We all understand that mining has an adverse impact on the environment, which is why it needs to be objectively and continuously monitored. To this effect, keeping an eye on the dynamics of land use and cover is necessary. Spaceborne observations have made it simpler to detect any changes the landscape may have undergone over time.
This blog post presents a remote sensing-based approach to assess the spatiotemporal impact of mining activities on land use/land cover (LULC). The methodology leverages the time series of satellite imagery for LULC mapping across various timestamps, enabling change detection analysis over multiple decades. It involves the following:
Random Forest classification for LULC mapping
A supervised machine learning algorithm, Random Forest classification, can be employed to classify the satellite imagery. This approach involves training the algorithm on labeled data sets encompassing diverse LULC classes such as settlements, water bodies, active mining areas, vegetation cover, and bare land. The trained model can then be used to classify unknown pixels within the imagery, generating detailed LULC maps for each chosen timeframe.
Change Detection Analysis for Impact Assessment
By performing a comparative analysis of LULC maps generated from different time periods, we can identify and quantify land cover alterations attributable to mining activities. This analysis can reveal the spatial extent of changes, including the expansion of mining areas, degradation of vegetation cover, and conversion of other land cover types.
Quantifying Land Degradation
Change detection maps focusing specifically on vegetation and mining areas are crucial for assessing land degradation. We can determine the degree to which mining operations have negatively impacted land health by examining the decline in Normalized Difference Vegetation Index (NDVI) values within or close to active mining zones.
Advantages of the Remote Sensing Approach
- Extensive Spatial Coverage: Satellite imagery offers a synoptic view, enabling the analysis of vast geographical areas encompassing the entire mining landscape and its surrounding regions.
- Cost-Effective Data Acquisition: Compared to traditional field-based surveys, remote sensing data acquisition proves significantly more cost-effective, particularly for large-scale and diachronic (spanning time) analyses.
- Temporal Monitoring Capabilities: Regularly acquired satellite imagery facilitates the monitoring of LULC changes in near real-time, allowing for timely assessments and potential interventions.
Conclusion
The remote sensing-based methodology provides a robust and efficient approach for assessing the environmental impact of mining activities. The data obtained complements traditional field observations and environmental monitoring to offer a comprehensive understanding of mining’s footprint on the landscape.
References
Cudjoe, M.N.M., Kwarteng, E.V.S., Anning, E., Bodunrin, I.R. and Andam-Akorful, S.A., 2024. Application of Remote Sensing and Geographic Information System Technologies to Assess the Impact of Mining: A Case Study at Emalahleni. Applied Sciences, 14(5), p.1739.
Hong, F., He, G., Wang, G., Zhang, Z. and Peng, Y., 2023. Monitoring of Land Cover and Vegetation Changes in Juhugeng Coal Mining Area Based on Multi-Source Remote Sensing Data. Remote Sensing, 15(13), p.3439.
