Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 275

ANNALS OF GEOPHYSICS, VOL. 51, N. 1, February 2008

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

Gerardo Colangelo (1), Vincenzo Lapenna (2), Antonio Loperte (2), Angela Perrone (2) and Luciano Telesca (2) (1) Dipartimento Infrastrutture e Mobilità, Regione Basilicata, Potenza, Italy (2) Istituto di Metodologie per l’Analisi Ambientale (IMAA, CNR), Tito Scalo (PZ), Italy

Abstract The results of a geoelectrical survey in the study of recent activation landslides in the Lucanian Apennine chain (Southern Italy) are discussed in this paper. During the last two years, after meteorological conditions which af- fected Southern Italy and in particular Basilicata Region, many landslides occurred in this area as reactivations of old movements. These reactivations seriously damaged buildings and infrastructure and they threatened the safety of the people living in the area. Taking into account the complexity and danger of the phenomena, some evacuation decrees for a few houses were adopted. In a short time and at low cost, by using the Mobile Labora- tory of IMAA for geophysical measurements, active geoelectrical investigations were carried out and data pro- cessing performed using innovative techniques for data inversion. The results represent a valid cognitive support to choose the most appropriate technical solution for strengthening of the slopes and an example of best prac- tice for the cooperation between the Civil Protection of Basilicata Region and IMAA-CNR.

Key words landslides Ð electrical resistivity tomog- region between November 2005 and March raphy 2006 with snow blanket changing in the range 0.30-1.00 m (www.sinanet.apat.it, 2005). These climatic conditions worsened the physical and 1. Introduction mechanical characteristics of the terrains out- cropping in the region. In particular, the intense Precipitation in Southern Italy during the precipitations increased their saturation degree 2005 winter was heavier than the norm. In par- and the pore pressures. The snow blanket made ticular, the months of December and February the slope heavy and changed the equilibrium of were characterized by an anomaly of about the strengths involved in its stability. Therefore, 50% more than the normal values (fig. 1). many dormant landslides, which affected the The Basilicata Region (Southern Italy) par- lucanian slopes in the past were reactivated. ticularly suffered from this meteorological situ- The main typologies of reactivation have been ation. Many snowy events also occurred in the earth-flow, translational or rotational slides. The new slides involved buildings and infra- structures constructed on the slopes. The risk for people and assets needed the intervention of the end users involved in risk management and, Mailing address: Dr. G. Colangelo, Dipartimento Infra- in particular, the inspection of the Regional De- strutture e Mobilità, Regione Basilicata, C.so Garibaldi, 85100 Potenza; e-mail: [email protected] partment of Infrastructure and Civil Protection (RDICP). In many involved areas and for many

275 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 276

G. Colangelo, V. Lapenna, A. Loperte, A. Perrone and L. Telesca

Fig. 1. Anomalies of accumulated precipitation 2005 (in percentage values) respect to the normal values 1961- 1990 (from www.sinanet.apat.it).

Fig. 2. Map of Basilicata region with location of Tito, Ripacandida and Castelluccio Inferiore areas involved in reactivation slides respectively after meteorological events of February 2005, March 2006 and October 2006.

276 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 277

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

families evacuation decrees were issued in or- lowed to reconstruct the geometry of landslide der to allow damage valuation. The complexity body, to outline the sliding surface and to locate of the phenomena and the need to obtain infor- areas characterized by high water content (De- mation quickly demanded the application of a moulin et al., 2003; Bichler et al., 2004; Per- non-invasive and fast field acquisition method. rone et al., 2004; Lapenna et al., 2005; Meric et Many examples of the Electrical Resistivity al., 2005; Godio et al., 2006). Tomography (ERT) method, applied for inves- Taking into account these considerations, by tigating landslides, are reported in the literature. using the Mobile Laboratory for chemical- In many cases the results of its application al- physical and geophysical measurements of the

Fig. 3. Geological-geomorphological map of Tito area (Frascheto district) close to the town of Potenza (cen- tral part of Basilicata Region).

277 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 278

G. Colangelo, V. Lapenna, A. Loperte, A. Perrone and L. Telesca

Institute of Methodologies for Environmental ed along the external thrust belt of the chain. It Analysis (IMAA) of CNR, some ERTs have is a complex strato-volcano whose pyroclastic been performed in the more damaged areas of products are the results of both explosive and the Basilicata region. In particular, three sites effusive activity occurring from the Middle have been highlighted by the RDICP (fig. 2). Pleistocene to the Upper Pleistocene (Serri et The Tito landslide, located at the western side al., 2001). The southern part of the region is of Potenza town, occurred in February 2005 af- characterized by the Pollino Ridge, formed by ter a strong snowy precipitation period. The Ri- Meso-Cenozoic rocks, that marks the boundary pacandida landslide, located in the northern between Basilicata and Calabria region (Schi- part of the Basilicata region, occurred in March attarella, 1998). 2006 after a strong rainy precipitation. The Castelluccio Inferiore landslide, located in the Pollino area at the southern part of the Basilica- 2.1. Test-sites ta region, occurred for the first time in Novem- ber 2005 but it reactivated after a strong rainy Tito area (fig. 3) is characterized by Sicilide event in October 2006. Units, in particular the Corleto Perticara For- The areas have been previously studied and mation (Fm). From a lithological point of view the landslide bodies have been mapped by the Corleto Perticara Fm is represented by alternat- technical staff of the RDICP. For each site, a ge- ing calcarenites, calcirudites and marly lime- ological and geomorphological map was drawn stones. and used to locate the geoelectrical measure- According to Pescatore (1988), the terrains ments. attributed to the Sicilide Units (Argille Varicol- For all the sites the results obtained helped ori, Corleto Perticara and Tufiti di Tusa) could the decisions of end users. In particular, the in- represent the Upper Cretaceous-Neogene por- formation from the ERTs were very useful in tion of the Lagonegro Basin Unit that deposited the phases of the valuation damage and slope in the axial part of the basin. stabilization planning. Ripacandida area (fig. 4) is characterised by Meso-Cenozoic substratum Unit of Vulture vol- cano. In particular, the lithology is represented 2. Geological setting by pale yellow stratified sandstones, gray siltites, and gravely silty clays with inter-layered 2.1. Regional grayey calcarenites (Serrapalazzo Formation- Upper Burdigalian-Serravallian). The Serra- The investigated areas are located in Basili- palazzo Fm is considered the distal portion of cata Region, along the axial zone of the south- the foredeep deposits in the Upper Tortonian. ern Apennine Chain. The latter is mainly com- Castelluccio area (fig. 5) is characterised by posed of sedimentary cover of platform and un-metamorphosed terrains of the Liguride Units deep water environments, scraped off from the (Paleogene-Jurassic) and overlying Miocene de- former Mesozoic Ligurian ocean, the western posits. In fact the terrains belonging to the Lig- passive margin of the Adriatic plate and from uride Units can be divided into two main groups: the Neogene-Pleistocene foredeep deposits of metamorphosed and un-metamorphosed ter- the active margin. From west to east, the main rains. The un-metamorphosed terrains, located Mesozoic domains are as follows: 1) the inter- in the investigated area, can be considered a nal oceanic to transitional Liguride-Sicilide Çbroken formationÈ related to subduction-accre- basinal domains (internal nappes), 2) the Apen- tion processes. It consists of an un-metamor- nine carbonate platform, 3) the Lagonegro- phosed succession containing tectonically em- Molise basins, and 4) the Apulian carbonate bedded blocks. Included within a pelitic-calcare- platform (Scrocca et al., 2005). ous-arenaceous succession, several lithotypes The northern part of the Basilicata Region is can be recognised such as ophiolytic suites with characterised by Mount Vulture Volcano, locat- their pelagic sedimentary, terrigenous sediments

278 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 279

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

Fig. 4. Geological-geomorphological map of Ripacandida area (Frascolla district) close to the Vulture volcano (northern part of Basilicata Region) (from Giannandrea et al., 2004).

279 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 280

G. Colangelo, V. Lapenna, A. Loperte, A. Perrone and L. Telesca

Fig. 5. Geological-geomorphological map of Castelluccio Inferiore (Amoroso district) close to Pollino area (southern part of Basilicata Region).

(Crete Nere Fm), cherty limestone and colcani- electrode spacing . The values of the apparent clastic turbidites containing andesitic detritus. resistivity for each profile are assigned at a def- inite depth and position. In a second step, it is necessary to transform 3. 2D Electrical Resistivity Tomography the apparent resistivity values obtained during the field survey into real resistivities of the sub- Electrical resistivity tomography (ERT) is a soil. The same processes are needed for depth geoelectrical method widely applied to obtain values. 2D and 3D high-resolution images of the resis- In this work, the algorithm proposed by tivity subsurface patterns in areas of complex Loke and Barker (1996) for the automatic 2D geology (Griffiths and Baker 1993). During the inversion of apparent resistivity data was used. field survey, ERT can be carried out by using The inversion routine is based on the smooth- different electrode configurations (dipole-di- ness constrained least-squares inversion (Sasa- pole, Wenner, Schulumberger, etc.) placed at ki, 1992) implemented by a quasi-Newton opti- the surface to send the electric currents into the misation technique. ground and to measure the generated voltage The subsurface is divided into rectangular signals. Technically, during an electrical resis- blocks, whose number corresponds to the num- tivity measurement, the electric current is in- ber of measurement points. The optimisation jected into the ground via two electrodes and method adjusts the 2D resistivity model trying the potential drop is measured between two oth- to reduce iteratively the difference between the er electrodes in line with current electrodes calculated and measured apparent resistivity (Sharma, 1997). In surveying, several electrical values. The Root-Mean-Squared (RMS) error profiles are made with various value of unit gives a measure of this difference.

280 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 281

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

Fig. 6a-c. a) Traces of the electrical resistivity tomographies (red lines) on Tito (Frascheto district) landslide; b) 2D Electrical resistivity tomography (10 m electrode spacing) carried out with direction parallel to the lon- gitudinal axis of the landslide; c) 2D Electrical resistivity tomography (5 m electrode spacing) carried out with direction parallel to the longitudinal axis of the landslide.

281 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 282

G. Colangelo, V. Lapenna, A. Loperte, A. Perrone and L. Telesca

4. Results of 2D-ERT and Discussion increase in the water content could cause an in- crease in the pore pressures then a weakening A knowledge of local geology, associated of the slope, the presence of this high conduc- with the high spatial resolution of the mea- tive nucleus could be one of the causes of the sumements, gave us an interpretative tool to ex- movement. plain the ERTs obtained for each case study of this work. All the tomographies show a resistivity con- 4.2. Ripacandida landslide trast between shallow conductive and deeper re- sistive material. The conductive material was A tomography with direction parallel to the associated with slide material which is reshuf- longitudinal axis of the landslide was carried fled and rich in water. The sliding surfaces were out on the Ripacandida landslide (fig. 7a) using located in a zone where the resistivity values a multielectrode system with 32 electrodes 5 m clearly change. Then, the dashed lines reported spacing and a dipole-dipole array. The central on each tomographies represent the higher part of the AAl ERT (fig. 7b), with topographi- probability sliding surface. cal correction, shows a weak resistivity contrast between a relatively high-resistivity zone (ρ> 60-100 Ω) and a relatively low-resistivity zone 4.1. Tito landslide (ρ< 15-20 Ω) at a depth of about 12-13 m. The low-resistivity zone has a lenticular shape Two ERTs along the same profile were car- which is likely to be associated with a high ried out in a direction parallel to the longitudi- clayey content and high saturation levels in the nal axis of the Tito landslide (fig. 6a) using a landslide. The deep relatively high-resistivity multielectrode system with 32 electrodes and zone is considered to reflect the bedrock which the dipole-dipole array. The first one (AAl) was is not involved in the landslide. The shallow carried out using an electrode spacing of 10 m high-resistivity nuclei may be associated with and it reached an investigation depth of about intercalations within the disturbed material. It is 20-25 m (fig. 6b). The second one (BBl) was worth noting that the low-resistivity zone is performed using an electrode spacing of 5 m limited on both sides of the landslide. reaching an investigation depth of about 15 m (fig. 6c). Both the tomographies were topo- graphically corrected. The AAl ERT shows a 4.3. Castelluccio landslide shallow (10-15 m thick) conductive layer, with electrical resistivity (ρ) values less than 20 Ωm, Two ERTs were carried out on the Castel- which covers a more resistive material (ρ> 40 luccio landslide (fig. 8a) using a multielectrode Ωm). The conductive layer could be associated system with 32 electrodes 5 m spacing and a with slide material, the resistive zone with the Wenner-Schlumberger array. Both the ERTs bedrock not involved in the movement. The were topographically corrected. The first (AAl) highest conductive nucleus (ρ< 10 Ωm) located was performed with parallel direction to the before the source area of the landslide, at a dis- longitudinal axis (fig. 8b) of the landslide, the tance ranging from 90 to 140 m from the profile second one (BBl) with a transversal direction origin, could be associated with an area charac- (fig. 8c). terized by high water content. The same resis- The first part of the AAl longitudinal ERT tivity distribution characterized the BBl ERT. shows conductive material, whereas higher resis- The higher shallow spatial resolution of the tive material characterizes the second part. The ERT allows a better geometrical definition of resistivity distribution suggests that the top of the the shallow conductive layer, which is associat- slope is characterized by high water content. ed with slide material. Moreover, the conduc- This hypothesis was confirmed by the in-field tive nucleus located before the source area ap- observations since during the measurements the pears more visible. Taking into account that an ground water table appeared on the surface.

282 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 283

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

The same tomography shows a very high could be associated with the gabionades resistive nucleus located at a distance of about which have been built to increase the stability 60 m from the origin. Taking into account the of the slope. The ERT shows that the depth of position and the resistivity values the nucleus the gabionades did not reach the sliding sur-

Fig. 7a,b. a) Traces of the electrical resistivity tomographies (red lines) on Ripacandida (Frascolla district) landslide; b) 2D Electrical resistivity tomography carried out with direction parallel to the longitudinal axis of the landslide.

283 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 284

G. Colangelo, V. Lapenna, A. Loperte, A. Perrone and L. Telesca

Fig. 8a,b. a) Traces of the electrical resistivity tomographies (red lines) on Castelluccio Inferiore (Amoroso dis- trict) landslide; b) 2D electrical resistivity tomography performed with direction parallel to the longitudinal ax- is of the landslide; c) 2D electrical resistivity tomography carried out with direction transversal to the landslide body.

284 Vol51,1,2008_DelNegro 16-02-2009 21:28 Pagina 285

2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy)

face; it is possible that during the intervention REFERENCES phase wrong sizes of the landslide were con- BICHLER, A., P. BOBROWSKY, P. BEST, M. DOUMA, J. sidered. HUNTER, T. CALVERT and R. BURNS (2004): Three-di- The BBl transversal ERT shows a chaotic mensional mapping of a landslide using a multi-geo- distribution of the resistivity values yielding physical approach: the Quesnel Forks landslide, Land- more information on the geometry of the inves- slide, 1, 29-40. DEMOULIN, A., A. PISSART and C. SCHROEDER (2003): On tigated body. In the central part of the tomogra- the origin of late Quaternary paleolandslides in the phy is again present a conductive nucleus which Liege (E Belgium) area, Int. J. Earth Sci., 92, 795-805. is associated with high water content material. GIANNANDREA, P., L. LA VOLPE, C. PRINCIPE and M. SCHI- ATTARELLA (2004): Carta Geologica del Monte Vulture, The high resistive nuclei could be related to scala 1:25:000 (Lac Litografia artistica cartografica, compact material embedded in clayey matrix. Firenze). GODIO, A., C. STROBBIA and G. DE BACCO (2006): Geo- physical characterisation of a rockslide in an alpine re- gion, Eng. Geol., 83, 273-286. 5. Conclusions GRIFFITHS, D.H. and R.D. BAKER (1993): Two-dimensional resistivity imaging and modelling in areas of complex Three landslides, which occurred in Basili- geology, J. Appl. Geophys., 29, 211-226. cata region after the meteorological events of LAPENNA, V., P. LORENZO, A. PERRONE, S. PISCITELLI, E. RIZZO and F. SDAO (2005): Case history: 2D electrical the 2005 year have been investigated using a resistivity imaging of some complex landslides in Lu- geoelectrical technique. In particular, the elec- canian Apennine (Southern Italy), Geophysics, 70 (3), trical resistivity tomography method was ap- B11-B18. plied to obtain information on the deep charac- LOKE, M.H and R.D. BAKER (1996): Rapid least-squares in- version of apparent resistivity pseudosections by qua- teristics of the landslide bodies such as sliding si-Newton method, Geophys. Prospect., 44, 131-152. surface location, thickness of the slide material, MERIC, O., S. GARAMBOIS, D. JONGMANS, M. WATHELET, etc. The ERTs also highlighted areas character- J.L. CHATELAIN and J.M. VENGEON (2005): Application of geophysical methods for the investigation of the ized by high water content; the increase in the large gravitational mass movement of Sechilienne, saturation degree and pore pressures in these ar- France, Can. Geotech. J., 42, 1105-1115. eas could have caused a weakening of the PERRONE, A., A. IANNUZZI, V. LAPENNA, P. LORENZO, S. slopes. PISCITELLI, E. RIZZO and F. SDAO (2004): High-resolu- tion electrical imaging of the Varco d’Izzo earthflow For each test site, the results obtained (Southern Italy), J. Appl. Geophys., 56 (1), 17-29. proved to be very useful to the RDICP to im- PESCATORE, T. (1988): La sedimentazione miocenica nel- prove the knowledge of the investigated phe- l’Appennino campano-lucano, Mem. Soc. Geol. It., 41, nomenon and to plan possible management ac- 37-46. SASAKI, Y. (1992): Resolution of resistivity tomography in- tivities. ferred from numerical simulation, Geophys. Prospect., In particular, the information obtained in 54, 453-464. Tito test-site helped to estimate the costs for a SCHIATTARELLA, M. (1998): Quaternary tectonics of the future consolidation planning. The results in Pollino Ridge, Calabria-Lucania boundary, Sourthern Italy, Geol. Soc. London Spec. Pubbl. 135, 341-354. Ripacandida test-site and, in particular, the lo- SCROCCA, D., E. CARMINATI and C. DOGLIONI (2005): Deep cation of the sliding surface allowed adequate structure of the Southern Apennine (Italy): thin skinned strengthening structures to be planned for the or thick-skinned?, Tectonics, 24, TC3005, doi: 10.1029/ slope; the area was completely improved. The 2004TC001634 SELLI, R. (1962): Il Paleogene nel quadro della geologia results in Castelluccio Inferiore highlighted that dell’Italia meridionale, Mem. Soc. It., 3, 735-789. the uncertainty in the landslide body sizes defi- SERRI, G., F. INNOCENTI and P. MANETTI (2001): Magma- nition could cause an underestimation or over- tism from Mesozoic to Present: petrogenesis, time- estimation of the adequate strengthening struc- space distribution and geodynamic implications, in Anatomy of an Orogen: the Apennines and Adjacent tures during the intervention planned phase. Mediterranean Basins, edited by G.B. VAI and I.P. Taking into account the cycle of landslides MARTINI (Kl Acad. Publ., Dordrecht, The Netherlands), emergency, the information obtained by the ap- 77-103. SHARMA, P.S. (1997): Enviromental and Engineering Geo- plication of the ERT method could give a valid physics (Cambridge University Press). contribution during the post-event phase which WWW.SINANET.APAT.IT (2005): Gli Indicatori del Clima in mainly regards damage evaluation. Italia nel 2005, I Anno, APAT.

285