A special joint symposium of ISPRS Technical Commission IV & AutoCarto in conjunction with ASPRS/CaGIS 2010 Fall Specialty Conference November 15-19, 2010 Orlando, Florida SIMULATING LANDUSE CHANGES DRIVEN BY A 3 rd BOSPHORUS BRIDGE I. E. Ayazli a, *, F. Batuk a , B. Kleinschmit b a Department of Geomatic Engineering, Yildiz Technical University, Faculty of Civil Engineering, 34220, Davutpasa, Istanbul, Turkey - (eayazli, batuk)@yildiz.edu.tr b Department of Geoinformation Processing for Landscape and Environmental Planning, Technische Universität Berlin, Institute of Landscape Architecture and Environmental Planning, 10623 Berlin, Germany – birgit.kleinschmit@tu- berlin.de Commission VI KEY WORDS: Spatio-temporal Geodatabases, Urban Sprawl, Remote Sensing, Cellular Automata, Simulation * Corresponding author. ABSTRACT: Istanbul, which joins two continents-Asia and Europe, is an important city with its cultural, natural and environmental heritages. In the past and today, increasing population has caused an enormous transportation problem. To overcome this problem, two bridges were built on the Bosphorus and a third bridge will be built on the north side of Bosphorus. Although zone plans made in the 1960s suggested the urban growth to east-west direction, urbanization was triggered to northward by Bosphorus Bridges. Main research questions of this study are: What is the influence of a third Bridge on urbanization processes in Istanbul? Which protected areas will be damaged? To find out the influences of 3 rd Bosphorus Bridge on the urbanization in Istanbul, four Landsat satellite images were classified. First bridge was built on the Bosporus in 1973 and the second bridge in 1988. Therefore, time periods of the images were determined as 1972, 1987, 2002 and 2009. SLEUTH simulation software was used to predict the urban sprawl of Istanbul for the year 2030. The result of the prediction map was shown that 28.88 % forest and 71.43 % agricultural areas and open spaces will be transformed to urban area. 1. INTRODUCTION Urban sprawl is unplanned and uncontrolled urban expansion. The European Environment Agency (EEA) has described sprawl “as the physical pattern of low-density expansion of large urban areas, under market conditions, mainly into the surrounding agricultural areas” (EEA, 2006). To determine urban sprawl, several theories have been developed such as the Monocentric City Model, the Tiebout Local Public Finance Model (Nechyba and Walsh, 2004), the Concentric Zone Theory, the Sector Theory, and the Multiple Nuclei Theory (Yu and Ng, 2007). However, urban sprawl is a dynamic system that contains physical, ecological and environmental parameters, all of which focus on economic and social factors. Therefore, complex systems are used to model urban sprawl (Clarke et al, 1997; Cheng, 2003). Several data are utilized for monitoring urban sprawl, such as satellite imagery, land use/cover maps, digital elevation model (DEM), digital terrain model (DTM), administrative boundary, topographic maps, aerial photos, ortophoto, geological maps, socio-economic data etc. (Cheng, 2003; Çelikoyan, 2004) CA is one of the simulation methods which runs subdivided cells of a regular lattice. Future state of each cell is determined by its adjacent cell’s state. CA has five main elements, including, space, state, neighborhood, transition rules and time. Each automaton is defined by a set of state , neighborhood of automaton and transition rules. Various software based on CA algorithm were created to simulate urban sprawl. SLEUTH is one of them and has been used in a lot of projects. SLEUTH, which was written in C programming language and working under UNIX, has two components; urban growth model (UGM) and land cover deltatron model (LCD). The UGM uses the standard gnu C compiler (gcc) and the LCD is embedded in the code and driven by the UGM (URL 1). Simulation model is created by four growth rules and five growth coefficients. In the Table 1, relations are shown between growth rules and growth coefficients. Each coefficient value must be between zero and 100 (URL 2 and Sevik 2006). Growth Rules Growth Coefficients Spontaneous Growth Dispersion, Slope New Spreading Center Breed, Slope Edge Growth Spread, Slope Road Influenced Growth Breed, Road Gravity, Slope, Spread Table 1: Growth Rules and Coefficients Self-modification is second level of growth rules. It “is prompted by an unusually high or low growth rate. The limits CRITICAL_HIGH and CRITICAL_LOW are defined in the scenario_file” (URL 3).