Enhancing visual attribute comprehension of urban heritage landscapes using combined GIS-based visual analysis methods

Abstract

Urban heritage landscapes, with their layered cultural and aesthetic values, require precise visual analysis to support conservation and planning. However, existing visual analysis methods are often fragmented and fail to fully capture their complex visual-spatial characteristics. To address this gap, this paper proposes a combined visual analysis framework that integrates four GIS-based visual analysis methods—cumulative viewshed (CV), visual magnitude (VM), field of view (FOV) analysis, and street-view image (SVI) segmentation. These methods were applied to the UNESCO World Heritage Site of West Lake in Hangzhou, China, to explore lake–city–landscape relationships, classify lakeside landscape types, and interpret the spatial composition of iconic viewpoints. Findings indicate that: (a) four zones with both high CV and VM values coincide with key architectural and scenic landmarks, suggesting intentional spatial design strategies, while half of the “Ten Scenic Places” are influenced by symbolic or experiential factors beyond visibility; (b) 37 landscape types were identified along lakeside roads, revealing areas where vegetation obscures potential lake views and where design trade-offs are evident; (c) only two of ten potential city-to-lake visual corridors remain unobstructed, pointing to unmanaged vegetation as a critical barrier; and (d) these insights inform targeted visual management strategies, including vegetation control, viewpoint activation, and circulation optimization. This study highlights the limitations of single-method approaches, such as SVI's insensitivity to topographic variation, and suggests that a multi-perspective integration of VAMs can yield deeper spatial insights and more actionable guidance for managing urban heritage landscapes.