Remote Sensing Applications in Population Estimation, Regression Analysis, and Urban Growth Simulation Modeling for a Middle Eastern City
Ph.D. (Doctor of Philosophy)
Department of Earth, Atmosphere and Environment
Due to a change in government regime and population migration, the City of Hillah in Iraq has been facing several urban issues, particularly population estimating and urban growth planning. Since the last census conducted in 1997, population in Iraq has been estimated based on the annual growth, without considering migration as factor of the population growth. Therefore, the aim of this dissertation study was to develop a population estimation method and an urban growth simulation method for the City of Hillah. Population data were estimated for Hillah, using the Normalized Difference Built-up Index (NDBI) derived from Remote Sensing (RS) and Urban Morphology (UM) model. The Dasymetric Mapping (DM) technique was used to estimate the population based on residential land use (LU). The City of Hillah is classified into five UM periods with each period having different characteristics. The UM model was used for Hillah to standardize housing sizes due to different urban forms evolving as a function of undocumented population growth since 1997. Population estimated for households for each Mahallah (the City of Hillah comprises planned urban administrative units known as Mahallahs, which is the smallest administration unit), a formal district within the city. Population estimates were then compared to census data for 1987 and 1997 for calibration after non-residential LU, and street and road areas were subtracted from the LU base map. The spatial pattern between the population distribution and human and physical factors was analyzed by applying regression models. Two different geographical methodologies (regional geography and systematic geography) were considered to understand the spatial pattern for the population distribution and the human and physical factors for the years 1998, 2008, and 2018. Three pixel regression models were applied to examine the spatial pattern, namely Ordinary Least Square (OLS), Spatial Lag Model (SLM), and Spatial Error Model (SEM). The population as the dependent variable was estimated by people per pixel, and LU and Land Cover (LC) as explanatory variables which were classified by INDEXs methods. The outcome results show the population distributed irregularly over that time and there is a negative correlation between the population and the explanatory variables. The urban growth was simulated for the City of Hillah for the period 1998-2018, and then urban growth was predicted for the next 20 years 2028 and 2038. LU and LC interpreted using a machine learning maximum likelihood classification method based on 30 m resolution to classify the LC for the city for the years 1998, 2008, and 2018. Cellular Automata (CA) modeling was used to simulate the urban growth with three neighbors window sizes (3x3, 5x5, and 7x7). Several factors were used from different urban growth theories, which were distance to Central Business District (CBD), distance to school, distance to hospital, distance to mosque, distance to main road, distance to main road entrance, and population densities for 1998, 2008, and 2018. The results indicate that 3x3 and 5x5 neighbors windows were the most accurate; therefore, 3x3 neighbors windows were applied to simulate the urban growth for 2028, and 5x5 neighbors windows were applied to simulate the urban growth for 2038. The spatial pattern was analyzed suggesting that infill and linear urban growth forms are the most common type of spatial patterns that occurred in the period 1998-2018. Findings also show that expansion and clustered urban growth form are the most common spatial pattern type to happen for the predicted urban growth simulation for 2018- 2038.
Alyasiri, Elaf Amer, "Remote Sensing Applications in Population Estimation, Regression Analysis, and Urban Growth Simulation Modeling for a Middle Eastern City" (2021). Graduate Research Theses & Dissertations. 6809.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.