As part of China's northern ecological fragile zone, the desert regions of Inner Mongolia face severe challenges from wind erosion, desertification, and the sustainable development of agro-pastoral systems. Conservation tillage practices - such as reduced tillage, no-tillage, and straw mulching - play a vital role in improving the ecological conditions of these areas. The integration of remote sensing (RS) and geographic information systems (GIS) provides essential technical support for evaluating their ecological impact. However, current research faces limitations including inadequate spatial resolution in traditional methods, poor adaptability of RS change detection algorithms to the complex spectral characteristics of desert environments, and a tendency to assess ecological impact using single-factor analyses. These issues hinder a comprehensive understanding of the overall effects of conservation tillage on agro-pastoral ecosystems. This study leverages GIS and RS technologies to address three core objectives: (1) to analyze the spatial distribution of conservation tillage practices in Inner Mongolia's desert areas and examine their correlation with topography and climate; (2) to apply the Fast Bag-of-Visual-Words (BOVW) algorithm to enhance the accuracy and efficiency of change detection in conservation tillage regions using RS imagery; and (3) to comprehensively assess the impacts of conservation tillage on desert agro-pastoral ecosystems from multiple dimensions, including soil properties, vegetation growth, water resource utilization, and ecosystem services. The study establishes a technical evaluation framework tailored to arid and semi-arid regions, providing a scientific basis for optimizing agricultural production layouts and ecological protection in Inner Mongolia, and offering methodological insights for sustainable development in similar regions.
Song et al. (Sun,) studied this question.