Fractal Modeling Of Urban Morphology

SHAROLYN ANDERSON; CLAUDIO DELRIEUX

Congreso; 2007 Annual Meeting of Association of American Geographers; 2007

Understanding urban morphology using mathematical techniques applied to remotely sensed data is essential in modeling urban growth patterns. This work is aimed to provide a new analytical approach for assessing urban morphology and growth models using fractal and multifractal estimators. We derived indicators of the growth patterns over three distinct locations. Urban morphology attributes are derived using ten years of DMSP-OLS composite nighttime lights imagery, over urban areas in the United States, Mexico, and Argentina. Two analytical methodologies were applied to these data. First, the contour of the light imaginary is regarded as an indirect measure of the border of the urban settlements. The shapes of the contours for the series of images are analyzed using monofractal dimension estimation for several luminance thresholds. Second, the light map itself can be regarded as an indirect estimation of the 2D population density function. Over this map we performed multifractal analysis using the of fractal dimension technique. The conclusion of this analysis shows different estimation parameters for each of the urban areas under study, while the parameters at each location show similar results along the ten years of data.