Landslide characterization using a multidisciplinary approach

(1)

aDepartment of Civil Engineering, University of Calabria, Via Pietro Bucci, Cubo 44/B, 87036 Arcavacata di Rende (CS), Italy

bDepartment of Civil and Environmental Engineering, 5731 Boelter Hall, University of California, Los Angeles, CA 90095-1593, USA

a r t i c l e i n f o

Article history:

Available online 8 January 2016

Keywords:

Permanent scatterer interferometry Electrical resistivity tomography Geotechnical investigations Inclinometer measurements Landslide characterization

a b s t r a c t

A large number of factors should be taken into account to understand landslide phenomena and to facilitate stabilization design especially for complex cases. In this paper the case history of the Gimigliano landslide in the Calabria region (southern Italy) is investigated by using a multidisciplinary combined- technique approach based on conventional geotechnical measurements and modern technologies. The first technique (with reference to inclinometer measurements) is usually affected by errors. Particular care is devoted to the data processing by usingad-hocmethodology to take these errors into account, with the purpose of ensuring a high reliability. For the second technique (electromagnetic sensing techniques and electrical resistivity tomography) convenient methodologies for data post-processing are used and herein presented. The combination of information taken from the different techniques allows the measurements obtained to be validated by conventional and modern approaches and the accuracy of each to be enhanced.

1. Introduction

A very large number of factors are involved in complex natural phenomena such as landslides. Geometrical aspects and their time- evolution, parameters of the geo-materials and hydrogeological conditions are critical factors that should be taken into account in order to have a credible understanding of the process. Slope stabilization planning needs an adequate preliminary landslide characterization. Owing to its complexity, in order to improve knowledge about the process, it is necessary to use data and information from different techniques and combine the obtained results.

Landslides are generally studied using ‘‘direct recognition” or geotechnical investigations. These investigations usually consist of boreholes, laboratory andin-situtests. The boreholes are then equipped with instruments such as piezometers, to define the groundwater conditions and inclinometers for the evaluation of deep displacements and then for the detection of the sliding surface. These standard geotechnical techniques are essential and indispensable because they provide direct information but have a limited spatial representativeness. Therefore they do not allow a complete spatial characterization of the landslide except by multi- plying their number and accordingly the cost. From this comes the

need to integrate the information from geotechnical investigations with what can be obtained by using other techniques suitable to extend it on a broad spatial and temporal scale. Several modern techniques can be useful for this purpose. Among others, it has been decided to use technologies that have been developed and used for investigating and monitoring landslides during the last few years.

A multidisciplinary approach that involves the combination of different techniques is presented and applied in this paper, by using standard geotechnical techniques and innovative geophysical and electromagnetic sensing techniques, for investigating the Gimigliano landslide (Italy). The main aim of this work is to show how the combination of these investigation approaches and techniques leads to a better and more accurate understanding of landslide phenomena.

2. The Gimigliano landslide

The village of Gimigliano (Fig. 1) is located in the Calabria region (southern Italy). The site object of the present study can be divided into two main areas within the town: the medieval village and a new building estate (Fig. 2). The area of Gimigliano is affected by the presence of several instability phenomena which are characterized by different types of landslides with different states, distributions, and styles of activity because of the complex geological settings, as shown on PAI (Piano per l’Assetto Idrogeo- logico)[1]landslide inventory map (Fig. 3).

⇑Corresponding author. Tel.: +39 0984 496527; fax: +39 0984 496526.

E-mail addresses: [email protected](E. Ausilio),[email protected] (P. Zimmaro).

(2)

In the winter seasons of 2009–2010 the activity of the instability phenomena showed an appreciable intensification which caused damage to buildings and infrastructures resulting in the evacuation of some buildings and a block on the viability of some roads. The basic stratigraphic sequence presents phyllites, metarenites and metaconglomerates intercalated by metalime- stones overlaying banks of metabasites alternated with serpentinites and ophicalcites that are part of the Monte Reventino Ophiolitic unit.

The investigations carried out for the characterization of the phenomenon known as Gimigliano landslide, are shown, com- mented, and used in this paper with particular reference to the new building estate which is characterized by the presence of a detrital–colluvial deposit. This phenomenon shows a complex situation in which minor landslides overlap, and in the central part cover, the main body of the landslide (Fig. 3).

3. Field tests and measurements

3.1. Permanent scatterer interferometry

Recently developed geodetic methods use synthetic aperture radar (SAR) images for generating digital elevation models (DEM) and monitoring ground deformations. These techniques are based on differences in the phase of waves returning to a moving plat- form (e.g. aircrafts or satellites). Over the years, several procedures have been developed and proposed: Differential SAR Interferome- try (DInSAR) [2], Permanent Scatterer SAR Interferometry (PSIn- SAR^TM) [3,4], Interferometric Point Target Analysis (IPTA)[5] and Spatio-Temporal Unwrapping Network (STUN)[6], among other.

In this section the results of PSInSAR^TM[7]are presented. This technique overcomes the main drawbacks and limitations of traditional interferometry (e.g. geometric and temporal decorrelation, atmo- Fig. 1.Location of the village of Gimigliano (c) in the Calabria region (b), Southern Italy (a).

Fig. 2.Overview of the village of Gimigliano and position of the boreholes.

(3)

spheric distortion). PSInSAR^TMis based on the use of stable reflec- tors (permanent scatterers) characterized by point-like behaviors, such as exposed rock formations, buildings, bridges and dams.

The use of techniques such as PSInSAR^TMis particularly suitable for urban areas[4]where man-made structures are used as permanent scatterers.

Permanent scatterers techniques are used for accurately evaluating local velocity field in the direction of an axis, called line of sight (LOS), which connects the satellite with the target permanent scatterer on the ground. In this study, historical motion rates and recent data are used to obtain a first evaluation of the deformation fields, along the LOS direction, that characterized the Gimigliano landslide. The historical data refer to a quite long period of observation, divided into two: 1993–2000 and 2002–2010.

These data were interpreted through the PS-InSAR^TMapproach[7], the satellite data used were taken from the satellite ERS-1 and ERS-2 datasets for the period 1993–2000 and from the ENVISAT dataset for the period 2002–2010. ERS-1, ERS-2 and ENVISAT are owned by the European Space Agency (ESA) and operate in the C-band.

The main results of the analysis for the period 1993–2000 are shown inFig. 4a. The significant displacements in this period are located in the new building estate of Gimigliano, which is characterized by a subsidence rate higher than 2 cm/year (red¹dots). In the period 2002–2010 (Fig. 4b) the scenario is different, the average displacements are smaller and there are two main areas affected by the movement, the new building estate and the medieval center. In particular the new estate is characterized by a subsidence rate of about 1.2 cm/year (orange dots), on the other hand it is possible to point out that there is a migration of the deformation field to the medieval center that in this period is characterized by a subsidence rate that is small but not negligible of about 0.3 cm/year (yellow dots).

With reference to newer data interpretation, Bianchini et al.[8], used the data from the German satellite TerraSar-X for evaluating the landslide-induced displacements in the village of Gimigliano.

The TerraSar-X satellite, is the first high resolution radar system that uses the X-band SAR, which provides high ground resolution Fig. 3.The landslide inventory map[1].

1For interpretation of color inFig. 4, the reader is referred to the web version of this article.

(4)

(up to 1 m for the spotlight imaging modes, used in this study, with a scene size of 10 km5 km) and a short repeat cycle (11 days).

Bianchini et al.[8]showed, that exploiting the data acquired in November 2010–October 2011, the displacements are concentrated in the new estate. From the analysis of these newer data it is possible to notice that the average velocity is higher compared with the previous periods and it is different for the several landslides forming the complex phenomenon of Gimigliano landslide.

These results also allow reassessing the status of activity of the several areas. The analyses performed by using this methodology allow to reconstruct: the dynamics of the investigated area through the values of the mean yearly velocities since 1993 up to 2011 and, therefore, an overview of the spatial distribution and temporal evolution of ground displacements of Gimigliano site.

3.2. Standard geotechnical techniques 3.2.1. Borehole logs

After the paroxysmal phase that occurred during the winter seasons 2009–2010 it has been decided to improve knowledge about the instability phenomenon related to the Gimigliano landslide by executing further and more detailed investigations useful to facilitate the risk mitigation analysis and design. The standard geotechnical investigations are driven by the information derived from electromagnetic sensing techniques, in this connection, three boreholes are drilled within the landslide main body. The boreholes S2 and S3 are located in the new estate area (Fig. 2), previ- ously characterized by the highest velocity of displacement and where buildings and infrastructures have suffered and are suffer- ing greater damage. Borehole S1 is located close to the border between the medieval center and the new estate, corresponding to the crown of the landslide. This area has been characterized by a migration of the deformation field, as pointed out in Sec- tion3.1. The depth range of the boreholes varies between 70 and 76.5 m, they are instrumented with inclinometers and piezometers. The stratigraphy of the area is clarified by the borehole logs, analyzing them it is possible to have a confirmation that phyllites, metabasites (e.g. gray and green schists), and ophiolitic rocks (e.g.

serpentinites) with different weathering grades are present.

The groundwater flow is strongly influenced by the different permeabilities of each lithostratigraphic unit. Furthermore, several manmade drainage systems (wells, underdrains, trenches not completely geometrically defined), built in the second half of the last century, are present making the situation even more complicated.

Piezometric measurements, during the period of observation, show water table elevations of about 8 m and 22 m, for the piezometers adjacent to boreholes S2 and S3 respectively (with fluctuations of about ±2 m). These measurements, because of the complexity of

the structural-geological setting and the presence of several water sources, are not sufficient for providing a robust hydrogeological characterization.

Along the depth of the boreholes, ten pressuremeter tests were performed to better define the mechanical properties of the rocks.

These results show a large variability of the parameters (pressuremeter elastic modulus and pressuremeter limit pressure), which is related not only to the change of stratigraphy but also to other factors within the same material layer, such as the grade of alteration and weathering and the groundwater level.

3.2.2. Inclinometer measurements

Boreholes S2 and S3 are equipped with inclinometers. Measure- ments are periodically performed by using a mobile inclinometer probe[9]. The initial readings (February 2, 2012) were made after the backfill was completed and the grout had cured. Two series of initial data sets are needed to verify repeatability and accuracy since these readings will serve as the baseline measurement sets.

Subsequent data measurements can be field checked to verify rea- sonableness of data, including comparisons with previous data sets and evaluating checksums and standard deviations to determine whether anomalies (random or systematic errors) are present.

The series of inclinometer data are affected by errors due to the irregular sensor performance, shortcomings in the system architec- ture, as well as disturbances due to the performance of manual measurements. Usually the errors are classified as random and systematic. Random errors can occur within the sensors, limiting the precision of the probe. Systematic errors occur due to human actions that affect the condition of the sensors–probe and the data collection procedure. Systematic and random errors can create the appearance of deformation where there is none.

Specific data processing procedures are then needed in order to check the reliability of measurements. Greater reliability is achieved by rejecting the readings with gross errors and estimating the accuracy of the horizontal displacements. Standard readings [10]taken in the opposite directionsA₁andA₃on planeA,B₁and B3on planeB(Fig. 5) may be expressed as:

A1¼Aþ

e

^sA1þ

e

^rA1 ð1Þ

A3¼ Aþ

e

^sA3þ

e

^rA3 ð2Þ

B3¼ Bþ

e

^sB3þ

e

^rB3 ð3Þ

B1¼Bþ

e

^sB1þ

e

^rB1 ð4Þ

whereAandBare actual values (in reading units) on planesAandB.

e

^sA1;

e

^sA3;

e

^sB1and

e

^sB3 are systematic errors on planesAandB, while

e

^rA1;

e

^rA3;

e

^rB1and

e

^rB3are random errors on planesAandB. The random errors are considered normally distributed.

Fig. 4.Map of the deformation (velocities in mm/year) for the ERS dataset, 1993–2000 (a); and the ENVISAT dataset, 2002–2010 (b). adapted from[7].

(5)

The first evaluation of the data is a review of the checksums, which consists of adding the two values obtained in diametrically opposite directions at the same depth. The checksum evaluation is performed on site upon completing each data set. Theoretically, the checksums should be zero because the readings have opposite signs. In practice, the checksums produce a constant value, where a low standard deviation would confirm data quality. Checksums are used to evaluate the data for possible errors. The checksums ideally are constant for all depth intervals in a data set. There are several factors that can cause checksums to vary, including casing groove condition and local variations, instrument performance (internal zero-offset of the probe), probe positioning accuracy, and operator inconsistencies–errors. Small variations do not indicate a problem since slight variations are nearly impossible to eliminate. Check- sum variations can become a concern if the standard deviation exceeds about 5–10 units of the mean checksum for the primary axis (A). If large checksum differences are localized to one depth, the data can be corrected based on the mean of the other checksums. However, if large checksums and variations occur in a dataset, the readings should be repeated until satisfactory checksums are achieved[11]. DifferencesD^A^andD^Band checksumsSA and SBof readings on planesAandBare given by:

D^A¼A1A3

2 ¼Aþ

e

^sA1

e

^sA3

2 þ

e

^rA1

e

^rA3

2 ð5Þ

DB¼B1B3

2 ¼Bþ

e

^sB1

e

^sB3

2 þ

e

^rB1

e

^rB3

2 ð6Þ

SA¼A1þA3

2 ¼

e

^sA1þ

e

^sA3

2 þ

e

^rA1þ

e

^rA3

2 ð7Þ

SB¼B1þB3

2 ¼

e

^sB1þ

e

^sB3

2 þ

e

^rB1þ

e

^rB3

2 ð8Þ

If systematic errors are due only to instruments bias (the small nonzero value the probe reads at vertical) and they do not change while measuring, i.e.

e

^sA1¼

e

^sA3¼

e

^sA ð9Þ

e

^sB1¼

e

^sB3¼

e

^sB ð10Þ

The expressions of the differences and the checksums become:

D^A¼Aþ

e

^rA1

e

^rA3

2 ð11Þ

D^B¼Bþ

e

^rB1

e

^rB3

2 ð12Þ

SA¼

e

^sAþ

e

^rA1þ

e

^rA3

2 ð13Þ

SB¼

e

^sBþ

e

^rB1þ

e

^rB3

2 ð14Þ

The Pearson’s test shows that differences and checksums are normally distributed with equal variances on the same plane (

r

²A,

r

²B) and may be written as:

A N ðA;

r

²AÞ ð15Þ

B N ðB;

r

²BÞ ð16Þ

SA N ðSA;

r

²AÞ ð17Þ

SB N ðSB;

r

²BÞ ð18Þ

where A and B are the mean values of distributions of differences while SA and SB are the mean values of checksums distributions.

Given the normal distributions of checksums, gross errors can be recognized by applying the Chauvenet criterion[12]. This technique allows a probability band to be defined, centered on the mean of the distribution. Any data points that lie outside this probability band can be considered to be outliers and thus removed from the data set. The procedure defines an acceptable scatter, in a statistical sense, providing a reliable approach for the rational deletion of outlier data. For example, the values of checksumA (SA) for measurement of March 1, 2012 in the casing S2 are reported inFig. 6where the mean value is plotted with the dotted line. Any values of checksums affected by gross errors, according to the above described Chauvenet criterion, were rejected. This procedure was repeated for every measure.

Accordingly, the inclinometer measurements checked with the above-described procedure showed that in the S2 and S3 boreholes, relative movements take place at a depth of 58–60 m and 46–48 m respectively. In particular, the cumulative displacements for the S2 casing is shown inFig. 7a together with the lithology (Fig. 7b).Fig. 7c shows that the displacements are located within the very altered black phyllites layer.

Fig. 6.ChecksumSAfor measurement March 1, 2012 in casing S2.

Fig. 5.Standard inclinometer casing nomenclature.

(6)

3.3. Electrical resistivity tomography

Electrical resistivity tomography (ERT) is a geoelectrical technique that belongs to the most applied geophysical methods for environmental studies. This technique consists in the evaluation of the apparent electric resistivity (SI unit of resistivity is the Ohm-meter), by applying electric current into the analyzed soil through two electrodes and measuring the resulting voltage (potential) between two other electrodes. The method is based

on a multi-electrode system, where each electrode is alternatively used as current and potential electrode. The electrodes are usually arranged in a collinear array. For engineering, environmental, and groundwater studies, the Wenner, Schlumberger, and dipole–

dipole arrays are the most commonly used. Each geometrical con- figuration has its own specific advantages and disadvantages. ERT is useful to reconstruct the geometry of shallow geological structures and to determine the physical properties of outcropping geo-materials. In particular ERT has been used, in the last few Fig. 7.Cumulative displacements at inclinometer S2 (a); lithology of the S2 borehole (b) and picture of the black phyllites layer (c).

Fig. 8.Traces of the electrical resistivity tomographies (adapted from[23].).

(7)

years, for the investigation of morphotectonics [13], weathering studies [14], detection of cavities[15]. Recently, this technique has also been applied for investigating landslides, because the use of this method can provide important information upon the identification of the sliding surface, of groundwater effect, of internal structure of a landslide[16–19].

For investigating the Gimigliano landslide, seven ERT high resolution surveys (one longitudinal and six transversal to the landslide body) were carried out, as indicated inFig. 8, with lengths which vary from 235 m up to 940 m and reaching an investigation depth ranging from 35 m to 140 m and with 5–10–20 m electrode spacing in order to increase data coverage and quality. The geoelectrical profiles, in which the topographic correction has been introduced, cover the landslide area almost completely. Particular attention was paid to the execution of the tests to take into account that the area is intensely urbanized. In this study, the ERT results were obtained, by utilizing the Wenner–Schlumberger array con- figuration that provides deep penetration, reliable stability and ability to detect both horizontal and vertical subsurface features [20,21]. High-resolution geoelectrical tomographies are developed using the algorithm based on the smoothness-constrained least- squares inversion implemented by a quasi-Newton optimization procedure[22].

The tomographies show a chaotic distribution of the resistivity values with high variations in both horizontal and vertical direction. In some profiles (Fig. 8), the resistivity values vary in the range 5–900Xm [23]. With reference on the longitudinal profile (T4), which is almost parallel to the main landslide, it is possible to point out that there are weakness planes with very low values of resistivity.

4. Characterization of the Gimigliano landslide

Several techniques and measurements are analyzed in this study. The results clearly show a very complex, even chaotic, geological setting. Owing to this complexity, the definition of the landslide characterization is very challenging. The combination of the above-described techniques and measurements, at least, allows a clear definition of the kinematic aspects of the landslide.

The A–A⁰ section (almost coincident with the T4 longitudinal tomography profile) is shown in Fig. 9 [24]. It is important to underline that the failure surface, drawn together with the stratigraphy derived from the S2, S3 boreholes, is located within the phyllites layer. Also the inclinometer measurements show deep movements, located in the same layer. Moreover, the information derived from the analysis of the electrical resistivity tomography, shows an accurate overview of the weakness plane that seems to be located within this layer (i.e. phyllites layer). These observations

involve qualitative considerations on the relative variation of the electrical resistivity values. All the measurements converge on the same characterization of the movements caused by the landslide. It is important to underline that the instable volumes are quite large.

5. Conclusion

A multidisciplinary approach was carried out for the investigations and the characterization of the complex Gimigliano landslide.

The geological and the geomorphological information and the results of the standard geotechnical investigations were integrated with modern techniques such as electromagnetic sensing techniques and electrical resistivity tomography.

The interferometric technique provides an accurate spatial distribution and a temporal evolution of ground displacements. The results show that, in the observation period 1993–2011, the phenomenon of Gimigliano landslide is marked by a displacement field that is highly variable in time and space. These considerations provide a confirmation that the phenomenon is complex and com- posite and that the landslide movements are an historical problem for the area. Although, the inclinometer data are available only for a period not covered by the PSInSAR^TMmeasurements, they give detailed information about deep displacements for the boring loca- tions. These considerations should be verified through further measurements with updated data.

The geoelectrical methods, which are characterized by low costs and fast field investigations, provide useful information for the identification of the sliding surface, of groundwater effect, of internal structure of the landslide especially when calibrated with standard geotechnical investigations such as stratigraphic borehole data and inclinometer measurements. The tomographies show a very complex distribution of the resistivity values with high variations in both horizontal and vertical direction. This is confirmed, and more clearly shown, by the lithology derived from the boreholes and the values of the pressuremeter parameters. The exam- ination of the lithology and the inclinometer measurements allows one to conclude that the shear band is concentrated in the phyllites. Combining this information with the ERT longitudinal profile (T4), it is possible to recognize the weakness plane for low resistivity values, which is coincident with the above- described phyllites layer.

All this shows that a multidisciplinary approach can be considered a powerful tool for investigating landslides, especially characterized by complex geological and stratigraphic settings. Indeed, the geotechnical investigations that are essential and indispensable for investigating landslides have a limited spatial representativeness. In order to have a better understanding of both spatial Fig. 9.A–A⁰section with the S2, S3 stratigraphy and the failure surface.

(8)

[1] PAI Piano Stralcio di Bacino per l’Assetto Idrogeologico Regione Calabria, Autorità di bacino Calabria, 2001.

[2] K. Gabriel, R.M. Goldstein, H.A. Zebker, Mapping small elevation changes over large areas: differential interferometry, J. Geophys. Res. 94 (1989) 2156–2202, http://dx.doi.org/10.1029/JB094iB07p09183.

[3] A. Ferretti, C. Prati, F. Rocca, Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry, IEEE Trans. Geosci.

Remote Sens. 38 (5) (2000) 2202–2212,http://dx.doi.org/10.1109/36.868878.

[4] A. Ferretti, C. Prati, F. Rocca, Permanent scatterers in SAR interferometry, IEEE Trans. Geosci. Remote Sens. 39 (1) (2001) 8–20,http://dx.doi.org/10.1109/

36.898661.

[5] C. Werner, U. Wegmüller, T. Strozzi, A. Wiesmann, Interferometric point target analysis for deformation mapping, in: Proceedings of IGARSS International Geoscience and Remote Sensing Symposium, Institute of Electrical and Electronics Engineers Inc. 7, 2003, pp. 4362–4364 (CD-ROM).

[6] B.M. Kampes, Radar Interferometry: Persistent Scatterer Technique, Springer, 2006,http://dx.doi.org/10.1007/978-1-4020-4723-7.

[7] G. Fortunato, F. Ferrucci, Interferometria satellitare, DEM ed analisi della dinamica altimetrica 1993–2010, Technical Report, 2012 (in Italian).

[8] S. Bianchini, F. Cigna, C. Del Ventisette, S. Moretti, N. Casagli, Monitoring Landslide-Induced Displacements with TerraSAR-X Persistent Scatterer Interferometry (PSI): Gimigliano Case Study in Calabria Region (Italy), Int. J.

Geosci. 4 (2013) 1467–1482,http://dx.doi.org/10.4236/ijg.2013.410144.

[9] Lagic s.r.l. Servizi per l’Ingegneria e la Geologia. Monitoraggio inclinometrico Comune di Gimigliano (CZ), Technical Report, 2014 (in Italian).

[10]J. Dunnicliff, Geotechnical Instrumentation for Monitoring Field Performance, Wiley, New York, 1993.

[11] L. Simeoni, L. Mongiovi, Inclinometer monitoring of the Castelrotto landslide in Italy, J. Geotech. Geoenviron. Eng. 133 (6) (2007) 653–666,http://dx.doi.org/

10.1061/(ASCE)1090-0241(2007), 133:6(653).

doi.org/10.1016/j.geomorph.2006.12.019.

[18] G. Colangelo, V. Lapenna, A. Loperte, A. Perrone, 2D electrical resistivity tomographies for investigating recent activation landslides in Basilicata Region (Southern Italy), Annal. Geophys. 51 (1) (2008) 275–285,http://dx.

doi.org/10.4401/ag-3048.

[19] C. de Bari, V. Lapenna, A. Perrone, C. Puglisi, F. Sdao, Digital photogrammetric analysis and electrical resistivity tomography for investigating the Picerno landslide (Basilicata region, southern Italy), Geomorphology 133 (1–2) (2011) 34–46,http://dx.doi.org/10.1016/j.geomorph.2011.06.013.

[20] T. Dahlin, B. Zhou, A numerical comparison of 2-D resistivity imaging with 10 electrode arrays, Geophys. Prospect. 52 (5) (2004) 379–398,http://dx.doi.org/

10.1111/j.1365-2478.2004.00423.x.

[21] M.E. Candansayar, Two-dimensional individual and joint inversion of three and four-electrode array dc resistivity data, J. Geophys. Eng. 5 (3) (2008) 290–

300,http://dx.doi.org/10.1088/1742-2132/5/3/005.

[22] M.H. Loke, R.D. Barker, Rapid least-squares inversion of apparent resistivity pseudosections by a quasi-Newton method, Geophys. Prospect. 44 (1) (1996) 131–152,http://dx.doi.org/10.1111/j.1365-2478.1996.tb00142.x.

[23] E. Rizzo, A. Giocoli, S. Piscitelli, G. Romano, M. Votta, Tomografie geoelettriche ad alta risoluzione (ERT) per la caratterizzazione del versante in frana presso il Comune di Gimigliano (CZ), Technical Report, 2012 (in Italian).

[24] E. Ausilio, P. Zimmaro, A combined technique approach for the study of a landslide in the Calabria region, in: 20th IMEKO TC4 International Symposium and 18th International Workshop on ADC Modelling and Testing Research on Electric and Electronic Measurement for the Economic Upturn Benevento, Italy, September 15–17, 2014.