MUSUBOU-AR: An Open-Source Geospatial AR Framework for Integrating Public GIS Data, Field Authoring, and Disaster Walking Tours

Abstract. Conventional hazard maps, public GIS layers, and disaster education materials are essential resources for community disaster risk reduction, but they do not always support in-situ spatial understanding during field-based training. This paper presents MUSUBOUAR, a publicly released open-source geospatial augmented reality framework that links public GIS resources, local scenario authoring, mobile AR visualization, and post-activity review for disaster walking tours. The system combines a map mode based on public and custom XYZ tiles, an AR mode for location-based visualization of hazards and disaster-related field resources, a webbased authoring environment for points, routes, and GIS layers, GPX-based field logging and review, and optional LiDAR-enhanced AR visualization for improving environmental alignment and ground-surface-aware rendering. As a representative field deployment, the framework was used in a joint disaster walking tour in the Hiro area of Kure City, Hiroshima Prefecture. In this paper, the case is used primarily to examine the field-deployment workflow of MUSUBOU-AR, including public hazard-map use, route and point authoring, AR scenario configuration, and on-site operation. A previously reported facilitator survey provides supplementary evidence that the activity supported situated geospatial interpretation among university student facilitators. Although the supporting survey evidence was limited by a small sample and no control group, the results suggest that MUSUBOU-AR can support reusable geospatial workflows for field-based disaster risk communication.