Abstract

Identification of traditional flood control facilities concealed in the riparian forest: a case study of the Echi River, central Japan

Unprecedented climate change has intensified flooding globally, with Japan experiencing frequent annual flood events. Traditional flood control facilities (TFCFs), levees made of stone, have historically played a crucial role in mitigating flood risks, supporting livelihoods, and enhancing local ecosystems. However, urbanization and the loss of floodplain functionality have contributed to the physical degradation and erosion of TFCFs. The rediscovery and evaluation of these structures are crucial for preserving cultural heritage, informing climate adaptation strategies, and enhancing disaster resilience. This study develops an integrated methodology to identify and analyze the Saruo structure, a type of TFCF, in Shiga Prefecture, Japan. A multidisciplinary approach was employed, integrating UAV-LiDAR, UAV-SfM, old maps, field surveys, and interviews. A high-definition digital terrain model (DTM) was developed to detect convex landforms consistent with Saruo. An 1874 historical map confirmed their spatial distribution, while field surveys and interviews provided complementary insights into their physical characteristics and historical functions. The findings revealed that some Saruo structures have undergone significant transformations owing to flooding, vegetation encroachment, and urban development, while others have persisted despite environmental changes. Additionally, discrepancies between residents’ recollections and modern landscapes highlight the limitations of memory-based reconstruction. Quantitatively capturing historical changes highlighted changes in the dimensions of Saruo owing to factors such as flooding, vegetation growth, and urban development. The results underscored the limitations of relying on individual methods while demonstrating the strengths of an integrated, multidisciplinary approach. This study emphasizes the potential of combining advanced remote sensing technologies with historical and community-based knowledge to rediscover and reinterpret forgotten TFCFs. Furthermore, this approach has broader applications for flood risk assessment, disaster education, and cultural heritage conservation.