Urban parks are essential part of urban green area providing important ecosystem services. Ants are one of the biodiversity component in these parks. Waste food consumption by ants is one of the ecosystem services closely related to public health and esthetic. We, however, have little information on artificial food removal by ants in urban landscapes. Our study addressed following questions: (1) How many ant species involved in food removal and which species consumes more? (2) Is there any effect of land modification on the food removal？(3) Are there relationships between ant species number and food removal？ We set artificial foods (potato chips, cookie, sausage) at 90 sites with three types of land cover (pavement, grassland, woodland) in 10 parks in Tokyo (urban and suburban area), measured food removal rates and identified ant species involved. In total 11 ant species were identified. Around 30% of food In the field was removed in 24 hours. Species number and food removal rates of ants were affected by land cover types; both were higher at woodlands than the others, but not different between urban and suburban parks. Food removal rates were affected by presence of two particular species, but not by species number. Our results suggest that species identity can be more important than species diversity for provision of ecosystem services.