Land Cover Change Detection in the Urban Catchments of Dar es Salaam, Tanzania using Remote Sensing and GIS Techniques

  • Philip Mzava Department of Water Resources Engineering, Water Institute, P. O. Box 35059, Dar es Salaam, Tanzania.
  • Joel Nobert Department of Water Resources Engineering, College of Engineering and Technology, University of Dar es Salaam, P. O. Box 35131, Dar es Salaam, Tanzania.
  • Patrick Valimba Department of Water Resources Engineering, College of Engineering and Technology, University of Dar es Salaam, P. O. Box 35131, Dar es Salaam, Tanzania.
Keywords: Maximum Likelihood, NDVI, Artificial Neural Network, Landsat, QGIS

Abstract

In this study, the Maximum Likelihood (ML) classification, Normalized Difference Vegetation Index (NDVI) and Artificial Neural Network (ANN) methods were applied to three (3) Landsat images collected over time (1979, 1998 and 2014), that contained historical land cover features for the urban catchments of Dar es Salaam. Five major land cover classes were identified, mapped, and the land cover changes investigated. The major land cover changes observed from post-classification comparisons of the classified images are: the forest land losing 17.09% of its area in the period 1979-1998 to other land covers, mainly turning to grassland, and from 1998 to 2014, 17.55% of the total study area turned to high and medium/low-density built-up areas. Growth in urban settlement and infrastructure was observed to be continuously increasing and the high and medium/low-density built-up areas are projected to cover 66.09% of the total area by 2030; this is an increment of 29.01% from 37.08% coverage in 2014. This shift in land cover was further validated by the results of the Normalized Difference Vegetation Index (NDVI) analysis which showed a similar trend (shift from thick vegetation towards barren land) from 1998 to 2014, with median NDVI values changing from 0.52 to 0.36 respectively. These land cover changes are most likely the results of activities related to the increase in total population, the influx of urban population and the growth of the economy.

Keywords: Maximum Likelihood, NDVI, Artificial Neural Network, Landsat, QGIS.

Published
2019-12-22
Section
Articles

Journal Identifiers


eISSN: 2507-7961
print ISSN: 0856-1761