Recent progress in the use of documentary data in landslide research: a review of approaches, sources, and methods

Landslides cause severe impacts on society, infrastructure, and the environment globally, and their occurrence in some regions is expected to rise due to climate change. Although the cumulative impacts of landslides do not reach the level of earthquakes or floods, their disperse occurrence in space and difficult prediction pose a fundamental challenge for landslide disaster risk reduction effort. Clearly, accurate information is needed both for understanding spatiotemporal occurrence of landslides and their social impacts and responses held by societies. Documentary data are among the key sources that enable compilation of regional landslide databases, allow to quantify the landslide impacts and describe both quantitatively and qualitatively causal chains leading to increased landslide risk and the societal responses to landslide events. In this respect, the documentary data fill the time gap between the landslide occurrence in the past environments studied by proxy data, and the present-day landslides, for which different monitoring and mapping techniques may be used. Over the last decades, important progress has been made in employing various documentary data for landslide research, and extending empirical evidence about advantages and limitations is available thanks to case studies from different environmental and institutional settings. The synthesis of this progress that would guide further research is missing though. The overall goal of this paper is to broaden the perspective on the use of documentary data in historical landslide research, which has so far too much concentrated around the landslide inventories. To do so, we present a scoping literature review with three main objectives. First, we present a classification of both quantitative and qualitative approaches and related research questions in historical landslide research, linking them to key challenges in landslide disaster risk reduction. Second, we review the types and content of available documentary data sources with special attention paid to sources that have been underresearched so far. Finally, we review the quantitative and qualitative methods used to analyse the content of documentary data. While doing so, we draw also from comparative evidence in historical climatology and hydrology in order to point to methods that may hold a potential, but have not been validated in landslide research yet. The paper concludes with identifying challenges and pathways for future research.