Modeling land price distribution using multifractal IDW interpolation and fractal filtering method

Abstract

Characterizing the spatial distribution of urban land price is essential for improving urban planning and management, as well as for effectively modeling and predicting changes in urban land use. Previous studies have shown that in using conventional geostatistics methods to characterize the local structure of land price, there is controversy regarding the effectiveness of interpolation. In this paper, a recently developed Multifractal Inverse Distance Weighted (MIDW) interpolation method is applied to characterize the spatial structure of land price, and a spectrum analysis method (S–A) based on a fractal filtering technique is applied to separate the singularity from the background of land price distribution; these methods are applied to a study site in the city of Wuhan (China). It is shown that the MIDW interpolation method is a valid and effective alternative for characterizing land price distribution by comparison with ordinary IDW and Kriging methods. Based on deviation and parameters, the results of the MIDW method not only fit better with the surveyed values, but they also incorporate both the singularity and spatial association in data interpolation. The singularity of land price, which could be attributed to local special landscapes, such as the Yangtze River and East Lake, was successfully separated from its background by the S–A method. The background, which represents the overall spatial trend of land price distribution, was reclassified by the fractal concentration–area method. The derived singularity and background will better aid the decision-making process for urban planning.