Geological and meteorological disasters such as landslides, debris flows, avalanches, and extreme weather events have significant impacts on human society. These disasters not only threaten lives and property but also damage transportation infrastructure, affecting the delivery of relief supplies and evacuation of personnel. In disaster-prone areas, road networks are often sparse due to geological and climatic constraints, characterized by long distances between road segments, few traffic nodes, and limited alternative routes. Constructing an effective dual-layer topological model for sparse road networks in such regions is crucial. This study proposes a dual-layer topological model comprising a backbone layer and a local layer. Using directed graphs to represent the topology of sparse road networks, combined with complex graph theory and dynamic attribute data, the model accurately describes the state of the road network and its changes under disaster conditions. The model aids in disaster risk management and emergency route planning, providing robust tools and perspectives for sustainable development and safety in disaster-prone areas. Case studies validate the model’s effectiveness in managing and responding to geological and meteorological disasters.