Taxonomy-driven exposure mapping for earthquake risk assessment in the active tectonic setting of Eastern Indonesia

Abstract Background Eastern Indonesia is located in a highly tectonically active region and remains vulnerable to significant earthquake impacts. However, sector-specific exposure and loss assessments incorporating building taxonomy are still limited. This study develops a taxonomy-based exposure and seismic loss assessment framework to support risk-informed mitigation planning. Methods Spatial datasets of educational facilities, healthcare infrastructure, and residential settlements were integrated with asset valuation and construction typologies adapted from the GED4ALL framework. The assessment was conducted under 100-, 200-, and 500-year earthquake return period scenarios. Building classifications include Masonry Unreinforced (MUR), Masonry Confined (MCF), Light Wood (LW), and reinforced concrete structures. Results The results reveal pronounced spatial and sectoral variability in structural vulnerability, primarily influenced by building taxonomy and proximity to major fault systems. Educational facilities in Central and North Sulawesi are predominantly MCF, whereas Maluku and North Maluku exhibit a higher proportion of more fragile MUR structures. In the healthcare sector, community health centers in West Papua and Southwest Papua are largely LW, while reinforced concrete buildings are concentrated in urban hospitals. Residential settlements in rural and coastal areas remain dominated by non-engineered MUR, MCF, and LW typologies. Economic losses increase substantially with longer return periods. Under the 500-year scenario, educational facilities incur the highest losses, particularly in North Sulawesi (USD 346 million), Maluku (USD 316 million), and Central Sulawesi (USD 306 million). Conclusions The findings confirm that building taxonomy is a critical determinant of seismic vulnerability and loss distribution across sectors and regions. The integration of spatial exposure data with taxonomy-based structural classification enables more precise identification of high-risk provinces and infrastructure types. This approach provides a robust analytical basis for prioritizing retrofitting strategies, allocating disaster risk reduction investments, and strengthening earthquake resilience in Eastern Indonesia.