Postdoctoral Fellowship: EAR-PF: Erosion on the Mountaintop Removal Coal Mining Landscapes of Central Appalachia

A massive transition to renewable energy is underway. This transition is vital for avoiding the worst impacts of climate change. Even so, infrastructure from the fossil fuel past will remain on the land long-term.

The mountaintop removal coal mines in Central Appalachia have an uncertain future in terms of how they will erode, potentially causing hazardous conditions for property, people, and animals. This research project will address this problem by developing a way to automatically detect landslides and other forms of erosion, giving the first regional understanding of the degradation of former mountaintop removal coal mines.

Also, for the first time, this research project seeks to understand whether these former mined lands are eroding more quickly than unmined areas in the same region, which is an important consideration for both stream habitats and downstream infrastructure. Not only will the project answer these questions of regional interest but advance the techniques scientists use to study the long-term impacts of human disturbance on the environment.

The programming code associated with this project will be shared through the CodeWV program, which provides high-quality training and curriculum materials to secondary-school teachers and students in underserved communities in West Virginia. Also, through the WVUteach network, the research will be shared in secondary-school classrooms, highlighting the connection between Earth science and computer science.

In the United States, the geomorphic impacts of mountaintop removal coal mining (MTR) and the resulting ‘reclaimed’ landscapes have garnered limited scientific attention from within the surface processes community despite the area impacted (~5900 km2), encompassing portions of West Virginia, Kentucky, Virginia, and Tennessee. These anthropogenic landscapes can degrade over time via natural processes, such as gullying and landsliding, overtaking and incorporating designed drainage systems.

This research project will document geomorphic impacts at both the process and regional scale through leveraging big data repositories of high-resolution, multi-temporal, lidar-derived topographic data. The first task will be the regional-scale topographic differencing of lidar data for MTR-impacted areas to observe process-scale erosion and deposition.

The second task will be to leverage these observations for the training of deep-learning models which will be applied to MTR-impacted areas with only a single generation of lidar coverage. The third task will be a field campaign to obtain cosmogenic radionuclide concentrations in stream sediments from mined and unmined catchments to understand the alteration in sediment dynamics and to parameterize long-term landscape evolution models.

As similar engineering practices used to design reclaimed MTR landscapes will most likely be utilized when reclaiming surface mines generated during the anticipated boom in critical minerals used for the transition to renewable energy, the methods employed in this work and the resultant insights could be applied to these future anthropogenic landscapes. More generally, the project could lead to fundamental insights on optimal deep-learning inputs and architectures with regards to automated geomorphic mapping.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.