From theory to practice: optimisation of available information for landslide hazard assessment in Rome relying on official, fragmented data sources

Abstract

The definition of landslide hazard is a step-like procedure that encompasses the quantification of its spatial and temporal attributes, i.e., a reliable definition of landslide susceptibility and a detailed analysis of landslide recurrence. However, available information is often incomplete, fragmented and unsuitable for reliable quantitative analysis. Nevertheless, landslide hazard evaluation has a key role in the implementation of risk mitigation policies and an effort should be done to retrieve information and make it useful for this purpose. In this research, we go through this topic of optimising the information available in catalogues, starting from landslide inventory review and constitution of a boosted training dataset, propaedeutic for susceptibility analysis based on machine learning methods. The temporal recurrence of landslide events has been approached here either through the definitions of large-scale quantitative hazard descriptors or by analysis of historical rainfall (i.e., the main triggering factor for the considered shallow earth slope failures) databases through the definition of rainfall probability curves. Spatial and temporal attributes were integrated, selecting potential landslide source areas ranked in terms of hazard. Data integration was also pursued through persistent scatterer interferometry analysis which pointed out areas of interest within potential landslide source areas featured by ongoing ground movement. The consequential approach led to the definition of the first hazard product of the city of Rome at a local scale functional for advisory purposes or the statutory level, representing a thematic layer able to orient the risk managers and infrastructure stakeholders.

The Prediction of Cross-Regional Landslide Susceptibility Based on Pixel Transfer Learning

“…Distance to tectonic lines (Figure 4f) is a commonly used geological influencing factor for landslide susceptibility assessments [29]. Within a certain range, rocks closer to the tectonic line have higher discontinuity and looser soil bodies and so are more likely to lead to landslides.…”

The Open Landslide Project (OLP), a New Inventory of Shallow Landslides for Susceptibility Models: The Autumn 2019 Extreme Rainfall Event in the Langhe-Monferrato Region (Northwestern Italy)

“…Regarding the response variable, we extracted the highest point along the crown (LIP-Landslide Identification Point) from each polygon in our landslide inventory [65,69,70] and assigned a value of 0 or 1 to each pixel based on the absence or presence of landslides, respectively.…”

Analysis of Geological Multi-Hazards in an Urban District

“…Many studies have focused on hazard analysis in the city of Rome, some at the municipal scale [21][22][23][24][25][26], while others are site-specific, single-hazard analyses [27][28][29][30][31][32]. In particular, the susceptibility to pluvial floods and sinkholes has been explored in urban areas and in the Italian territory by many authors [33][34][35][36][37][38][39][40][41].…”