CONCEPTS - LANDSLIDE RISK INDEX
​
Disaster case – A disaster case is a unique hydro- geomorphologic occurrence (flood or landslide), which fulfils the DISASTER database criteria and is related to a unique spatial location and a specific period of time people (Zêzere et al., 2014). The DISASTER database provides detailed information about each disaster case, such as its type (flood or landslide), date of occurrence, publishing date of the newspapers, location, and number of human damages (casualties, injured, missing, evacuated and displaced people).
Landslide hazard – is the probability of occurrence of a potentially damaging landslide within a specified period, within a given area and a given magnitude. In this study landslide hazard assessment at the municipal scale considers the spatial propensity to landslides (SUSCL) and the weather and climate events index (WCE) as an indicator of the frequency of rainfall events which may generate landslides. Landslide hazard variables (WCE and SUSCL) were empirically weighted assuming that the landslide susceptibility has a major importance on landslide hazard than the climatic triggering conditions, using the following equation:
H = (WCE*0.25) + (SUSCL*0.75)
A landslide hazard value was computed per each municipality using the area with positive values of the susceptibility Information Value (>0) and the weather and climate events index, and the results were further normalized to the interval [0-1].
​
Exposure – The assessment of exposure to landslides at the municipal level was based on two main exposed elements: population and buildings, expressed with two variables (population density and number
of buildings). Other variables were tested to characterize exposure such as number of residents, rate of population change and the density of buildings but, statistically, we have found significant collinearity between these variables. Additionally, the country is characterized by the concentration of population and economic activities along the coastal zone and a depopulation of the innermunicipalities, with a concentration of population and economic activities in the district capitals, which is satisfactorily represented by the population density and the number of buildings.
The population density (PD, no. of inhabitants/km2) represented exposure of people to landslides per municipality in 2011. The number of buildings (NB) per municipality represents the exposure of buildings to landslides. The municipal exposure was computed using the same weight for both variables:
E = PD x 0.5 + NB x 0.5
The weighting assigned to the PD and NB were tested according to a sensitivity analysis based on the Cronbach Alpha and the Lamdba-2 reliability tests. Additionally, the best statistical results were combined with the spatial distribution of the exposure results to select the best weighting combination with statistical and spatial significance.
Exposure results per municipality were further normalized to the range [0, 1].
Physical vulnerability of buildings – Vulnerability is the predisposition of the exposed elements to be affected by the impacts of landslides. In this study, we focus on the physical vulnerability (PV) of the built environment and the estimated degree of loss due to landslides, expressed between 0 (without loss) and 1 (total loss). The physical vulnerability assessment at the municipal level considered the following building features: construction technique and construction materials (CT); reinforced structure (RS); number of floors (NF) and conservation status (CS). Each building feature is divided in a set of building feature classes obtained from the Census data for 2011 (INE, 2011). The physical vulnerability of buildings was assessed for each municipality as the weighted average considering both building feature class weight and building feature weight. Buildings physical vulnerability results were further normalized to the range [0, 1].
In the steps described below, the statistical individuals are the 278 municipalities (not the buildings).In a first step, the aggregate vulnerability of each building feature (AVBF) was computed for each municipality, as follows:
AVBF = SUMVBFC/NB
where AVBF is a score representing the aggregate vulnerability of a particular building feature (e.g. construction material), SUMVBFC is the vulnerability score of a building feature class x multiplied by the number of buildings within building feature class x,and NB is the total number of buildings in the municipality.
Finally, the buildings physical vulnerability (PV) of each municipality was computed as the weighted average of the aggregate vulnerabilities of the building features obtained in the previous step, as follows:
PV= (CMx0.6)+(RSx0.2)+(NFx0.2)+(CSx0.1).
Buildings physical vulnerability results were further normalized to the range [0, 1] using the max-min method (Eq. (1).
​
Landslide risk index (LRI) – Landslide risk is the probability of occurrence of a landslide that causes direct and indirect impacts on people, properties and infrastructure. Expressing landslide risk at the municipal level as a landslide risk index (LRI) implies losing a direct probabilistic relation between the physical processes in a given exposed element with a given vulnerability. The LRI is a dimensionless and comparable measure of landslide risk, calculated to the municipal level and valid for the entire municipality’s territory. LRI is the product of Hazard, expressed by the weather and climatic forcer, and the landslide susceptibility (H), Exposure of population and roads to flood hazard (E) and Physical Vulnerability of buildings (PV) expressed in the following equation:
LRI = (Hazardâ…“)*(Exposureâ…“)*(Physical Vulnerabilityâ…“)
This enunciation of LRI – considering the due differences in regard to scale, risk components and input data – is based on the INFORM risk index (De Groeve et al. 2014). The landslide risk index (LRI) enables to compare hazard, exposure and physical vulnerability of buildings between municipalities and to rank them according to each component of landslide risk.
​
REFERENCES
Pereira, S.; Santos, P.P.; Zêzere, J.L.; Tavares, A.O.; Garcia, R.A.C.; Oliveira, S.C. (2020) A landslide risk index for municipal land use planning in Portugal. Science of The Total Environment, Volume 735, 15 September 2020, 139463. https://doi.org/10.1016/j.scitotenv.2020.139463
Zêzere, J. L., Pereira, S., Tavares, A.O., Bateira, C., Trigo, R.M., Quaresma, I., Santos, P.P., Santos, M., Verde, J., 2014. Disaster: a GIS database on hydrogeomorphologic disasters in Portugal. Nat. Hazards 72, 503–532. https://doi.org/10.1007/s11069-013-1018-y.
​