Period Change Analysis on Abandoned Hobet Mine (West Virginia)

Using Landsat satellite imagery spanning 35 years (1985-2020), we analyzed landscape changes at West Virginia's abandoned Hobet Mine. Through unsupervised classification and change detection, we documented a 13.3% decrease in mine area as natural revegetation occurs. This analysis helps understand post-mining recovery rates and can inform restoration prioritization decisions across Appalachia.

North america has experienced economic development accompanied with an inherent need of raw materials for fabrication and industry. Coal is one of these resources, which can be found in abundance in Appalachia. Amongst the methods of coal extraction, mountain-top removal is a very efficient method of natural resource extraction, with the draw of considerable environmental impact to the area.
Using Geographic Information Systems (GIS), I aimed to better understand the extent the environmental impact these kinds of natural extraction operations would have on the land. Improving our understanding of the environmental impact of different approaches to natural resource extracting is essential to optimize mining operations while minimizing the impact of this industry on the local landscape.

For this project, we used Landsat 7, 8 and 9 imageries. The date range used spans from 1985 to 2020. Landsat was the best imagery source for this project because it provided high spatial resolution as well as vast past data, which was essential for this project.

Start:

1985-01-01 → 2005-12-31

End:

2014-01-01 → 2020-12-31

Annual Time Series Graph

An annual time series graph was generated in order to understand the difference between bands values over time. The year chosen for this annual time series graph was 2019-2020.

False Color Image Composite

A false image color composite of the 'before' (1986-07-05) and 'after' (2020-12-09) were generated to better understand the change in the major features of the study area. LS5 and LS8 were used for these composites (Path = 19; Row = 34). An infrared (CIS) false image composite was chosen.

Two Period Change Analysis (with unsupervised classification)

We had to reclassify spectral classes generated by the unsupervised classification model into informational classes (Forest1 = 100, Forest2 = 101 and Mine = 102). Once this was done. The informational classes Forest1 and Forest2 were merged. The area of these informational classes (Forest and Mine) for the 'before' and 'after' of the two-period change analysis. These area values can be found in Table 1. The map denoting the change between the 'Start' and 'End' can be found in Figure 5.

Annual time series analysis of study area (Figure 2) shows that the vegetation and the mine do not overlap in terms of NDVI. The false color composite (Figure 3). Shows little change in the study site.

Table 2 suggests that after the abandonment of the Hobet mine the area of the mine decreased by 13.3 % in area. Over time, nature has slowly taken the mine .

We believe exploring the literature on the past ecology of the area and accounts of the changes that may have taken place over the operation of the miens and conducting new ecological surveys might lead us to better understand how reduction of habitat for certain species might affect their populations.

Better understanding the change in the landscape of the Hobet mine is imperative to decide if restoration of the area is needed and it might lead to which approaches might be best. Due to the high costs of restoration and the reduced impact of the former mine to the landscape every year, it may not be a site of interest for immediate restoration, however, understanding other variables such as chemical contamination might change this conclusion. GIS is not the whole picture of this site. .