The mining sector faces a critical challenge. It needs to balance resource extraction with responsible environmental practices. This is more prominent in the face of climate change and growing concerns about environmental degradation. It is now essential to manage natural resources sustainably. Here, innovative approaches are urgently needed to ensure environmentally responsible mining operations.
One such approach lies in harnessing the power of remote sensing technology. This article explores how remote sensing can address a specific problem in the mining sector: automatic detection and monitoring of mining excavation pits, particularly aggregate quarries.
The satellite-based multispectral remotely sensed image comes in multiple bands. Each band corresponds to a particular wavelength region, such as Blue, Green, Red, Near-Infrared, Short-Wave Infrared, etc. These bands, or a subset from these bands, can be utilized for detecting various land cover elements, such as vegetation, surface water, barren land, etc. The same approach is applied to differentiate exposed soils and their surrounding land cover classes in a remote sensing image.
Exposed soils have a characteristically high reflectivity in the Short-Wave Infrared 2 (SWIR2) band of the electromagnetic spectrum. Most other land cover types lack this strong reflectance. Hence, it is beneficial to essentially utilize SWIR2 band for differentiating exposed soil from other classes. This has allowed for the development of specialized indices to distinguish exposed soils from surrounding areas. These indices typically leverage the difference between SWIR2 and visible channels and employ thresholding to arrive at crisp boundaries.
The Technical Challenge: Differentiating between spectral signatures
Unfortunately, a technical hurdle remains: the spectral response of exposed soils can be very similar to that of the actual mining pit itself. This spectral similarity makes it difficult to differentiate between the two using traditional single-thresholding techniques.
The Solution: A Multi-Faceted Approach
- Advanced Indices and Multi-Thresholding: We can get a more accurate picture of quarry boundaries by using data from the Sentinel-2 satellite and specially derived indices along with multi-thresholding methods instead of single thresholds. This multi-thresholding approach leverages the strengths of different spectral bands to create a more detailed picture.
- Including Synthetic Aperture Radar (SAR) Data: Combining Sentinel-1 and Sentinel-2 data allows for a more comprehensive analysis. SAR data can be integrated using data fusion techniques. The combined picture would use Sentinel-2’s spatial resolution to show edges and Sentinel-1’s texture data to tell the difference between spectral responses that are similar.
- Machine Learning and Computer Vision: Alternatively, computer vision algorithms can be trained to automatically detect quarries in satellite imagery. By feeding the algorithms with large datasets of labeled quarry images, they can learn to recognize the unique visual signatures of these sites.
Benefits
The successful implementation of these remote sensing techniques offers a multitude of benefits, including:
- Rapid Quarry Detection and Monitoring: Automatic detection allows for swift identification and monitoring of quarries, even in regions with limited control.
- Up-to-Date Quarry Records: Regular monitoring helps maintain current quarry records, facilitating responsible management.
- Identification of New Exploitations: New, unauthorized mining activity becomes easier to detect.
- Monitoring of Closed Sites: Abandoned or closed mines can be monitored for potential environmental issues.
- Cost and Time Savings: Remote sensing eliminates the need for extensive field work, saving time and resources for government agencies.
- Data for Analysis and Studies: The data obtained can be used to create crucial databases for further analysis and studies on mining practices and their environmental impact.
By embracing remote sensing technologies, the mining sector can move towards a more sustainable future, ensuring responsible resource extraction that minimizes environmental impact.
References
Jabłońska, K., Maksymowicz, M., Tanajewski, D., Kaczan, W., Zięba, M., & Wilgucki, M. (2024). MineCam: Application of Combined Remote Sensing and Machine Learning for Segmentation and Change Detection of Mining Areas Enabling Multi-Purpose Monitoring. Remote Sensing, 16(6), Article 6. https://doi.org/10.3390/rs16060955
López-Acevedo, F. J., Herrero, M. J., Escavy, J. I., & Peláez Fernández, M. A. (2024). Identification of Aggregates Quarries via Computer Vision Analysis as a Tool for Sustainable Aggregates Management and Land Planning. Sustainability, 16(8), Article 8. https://doi.org/10.3390/su16083099
Michałowska, K., Pirowski, T., Głowienka, E., Szypuła, B., & Malinverni, E. S. (2024). Sustainable Monitoring of Mining Activities: Decision-Making Model Using Spectral Indexes. Remote Sensing, 16(2), Article 2. https://doi.org/10.3390/rs16020388
