Author ORCID Identifier

https://orcid.org/0000-0002-4443-4631

Semester

Spring

Date of Graduation

2026

Document Type

Dissertation

Degree Type

PhD

College

Davis College of Agriculture, Natural Resources and Design

Department

Not Listed

Committee Chair

Stefania Staniscia

Committee Member

Michael Strager

Committee Member

Aaron Maxwell

Committee Member

Vincenzo Cribari

Abstract

The objective of this dissertation was to develop a comprehensive, data-driven spatial framework for characterizing the complex cultural landscape of the National Coal Heritage Area (NCHA) in West Virginia. By transitioning away from traditional, heuristic spatial mapping, this research integrates advanced spatial statistics, machine learning, and GIS-based methodologies to objectively quantify the physical, visual, and cultural dimensions of the post-mining environment. The research is structured around three interconnected empirical studies, each addressing a specific scale of the Landscape Character Assessment (LCA) framework to support heritage conservation and sustainable spatial planning. The first paper focused on landform classification, developing an automated deep learning (DL) framework to detect and delineate relict surface mining features obscured by dense forest canopy. Utilizing high spatial resolution light detection and ranging (LiDAR)-derived land surface parameters (LSPs), the study compared different DL (U-Net) configurations. The results revealed that, with optimization, a simpler Base U-Net architecture successfully bypassed the ‘complexity trap,’ outperforming deeper network variants (such as ResU-Net and DiU-Net) to achieve an overall accuracy of 81.4% in identifying relict surface mining features. The second paper shifted to the perceptual dimension to evaluate the route-based visibility of historic features along the Coal Heritage Trail within the NCHA. Based on the principle of visual magnitude (which measures how large and dominant an object appears to an observer), it introduces the Historic Visibility Index (HVI), a novel spatial metric that mathematically fuses a structure’s physical mass, historical significance, and an inverse-square distance decay function to quantify the true visual prominence of heritage assets. This establishes a measurable, data-driven indicator for formal Visual Impact Assessments (VIA), equipping planners to protect and manage historic visualscapes within complex topography. The third paper synthesized multi-layered environmental, geomorphic, and cultural variables to objectively classify the region. By applying a spatially constrained machine learning algorithm, Geographical Self-Organizing Maps (Geo-SOM) combined with Hierarchical Agglomerative Clustering (HAC)- the research successfully delineated 10 distinct Landscape Character Types (LCTs). This approach explicitly enforced spatial contiguity, capturing the intricate natural and industrial dichotomy of the Appalachian landscape without the fragmentation caused by traditional clustering methods. In conclusion, this dissertation establishes a reproducible, objective baseline of the regional landscape, a necessary foundation to justify conservation efforts and guide physical planning. The research identifies this quantitative characterization as the essential first step in a broader, iterative process. While true landscape quality is inextricably bound to human perception, this data-driven framework provides the analytical framework that enables future enrichment through humanistic, experience-based analysis and participatory community engagement, eventually capturing and sustaining the full cultural realities of historically industrialized environments.

Download

Share

COinS