Abstract: Spatial structure of regional tourism and temporal distribution of tourist flow are important for tourism, but few international or domestic researchers have examined the correlation of these two factors. The tourist flows are a typically intense short-term flow in Golden Weeks in China. We sampled 7 typical scenic spots from Kunming, Dali, Lijiang and Xishuangbanna (the four key tourism cities of Yunnan): Stone Forest, Kunming Expo Garden (KEG), Three Pagodas, Mt. Yulong, Tai Garden, Xishuangbanna Tropical Botanical Garden (XTBG) and National Forest Park. Yunnan Provincial Holiday Tourism Information Center issued daily reports for tourist flows of major tourist areas during Golden Week of "Labor Day" and "National Day". These data sets are used for this research. Except for KEG, the temporal distribution of tourist flow has no significant variation between different Golden Weeks at the same spot for the other six spots, which are tested by skewness index G, correlation analysis, and Person coefficient. Temporal distribution is a one-peak curve. There is an obvious difference between the timing for the peak of tourist flow and G index in one Golden Week. The peak occurs in the early Golden Week at Stone Forest, and it has the lowest G value at this place. The peak value appears in the middle of Golden Week at Three Pagodas, Tai Garden and XTBG, and a slightly higher G value is observed at these places respectively. The peak values at Mt. Yulong and National Forest are in the late Golden Week, but the highest G value is found at these two places. Results show that temporal distribution for short term tourist flows significantly corresponds to spatial structures of regional tourism. G value and the timing for the peak of tourist flow in Golden Weeks for these spots significantly correspond to spatial structure of tourism in Yunnan. The response mechanism follows these rules: the geographical location of regional tourism resources affects tourists choosing their destination places and the number of tourists; the spatial structure of transportation is the major factor that guides tourist flows. Limited by the holiday schedule, temporal distance from the tourist transport center to each destination is the major reason affecting temporal distribution of tourist flow. The longer the temporal distance is, the later the peak of tourist flow appears and the higher the G value is. And, temporal distribution may have multiple responses to spatial structures of regional tourism with different spatial scales in certain spots.

Key words: short term tourist flows, temporal distribution, tourism spatial structure, Golden Week, response, Yunnan