Skip to main content

Land use/land cover change evaluation using land change modeller: A comparative analysis between two main cities

Dr Kursah, Matthew B.
Senior Lecturer
  mkursah@uew.edu.gh

Authors
Tarawally, M., Wenbo, X., Weiming, H., Mushore, T. D., & Kursah, M. B.
Publication Year
2019
Article Title
Land use/land cover change evaluation using land change modeller: A comparative analysis between two main cities
Journal
Remote Sensing Applications: Society and Environment
Volume
16
Page Numbers
100262
ISSN
2352-9385
Abstract

Owing to the proven capability of remotely sensed data in the extraction and analysis of land use land cover (LULC) change, Landsat ETM+ and OLI imagery of 2000 and 2015 have been used in this research to carry out the LULC change comparative analysis in Bo and Freetown, two major urban areas in Sierra Leone, Africa. The supervised imagery classification with maximum likelihood algorithm method was adopted for the extraction of LULC categories. To demonstrate our idea effectively, we used the land change modeller integrated into IDRISI Selva software package to quantify and map the changes of each LULC category. Employing an error matrix table and estimator of Kappa statistics (Khat), we achieved overall accuracy and Khat greater than 80% for both cities and class level accuracies were also achieved as greater than 70%. The LULC change statistics show dynamic characteristics of LULC in the areas where maximum fluctuation was observed in dense vegetation category in Bo and agricultural land in Freetown. The built-up area shows a continuous increasing trend in both cities. Results of our analysis demonstrated that dense vegetation increased by 1024 ha (ha) in Bo whereas it reduced by 3807 ha (ha) in Freetown between 2000 and 2015 study years. Likewise, agricultural land increased by 545 ha (ha) in Bo and decreased by 9145 ha (ha) in Freetown during the same period. It is worth noting that the built-up area increased in both cities as 1326 ha (ha) and 8543 ha (ha) were recorded in Bo and Freetown, respectively. The spatial trend of LULC transition reveals that most of the transition has been occurring in the central part of Bo; whereas the transition occurs in the northern and southern parts in Freetown. However, both cities witness transition at the southern part with regards to dense vegetation category. These findings could assist in making policies for the efficient use of natural resources leading to the development of sustainable urban environments.

© 2019 University of Education, Winneba