迈向滑坡数据标准：弥合滑坡易发性建模和预警系统的差距

Toward a Standard for Landslide Data Bridging Gaps in Landslide Susceptibility Modelingand Early Warning SystemsPublic Disclosure Authorized Priscilla NiyokwiringirwaTjark GallAbhas K. Jha Urban, Disaster Risk Management,Resilience and Land Global DepartmentFebruary 2026 Policy Research Working Paper11324 Abstract Landslides claim more than 4,000 lives annually and leadto approximately US$20 billion in economic losses. How-ever, landslide hazard, risk assessment, and early warningsystems remain constrained by fragmented, inconsistent,and incomplete data. This study addresses the global datagap by proposing a standardized, interoperable frameworkfor documenting landslide events across countries. Usingopen-access data and machine learning–based susceptibilitymodeling in Nepal, the paper assesses the limitations ofexisting inventories in terms of spatial resolution, temporalupdates, and missing attributes such as triggers, volumes,impacts, and soil-geotechnical properties that are criticalfor hazard prediction and risk modeling. These deficienciesreduce the accuracy and transferability of models, limiting the effectiveness of early warning and risk mitigationstrategies. To fill this gap, the paper proposes a tiered datastandard aligned with the International Organization forStandardization 19115, the Open Geospatial Consortiumstandards, and the Sendai Framework indicators, enablingscalable, consistent reporting of landslide events. The frame-work improves data completeness and model performanceand supports risk-informed decision-making. The WorldBank is well-positioned to operationalize this standardthrough its extensive portfolio of landslide mitigation anddisaster risk reduction programs. Institutionalizing such aframework can improve global coordination, reduce disasterlosses, and protect vulnerable communities. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about developmentissues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry thenames of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely thoseof the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank andits affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Toward a Standard for Landslide Data:Bridging Gaps in Landslide Susceptibility Modeling and Early Warning Systems Priscilla Niyokwiringirwa, Tjark Gall, and Abhas K. Jha JEL classification:Q54 Keywords:Landslide Risk Reduction; Disaster Risk Management; Disaster Resilience; EarlyWarning Systems 1.How Current Landslide Inventories Limit Resilient Development Landslidespose a serious global threat to lives,livelihoods,and infrastructure,withdisproportionate impacts in developing regions. Landslide types (including mudslides, deep-seated landslides, rock falls, and debris flows1) cause thousands of fatalities and billions ineconomic losses each year. Conservative estimates indicate annual losses of around US$20 billionand a cumulative death toll of over 110,000 people since 1900. Notably, the burden falls heavilyon lower-income countries in Asia, Africa, and Latin America, where vulnerable populations settleon steep slopes and disaster preparedness is limited.2For example, the 2017 mudslide inFreetown, Sierra Leone, devastated an informal hillside community and killed over 1,100 people.Tropical regions such as the Himalayas, the Andes, and Southeast Asia experience frequentlandslides triggered by intense monsoon rains and cyclones, conditions exacerbated by climatechange.3 In Asia alone, particularly along the Himalayan Arc, in China, India, and the Philippines, thousandsof fatal landslides have been recorded in recent decades. Communities in these areas often lackresilient infrastructure. Landslide disasters can wipe out roads, bridges, schools, and hospitals inan instant, undermining years of development progress.4–8 Landslides are often localized and triggered by a combination of factors, including intense rainfall,earthquakes, anthropogenic changes to terrain, and slow processes such as weathering andhydrological shifts.9–12In many regions, especially the Himalayan belt, Andean mountains, andSoutheast Asian hills, the risk of landslides is compounded by population growth, urbanexpansion into steep terrains, deforestation, and infrastructure development. When thesefactors converge, landslides can lead to disastrous consequences, wiping out entire communitiesand severing vital transportation and communication links.13,14 Effective mitigation strategies start with susceptibility mapping, hazard and risk assessment, andreal-time early warning systems that rely on high-quality landslide data. In this case, susceptibi