Identifying the urban-rural fringe using wavelet transform and kernel density estimation: A case study in Beijing City, China

The urban-rural fringe, known as the region located between the urban and rural areas, is the frontier of urban expansion against rural reservations. Identifying this particular region precisely, which was usually simplified by researchers, is the most important prerequisite in studies related to urban-rural patterns. In this study, we proposed a new model, combining wavelet transform and kernel density estimation, to identify the urban-rural fringe based on land use data. After testing the model using Beijing City as a case study, it is proved that the model is able to delineate the boundaries of urban-rural fringe precisely with respect to different landscape patterns at different regions (central urban area, urban-rural fringe area, and outer rural area). Furthermore, due to the advantage of the self-adaptive-bandwidth kernel density estimation, the model can also distinguish some of the satellite towns from the central urban area and outer rural area with the boundaries of urban-rural fringe. (C) 2016 Elsevier Ltd. All rights reserved.