Objective The pressure of accelerated urbanization has caused rural landscapes to face the great challenges of reconstruction, homogenization and extinction, with rural landscapes gradually losing their diversity and regional characteristics. Compared with lowland villages, mountainous villages are constrained by special topography, rich in resources but backward in construction, and have always been the key and difficult areas in China’s rural revitalization work, and the identification and management of their landscape characters are particularly urgent. As a complex landscape system with multiple levels and diversity, the mountainous villages need a more comprehensive cross-scale tool to guide and manage the process of landscape change and the protection of local landscape characters. Landscape character assessment (LCA) is a mainstream method to identify landscape character, but its results often present a static, nodal and objective perception. The collective memory theory focuses on the dynamic characteristics of landscape evolution and is able to identify more subtle differences in landscape characters, compensating for the shortcomings of LCA in the temporal and human dimensions. Therefore, this research aims to 1) integrate LCA and collective memory theory to construct a theoretical and technical framework for multi-scale identification of mountainous rural landscape characters; 2) integrate quantitative analysis of LCA and qualitative interpretation of collective memory to solve the problem of spatialization of narrative memories; and 3) validate the scientific validity and effectiveness of the identification system through empirical evidence in the Wuling Mountain area.

Methods The research proposes to construct a multi-scale mountainous rural landscape character identification system. 1) Theoretical framework: Integrate LCA and collective memory theory to establish a systematic framework of “definition scope − element selection − classification and description − practical application”, realizing the layer-by-layer cognition and description from macroscopic to microscopic perspective. 2) Technical path: Establish the operational path of “data collection − character recognition − character translation”. The “extraction of memory carriers − spatial coordinate localization − generation of vector data” approach is used to realize the spatialization of collective memory. At the regional scale (Chongqing section of the Wuling Mountain area), natural elements such as elevation, and land cover are selected to identify landscape character zones. At the local scale (Qianjiang District), cultural elements such as land use, and characteristic villages are selected to identify landscape character categories. At the site scale (Taiji Township), traditional settlements, collective memory, and other cultural and culturally related elements are selected to identify landscape character units and elements. At the same time, the surviving condition of the character elements is marked.

Results The complexity and cultural diversity of the geographic environment in the Chongqing section of the Wuling Mountain area provide an ideal background for character identification. A multi-scale mountainous rural landscape character identification system is constructed and successfully applied to different scales (Chongqing section of the Wuling Mountain area − Qianjiang District − Taiji Town), which provides a regional reference for the identification and protection of mountainous rural landscape character. The identification system identifies 10 rural landscape character zones in the Chongqing section of the Wuling Mountain area at the regional scale, 8 rural landscape character categories in Qianjiang District at the local scale, and 6 rural landscape character units in Taiji Towns at the site scale. Cases such as the memory of natural beliefs (land temple) and the memory of the construction of public facilities (academy ruins) reflect the deep coupling of landscape character and the collective memory of the villagers. At the same time, marking the surviving condition of the character elements (including the disappearing elements) helps to explain the historical development of landscapes.

Conclusion Mountainous rural landscapes have both natural and humanistic attributes, and public participation and raising awareness of the cultural value of landscapes is a key part of rural landscape planning. This research analyzes the theoretical framework, technical path and practical application of the identification of mountainous rural landscape characters at multiple scales, and draws the following main conclusions. 1) The technical process of LCA can provide a spatial anchor point for collective memory, while collective memory can inject an interpretive dimension into LCA, and the two are complementary to help understand rural landscapes in a more comprehensive way. 2) It realizes the deep integration of the quantitative analysis of LCA and the qualitative interpretation of collective memory, clarifies the path of memory carrier materialization, and solves the problem that narrative memory can hardly interface with LCA technology. 3) The case of Wuling Mountain area proves that the identification system can effectively integrate geographic diversity, cultural characteristics and villagers’ perception, and realizes precise identification of landscape character through the synergistic collaboration of top-down (systematic delineation dominated by LCA) and bottom-up (local cognition driven by collective memory) identification, and provides a tool support for the national rural landscape planning and protection. In the future, the results of character identification should be transformed into planning actions through differentiated planning decisions, dynamic protection mechanisms, participatory governance, etc., so as to respond to the real dilemmas of rural planning.