Environmental degradation is a significant concern in Moroccan mining sites due to their impact on air quality, water resources, and soil health. A mixed-methods approach was employed, integrating laboratory testing with field deployment of prototype IoT systems. Sensor calibration procedures were standardised using a linear regression model for pollutant concentration prediction (y = β0 + β1x + ε, where β0=5 μg/m³, β1=2 μg/m³). The uncertainty in sensor readings was estimated to be ±3%. Field trials indicated that the sensors detected a 70% higher concentration of particulate matter compared to traditional manual sampling methods, highlighting their superior detection capabilities. The developed IoT systems have demonstrated robust performance and reliability under field conditions, providing valuable insights for environmental management in mining sites. Further research should focus on integrating machine learning algorithms to enhance predictive analytics and expand the network coverage across different mine environments.
Abdelaziz Benmoussa (Mon,) studied this question.