Abstract: The downtown area of Shanghai has rich tourism resources and numerous commercial, historical and cultural blocks, which attracts a lot of tourists every year. This paper touched on the method of mining data by grabbing the POI data from the social network Panoramio. According to the first law of geography, the paper aims to explore the tourists POI spatial static distribution and temporal and spatial dynamic variation by Arcgis, spatial trend surface modeling and so on. The study shows that the number and density distribution of POI is extremely uneven. The high level of the tourism area, large commercial, entertainment area, historic district area presents high degree of concentration, while the areas with dense residential areas gain very limited attention of oversea tourists. The number and distribution of tourists POI also changed greatly by month and time. The study aims to explore tourists' preference, provide guidance of the distribution and management of tourism services and tourist grooming.