Spatiotemporal studies of landscape pattern evolution in traditional villages are beneficial for addressing complex urbanization and global climate change. Using the traditional villages of Jiaxian and Linxian in the Jinshaan Gorge of the Yellow River Basin, this study employed a three-dimensional (3D) analysis involving three spatial scales (macro, meso, and micro), temporal scales (past, present, and future), and variables (humanity, society, and nature) based on the methods of spatiotemporal data analysis (SDA), geographic information system, remote sensing, and landscape pattern index (LPI) by Fragstats. On the macro scale, a significant turning point in ecological conservation awareness was indicated by LPI and SDA. Urban and rural construction land continuously increased because of urbanization. Plowland, grassland, and woodland were the main influencing factors in the evolution of rural settlements, with a 0.42% cumulative transformation rate. On the meso scale, the interactions and mutual promotion of mountain and aquatic environments, aquatic facilities, agricultural production, and cultural heritage have shaped the socioeconomic dimensions of evolution. On the micro scale, with urbanization, some traditional humanistic spaces have lost their original functions. A novel spatiotemporal-variable quantitative model explored the spatiotemporal evolution characteristics of human–land coupling, which can be used for the sustainable development of river basins worldwide.