Including informal housing in slope stability analysis – an application to a data-scarce location in the humid tropics

Abstract

<jats:p>Abstract. Empirical evidence from the humid tropics shows that informal housing can increase the occurrence of rainfall-triggered landslides. However, informal housing is rarely accounted for in landslide hazard assessments at community or larger scales. We include informal housing influences (vegetation removal, slope cutting, house loading and point water sources) in a slope stability analysis. We extend the mechanistic model CHASM (Combined Hydrology and Stability Model) to include leaking pipes, septic tanks, and roof gutters. We test CHASM+ in a region of the humid tropics, using a stochastic framework to account for uncertainties related to model parameters and drivers (incl. climate change). We find slope cutting to be the most detrimental construction activity for slope stability. When informal housing is present, more failures (+85 %) are observed in slopes with low landslide susceptibility and for high intensity, short duration precipitations. As a result, the rainfall threshold for triggering landslides is lower when compared to non-urbanised slopes, and comparable to those found empirically for similar urbanised regions. Finally, low cost-effective low regrets mitigation actions are suggested to tackle the main landslide drivers identified in the study area. </jats:p>

A mechanistic approach to include climate change and unplanned urban sprawl in landslide susceptibility maps

“…In this paper, we therefore suggest a new method that smartly uses a fast hillslope-scale mechanistic model to generate regional susceptibility maps under a wide range of climate change and -for the first timeinformal urban expansion scenarios. In a previous study (Bozzolan et al, 2020), we have shown that the joint effect of different climate and localised urban construction activities can lead to significantly different slope stability responses. We did this by using a mechanistic model that can represent dynamic hydrological changes due to changes in climate (rainfall), vegetation and localised urban construction activities.…”

“…properties are described by 32 input factors, which were all varied according to their probability distribution (Table 1). The vegetation properties (defined in Table 2) were kept fixed (e.g., root depth always 1 m), whereas urban properties were treated as random variables (bottom of Table 1) in accordance with the analysis performed by Bozzolan et al (2020). The soil properties are described for the three soil layers considered: weathered residual soils and colluvium overlying decomposed rock and volcanic bedrock (as typical in the study site, see section 2).…”

“…In particular, vegetation is represented through rainfall interception, evapotranspiration, root water uptake, vegetation surcharge, and increased permeability and soil cohesion due to the root network (see Wilkinson et al, 2002b). The new extended version CHASM+ (Bozzolan et al, 2020) also includes surface urban water management, through the representation of roof gutters on houses, leaking superficial pipes and buried septic tanks.…”

“…Through Monte Carlo simulations, we generated a large library of synthetic hillslopes representative of the study area (the Caribbean island of Saint Lucia) and we analysed their stability under a wide range of different rainfall, vegetation and urban conditions. Differently from Bozzolan et al (2020), we sampled the hillslope's geometric, geotechnical and soil properties to represent the variability of slopes across a region (the island of Saint Lucia) instead of the uncertainty about one specific location. By creating a sufficiently large catalogue of model runs, we could then map back the landslide predictions across that region, and so produce landslide susceptibility maps for many different storm events and land cover scenarios.…”

Shallow landslides and vegetation at the catchment scale: A perspective

“…In particular, simplified slope-scale analysis of shallow landsliding has been applied effectively as a tool for disaster relief and mitigation particularly to support the risk assessment of infrastructure (CHASM; Thiebes et al, 2014). The applications of these models can be made increasingly flexible through the use of search algorithms to determine the most likely failure planes which makes them important tools for decision support at the slope scale (Bozzolan et al, 2020).…”