| Title | A deep learning approach to the detection of gossans in the Canadian Arctic |
| |
| Author | Clabaut, É; Lemelin, M; Germain, M; Williamson, M -C ; Brassard, É |
| Source | Remote Sensing vol. 12, issue 19, 3123, 2020 p. 1-16, https://doi.org/10.3390/rs12193123  Open Access |
| Image |  |
| Year | 2020 |
| Alt Series | Natural Resources Canada, Contribution Series 20200423 |
| Publisher | MDPI |
| Document | serial |
| Lang. | English |
| Media | paper; on-line; digital |
| File format | pdf; html |
| Province | Northwest Territories; Nunavut |
| NTS | 26; 27; 36; 37; 38; 39; 45; 46; 47; 48; 49; 55; 56; 57; 58; 59; 65; 66; 67; 68; 69; 75; 76; 77; 78; 79; 85; 86; 87; 88; 340; 560 |
| Area | Canadian Arctic Islands; Axel Heiberg Island; Baffin Island; Melville Peninsula; Southamnpton Island; Wager Bay; Baker Lake; Coronation Gulf; Napaktulik Lake |
| Lat/Long WENS | -116.0000 -64.0000 81.0000 63.0000 |
| Subjects | surficial geology/geomorphology; economic geology; geophysics; Science and Technology; Nature and Environment; gossans; mineral deposits; sulphide deposits; mineral exploration; exploration methods;
remote sensing; satellite imagery; mapping techniques; models; Methodology; Artificial intelligence; machine learning |
| Illustrations | schematic cross-sections; location maps; tables; schematic representations; flow diagrams; models; plots; satellite imagery; spectra |
| Program | GSC Central Canada Division |
| Released | 2020 09 23 |
| Abstract | Gossans are surficial deposits that form in host bedrock by the alteration of sulphides by acidic and oxidizing fluids. These deposits are typically a few meters to kilometers in size and they
constitute important vectors to buried ore deposits. Hundreds of gossans have been mapped by field geologists in sparsely vegetated areas of the Canadian Arctic. However, due to Canada's vast northern landmass, it is highly probable that many
existing occurrences have been missed. In contrast, a variety of remote sensing data has been acquired in recent years, allowing for a broader survey of gossans from orbit. These include band ratioing or methods based on principal component analysis.
Spectrally, the 809 gossans used in this study show no significant difference from randomly placed points on the Landsat 8 imageries. To overcome this major issue, we propose a deep learning method based on convolutional neural networks and relying
on geo big data (Landsat-8, Arctic digital elevation model lithological maps) that can be used for the detection of gossans. Its application in different regions in the Canadian Arctic shows great promise, with precisions reaching 77%. This first
order approach could provide a useful precursor tool to identify gossans prior to more detailed surveys using hyperspectral imaging. |
| Summary | (Plain Language Summary, not published)
Gossans are highly weathered, iron-rich soils overlying bedrock. These surficial deposits form by the alteration of sulphides by acidic and oxidizing
fluids. Hundreds of gossans have been mapped by geologists in sparsely vegetated areas of the Canadian Arctic. In this paper, we propose that a deep learning approach based on Geo Big Data can be used for the detection of gossans. This first order
approach to the remote predictive mapping of gossans provides a useful precursor tool for detailed surveys targeting critical minerals. |
| GEOSCAN ID | 327241 |
|
|