helpful for its optimization and maintenance. After analyzing the topological characteristics of OSNs, this paper constructs connectivity model based on temporal-spatial graph theory, which is employed to describe the topological evolution law. According to the message reachability, temporal distance and topological distance are defined based on message transmission temporal and spatial characteristics. The correlation between them is also analyzed by statistical product and service solutions (SPSS), and the results show that there is no obvious correlation. Thus temporal distance and topological distance of each snapshots are employed to construct overall network connectivity. Experimental results show that the proposed model can better depict the network connectivity of OSNs, and comparing with the connectivity model based on Katz centrality, it can represent the connectivity changes better.