Please use this identifier to cite or link to this item:
|Title:||Urban land use change analysis and modelling: a case study of Setubal-Sesimbra, Portugal|
|Author:||Araya, Yikalo Hayelom|
|Series/Report no.:||Master of Science in Geospatial Technologies;TGEO0015|
|Abstract:||In this paper urban land use change analysis and modeling of the Concelhos of Setúbal and Sesimbra, Portugal is accomplished using multitemporal and multispectral satellite images acquired in the years 2000 and 2006 and other vector datasets. The LULC maps are first obtained using an object-oriented image classification approach with the Nearest Neighbour algorithm in Definiens. Classification is assessed using the overall accuracy and Kappa measure of agreement. These measures of accuracies are above minimum standard accepted levels. The land use dynamics, both for pattern and quantities are also studied using a post classification change detection technique together with the following selected spatial/landscape metrics: class area, number of patches, edge density, largest patch index, Euclidian mean nearest neighbor distance, area weighted mean patch fractal dimension and contagion. Urban sprawl has also been measured using Shannon Entropy approach to describe the dispersion of land development or sprawl. Results indicated that the study area has undergone a tremendous change in urban growth and pattern during the study period. A Cellular Automata Markov (CA_Markov) modeling approach has also been applied to predict urban land use change between 1990 and 2010 with two scenarios: MMU 1ha and MMU 25ha. The suitability maps (change drivers) are calibrated with the LULC maps of 1990 and 2000 using MCE and a contiguity filter. The maps of 1990 and 2000 are also used for the transition probability matrix. Then, the land use maps of 2006 are simulated to compare the result of the “prediction” with the actual land use map in that year so that further prediction can be carried out for the year 2010. This is evaluated based on the Kappa measure of agreement (Kno, Klocation and Kquanity) and produced a satisfactory level of accuracy. After calibrating the model and assessing its validity, a “real” prediction for the year 2010 is carried out. Analysis of the prediction revealed that the rate of urban growth tends to continue and would threaten large areas that are currently reserved for forest cover, farming lands and natural parks. Finally, the modeling output provides a building block for successive urban planning, for exploring how and|
|Description:||Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies|
|Appears in Collections:||NIMS - MSc Dissertations Geospatial Technologies (Erasmus-Mundus)|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.