Spatial distribution of block falls using volumetric GIS–decision-tree models

Abstract

Block falls are considered a significant aspect of surficial instability contributing to losses in land and socio-economic aspects through their damaging effects to natural and human environments. This paper predicts and maps the geographic distribution and volumes of block falls in central Lebanon using remote sensing, geographic information systems (GIS) and decision-tree modeling (un-pruned and pruned trees). Eleven terrain parameters (lithology, proximity to fault line, karst type, soil type, distance to drainage line, elevation, slope gradient, slope aspect, slope curvature, land cover/use, and proximity to roads) were generated to statistically explain the occurrence of block falls. The latter were discriminated using SPOT4 satellite imageries, and their dimensions were determined during field surveys. The un-pruned tree model based on all considered parameters explained 86% of the variability in field block fall measurements. Once pruned, it classifies 50% in block falls’ volumes by selecting just four parameters (lithology, slope gradient, soil type, and land cover/use). Both tree models (un-pruned and pruned) were converted to quantitative 1:50,000 block falls’ maps with different classes; starting from Nil (no block falls) to more than 4000 m³. These maps are fairly matching with coincidence value equal to 45%; however, both can be used to prioritize the choice of specific zones for further measurement and modeling, as well as for land-use management. The proposed tree models are relatively simple, and may also be applied to other areas (i.e. the choice of un-pruned or pruned model is related to the availability of terrain parameters in a given area).

Introduction

Block falls represent the most intensely geomorphic process of the cliff zone in mountainous areas (Luckman and Fiske, 1995). They are widely distributed in the Mediterranean region (including Lebanon), affecting life and properties severely. In Lebanon, estimations over the last 40 years indicate that mass movements (including block falls) cost 10–15 millions of US dollars per year causing numerous fatalities and injuries (Khawlie, 2000). Concrete examples can give an idea about the importance of mass movement occurrence damages in this small country (Abdallah, 2007). A mass movement in Mazrait Bani Assaf (central Lebanon) caused tremendous damages, i.e. complete destruction of several farmlands, partial destruction of the irrigation system, and occurrence of some fractures in the earth next to some houses with huge resettlement of the ground at the main road access to the village. On 18 March 1992, a huge movement occurred in Hazerta near Zahlé (in the Bekaa valley) which destroyed several houses and 20 people died. On 25th March 2000, a mass movement destroyed 10 houses and several ski resorts in Aayoun Orghouch (Mount Lebanon). Moreover, dense road networks often pass through rugged and unstable topographic features and hence are affected by block falls with high impact on the population and environment. Protection plans to prevent block falls are limited and are often not adequate to meet general safety requirements.

A number of methodologies for mass movement hazard/susceptibility zonation over large areas exist in the literature, which are belonging to qualitative, semi-quantitative and quantitative GIS-assisted approaches. Each of these approaches has its value for certain applications and disadvantages for other objectives, given the specific conditions of the Mediterranean environments. In this context, the GIS technology has been applied to examine landslide hazard in several Mediterranean countries, such as Jordan (Al-Homoud and Masanat, 1998, Husein et al., 2000); northern Spain (Duarte and Marquinez, 2002); Tuscany, Italy (Luzi et al., 2000), central Italy (Carrara et al., 1991), Southern Italy (Budetta et al., 1994, Refice and Capolongo, 2002), and Fabriano, Italy (Luzi and Pergalani, 1996).

Studies of rockfalls are often based on field surveys (Selby, 1980, Kirkaldie, 1985, Romana, 1985, Bieniawski, 1993, Habibagahi and Katebi, 2004, Lindsay et al., 2001), and hazard is estimated either by an empirical assessment of the susceptibility to failure, or by the calculation of a safety factor derived from models of rock mechanics (Carere et al., 2001, Crosta et al., 2001, Irigaray et al., 2003). Very few statistical/deterministic GIS models have been developed for predictive susceptibility modeling of rock falls at regional scale (Abbot et al., 1998, Ayala-Carcedo et al., 2003, Marquinez et al., 2003, Baillifard et al., 2004, Calcaterra et al., 2004, Guzzetti et al., 2004). They are essentially classifying a given area into units according to hazard degree based on information obtained from the intercorrelation between landscape factors and mass movement distribution. Our approach provides a systematic decision-tree model for quantifying and predicting volumes of block falls in a representative area of Lebanon, based on analyzing terrain parameters, in the frame of geographic information systems (GIS). The proposed model comprises a set of rules to classify (predict) a dependent target variable (volume of block falls in m³) using the values of independent predictor variables (terrain parameters). These predictors are both continuous (e.g. elevation) and categorical (e.g. lithology). The obtained block fall maps from converting various decision-tree models, at 1:50,000 cartographic scale, can serve as geological inventories useful for land-use management and environmental decision-making.