Bare soil detecting algorithms in western iran woodlands using remote sensing

Abstract

This study presents the Modified Bare Soil Index (MBI), an innovative remote sensing instrument that employs Short Wave Infrared (SWIR) and Near Infrared (NIR) wavelengths sourced from Landsat 8 satellite observations. Its purpose is to refine the identification and distinction of bare soil within woodland regions. Addressing the limitations of existing methods, MBI represents a significant advancement in land cover change detection. Through a focused study in the Sardasht forests of western Iran, the MBI demonstrates superior accuracy, achieving an overall accuracy of approximately 96 % with a Kappa coefficient exceeding 0.94, surpassing traditional bare soil indices. This innovation improves the classification of land surfaces and also aids in the efficient monitoring and safeguarding of changes in land cover. The MBI's robust performance underscores its potential in smart forestry, enabling more precise identification and surveillance of bare soil areas. This advancement is pivotal for ecological studies and resource management, offering a practical solution to the challenges of soil erosion and land degradation. By integrating advanced spectral analysis with satellite imagery, the MBI sets a new standard in remote sensing techniques, offering vital insights for environmental conservation and agricultural practices.