The exposure factor in the analysis of geological riskNatural risks
APPLICATION TO ROCKFALLS IN THE MONTAÑA DE MONTSERRAT
The UN argues that today’s soaring disaster risk stems above all from the rapid increase in exposure. Given the widespread social interest in visiting certain sites, there is therefore now a pressing need for greater knowledge and quantification of the exposure to potentially hazardous natural events, in order to minimise damage to property and harm to people. The high human vulnerability of some sites cries out for a sound analysis of the exposure factor as an essential step in evaluating the geological risk and then deciding accordingly on the best mitigation measures in each case.
By SARA FONTQUERNI. MSc in geology specialising in geological risks (UB-UAB). e-mail: email@example.com.
JOAN M. VILAPLANA. Professor of Universitat de Barcelona (UB). Doctor in Geology (UB). e-mail: firstname.lastname@example.org.
MARTA GUINAU. Professor of Universitat de Barcelona. Doctor in earth sciences (UB). MANUEL J. ROYAN. MSc in Geology specialising in geological risks (UBUAB).
GRUPO DE INVESTIGACIÓN RISKNAT. Department of geodynamics and geophysics (UB), School of Geology, Barcelona.
The Montaña de Montserrat is a massif lying 50 kilometres northwest of Barcelona between the districts of Anoia, Baix Llobregat and Bages. On top of the massif sits the Benedictine shrine and monastery dedicated to the Virgen de Montserrat (Figure 1). The monastery and grounds can be reached by road, rack railway, cable car or on foot.
The Montaña de Montserrat is a unique geographical unit in the natural environment of Catalunya, standing out in its own right due to its particular geological and geomorphological features. These features make it a unique massif in the whole world, especially in terms of its singular and striking shapes. The massif was officially listed as a parque natural (nature park) in 1987 to ensure proper conservation. It is run and managed by the Montaña de Montserrat Board (Patronato de la Montaña de Montserrat: PMM). Pilgrims, day-trippers, rock climbers and many tourists flock to the monastery and the nature park as a whole (Parque Natural de la Montaña de Montserrat: (PNMM).
Geologically, the Montserrat massif forms part of the River Ebro catchment area, lying at the SE corner and backing onto the Catalan Coastal Range (Figure 2). The rocks making up the massif are conglomerates with lutite and sandstone intercalations and subhorizontal stratification. This material is affected by a subvertical network of three main fracture families, mainly jointed in form.
The slopes and cliffs of the Parque Natural de la Montaña de Montserrat are very prone to rockfalls, each one usually over 1000 m3 in volume. These rockfalls pose a significant geological risk to existing infrastructure and all visitors to the Barcelona massif. In 2011 over two million people visited the Montserrat monastery and its grounds. On the most crowded days the rack train alone can clock up 4000 passengers, this figure falling to 400 on off-season days.
The slopes and cliffs of Montaña de Montserrat are highly prone to rockfalls, posing a grave geological risk to existing infrastructure and all visitors to the site
In the last five years rockfalls have occurred in some of the most frequently visited parts of the site. The monastery itself (see Figure 3), the main access road (see Figure 4), the carpark and the rack train have all been seriously affected. This has prompted the Catalan Geology Institute (Institut Geològic de Catalunya: IGC) and the RISKNAT research group from the Universitat de Barcelona (UB) to look into this problem and search for solutions, carrying out various studies of the massif’s rockfalls.
The Origin of the Montaña de Montserrat
BRIEF GEOLOGICAL HISTORY
According to Gibert, J.M. et al. (2007), the massif of Montserrat began to be formed during the Bartonian-Priabonian stage (40-34 Ma), due to upthrusting of the Catalan Coastal Range on the southern edge of the Ebro catchment area. A large delta running SE-NW laid down sediment in the current area of Montserrat, with identifiable features of a conglomerate proximal fan-delta facies up to 1300 m thick alternating with beds of sandstone and lutite (distal alluvial fan). Vilaplana d’Abadal and Busquets (2007) .
In the Rupelian stage (34-29 Ma), according to Alsaker et al. (1996), the weight of the sediment itself began to fracture the whole set, leading to a system of parallel joints running NNE-SSW. In the middle and upper Oligocene series (30-23 Ma) the same author identifies three tectonic fracturation episodes related to the compressive phase of the Catalan Coastal Range:
- Reactivation of the NNE-SSW joint system, some of them behaving as faults, generally with little shift.
- Conjugated joint system running NE-SW and NW-SE, resulting from N-S compression.
- And finally a third system of joints running WNW-ESE and hence perpendicular to the first system(NNE-SSW).
Between the Upper Oligocene and Lower Miocene (25-20 Ma) the massif rose, coinciding in time with the start of an extensional phase of the Iberian Peninsula.
This brought about a second reactivation of the first joint system (NNE-SSW), some of the joints behaving as faults, albeit with little shift.
THE GEOMORPHOLOGICAL MODELLING
The raising of the massif exhumed it and exposed it to meteorological agents, which have weathered and eroded it during the last few millions of years. Due to the set of existing geological factors (joint systems, lithology, carbonated matrix of the conglomerates, etc.), the surface run-off and subterranean water has karstified the conglomerates and modelled the relief in a very idiosyncratic way, giving rise to the trademark shapes of the massif (steepling cliffs, needle crags, pinnacles, channels and corridors). This relief makes Montserrat a geological landscape of high environmental value, internationally recognised (see Figure A).
The carbonated cement that held together the rocks slowly dissolved away due to surface and subterranean water run-off. Unlike karstic massifs in limestone rocks Montserrat has no great networks of underground running water (Vilaplana d’Abadal and Busquets (2007) ). The dissolution of the cement had two main morphological effects:
1. The water running through the joints eroded the fissures, opening them up and separating them into individualised masses of long and rounded rock.
2. Water run-off down the surface of the walls freed the rocks from their matrix and exposed great slabs of rock, resulting in abundant fragments of unstable rock liable to rockfalls.
The great majority of the technical studies carried out have tried to find out how all these phenomena work in zones where accidents have occurred, with the aim of then designing and making defence structures. Until now there has been no in-depth study of the geological risk posed by rockfalls throughout the whole nature park.
Rockfalls: a geological risk to be taken into consideration
Rockfalls are one of the commonest geomorphological phenomena in steep mountainous areas (Copons and Vilaplana, 2008). A rockfall is in fact a very frequent occurrence and is considered to be the quickest moving landslide (Varnes, 1978), sometimes generating very high impact energies (Agliardi and Crosta, 2003 and 2009)[8 and 9].
Natural hazard is considered to be the probability or possibility of a potentially hazardous natural phenomenon of a given scale occurring in a specific site and a given timespan. Natural risk (in this case geological risk) is evaluated in terms of the likelihood of damage due to a natural phenomenon occurring in a specific site and in a given timespan (Vilaplana and Payàs 2008) .
There is a host of rockfall studies (Frattini et al. 2008  and Stoffel et al. 2005 ) but very few looking into the exposure and vulnerability to this phenomenon (Ferlisi et al. 2012 ).
Most of the existing studies concentrate on the hazard factor and only some assess the risk, assigning values of exposure and vulnerability (Corominas et al. (2005) ). Very few studies assess exposure. Some authors, however, do try to define and quantify it, such as Peduzzi et al. (2002), who define the term physical exposure as the frequency of a hazard multiplied by the population living in the exposed area. Other authors like Bell and Glade (2004) group vulnerability and exposure, defining them as risk items.
Vulnerability is the degree of loss of an element in the zone affected by the phenomenon (Fell et al. (2008) ). The main damage may be structural, functional or social. Rockfall vulnerability evaluation procedures are still in development stage (Mavrouli and Corominas (2010) ). Even so, two main factors determine the amount of rockfall damage: impact intensity and the nature of the affected element (Corominas et al. (2005)) .
The frequency of a hazardous event can be expressed cartographically by means of frequency zoning maps. These demarcate geographical zones and are classed according to the frequency of the phenomenon within each one. Various Swiss working groups have drawn up rules for analysis and classification of landslip frequency (Heinimann et al., 1998 ; Lateltin, 1997 ; Raetzo et al.,2002  ). Raetzo et al. (2002) prefer to use the term probability rather than frequency or return period. Corominas et al. (2003)  work with the concept of return period in years, defining this as the mean expected time between two consecutive events of similar magnitude.
A rockfall occurs when one or several fragments of rock (blocks) come away from a steep slope or cliff and break up as they topple downhill in free fall and/or by bouncing or rolling along. Areas affected by rockfalls are broken down into three zones: the release zone (where it comes away from the cliff or slope); the transit zone over which it topples, bounces and breaks up, and the deposit zone where the rock fragments finally come to rest as scree (Figure C).
Basically there are three sets of factors that impinge directly on the mobilisation of the blocks on the rocky slopes:
- Intrinsic factors of the rocky mass that determine the sizes and starting points of the rockfalls, such as the presence, disposition and geometry of any discontinuity planes.
- Rock degrading factors, corresponding to the external factors of the slope or cliff from which the rockfall breaks away and breaks up. These intervene in degradation of the discontinuity planes and in the alteration of the slope or cliff surface. Examples might be the presence of standing or running water, the occurrence of frosts, changes of temperature due to insolation or forest fires.
- Factors that trigger the breakage and finally destabilise the blocks, such as certain rainfall episodes producing abundant seepage of water in the discontinuities, frost action on fissures, seismicity, gales, the passage of animals, etc.
The main objective of this study is to assess and map rockfall exposure of the vulnerable elements in the Parque Natural de la Montaña de Montserrat. This overarching objective embraces the following specific goals. Firstly, we make a methodical proposal for assessing rockfall exposure within the analysis of geological risk. We consider this to be a fundamental and groundbreaking input of this study. Secondly, we apply the abovementioned methodology to the specific case of the Montaña de Montserrat, taking in a great part of the nature park and focusing the study on the areas containing the main exposed vulnerable elements. This second specific objective calls in turn for a three-tier study. First of all it is necessary to identify, list and classify by type the various vulnerable elements (permanent/ temporary, structural/cultural /human/ etc). Then an analysis has to be made of the exposure factor. Finally, to visualise the results on the ground, the degree of exposure of the vulnerable elements has to be mapped.
Methodology developed for this study
Risk can be estimated as the hazard multiplied by exposure and vulnerability in a given period of time. For rockfalls Varnes (1978)  has already put forward the following equation:
R = H x E x V x C (1)
H: hazard or probability of a rockfall of a given magnitude.
E: exposure of an element or set of elements at risk of a rockfall.
V: Vulnerability of the exposed elements.
C: cost of the exposed elements.
The hazardousness of the rockfall is defined in terms of the frequency and its intensity or magnitude (energy). The methodology has been designed on the assumption that the vulnerability (susceptibility to damage) bears a direct relation with the energy of the rockfall, and the exposure of the vulnerable elements is also directly related to the frequency (possibility of encounter between the phenomenon and the element).
The main objective of this project is to assess the degree of rockfall exposure and a methodology is put forward to assess the exposure as the product of frequency multiplied by an exposure function of the vulnerable element.
E = F x f (ev) (2)
E: exposure for a type of vulnerable element.
F: frequency of a potentially hazardous phenomenon.
f(ev): exposure function, specific for each type of vulnerable element.
The frequency of a hazardous phenomenon is mapped in this study by means of frequency zoning maps to demarcate geographical zones and break them down into classes (high, medium, low, very low).
The exposure function, f(ev) takes into account the particular characteristics of the element under study, such as the speed at which the element might cross the hazardous zone or the number of vehicles running on a thoroughfare.
|Element type||Specific elements taken into account|
|Moving vehicles||– Cars, buses and motorbikes running on a road.
– Rack train, funicular railway and trains running on a track.
|People in moving vehicles||– People in cars, buses and motorbikes running on a road.
– People in rack train and funicular running on a track.
|People on paths||– People walking on a path (day trippers).|
|Permanent elements||– Permanent buildings (monasteries, dwellings, hermitages, train and funicular stations, industrial buildings, equipment, water tanks, etc.).|
|People in buildings||– People occupying permanent buildings.|
The proposed exposure-quantification method on a 1:25,000 scale takes into account the spatial distribution of the frequency factor and the particular characteristics of the vulnerable element, giving a weight to both terms. The proposed method therefore bears a conceptual resemblance to that put forward by Peduzzi et al. (2002) , who define physical exposure as the frequency of a hazard multiplied by the exposed population living in that area.
Calculation of the degree of exposure is based on the six-phase flow diagram shown in Figure 5:
- Identification of the potentially affected element under study.
- Rockfall frequency zoning in the study area.
- Design of the specific exposure function for the element under study.
- Obtaining the index of exposure, which is the product of the frequency of the event and the exposure function of the vulnerable element. This is done by means of GIS analysis using ESRI’s ArcGis software.
- Obtaining the degree of exposure, grouping in categories the numerical values of the index of exposure obtained for the whole set of elements under study.
- Drawing up the exposure zoning map.
The following subsections of this chapter run through each of these steps in turn.
a. Vulnerable elements broken down into types
Once the vulnerable elements existing in the study zone have been identified, they are classified in terms of their spatial morphology and presence over time. From this first classification an identification is then made of the particular properties of the vulnerable elements and those sharing similar characteristics are regrouped (Table 1). Exposure is calculated first in terms of the set of elements without people and then, at a later stage, these elements with people.
b. Rockfall frequency zoning
In the methodology development phase, and taking into account the working scale of 1:25,000, it was decided to phase into the process the existing information on the 1:25:000 scale (MPRG25M) Map for the Prevention of Geological Risks drawn up by IGC, where hazard is defined as a relation between rockfall magnitude and frequency/activity.
This project puts forward a rockfall exposure assessment method based on the frequency of the hazard multiplied by an exposure function of the vulnerable element
Factors considered in drawing up the IGC’s MPRG25M were not only the release or starting frequency (of the break-off and fall of the block) but also its range. The likelihood of a block reaching a given point of the transit zone falls with increasing distance from the release zone. In the release zone and close to the cliff face the percentage of trajectories will be higher for all block volumes and the frequency with which this area is reached will be high and similar to the release zone itself. Conversely, the further we get from the cliff face, the less likelihood of any but the biggest blocks reaching this spot and the frequency will be correspondingly lower.
Degrees of frequency have been assigned to the polygons created in MPRG25M by IGC, according to the method used for drawing it up, the field information culled by Geocat, knowledge of the area and in-situ observations. Four different degrees of frequency have thus been defined: high, medium, low and very low (see Figure 6).
Adopting frequency zoning from a previous study has constrained application of the developed methodology. A definition has also been made of the limits of the degree of exposure of the elements studied since they assume some pre-established starting conditions: the hazard zoning defined in MPRG25M.
c. Exposure function of the vulnerable element f (ev)
The exposure function of the vulnerable element, f(ev), takes into account the exposure-affecting characteristics of the element under study; this function is unique for each type of element. Figure 7 shows the f(ev)s drawn up for calculating the 1:25,000 scale index of exposure.
d. Index of exposure: calculation with SIG
The index of exposure is a number resulting from multiplying rockfall frequency by the vulnerable element’s exposure function. This value is obtained from an analysis using geographical information systems (GIS), assigning to each frequency category a number and the value of the element’s exposure function.
IE = F x f (ev) (3)
IE: index of exposure for a type of vulnerable element.
F: frequency. f(ev): exposure function for a given type of vulnerable element.
The numerical assignation of the rockfall frequency is based on a geometric progression with common ratio 2, as follows (very low, low, medium and high): 1, 2, 4 and 8.
e. Degree of exposure and representation of the results
The exposure function of the vulnerable element, f(ev), takes into account the exposure-affecting characteristics of the element under study; this function is unique for each type of element
The degree of exposure is defined to map the exposure factor by zones. The degree of exposure is obtained by classifying the exposure indices into categories. Once the index of exposure (IE) has been calculated, therefore, the values are grouped into four categories corresponding to the low, medium, high and very high degree of exposure. Each category corresponds to a degree of magnitude of the index of exposure. Thus, the lowest IE magnitude tallies with the low degree of exposure, increasing one degree for each order of magnitude. The results of applying this methodology to the study zone bear out the suitability of this exposure-index and degree-of-exposure assignation system.
Once the degree of exposure limits have been defined, the polygons obtained from the GIS analysis are thus classed. This step enables us to map the degree of exposure by zones.
Although the index of exposure is numerical, the degree of exposure obtained from applying the developed methodology should be taken as qualitative and relative.
The impact of rockfalls
Newspapers have reported various rockfall problems in Montserrat. Below is a selection of excerpts published in the newspaper La Vanguardia (Figure D).
Vulnerable elements considered and materials used
UNISDR’s global assessment report on disaster risk reduction (2009)  argues that today’s soaring disaster risk is due above all to the rapid increase in exposure. Given the widespread social interest in visiting certain sites, therefore, there is now a pressing need for greater knowledge and quantification of the exposure to potentially hazardous natural events, in order to minimise damage to property and harm to people.
There are many varied vulnerable elements on the site and the methodology has therefore been designed in view of the sensitivity and vulnerability of the items present in the study zone. We have been able to glean the minimum necessary information on these items, to be able to gauge their degree of exposure (see Figure 8). The local items considered in this study are the following: the rack train, the funicular railways, vehicles travelling along the roads, people walking along paths and the permanent immobile elements (monasteries, dwellings, hermitages, train and funicular stations, industrial buildings, equipment water tanks, etc).
This project puts forward a methodology for gauging the degree of exposure of the site elements at the
PNMM (Table 1) and of the people using them. The various analysis procedures designed herein have therefore been applied to the assessment of the degree of exposure to rockfalls, bearing in mind not only the main paths, the rack train, the funicular railways, the road traffic, the grounds and associated buildings but also all the people occupying or using these items. This article, however, shows only the result of applying the methodology on the access roads to Montserrat Monastery (BP-1121 and BP-1103) and on the two main thoroughfares of the Llobregat axis (C-55 and C-58).
BP-1121 and BP-1103 are winding two-lane (one in each direction) mountain roads leading up to the monastery from the main communication thoroughfares (see Figure 9): BP-1121 from the C-55 road on the eastern side of the massif and BP-1103 from El Bruc (A-2) on the western side or from Sant Salvador de Guardiola (C-37) on the northern and NE side.
The BP-1121 road tends to have more traffic than the BP-1103. The average daily traffic (ADT) (intensidad media diaria: IMD) is 700 vehicles on the BP-1121 and 300 on the BP-1103. On the contrary the C-55 and C-58 roads, for the most part with a single lane in each direction, carry very heavy traffic, often more than 15,000 vehicles a day, most of which do not end up driving up to the monastery.
The ADT (IMD) figures were taken at five points of the four roads considered within the study zone. The vehicle counting stations on the monastery access roads are manned only temporarily, recording only one reading a year, so the sample is much too small to be representative. Due to the limited figures to hand, it was decided to take traffic figures from the entrance to Montserrat monastery carpark. This gives us a rough idea of vehicle traffic in the various scenarios but it must be borne in mind that the actual number of vehicles will be higher because not all vehicles driving over the Barcelona massif end up at the monastery and enter its carpark. On the basis of an estimate made by the Catalan Railway Authority FGC (see FGC (2003), 70% of the vehicle entries have been attributed to the BP-1121 road and the remaining 30% to the BP-1103.
The calculation of the number of people travelling in the vehicles running on the roads should ideally be based on accurate vehicle occupancy figures. According to the PMM, the mean occupancy figures of the vehicles entering the monastery carpark are as follows: 1.8 for motorbikes, 3.4 for cars and 43.9 for buses. Actual or estimated vehicle occupancy figures for the roads C-55 and C-58 are not available, so the decision was taken to focus on the people travelling inside vehicles on the BP-1121 and BP-1103 roads. The main materials used for this project are detailed in Table 2.
|Mapa de Prevenció als Riscos Geològics a escala 1:25000 (MPRG25M).||Local geological hazard map||Institut Geològic de Catalunya (IGC) en colaboración con la empresa Geocat||IGC (2011a), IGC (2012), IGC (2011b) and IGC web IGC|
|Espais Naturals de Protecció Especial (ENPE)||Base cartography in digital format||Departament de Territori i Sostenibilitat de la Generalitat de Catalunya||Web of Nexus Geografics|
|1:25.000 scale Mapa Geològic de Catalunya||Base cartography in digital format||IGC and ICC||ICC Web via the Vissir3 map viewer GGeo-information used in shape format and with ETRS89reference system.|
|1:5000 scale topographical base.|
|1:10,000 scale topographical base|
|1:5000 scale orthophotographs|
|2x2 ground elevation model (ASCII) from the LIDARCAT project||ICC||Loaned by ICC to RISKNAT.|
|ADT ( IMD) stations||Location of the ADT ( IMD) vehicle counting stations/td>||Web of the Departament de Territori i Sostenibilidad of the Generalitat de Catalunya, but the daily recorded figures were furnished by DGC|
|Entrance of vehicles to the monastery carpark||The Patronat de la Muntanya de Montserrat (PMM) has provided daily vehicle counts of vehicles entering Montserrat Monastery carpark|
Application of the methodology to the study zone: results
For application of the methodology to the study zone, three common scenarios have been considered for the categories of road vehicles and people travelling in road vehicles. The selected scenarios represent minimum, normal and peak PNMM-visiting conditions and are based on the figures compiled for specific days, on the understanding that they reflect characteristics of a typical day within that scenario.
The advantage of defining a scenario is that the degree-of-exposure maps give us snapshots of actual site use. By comparing different scenarios we can see how these vulnerable elements might show a different degree of exposure in each particular scenario.
The f(ev) values and rockfall frequency serve as the basis for drawing up the index of exposure, IE, which is used to define four degrees of exposure: low, medium, high and very high.
The three considered scenarios – A, B and C – represent visitor influx in terms of a typical day of minimum visits, a normal day and a peak day, respectively. Scenario A (31 January 2011) usually occurs on weekdays during the tourism off season; B (27 July 2011) is habitual on weekends during the low or medium season and week days during the high season; and C (12 October 2011) reflects a peak scenario with record numbers of visitors flocking to the monastery grounds.
The choice of days to assign to this typical day was made from an analysis of the figures compiled for the last five years (2008-2011). The figures for the typical days can be used for calculating foreseeable scenarios.
Road Fact File
Information on the number of users of the various assessed elements has varying levels of confidence. Figures on car entries into the monastery carpark are very trustworthy because its managers keep a thoroughgoing check on things. Conversely, as already pointed out, the calculation of the number of vehicles travelling along the BP-1121 and BP-1103 roads is an estimate based on cars entering the monastery carpark. This therefore understates the real number of vehicles, excluding as it does the cars that park outside the monastery grounds or all those vehicles that enter the nature park without actually visiting the monastery grounds at all. On the other hand, the carpark-information has allowed us to distinguish between different classes of vehicles (motorbikes, cars and buses) and thus make a more realistic estimate of the number of people travelling in the cars running on the roads.
The assessed scenarios do not take into account the fact that on peak visit days the monastery access roads are jammed with vehicles coming to a complete stop for considerable periods of time. In this case the exposure indices of vehicles and people would be maximum because the driving speed would be close to 0 km/h.
Rockfall risk exposure of road vehicles
Analysis of the degree of exposure of vehicles running on the roads is mapped by road sections. A total of 26 rockfall-prone sections have been identified with a total length of 13.91 kilometres in horizontal projection. Figure 10 shows the zoning of the degree of exposure for this element and scenario C.
One of the applications of the developed methodology is assessment of the exposure of vehicles and their occupants on the access roads into Parque Natural de la Montaña de Montserrat
The degree-of-exposure maps show two types of fundamental information (one zonal and one linear) to help grasp the rockfall problem. Firstly the general map base colour tells us the zones classed with different degrees of frequency; this gives us a good initial idea of the hazard. Secondly the colouring of the various road segments or sections tells us the degree of rockfall exposure. This in turn gives us a working idea of the geological risk. The degree of exposure is expressed on the maps by traffic-light colouring of the section under assessment. Vehicles running along a section of the road coloured orange, therefore, are exposed to a high degree of rockfall risk. The same goes for the degree of exposure of vehicle occupants.
Figure 11 shows the percentage degree of exposure of road vehicles in terms of the number of sections in each category. In scenario C the low degree of exposure falls to 0.1%, the medium degree holds pretty steady on scenario B at 4.3%, while the high and very high degrees of exposure change to 50.5% and 45.1% respectively. The graph shows that the high and very high degrees of exposure account for over 85% of the length in the three scenarios; these high percentages tell us that the greater part of the access roads to the massif have a high exposure level.
As regards the results for this vulnerable element, it should be borne in mind that the very high degree of exposure in scenario A and B obtains only on the C-55 road, running through the Llobregat valley, due to the heavy traffic on this road, while on the BP-1121 Montserrat monastery access road from Monistrol and for scenario C there are sections of a very high degree of exposure. These occur where there is a medium or high rockfall frequency and with heavy traffic conditions. On the monastery access road from El Bruc (BP-1103) the maximum degree of exposure is high. The BP-1103 road does not cross the high rockfall frequency zone and traffic is lighter than on the BP-1121 and much less than on the C-55.
Rockfall risk exposure of people travelling in road vehicles
This part of the exposure analysis is crucial for obtaining a good snapshot of the geological risk stemming from the high human vulnerability of the elements analysed. The results of the analysis of the degree of exposure of people travelling in road vehicles are shown in figures 12 and 13. In this case 11 road sections have been obtained with a total length of 9.03 kilometres in horizontal projection of the BP-1103 and BP-1121 monastery access roads from Monistrol and El Bruc, respectively.
Of the three scenarios only A shows up sections with all degrees of exposure. In scenario B, where there are no low exposure sections, the very high degree of exposure weighs in with 81.2 % in length exposed while the medium and high degree of exposure account for 18.2% and 36.4%, respectively. The results for the degree of exposure of people travelling in road vehicles are very significant, since the sum of the high and very high degree of exposure in all scenarios represents at least 97.3% of the total exposed road length. These results give a good idea of the rockfall risk on the studied roads.
Exposure analysis is crucial for obtaining a good snapshot of the geological risk stemming from the high human vulnerability of the elements analysed
This article presents the methodology developed for analysing the degree of exposure at 1:25,000 scale and the results of applying this methodology to a significant part of the Parque Natural de la Montaña de Montserrat. First of all the vehicles running along these roads are considered in themselves and then with their occupants and passengers. The development of this methodology forms part of a more far-reaching study currently underway, which centres on the analysis of rockfall exposure at different scales and in varying natural and social contexts.
This study puts forward a methodology for quantifying exposure as a product of rockfall frequency and an exposure function of the vulnerable element at scale 1:25,000, even though the working scale has been 1:10,000.
The degree of exposure calculation methodology can be broken down into six phases:
- 1. Identification, classification and listing of the potentially threatened elements.
- 2. Rockfall frequency zoning in the study area.
- 3. Design of the specific exposure function for the element under study.
- 4. Obtaining the index of exposure as a product of the rockfall frequency and an exposure function of the vulnerable element, doing so by means of GIS analysis using ESRI’s ArcGis 10.0 software.
- 5. Obtaining the degree of exposure, grouping in categories the numerical values of the index of exposure obtained for the whole set of elements under study.
- 6. Drawing up the exposure zoning map.
The types of vulnerable elements considered in the overall study are: moving vehicles, people travelling in moving vehicles, people on paths, permanent items and people in buildings. Each defined type groups all those elements that share the same characteristics and an exposure function has been designed for each one of them. Exposure has been calculated by considering first of all the set of items separately without taking into account the people occupying or using them and later including the same items with said people.
The frequency of a hazardous phenomenon can be expressed cartographically by means of frequency zoning maps. These demarcate geographical zones and are classed in degrees according to the frequency of the phenomenon in question. Frequencies were assigned according to the methodology applied in drawing up the MPRG25M, the field information culled by Geocat, knowledge of the area and in-situ observations. Four different degrees of frequency have thus been defined: low, medium, high and very high. These degrees are relative to each other and are not quantified. To calculate the index of exposure, each frequency category was assigned a number as follows: very low: 1, low: 2, medium: 4 and high: 8).
Adopting frequency zoning from a previous study has constrained application of the developed methodology. A definition has also been made of the limits of the degree of exposure of the elements studied since they assume some pre-established starting conditions: the hazard zoning defined in MPRG25M.
The exposure function, f(ev) is unique for each type of element. It takes into account the particular characteristics of the element under study, such as the speed at which the element might cross the hazardous zone or the number of vehicles running on a road. This project puts forward an exposure function for each vulnerable element assessed.
The f(ev) values obtained and the frequency then served as the basis for drawing up the index of exposure IE, which defines four degrees of exposure: low, medium, high and very high. Each one tallies with an order of magnitude of the index of exposure. Results are displayed on 1:25,000 maps.
The degree-of-exposure maps show two types of fundamental information (one zonal and one linear) to help grasp the rockfall problem. Firstly the general map base colour tells us the zones classed with different degrees of rockfall frequency; this gives us a good initial idea of the hazard. Secondly the colouring of the various road segments or sections tells us the degree of rockfall exposure. This in turn gives us a working idea of the geological risk.
Apart from defence structures, implementation of non-structural defence strategies comes across as the best risk mitigation policy
This article presents the results of applying the developed methodology to moving vehicles with or without people in three PNMM scenarios (A-minimum visits, B-normal visits and C-peak visits).
One of the most notable results of applying the methodology to the study zone is the high percentage of road lengths with high and very high degrees of exposure both for the vehicles themselves and their occupants. This part of the exposure analysis is crucial for obtaining a good snapshot of the geological risk stemming from the high human vulnerability of the elements analysed.
The methodology developed herein is capable of reflecting the real rockfall exposure situation of the site in question in defined scenarios. Nonetheless the results do have to be taken with some caution until such time as the application of the method is improved and the degree of confidence is increased.
Application of this methodology has nonetheless borne out the various working hypotheses, whereby the degree of exposure varies on the site according to the items exposed and the scenarios posed. It has also proven that the starting data and working scale constrain the results.
This is a general and synthetic study of rockfall exposure in the massif of Montserrat. It is important to bear firmly in mind here that the calculations of the degree of exposure have been made in terms of natural hazardousness without taking defence constructions into account. Thus, in areas where rockfall mitigation measures have been put in place the real exposure is likely to be lower than that calculated herein. Quantification of what we might call «protected exposure» would call for a specific study at the right scale, working from the residual hazard still existing in areas with defence constructions.
The results of this study give us good insights into the best rockfall risk mitigation strategy in the PNMM. It is clear that, apart from the necessary defence structures, already deployed in some cases, implementation of non-structural defence strategies comes across as the best medium- and long-term risk mitigation policy. In the case of the PNMM, rethinking mobility management in access routes to the massif and, especially, to the monastery, would no doubt help to reduce the geological risk.
This study has been financed by a FUNDACIÓN MAPFRE research grant. RISKNAT and the project authors wish to express their thanks for the aid obtained from the following institutions and persons: Geocat: Marc Janeras and Judit Pons; FGC: Ferran Gallego, Iván Pascual and Francesc Ludeña; Parc Natural de la Muntanya de Montserrat: Jordi Calaf and a special mention to the memory of Lluís Baciero; Abadía de Montserrat: Ramon Oranies; DGC: Eugenia Álvarez; PMM: Jesús Alcantarilla; IGC: Pere Martínez; and ICC: Armand Güell.
BY WAY OF A GLOSSARY
Taken from Vilaplana and Payàs (2008) :
Natural disaster / natural catastrophe. An event generated by a natural hazard causing intense alterations to people, property, services and the environment, overwhelming the response capacity of the affected community.
Exposure. This indicates the location of the set of elements that occupy and/or use the territory potentially affected or threatened by a given natural hazard (when we speak of territorial elements we are referring to persons, buildings, communication networks, diverse infrastructure and in general, different uses of the land). Natural hazard or threat. A potentially destructive natural phenomenon: earthquakes, volcanic eruptions, landslides, floods, storms, etc.
Natural hazardousness. The likelihood of a natural hazard of a given magnitude occurring in a specific site and in a given time span.
Natural risk. Likelihood of damage caused by a natural phenomenon in a specific site and in a given time span. Natural risk is understood as a product of hazardousness due to the vulnerability of the elements exposed.
Vulnerability. Vulnerability expresses the percentage of the economic and/or social value of the exposed elements that is likely to be lost in a given natural phenomenon (also known as the degree of potential losses expressed from 0 to 1).
DGC: Direcció General de Carreteres del Departamentde Territori i Sostenibilitat de la Generalitat de Catalunya. (Catalan Road Traffic Authority)
Diba: Diputació de Barcelona. (Provincial Council of Barcelona)
FGC: Ferrocarrils de la Generalitat de Catalunya. (Catalan Railway Authority)
Geocat: Geocat Gestió de Projectes S.A.
GIS: Geographical Information System.
ICC: Institut Cartogràfic de Catalunya. (Catalan Cartographic Institute)
IGC: Institut Geològic de Catalunya. (Catalan Geological Institute)
MPRG25M: Mapa per a la Prevenció dels Riscos Geològics (Map for the Prevention of Geological Risks) at scale 1:25,000.
PMM: Patronat Muntanya de Montserrat. (Muntanya de Montserrat Board)
PNMM: Parc Natural de la Muntanya de Montserrat. (Nature Park of Muntanya de Montserrat)
RISKNAT: Grupo de investigación consolidado de Riesgos Naturales. (Consolidated Natural Risks Research Group)
UB: Universitat de Barcelona. (Barcelona University)
UNDRO: United Nations Disaster Relief Co-ordinator.
UNISDR: United Nations International Strategy for Disaster Reduction.
PARA SABER MÁS
- Gibert, JM; López-Blanco, M; Ramos, E. (2007) Presencia de la incoespecie Dactyloidites ottoi en el complejo de abanico costero de Sant Llorenç del Munt (Eoceno, Cuenca del Ebro, NE de España). Geogaceta, 2007 (41).
- Vilaplana d’Abadal, M; Busquets, P. Itineraris d’interés geològic de Catalunya. Fitxa 224, Montserrat.Generalitat de Catalunya. Departament de Medi Ambient i Habitatge, 2007.
- Alsaker, E; Gabrielsen, H; Roca, E. The significance of the fracture pattern of the Late- Eocene Montserrat fan-delta, Catalan Coastal Ranges (NE Spain). Tectonophysics, 1996 (266), 465-491.
- Martínez, A. Aproximació a la geologia de Montserrat. CEC, Barcelona. Muntanya, 2006 (864).
- IGC (2011). Reconeixement preliminar de la paret rocosa que domina el recinte del Monestir de Montserrat. Institut Geològic de Catalunya, Codi AP-040/11, Barcelona.
- Copons, R.; Joan M. Vilaplana, J.M. Rockfall susceptibility zoning at a large scale: From geomorphological inventory to preliminary land use planning. Engineering Geology, 2008 (102 (3-4)): 142-151.
- Varnes, D.J. (1978): Slope movement types and processes. In Schuster R.L. & Krisek R.L. & Krizek R.J Ed. Landslides, analysis and control Transportation Research Board Sp. Rep, 176, Nat. Acad. of Sciences.
- Agliardi, F; Crosta, G. B. High resolution three-dimensional numerical modelling of rockfalls. International Journal of Rock Mechanics and Mining Sciences, 2003 (40) 455-471.
- Agliardi, F., Crosta, G.B. i Frattini, P. Integrating rockfall risk assessment and countermeasure design by 3D modeling techniques. Natural Hazards and Earth System Sciences, 2009 (9) 1059-1073.
- Vilaplana, J.M.; Payàs, B. RiskCat. Els riscos naturals a Catalunya. Informe executiu / Los riesgos naturales en Cataluña / Natural Risks in Catalonia. Informes del CADS nº6, Dept. de la Presidència, 2008 , 228 pp.
- Frattini, P; Crosta, G; Carrara, A; Agliardi, F. Assessment of rockfall susceptibility by integrating statistical and physically-based approaches. Geomorphology, 2008 (94) 419- 437.
- Stoffel, M; Schneuwly, D; Bollschweiler, M; Lièvre, I; Delaloye, R; Myint, M; Monbaron, M. Analyzing rockfall activity (1600-2002) in a protection forest – a case study using dendrogeomorphology. Geomorphology, 2005 (68) 224-241.
- Ferlisi, S; Cascini, l; Corominas, J; Matano, F. Rockfall risk assessment to persons travelling in vehicles along a road: the case study of the Amalfi coastal road (Southern Italy). Natural Hazard, 2012, (2 -col. 62) 691-721.
- Corominas, J; Copons, R; Moya, J; Vilaplana, J.M; Altimir, J.; Amigó, J. Quantitative assessment of the residual risk in a rockfall protected area. Landslides Springer-Verlag GMBH, 2005 (2) nº 4: 343 – 357.
- Peduzzi, P; Dao, H; Herold, C. (2002). Global Risk And Vulnerability Index Trends per Year (GRAVITY). Phase II: Development, analysis and results. De «GRAVITY-Team» United Nations Environment Programme Global Resource Information Database – Geneva (UNEP/DEWA/GRID-Geneva) per United Nations Development Programme Bureau of Crisis Prevention & Recovery UNDP/BCPR (formerly ERD). 62 pp.
- Bell, R; Glade, T. Multi-hazard analysis in natural risk assessments. Department of Geography, University of Bonn, Germany, 2004, 10 pp.
- Fell, R; Corominas, J; Bonnard, C; Cascini, L; Leroi, E; Savage, W.Z. Guidelines for landslide susceptibility, hazard and risk zoning for land use planning. Engineering Geology, 2008 (102) 85-92.
- Mavrouli, O; Corominas, J. Rockfall vulnerability assessment for reinforced concrete buildings, Natural hazards and Earth System Sciences, 2010 (10), pp. 2055-2066.
- Heinimann, H; Hollesten, K.; Kienholz, H; Krummencacher, B; Mani, P. (1998). Methoden zur Analyse und Bewertung von Naturgefahren. Bundersamt für Umwelt, Walt und Landschaft (BUWAL), Berna, 248pp.
- Lateltin, O. (1997). Recommandations: prise en compte des dangers dus aux mouvements de terrain Dans le cadre des activités de l’aménagement du territoire. OFAT, OFEE and OFEFP. Berne, Switzerland, 42 pp.
- Raetzo, H; Lateltin, O; Bollinger, D; Tripet, J.P. Hazard assessment in Switzerland- Codes of practice for mass movements. Bull. Eng. Geol. Env., 2002 (61) 263-268.
- Corominas, J; Copons, R; Vilaplana, J.M; Altimir, J; Amigó, J. Integrated landslide susceptibility analysis and hazard assessment in the Principality of Andorra. Natural Hazard, 2003 (30) 421-435.
- UNISDR – ONU (2009) Global assessment report on disaster risk reduction. United Nations International Strategy for Disaster Reduction Secretariat, 207 p.
- FGC (2003). Dossier tren-cremallera. Juny 2003. Ferrocarrils de la Generalitat de Catalunya.