Modeling Spatial and Temporal Urban Growth Patterns and trends by using Geospatial technologies: (The Case of Modjo Town)

Abstract

A rapid increase in the world’s population over the last century has triggered the transformation of the earth surface, especially in urban areas, where more than half of the global populations live. Ethiopia is no exception and a high population growth rate, coupled with economic development over the last three decades, has transformed the Modjo Town into a hot spot for massive urban growth .Although urbanization presents several opportunities, the environmental and social problems cannot be underestimated. Therefore, the need to estimate the rate and extent of land use/land cover changes in the area and the main drivers of these changes is imperative. Geographic Information Systems (GIS), Remote sensing, IDRIS SELVA and ENVI techniques provide effective tools in studying and analysis land-use/land-cover change over space and time. A supervised and support vector machine algorithm (SVMAs) was used for multi_temporal land satellite classification. The change detection of multiple Landsat images was conducted to map and analyzed the extent and rate of land use/land cover change in the ModjoTown between 1990 and 2019. Subsequently, the urban growth extent of the area was forecasted for the year 2030, 2040 and 2050 on the land change modeler using MultiLayer Perceptron neural network, with combination of markov chain method, together with drivers representing proximity, biophysical, and socio-economic variables and finally the relationship between population growth and urban built-up area expansion were analyzed using regression analysis. Overall, the results show there was the extensive LULC changes in Modjo town over the study Period (29 years). Agricultural land was dominant class at the beginning of the study period and until this research conducted but declined with the continuous increase of urban land area. It can be seen that the rate of annual urban expansion greater by far, from the rest land use classes. The coverage of agricultural land was 88.25% of the total area at the beginning of the study timeframe. In the following 29 years, it dramatically dropped to about 66.34%. By contrast, the proportion of urban class was 0.93% in 1990, this figure rose to just over 15.94% in 2019, showing an increase of more than sixteen folds of its initial spatial share. It showed 55.42% of average annual expansion across the study period. Temporally, the rate of urban built-up land expansion was comparatively low in the first ten years between 1990 and 2000, which showed nearly 0.9% increase, then it peaked about 15.94% of the total area in the last year. On the other hand, vegetation class showed a remarkable increase in spatial cover between 1990–2000 has also contributed to the increase of its share of the study area from 6.67–9.01% in the first and last years of the study period, respectively. While the areal share of the barren land decreased in the first 20 years during the study timeframe however it is increased dramatically in the last 10 years. In the next coming 30 years it will be expected that urban built-up land area is increased from 15.94% to 29.46%,32.78% and 34.84% of the total area coverage according to prediction at 2030,2040 and 2050 respectively. As the study reveals also relationship between population growth and urban built-up area expansion were positive correlation. The success of this research is generating a future urban land-use map for 2030, 2040 and 2050, together with the other significant findings, demonstrates the usefulness of spatial models as tools for sustainable city planning and environmental management, especially for urban planners in developing countries like Ethiopia.