Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 161 (2016) 380 – 387
World Multidisciplinary Civil Engineering-Architecture-Urban Planning Symposium 2016, WMCAUS 2016
Landslide Susceptibility Mapping Using a Fuzzy Approach Giovanni Leonardia,*, Rocco Palamaraa, Francis Ciriannia a
University of Reggio Calabria, DICEAM, Via Graziella, 89124 - Reggio Calabria, Italy
Abstract The present paper proposes a new methodology to characterize the landslide susceptibility of Reggio Calabria territory. The values obtained were classified into five categories and exported into GIS environment to produce a landslide susceptibility map. The principal objective of the proposed study is to identify the sections of the road network exposed to landslide hazards starting from the susceptibility map. To this aim, a fuzzy system was implemented for the assessment of the landslides susceptibility of the considered transport network. © Authors. Published by Elsevier Ltd. This ©2016 2016The The Authors. Published by Elsevier Ltd.is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of WMCAUS 2016. Peer-review under responsibility of the organizing committee of WMCAUS 2016 Keywords: Lifelines; Fuzzy logic; Landslide; Susceptibility map; Roads; Reggio Calabria;
1. Introduction Despite tremendous progress in science and technology, natural hazard considerably affects the socio-economic conditions of all regions of the globe. Natural hazards as earthquakes, landslides and floods represent the most common hydrogeological instability realizations. Landslide susceptibility assessment can be tricky because of the difficult evaluation of both the spatial and temporal distribution of past events for large areas mainly due to limitations and gaps of historical records and geographic information. For these reasons, the Geographic Information Systems (GIS), which allow to analyze and manage a considerable amount of information, have been more often used to evaluate the landslide susceptibility [1]. The main objectives of the present work can be synthesized as follows:
* Corresponding author. Tel.: +39-0965-1692207. E-mail address:
[email protected]
1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of WMCAUS 2016
doi:10.1016/j.proeng.2016.08.578
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
1.
2.
381
to create a landslide susceptibility map for the province of Reggio Calabria using the GIS software by means of the weighted combination of various factors such as the slope, the lithology, the elevation, the rainfall and the land use; the determinations of the network infrastructures to mainly pay attention to, by applying the fuzzy logic rules “if-then” to the previously obtained landslide susceptibility map.
2. Methodology 2.1. Case study The considered study area is the territory of the Province of Reggio Calabria (Italy) that with 97 cities and towns and with a total population of 550000 inhabitants is the Province of Calabria with the highest population density. Almost all the territory of Calabria is subject to phenomena of hydrogeological and seismic risk. The province has a territorial extension of 3183 km2, of these 1685 km2 (52.95%) are represented by hilly terrain, 1.275 km2 (40.07%) are mountainous and the remaining 223 km2 (6.97%) are represented by land lowland. The territory of Calabria is geologically young, and often subject to natural modifications. The hydro-geological disaster (landslides, floods) is one of the risk factors to which Calabria is exposed. This is due, among other conditions, from the physical conformation of the region, and the climatic conditions. The landslide susceptibility zoning is based on data furnished by the Hydrogeological Layout Plan (Piano di Assetto Idrogeologico, PAI), containing the landslides occurred in the past together with an assessment of the areas with a potential to experience land-sliding in the future, but with no assessment of the landslides occurrence frequency (annual probability. The objective of the present work is to create a landslide susceptibility map for the province of Reggio Calabria by means of the weighted combination of various indices. The calculated indices are referred to factors such as the slope, the lithology, and the land use, evaluated by means of GIS devices (Fig.1); the landslide susceptibility values obtained have been divided into five classes: very low, low, moderate, high, very high. The network infrastructures resulting particularly relevant during emergency and which need plans aiming at reducing the landslide risk have been highlighted by means of the superposition of the main network infrastructures on the obtained landslide susceptibility map and the use of the fuzzy-logic rules “if-then”[2]. 2.2. Landslide susceptibility zoning map The landslide susceptibility assessment can be tricky because it is very difficult to evaluate both the spatial and temporal distribution of past events for large areas mainly due to limitations and gaps of both historical records and geographic information. Landslide susceptibility assessment can be considered as the initial step towards a landslide hazard and risk assessment. In the proposed study, areas with different classes of landslide susceptibility are marked with different colours (from green, which indicates very low susceptibility, to red, standing for very high susceptibility). 2.3. Data description and indices determination To produce the landslide susceptibility map, a total of 5 inputs were selected for the model, considering the main characteristics of the landslides: the slope, the lithology, the elevation, the rainfall and the land use. Each factor has been characterized into classes whose weight has been determined based on the relevance resulting from the analysis of the landslide areas map identified by the PAI. The analysis has been carried out by means of GIS devices which allow to analyze easily a considerable amount of data and to convert the map pixels into data sets.
382
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
Fig. 1. Characteristic factors.
The weight of each class (Table 1) has been determined by the examination of the value attained by each factor class in the landslide areas identified by the PAI in order to evaluate the relevance of the considered factor class with respect to the past landslides events occurred in the territory of the province. The importance of the five factor has been instead determined using the multicriteria analysis method AHP [3] considering the slope, the elevation, the rainfall as main factors. The resulting index has been finally obtained (Table 2) by multiplying the weight of each factor class and the weight of the same factor gathered from the AHP method. Table 1. Characteristic factors weights. Factor
SLOPE
LAND USE
ELEVATION
Factor class
Weight
<8° 9 – 15° 16-25° >26° Forests Rural areas Superficial areas 0-150 151-300 301-600 >601
0.29 0.26 0.24 0.22 0.12 0.77 0.12 0.25 0.35 0.30 0.09
Factor
RAINFALL
LITHOLOGY
Factor class
Weight
<850 850-1200 1200-1800 >1800 Rocks Clays and sand stones formations Flysch
0.35 0.31 0.25 0.09 0.29 0.57 0.15
The landslide susceptibility is a quantitative or qualitative assessment of the classification, volume (or area), and the spatial distribution of existing landslides or potentially may occur in an area. It can be used to create the map of danger and / or in combination with landslide-prone items inside the sensitive area, is the information for policy makers and the general public, but, The landslide susceptibility mapping has been performed through the weighted sum and the product between the calculated indices.
383
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
Table 2. Resulting indices.
SLOPE
LAND USE
ELEVATION
<8° 9 – 15° 16-25° >26° Forests
Class weight 0.29 0.26 0.24 0.22 0.12
Rural areas
0.77
Superficial areas 0-150 151-300 301-600 >601
Factors weight 0.44
Resulting index 0.126 0.113 0.105 0.095 0.013
RAINFALL
0.036
0.05
LITHOLOGY
0.12
0.005
0.25 0.35 0.30 0.09
0.079 0.110 0.094 0.029
0.31
<850 850-1200 1200-1800 >1800 Rocks Clays lands and stones formations Flysch
Class weight 0.35 0.31 0.25 0.09 0.29 0.57
Factors weight 0.14
0.05
0.15
Resulting index 0.049 0.044 0.036 0.012 0.015 0.030 0.008
The data obtained for the susceptibility evaluation have been classified into five categories: Very low, Low, Moderate, High, Very high. According to the results, the 22% of the study area is classified as very highly susceptible, the 36% as highly susceptible, the 20% as moderately susceptible, the 17% as lowly susceptible and the 5% as very lowly susceptible (Table 3). The above-mentioned results highlight that the 58% of the whole territory of the province is affected by a high landslide susceptibility value. From the obtained map (Fig. 2) it is possible to notice that the highly and very highly susceptible zones are the hills, i.e. the areas between the coasts and the central part of the province. These areas are particularly relevant because of the connections within the various internal urban areas and the main towns and services located along the coasts. Further, the landslide susceptibility evaluation procedure adopted result to be in agreement with the work performed by the Calabria Region Basin Authority: the 80% of the highly and very highly susceptible areas coincide to the landslides areas identified by the PAI maps. Table 3. Landslides susceptibility evaluation. Landslide susceptibility
Index
Very low
0.05
Low
0.17
Moderate
0.20
High
0.36
Very high
0.22
Susceptibility Very low Low Moderate High Very high
Fig.2. Landslides susceptibility map.
3. The importance of lifelines Lifeline is a term that denoting those systems necessary for human life and urban function, without which large urban regions cannot exist. A lifeline is a large spatial system that perform crucial functions for post-natural hazard activity and for the social life of the community. Lifelines are, therefore, the networks, which are developed on the entire territory to relate and connect the various settlements and points of interest of the different subsystems. They guarantee the essential services necessary for the functioning and the survival of the communities (transports, energy, telecommunications, water and sanitary networks). We can define them as the set of structures, infrastructures and
384
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
services regarded as indispensable for the maintaining or protection of the life of the given systems. This is why nowadays we refer to Engineering of lifelines [4]. In this term we address all knowledge and methodologies to design infrastructures in the system which have been planned to reduce and minimize the exposure and susceptibility of infrastructures, also as an outcome of the use of new technologies. Lifelines engineering doesn’t have to be referred exclusively to natural disasters, as earthquakes, but in general, to any kind of emergency due to a generic human or natural hazard or disaster: meteorological or hydro-geological events, fires, floods, toxic and industrial accidents, hazardous materials transportation [5], etc. To design, adopting the criteria for the maintenance of lifelines means to have greater guarantees of reliability and efficiency in any condition of emergency [6]. These aspects of the design and management of infrastructures have come to the public attention with a stronger appeal only in recent years. Recent experiences have highlighted the extreme importance of the lifelines functioning in the conditions of emergency, which follow a catastrophic event [7; 8]. After natural hazard, the proper operation of all vital systems is necessary, for instance hospitals for medical attention of the wounded and highways for communication and assistance for victims. Unfortunately, lifelines elements and infrastructures are inherently complex, and as most man-made constructions, often prone to natural hazard. Landslides are the most likely natural disaster that would lead to major lifeline disruption, in particular the interruption of road infrastructure. 3.1. Transportation lifelines The national transport system has its backbone, mainly, on infrastructures, which are part of the lifelines system. Quite simply, if we consider the road and rail network, or the urban mobility systems, as underground metro and tram lines. The main characteristic, which differentiates transport infrastructures from other types of lifelines, is the human access, for individual use, and in general for all needs of mobility. In the case of transport systems, there is a direct exposition for the population, as in transports the users are the people served by the transport infrastructures. Therefore, the exposition to a risk, or hazard, of the transport infrastructure, implies an exposition to the risk of the users of the system [9; 10]. This is an aspect, which assumes primary value in the assessment of the risk for the population of an area, because the principal object of the exposure analysis is the population, which could be hit by the event. The network infrastructures represent thus elements highly subjected to various natural hazards and very relevant for the daily activities and for the emergency management. It is noteworthy that the potential management and reparation operations of other lifelines (fuel and electricity lines) used the road network [11]. 3.2. Road transportation lifelines in the Province of Reggio Calabria The road network of the Province of Reggio Calabria is composed by 1850 km of roads (Fig. 3). Some infrastructures have high traffic flows, with a high rate of HGV, and in other cases sporadic traffic: the road use varies with the seasonal activities, and the variability is particularly significant in the areas of tourist attraction. The strategic roads, which are the backbone of the transport lifelines, have been classified in function of traffic flows, served population and relevance in the national and regional transport system. This classification of the importance of the primary transport network of the Province of Reggio Calabria, is based on the level of service in case of calamity, the traffic flows and the served population defining three sets (Fig.4): 1) strategic transport ways; 2) transversal access ways; 3) secondary transport ways; The principal aim of this study is to identify the sections of the considered road network exposed to landslide hazards starting from the susceptibility map through the application of the Fuzzy logic theory [12; 13]. So a fuzzy logic system with 2 input, 15 rules “if-then” and 1 output representing the “level of attention” has been implemented in Matlab environment.
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
385
3.3. Application of the fuzzy logic model The Fuzzy logic was introduced by Zadeh [14] as an extension of classical set theory and is built around the central concept of a fuzzy set or membership function. Fuzzy set theory, the proper name for this theory, enables the processing of imprecise information by means of membership functions, in contrast to Boolean characteristic mappings. Assigning 0 to false values and 1 to true ones, fuzzy logic also allows in-between values.
Fig.3. Principal road networks.
Fig.4. Road network considered.
The fuzzy logic model was designed as one-level hierarchical structure with two inputs and one output. The input number corresponds to the linguistic variables (indicators), while the output represents the “Level of attention” evaluation for the considered network infrastructures. In the present study, the “landslide susceptibility” and the “infrastructure relevance” have been considered as inputs, whereas the “level of attention” has been obtained as output. In the fuzzy-logic system proposed (Fig. 5) the input variable 1, i.e. “landslide susceptibility”, has been represented by means of five membership functions which are “very low”, “low”, “moderate”, “high”, “very high”, and the input variable 2, i.e. “infrastructure relevance”, has been described by means of three membership functions which are “strategic road”, “connection road”, “service road” associated to the corresponding importance levels “high”, “moderate” and “low”.
Fig.5. Fuzzy logic model used for the analysis.
386
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
The input variable 1 has been evaluated in the interval [0÷0.4] by means of the trapezoidal-shaped membership function, while the input variable 2 has been evaluated in the interval [0÷10] by means of the triangular-shaped membership function. The output has been described by means of three fuzzy membership functions, i.e. “high”, “moderate” and “low”, in the interval [0÷10] using triangular-shaped functions. Fifteen fuzzy rules “if-then” have been set in order to evaluate the “level of attention”, according to the following logic: - If “susceptibility” is LOW and “infrastructure relevance” is MODERATE then “level of attention” is LOW; - If “susceptibility” is MODERATE and “infrastructure relevance” is MODERATE then “level of attention” is HIGH; - If “susceptibility” is HIGH and “infrastructure relevance” is MODERATE then “level of attention” is HIGH. 4. Results and conclusions From the obtained results (Fig. 6) it is possible to notice that the most relevant roads in terms of connection, as the highway A3, the road SS106 and other roads linking the Ionian Coast to the Tirrenic Coast, are those displaying the highest level of attention. This result highlights the weakness of the network system of the province of Reggio Calabria. Many of the infrastructures correspondent to a high level of attention, in fact, constitute the only connection for very large areas. The analysis performed stresses how not deferrable are the interventions, both active and passive, to mitigate the risk level on the identified lifelines, pointing out that the infrastructures represent the only way to reach the areas affected by a natural hazard and to provide first aid to the population.
Fig.6. Level of attention of the road infrastructures considered for the analysis.
References [1] Carrara, A., Cardinali, M., Detti, R., Guzzetti, F., Pasqui, V., & Reichenbach, P. 1991. GIS techniques and statistical models in evaluating landslide hazard. Earth surface processes and landforms, 16, (5), 427-445. [2] Zlateva, P., Pashova, L., & Stoyanov, K. 2011. Fuzzy logic model for natural risk assessment in SW Bulgaria. In 2nd International Conference on Education and Management Technology (Vol. 13). Singapore: IACSIT Press. [3] Nithya, S. E., Prasanna, P. R., & Eswaramoorthi, S. 2012. Landslide Suceptibility Zonation using Fuzzy Logic for Kundahpallam Watershed, Nilgris. European Journal of Scientific Research, 78, (1), 48-56. [4] Bakir, S. A., Elms, D. G., & Lamb, J. 1994. Risk management and lifeline engineering. 12th World Conference on Earthquake Engineering, 1-8. [5] Barilla, D., Leonardi, G., & Puglisi, A. 2009. Risk Assessment for Hazardous Materials Transportation. Applied Mathematical Sciences, 3,
Giovanni Leonardi et al. / Procedia Engineering 161 (2016) 380 – 387
(46), 2295-2309. [6] Cirianni, F., Leonardi, G., & Scopelliti, F. 2008. A Methodology for Assessing the Seismic Vulnerability of Highway Systems. AIP Conference Proceedings, 1020, 864-871. [7] Casari, M., & Wilkie, S. J. 2005. Sequencing Lifeline Repairs After an Earthquake: An Economic Approach. Journal of Regulatory Economics, 27, (1), 47-65. [8] Chang, S. E., & Nojima, N. 2001. Measuring post-disaster transportation system performance: the 1995 Kobe earthquake in comparative perspective. Transportation Research Part A: Policy and Practice, 35, (6), 475-494. [9] Buonsanti, M., & Leonardi, G. 2013. 3-D Simulation of tunnel structures under Blast loading. Archives of Civil and Mechanical Engineering, 13, (1), 128–134. [10] Buonsanti, M., Leonardi, G., & Scopelliti, F. 2011. 3-D Simulation of shock waves generated by dense explosive in shell structures. Procedia Engineering, 10, (2011), 1550-1555. [11] Cirianni, F., Fonte, F., Leonardi, G., & Scopelliti, F. 2012. Analysis of lifelines transportation vulnerability. Procedia-Social and Behavioral Sciences, 53, 29-38. [12] Ilanloo, M. 2011. A comparative study of fuzzy logic approach for landslide susceptibility mapping using GIS: An experience of Karaj dam basin in Iran. Procedia-Social and Behavioral Sciences, 19, 668-676. [13] Beaula, T., & J, P. 2013. Risk Assessment of Natural Hazards in Nagapattinam District Using Fuzzy Logic Model. International Journal of Fuzzy Logic Systems, 3, (3), 27-37. [14] Zadeh, L. A. 1965. Fuzzy sets. Information and control, 8, (3), 338-353.
387