Abstract: Based on five phases of Landsat satellite remote sensing imagery from 2000 to 2020, this study analyzed the spatial and temporal characteristics of land use and landscape patterns of giant panda habitats in the Xiangling area of southwestern China using the land use transfer matrix and the landscape pattern index; and quantified the impacts of human socio-economic development on the dynamics of landscape patterns of the habitats using grey correlation analysis. The results show that the total area of the forest patches in the giant panda habitat in this area experienced a process of "decline - growth - decline - growth" in the past 20 years. The actual habitat area of giant pandas decreased, while the potential habitat area increased. In 2000—2015, the landscape pattern within the habitat changed drastically, with large-scale shifts between woodland and built-up and unused land, and increased landscape fragmentation; in 2015—2020, there was less change in the landscape pattern within the habitat with the gradual conversion of other land types to woodland, a significant increase in the area of woodland, and a reduction in the degree of landscape fragmentation. Population growth, agricultural farming, grazing scale, energy transition, and transport roads are important factors by which of socio-economic development affects changes in habitat landscape patterns, and strongly contribute to the evolution of habitat landscape patterns. Increased control of anthropogenic socio-economic disturbance can promote habitat restoration and increase the actual area of future habitats. Finally, based on the results of the study, recommendations are provided for the future conservation and management of giant panda habitats in the Xiangling area.