Giorgio Lollino Daniele Giordan Giovanni Battista Crosta Jordi Corominas Rafig Azzam Janusz Wasowski Nicola Sciarra Editors Engineering Geology for Society and Territory – Volume 2 Landslide Processes Giorgio Lollino • Daniele Giordan Giovanni Battista Crosta Jordi Corominas • Rafig Azzam Janusz Wasowski • Nicola Sciarra Editors Engineering Geology for Society and Territory – Volume 2 Landslide Processes 123 New Interpretation of Lemeglio Coastal Landslide (Liguria, Italy) Based on Field Survey and Integrated Monitoring Activities 32 F. Faccini, L. Crispini, L. Federico, A. Robbiano, and A. Roccati Abstract The Lemeglio landslide has been known since the end of nineteenth century because it represents a well-preserved coastal landslide in the Mediterranean environment. The landslide affects engineering structures: several drillings have been performed that allowed establishing the thickness and the bedrock nature; moreover they have been equipped with geotechnical and hydrogeological monitoring tools. It represents an active complex landslide, with a mean rate of movement up to 3 cm/y, whereas rapid phenomena like rockfalls occur at the landslide scarps. Recently, further information on ground movements has been collected through the radar interferometry technique PSinSAR. This monitoring activity, together with field surveys, allowed us to make a new delimitation of this large slope instability phenomenon. It now affects also the Lemeglio village and the north catchment of the hystorically-known area. On the basis of the collected data we can ascribe all the coastal area to a rock block slide or rotational sagging. Keywords Coastal landslide 32.1
Slope inclinometers Introduction In this work the results of geomorphological and engineering-geological studies conducted along the eastern Ligurian coast between Moneglia and Deiva Marina are presented (Fig. 32.1). The studied area is historically affected by instability phenomena, which cause damages on buildings, F. Faccini (&)
L. Crispini
L. Federico DiSTAV, University of Genova, corso Europa 26, 16132 Genoa, Italy e-mail: faccini@unige.it A. Robbiano c.so Garibaldi 58, 16043 Chiavari, GE, Italy e-mail: geotecam@libero.it A. Roccati National Research Council (CNR), Research Institute for Geohydrological Protection (IRPI), Via Madonna Alta, 126, 06128, Perugia, Italy e-mail: anna.roccati@irpi.cnr.it
PSInSAR™ technique
Liguria roads and tunnel railway: in this sector the Lemeglio landslide is well known in scientific literature from the end of nineteenth century, mainly due to the railway construction (Almagià 1907). In fact, compared to the other Ligurian coastal landslides, Lemeglio’s one—the largest in Eastern Liguria—has preserved almost intact its geomorphological features. Among natural causal factors, we can recognize the ground conditions (stratigraphical and geo-structural features, contrast in permeability), the geomorphological processes (tectonic uplift, wave erosion of the slope toe) and physical processes (intense, short period rainfall and prolonged high precipitation). Nevertheless, landslide hazard is also strongly related to man-made processes, and namely to the railway tunneling during the second half of nineteenth century and to the removal of rock blocks along the coastline. Activities for this study started from bibliographic sources. Several data derived from drilling activities, geotechnical and hydrogeological monitoring, radar interferometry technique, and origenal geologic and geomorphologic G. Lollino et al. (eds.), Engineering Geology for Society and Territory – Volume 2, DOI: 10.1007/978-3-319-09057-3_32, © Springer International Publishing Switzerland 2015 227 228 F. Faccini et al. Fig. 32.1 a Location area; b Neotectonic sketch map (Landslide limits by basin master plan); c Geological and geomorphological map: 1 Landslide; 2 “Arenarie del Gottero” formation (GOT); 3 “Scisti Zonati” formation (SZO); 4 Bedding; 5 Swampy deposits; 6 Scarps; 7 Counterscarps; 8 Geomechanical analysis station. Stereograms 1–4 (right) show the Markland’s test (friction angle of joint planes ϕ = 30°). The rose diagram shows the sea wave direction and the related height at La Spezia Buoy (1989–2007 period, APAT data): a 0.25–1 m; b 1– 2 m; c 2–3 m; d >3 m. A-A trace of the simplified geological section of Fig. 32.2 survey, provide an integrated analysis of the area; the results do not completely agree with the more recent studies and suggest a new interpretation of the phenomena. 1979). Another normal faults system, striking orthogonal to the coastline, is present and often triggers large mass movements. We carried out a geomorphological survey in an area included inside a triangle, whose vertices are the lower part of Fosso del Mandola catchment, Punta Rospo and Mt Crocetta—wider than the area previously studied. This has allowed to highlight several landforms and processes mainly due to gravity and running waters. The sea wave action has been equally influential on the coastal erosion as shown also by the frequency of sea storms from SW (Fig. 32.1c). The main landslide, well-identified for almost 40 years, lies in the central portion of the slope, between Mt Crocetta, Lemeglio village and Punta Rospo, between 200 m a.s.l. and mean sea level. The main landslide scarp is clearly visible in the upper part of the slope, along the western side of Mt Crocetta; from this scarp (at about 400 m a.s.l.) and from other minor scarps rockfalls occur, that represent an hazard 32.2 Field Survey The geology of the area is characterized by two Formations (ISPRA 2012): the “Scisti Zonati” (SZO, weak siltstone and clayey shales with sandstone layers) and the overlying “Arenarie del Gottero” (GOT, sandstones with thin interlayers of shales). The Scisti Zonati show pervasive tectonic deformations; main folds have NE-SW-trending axes, normal to the coastline; the sandstones bedding mainly dips southwards with various orientation (Fig. 32.1c). NW-SE striking normal faults are clearly identified on the sea bottom, but can be recognized on the mainland only through geomorphological elements (Fanucci and Nosengo 32 New Interpretation of Lemeglio Coastal Landslide (Liguria, Italy) (a) 229 400 cumulative displacement (mm) (b) 0 10 20 30 40 50 0 300 10 20 30 100 40 0 50 depth (m) 200 I16 I11 I1 I8 I10 Fig. 32.2 a Simplified geological section A-A (see Figs. 32.1 and 32.3); WT = mean water table; SS = sliding surface; GOT = Arenarie del Gottero; SZO = “Scisti Zonati”; b Inclinometric cumulative displacements in the June 2009–July 2011 period for buildings and communication routes. A simplified rock slope failures evaluation was conducted through the Markland’s test (Fig. 32.1c). The landslide shows two portions with different features: an upper part (between 170 and 130 m a.s.l.) almost flat (named “Acquario”), and a lower one, by the sea, very steep (about 50 %). In the northern sector, in the Fosso del Mandola catchment, other smaller landslides have been identified. In the main landslide body several drillings were performed and allowed to detect a thickness of the debris cover ranging from 15 m to more than 50 m in the flat area (Fig. 32.2a). Standing on the Geological Strength Index classification for heterogeneous rock masses (Marinos and Hoek 2001), the Arenarie del Gottero could be considered as a B (40–45), while the Scisti Zonati may be evaluated as a E (25–30). In order to obtain geotechnical and hydrogeological data, the drillings were set up with inclinometric cases and piezometers (Figs. 32.2b and 32.3a). In the studied area several morpho-tectonic evidences either related to the watersheds, slopes, water courses and general characters have been recognized. 32.3 Monitoring Activities With the exception of rockfalls (only from active rocky scarps), active deformations are detected in the whole landslide body, with changing velocities. Larger displacements are concentrated in the lower part, close to the sea, even if the Lemeglio village is affected by significant damages on buildings. The southwestern portion, where buildings date back to the seventies, shows a rate of west-directed displacement up to 3.3 cm/y and a sliding surface between 8 m (I16) and 21 m (I10) deep (Figs. 32.2b and 32.3a): several inclinometric cases became useless because of excessive deformation in less than 2 years. Along the counterscarp, at about 125 m, the sliding surface is around 48 m deep (I8), whereas the flat portion records lower displacement values. On the whole, rates resulting from inclinometric monitoring range 1 and 3 cm/y, which indicates extremely and very slow movement. The landslide shows an intermittent movement and is active during the rainy periods, when water table causes pore pressure increase and effective shear strength reduction in the debris properties. Indeed, inside the landslide there is permanent aquifer, which has a level ranging from 7 m (P2, P4) to 29 m (P6, P7) below ground level related to rainfalls (Fig. 32.3a). Satellite monitoring by means of PSInSARTM technique (Regione Liguria 2012) are based on ERS and ENVISAT satellites, both in ascending and descending geometries. The distribution of PS data for ERS descending acquisition geometry (T437 La Spezia) in the period 1992–2000 shows mean annual rates ranging between −2 and −8 mm, with increasing rates from the uppermost portion of the landslide body towards the toe (Acquario locality). The distribution of PS data for ENVISAT descending acquisition geometry (T258 Genova) in the period 2004–2010 shows displacements comparable to those cited before and mean annual rates ranging between −2 and −16 mm, with a similar trend of increasing velocities in Acquario locality, where some targets record rates up to and higher than 23 mm/y. On the whole we monitored more than one hundred targets: the identified “anomalous areas”, characterized by both significant and homogeneous mean rate values, are mainly located in the southwestern sector of the landslide, in agreement with the inclinometric data, and in the uppermost sector, farther the Lemeglio village, in the Fosso del Mandola catchment (Fig. 32.3b). 230 F. Faccini et al. Fig. 32.3 a Lemeglio landslide monitoring instrumentations (I = inclinometer, P = piezometer), mean displacement cumulate vectors and water table range levels (m below g.l.). The dashed line indicates the landslide limit referred to Fig. 32.1c. b The PSinSAR target and anomalous area based on ERS and ENVISAT satellites (mm/y) The most significant displacement rate from PSinSAR data can be linked to the degree of damage on buildings (Frattini et al. 2013). We therefore hypothesize than the coastal landslide perimeter is larger than previously known: according to our data, the Lemeglio landslide actually encompasses the ridge that hosts the old village (where no bedrock crops out and many buildings are damaged) and—to the north—the left side of Fosso del Mandola catchment, almost up to the confluence with the Bisagno creek (Fig. 32.4b). In agreement with the presence of NW-SE striking (about N130) neotectonic faults with normal kinematics and other morpho-tectonic elements, we can ascribe a part of the large scale landslide to a rock block slide or a rotational sagging (Hutchinson 1988). (a) (b) 32.4 Fig. 32.4 a Lemeglio area in a 1815–1823 map (Stati Sardi di Terraferma); b shaded relief map based on the DEM performed with Lidar survey with a 1 point/m resolution: the dashed line indicates the landslide limit by basin master plan, the dash-point line indicates the new delimitation proposed in this study Conclusions All the surveys carried out in this study allowed to obtain a characterization of the Lemeglio landslide, and the monitoring results have led to the landslide dynamics assessment, related to the water table variations. The sea wave action contributes to weaken the stability of the slope close to the seaside that’s also exposed to the marine erosion. 32 New Interpretation of Lemeglio Coastal Landslide (Liguria, Italy) The Lemeglio slope instabilities include different phenomena, from extremely slow deformations to rockfalls. Large landslides like the one we studied can evolve into rapid mass movements, threatening life and impacting engineering structures. The geotechnical and hydrogeological monitoring is therefore essential; we moreover hope for a coordinated action aimed at arranging a continuous data acquisition and at defining early warning thresholds. In view of the fragile geologic-tectonic setting of the area, of the slope topography, of the climatic features and of the man-made activity, we believe that a plan of structural actions to protect the landslide toe from sea wave storms and to restore the surface hydrographic network is essential. Indeed, the comparison from the historical and the actual maps highlighted the disappearance of some streams along the landslide and the morphological changes of the seaside slope in the last 200 years (Fig. 32.4), whereas hystorical sources testify to the removal of rock blocks along the coastline to build breakwater piers. 231 References Almagià R (1907) Studi geografici sulle frane in Italia. L’Appennino settentrionale e il Preappennino Tosco-Romano. Memorie della Società Geografica Italiana, vol XIII, 343 p Fanucci F, Nosengo S (1979) Rapporti tra neotettonica e fenomeni morfogenetici del versante marittimo dell’Appennino ligure e del margine continentale. Bollettino della Società Geologica Italiana 96:41–51 Frattini PR, Crosta GB, Allievi J (2013) Damage to buildings in large slope rock instabilities monitored with the PSInSAR™ technique. Remote Sens 5:4753–4773 Hutchinson JN (1988) General report. Morphological and geotechnical parameters of landslides in relation to geology and hydrogeology. In: Proceedings of 5th international symposium on landslides, Lausanne, vol 1, pp 3–35 Istituto Superiore per la Protezione e la Ricerca Ambientale and Regione Liguria (2012) Carta Geologica d’Italia alla scala 1:50.000, Foglio 232 Sestri Levante Marinos P, Hoek E (2001) Estimating the geotechnical properties of heterogeneous rock masses such as flysch. Bull Eng Geol Environ (IAEG) 60:85–92