The identification of geographic trends of an influenza pandemic is important for analyzing its social epidemic factors. We performed spatiotemporal analyses focusing on the metropolitan areas in Japan by using the influenza-like illness (ILI) sentinel surveillance data for the pandemic (H1N1) 2009 and seasonal influenza. The epidemic curves and spread features expressed by the kriging method of geographic information system (GIS) and correlations between reported cases and demographic data were analyzed. The incidence of pandemic (H1N1) 2009 increased gradually at the beginning and showed more sporadic epidemic features compared to seasonal influenza. However, there were coincidental locations of patient clusters affected by the seasonal influenza, with a significant coefficient for the total sentinel reported cases (r = 0.71, P < 0.01). This suggested similar patterns of the epidemic over seasons. Patient clusters tended to be located in suburban areas, and there seemed to be stronger relationships between epidemics and higher ratio of larger families (with r = 0.26–0.35, P < 0.01, between ratio of families having more than 3 members and total reported cases in Tokyo and Nagoya areas). Whether populous areas had a greater probability of maintaining the epidemic patterns needs to be determined. Nonetheless, the patterns found in this study can be useful for further analyses for epidemic modeling and designing relevant controls.