Charles Explorer logo
🇬🇧

Land cover changes in the extent of surface mining and reclamation in Northwest Bohemia using a 1984-2017 landsat time series

Publication at Faculty of Science |
2019

Abstract

Surface coal extraction and reclamation of disturbed land represent the dominant driver of land-cover changes in Northwestern Bohemia. Over many decades, the forested and agricultural landscape has been changed into an industrial one and the central area almost completely lacks environmentally sustainable features.

In the last few decades as the number of reclamation projects has increased, the diversity of the landscape has also increased, as has the proportion of ecologically valuable elements, such as forests, meadows and water bodies, which can approximate the original landscape with a greater degree of ecological stability. More accurate quantification of the extent of mining activities and reclamation projects is important for exploration of environmental impacts on the ecosystem.

The Landsat images for the 1984 - 2017 period were used to map the extent of surface mines, with special emphasis on reclamation of selected sites. Land cover from former reclaimed mines is often converted to forest and agricultural land, and also to water bodies.

Image processing techniques extended by the calculation of vegetation indices (NDVI, EVI, NDMI, SAVI and MSAVI) are used to track landcover changes over several decades. The values of the vegetation indices from 18 Landsat images were calculated for each cell of the selected sites.

The sites represent a sample of artificial forestation (36 cells) and a sample of spontaneous vegetative growth (100 cells). Based on model calculations, time changes of vegetation indices related to artificial forestation indicate lower variability compared to time changes related to spontaneous vegetative growth.

Finally, NDVI growth curves for each site are used for model predictions based on nonlinear regression. The contribution also describes GIS and MATLAB capabilities for assessment and modeling of land-cover changes, and their utilization in decision-making procedures for similar regions. (C) 2019 Nova Science Publishers, Inc.