1. An Exploration of False Positives Following Multispectral Analysis of Mine Tailings Extent.
- Author
-
Jewell, Daniel A., Campbell, Linda M., and White, H. Peter
- Subjects
- *
HISTORICAL maps , *CANADIAN history , *REMOTE-sensing images , *HISTORIC districts , *COASTAL wetlands , *ARSENIC - Abstract
Satellite imagery can be analyzed to offer a preliminary regional assessment of mine tailings indicators, enabling identification before performing in-depth fieldwork. Nova Scotia, Canada, still retains mine tailings produced in the 1860s to the 1940s in 64 historic gold districts, which exceed soil guidelines for arsenic (As) and mercury (Hg) levels. Tailings data often relies on historical maps, which may not accurately depict the current extent due to wind and rain transportation. A classification model was designed to analyze multispectral Sentinel-2 images and indicate pixels potentially comprising tailings. This classifier had an overall F1-Score about 0.7 for most methods tested, but this accuracy was not consistent across all land cover types. In particular, wetland and coastal areas seemed to generate a high number of false positives. In this study, these potential false positives are investigated to better understand the model's confusion and improve future results. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF