Application of remote sensing technique for developing of quality assessment model for the reclamation areas of Mae Moh mine

Abstract

Mae Moh lignite coal mine, the most extensive open-pit mining in Thailand, has undergone significant land use changes, transforming forested areas into mining and dumping sites. This transformation has notably altered the forest ecosystem, necessitating the implementation of mine rehabilitation activities in tandem with ongoing mining operations. These activities aim to restore the environment to a state that closely resembles its original condition. The monitoring and assessment of the quality of the reclaimed areas are of paramount importance. However, the current methods for field data collection, which predominantly rely on manual labor and are time-intensive, present considerable challenges due to the vast expanse of the reclamation areas. This thesis is dedicated to exploring methodologies for the qualitative assessment of reclaimed areas, emphasizing minimizing the time required for such evaluations and facilitating assessments across mining areas. Remote sensing has been applied through the analysis of aerial photography, which is rapidly sourced from satellites or cameras with multispectral sensors. Utilizing the spectral characteristics inherent in photos. The research investigates the correlation with the Vegetation Index, effectively distinguishing between areas with and without biomass coverage. The Normalized Difference Vegetation Index (NDVI) is employed within the study area for this purpose. Areas abundant in vegetation are characterized by a higher reflectance in the Near-Infrared (NIR) spectrum as opposed to the red (RED) spectrum within the visible range. The NDVI values are instrumental in indicating fluctuations in vegetation cover, thereby serving as a valuable tool in monitoring both the augmentation and diminution of vegetation and assessing the prevailing environmental conditions. The research found that aerial photographs obtained through Unmanned Aerial Vehicles (UAVs) provide more precise Normalized Difference Vegetation Index (NDVI) values compared to data from Landsat-8 and Sentinel-2 satellites. The coefficient of variation (%RSD) in NDVI ranged from 16.58% to 40.78%. To establish a correlation with NDVI values, various climatic variables were considered and categorized based on Thailand's primary seasons, consisting of the rainy and dry seasons. These climatic variables encompassed parameters such as rainfall, relative humidity, temperature, elevation, and other vegetation indices like the Leaf Area Index (LAI) and the Soil Adjusted Vegetation Index (SAVI). Among the derived equations, the one representing teak trees during the dry season exhibited the highest R-squared value, at 0.955. This equation can be expressed as follows: NDVI = 1.865SAVI - 0.229LAI + 0.001R + 0.003RH + 0.005T - 0.272. This equation holds the potential for estimating above-ground biomass and evaluating carbon sequestration in forest restoration areas, thereby aligning with the goals of achieving carbon neutrality and net zero emissions as outlined by the Electricity Generating Authority of Thailand (EGAT) for the year 2050.