Monitoring the extent of reclamation of small scale mining areas using artificial neural networks
Loading...
Date
OCTOBER 2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Small scale mining is widespread mainly in developing and underdeveloped countries.
It causes environmental degradation although it is a source of livelihood for several
people. Reclamation is needed to restore mined areas to an acceptable condition.
Artificial neural networks (ANN) are also being used recently for soil analysis, land
use/land cover analysis, etc. due to the increased availability of Landsat data. There
have been studies on various aspects of small scale mining, reclamation and artificial
neural networks but this research focused on using artificial neural networks to monitor
reclamation activities in small scale mining areas. Data used for analysis included
Landsat satellite images of study area (2007, 2011 and 2016), ground truth data and
shapefile of the study area. Two ANN classification methods, Unsupervised Self –
Organized Mapping (SOM) and Supervised Multilayer Perceptron (MLP), were used
for the classification of the satellite images. Normalized Difference Vegetation Index
(NDVI) change maps and class mask maps were also generated in order to help confirm
where actual change and to what extent it had occurred. Results of the study indicated
disturbance and revegetation in the study area within the 9 –year period. The
Barelands/mined areas class increased by 60.4% and a decrease in the vegetation class
by 18.7% from 2007 to 2011. NDVI maps, NDVI change maps, class mask maps and
maps showing reclaimed areas, disturbed and undisturbed areas together with statistical
information obtained from the classification results, confirmed the extent to which the
reclamation activities had gone. There was evidence of revegetation from 2011 to 2016
with the Barelands/Mined Area class decreasing by 51.7% and the vegetation class
increasing by 3.9%. There was also an increase in the settlement class by 87.3%. The
research concludes that the application of ANN be strongly encouraged for image
classification and mine reclamation monitoring activities and studies in the country
Description
A thesis submitted to The Department of Geomatic Engineering, Kwame Nkrumah University of Science and Technology in partial fulfilment of the requirements for the degree of Master of Science in Geomatic Engineering,