Engineering Geology 108 (2009) 161–168

Contents lists available at ScienceDirect

Engineering Geology

j ourna l homepage: www.e lsev ie r.com/ locate /enggeo

Review

Contribution of electrical resistivity tomography to the study of detrital aquifersaffected by seawater intrusion–extrusion effects: The river Vélez delta(Vélez-Málaga, southern Spain)

J. Martínez a,⁎, J. Benavente b, J.L. García-Aróstegui c, M.C. Hidalgo d, J. Rey d

a Dpto. Ingeniería Mecánica y Minera, Escuela Politécnica Superior de Linares, Universidad de Jaén, 23700 Linares, Jaén, Spainb Instituto del Agua, Universidad de Granada, Ramón y Cajal, 4, 18071 Granada, Spainc Dpto. de Investigación en Recursos Geológicos, Instituto Geológico y Minero de España, 30008 Murcia, Spaind Dpto. de Geología, Escuela Politécnica Superior de Linares, Universidad de Jaén, 23700 Linares, Jaén, Spain

⁎ Corresponding author. Tel.: +34 953648506.E-mail addresses: jmartine@ujaen.es (J. Martínez), jb

j.arostegui@igme.es (J.L. García-Aróstegui), chidalgo@ujjrey@jaen.es (J. Rey).

0013-7952/$ – see front matter © 2009 Elsevier B.V. Aldoi:10.1016/j.enggeo.2009.07.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 14 January 2009Received in revised form 7 July 2009Accepted 12 July 2009Available online 17 July 2009

Keywords:Electrical resistivity tomographyDeltaic depositsSeawater intrusionDetrital coastal aquifersVélez River

The coastal aquifer of the Plio-Quaternary delta sediment deposits of the Vélez river (province of Málaga,Spain) presents a highly irregular basement morphology and widely varying fill thickness (10–80 m betweenneighbouring sectors). The basin, which is tectonically controlled, is filled with lutite facies alternating withchannel-filling rudites. This detrital aquifer is affected by seasonal seawater intrusion–extrusion processesdue to increasing withdrawal of groundwater for human consumption and irrigation during dry periods.A study was performed to improve the hydrogeological knowledge of this coastal aquifer system. The studyexamined themorphology of the impervious substratum, the facies distribution and the position of the seawaterwedge. For this purpose, an Electrical Resistivity Tomography (ERT) geophysical technique was used and thetomographic datawere calibratedusing geological observations and borehole studies. An analysiswas carried outto compare the direct information obtained from the 35 boreholes with the indirect data corresponding to thefour electrical tomography profiles. In the study, over 9660 resistivity data points were processed.The ERT profiles perfectly corroborated the information derived from the boreholes. The profiles made it possibleto detect thickness changes, lithological changes and the presence of faults. Moreover, from a hydrogeologicstandpoint, this research technique is capableof detecting the positionof thephreatic level and, in coastal aquiferssuch as the one examined in this study, the possible horizontal or vertical penetration of seawater intrusion.Therefore, the electrical geophysical prospecting based on ERT can be highly useful in areas lacking sufficientgeological information and/or mechanical borehole data.

© 2009 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1612. Geological and hydrogeological context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1623. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1644. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1675. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

1. Introduction

In studying the thickness and geometry of depositional systems, acommon procedure is to make use of information from geological

enaven@ugr.es (J. Benavente),aen.es (M.C. Hidalgo),

l rights reserved.

research, drilling and exploitation boreholes. However, these meth-ods are expensive and time consuming, preventing their use on alarge scale. Moreover, these types of data are spatially limited. Incontrast, geophysical measurements can provide a less expensiveway to improve the knowledge of a set of boreholes (Maillet et al.,2005). For this reason, in many cases, geophysical prospectingtechniques can provide complementary data that enable geologicalcorrelation, even in sectors where there are no data from boreholes(Gourry et al., 2003; Colella et al., 2004; Maillet et al., 2005;

Fig. 1. Geographic situation of the Velez river aquifer and location of the studied deltaic sector. The position of the analysed mechanical boreholes and the spatial distribution of thefour electrical tomography profiles are also included.

162 J. Martínez et al. / Engineering Geology 108 (2009) 161–168

Sumanovac, 2006; Massey and Taylor, 2007; Naudet et al., 2008).Borehole can provide direct information about the subsoil, but isspatially localised. However, indirect geophysical methods generatecontinuous data throughout a given profile. Both aspects are ofparticular interest in this study because block tectonics give rise tosignificant lithological variations between nearby areas. An accuratedetermination of the fill geometry is important from a hydrogeolo-gical standpointwhen assessing availablewater reserves. It also helpsin the understanding of the spatial relations between fresh, brackishand salt water, which commonly coexist in coastal aquifers.

In the last decade, a renewed interest in the geoelectricalmethodhasbeen observed due to the development of multi-electrode arrays, fastacquisition systems, and new inversion algorithms. As a consequence,this method has been re-applied to a wide spectrum of geologicalstudies. However, relatively few electrical resistivity investigations havebeen performed in fluvial environments totally or partially saturatedwith salt or brackish water. Electrical Resistivity Tomography (ERT) hasbeen applied in a sand-infilled paleochannel located in the Rhône Delta(Maillet et al., 2005). Massey and Taylor (2007) used ERT in a study ofthe coastline of South Devon, South-west England, which is believed tobe subsiding rapidly. However, ERT has rarely been used for complextectonic reconstruction and control of saline intrusions.

This study set out to determine the reliability of indirect dataobtained by ERT. To do so, these data were compared with thoseobtained from a well-documented depositional system, for whichinformation was available from both boreholes and indirect methods.The system in question is that of the delta sediments of the Vélez river,near the town of Torre del Mar (SE Spain) (Fig. 1). A comparison wasmade with data from the boreholes in the area, on some occasionscomplemented with piezometric, physical and chemical data on thegroundwater, and information from electrical tomography profiles. Theaim of this study was to test the efficiency of the ERT technique indetermining the geometry of complex depositional systems affected byseawater intrusion while providing knowledge about the hydrogeolo-gical setting of the detrital aquifer of the Vélez river basin.

Fig. 2. Stratigraphic columns from mechanical boreholes located in Fig. 1. The basement colevels led to the identification of an upper unconfined and a lower confined aquifer.

2. Geological and hydrogeological context

The Vélez river basin, with a surface area of 610 km2, is located inthe south-eastern part of the province of Málaga (Spain) on theMediterranean coast (Fig. 1). The southern sector consists of alluvialdeposits with an area of 20 km2 which form a delta at the river mouth.The alluvial deposits are assumed to be related to the fluvial activity ofthe Vélez river and its tributaries. In lithological terms, these depositsare made up of gravel, sand and silt of varying thicknesses, which mayexceed 80 m (Lario et al., 1995; García-Aróstegui, 1998).These fluvial-deltaic materials comprise a detrital coastal aquifer that is of greatinterest because of its high degree of permeability and relativelyimportant volume of reserves. In the delta sector, the aquifer (likeothers on the Spain Mediterranean coast) has a dual-layer arrange-ment. The lower level is confined, while the upper one is phreatic.Between the two, there is a layer mainly made up of silt. The siltdisappears upstream so that this dual-layer arrangement gives way toa single unconfined aquifer. The Palaeozoic substrate of the aquifer isconstituted of metapelitic materials from the Alpujarride Complex(Internal Zones of the Betic Cordilleras) and the Benamocarra Unit(Elorza et al., 1981), forming a monotonous series of schists withbluish–grey toned staurolite. Sporadically, the substrate containsmarine Mio-Pliocene materials, in which three facies can bedistinguished: sands, marly clays and conglomerates. Fig. 2 containsstratigraphic columns elaborated from the borehole data.

The aquifer is located in an area with high water demand foragriculture and human consumption. The demand increases duringthe summer months because of tourism. Prior hydrogeological studieshave been completed, and some have involved mechanical prospec-tion boreholes and the application of geophysical techniques.Furthermore, pumped extraction of water from the aquifer duringperiods of scarce recharge has provoked an inland advance of severalkilometres of the transition zone between fresh and salt water, leadingto the salinisation of some wells in the delta sector; however, thisprocess has been found to be reversible after episodes of significant

nsists of Palaeozoic metapelitic materials. The presence of the impervious silt and clay

163J. Martínez et al. / Engineering Geology 108 (2009) 161–168

164 J. Martínez et al. / Engineering Geology 108 (2009) 161–168

precipitation and the consequent reduction in the pumping/rechargeratio.

Fig. 3 illustrates the above-described situation, showing the resultsof various vertical samplings of the electrical conductivity of the waterin a borehole located approximately 1 km from the coastline, drilledinto the lower sector of the aquifer (piezometer IGME-2; see locationin Fig. 1). The change in the water salinity is apparent. The valuesrange from those representative of fresh water to others typical of amarine environment. The latter values were recorded after animportant period of water scarcity from 1993 to 1995. However, in1996, rainfall levels were high, and the tests were repeated in 1997.The rain produced a significant degree of recharge and reduced therate of pumping from the aquifer. Therefore, the salt water in the(salinised) lower level of the aquifer retreated towards the coast. Thissituation has remained constant to date. In mid-2007, the electricalconductivity in the lower aquifer was around 1000 µS/cm. However, inthe upper aquifer, brackish water (2–5 mS/cm) has been found insome sectors. This finding has been interpreted as being due to therecycling of irrigation water.

The hydrodynamic influence on the marine intrusion–extrusionsituation is very apparent in Fig. 4, which shows the conductivityvalues of the water in the lower aquifer (at a depth of 35 m) and thepiezometric levels at the IGME-2 point. It can be seen thatconductivity values close to those considered typical of a marineenvironment were recorded in association with negative piezometriclevels during the period 1993–95. Moreover, a similar piezometricsituationwas observed in late 1989. The recovery of piezometric levelsafter 1997 has been maintained to date. This finding is associated witha reduction in water salinity due to the extrusion of the sea water thathad previously entered the aquifer.

This delta sector, where the geometry is complex (see Fig. 5) andaffected by the processes of marine intrusion–extrusion, was chosenas a suitable area in which to carry out an analysis of the effectivenessof electrical tomography as an instrument for distinguishing theimpervious basement and for identifying the presence of salt waterwithin the aquifer.

3. Materials and methods

Electrical resistivity surveys have progressed from conventionalvertical soundings to techniques such as ERT, which provides two andeven three-dimensional high-resolution electrical images of thesurface (Colella et al., 2004).

Fig. 3. Electrical conductivity logs carried out on different dates at the IGME-2 piezometer shosalinisation reached during the summer of 1995, the rainy period beginning in 1996 producwedge. This flushing process has continued through the present day.

This electrical geophysical prospectingmethod consists of determin-ing the distribution of a physical parameter that is characteristic of thesubsoil (the resistivity) on the basis of a very large number ofmeasurements of apparent resistivity made from the ground surface(Telford et al., 1990; Store et al., 2000). By measuring the electricalpotential difference induced by an electrical current passing through amaterial, the resistance of that material can be calculated. Variations ingeoelectric behaviour make it possible to obtain 2D profiles. Thistechnique constitutes one of the most effective non-destructiveinstruments available for studying and characterising discontinuitiesin the subsoil (Sasaki, 1992; Store et al., 2000; Naudet et al., 2008). Theacquisition can be carried out using diverse electrode configurations(e.g., dipole–dipole, Wenner, Schlumberger) that are positioned on asurface in order to inject the electric currents into the ground andmeasure thevoltage signals generated. The apparent electrical resistivityis then calculated using a predetermined geometrical constant for thedevice employed.

In the present study, the Wenner–Schlumberger array was used.The apparent resistivity for the array is given by ρa=πn(n+1)aR,where R is the resistance, a is the spacing between the P1 and P2potential electrodes and n is the ratio of the distances between thecurrent and potential electrodes. The depth study range increaseswith increasing space between the current electrodes.

The electrical tomography equipment used in this study is theRESECS model manufactured by Deutsche Montan Technologie(DMT). It has an integrated computer and features 112 addressableelectrodes connected via a single seven-core cable. The power sourceis 250 W and 2,5 A, which generates impulses of 880 V. Thetransmitter, receiver and power supply are built into the RESECSresistivity meter. Unlike other equipment employed in the previousstudies (Maillet et al., 2005; Massey and Taylor, 2007), the RESECSresistivity meter allows the management of a higher number ofelectrodes and the obtainment of more data in each profile. Thisallows for longer profiles without the use of overlapping profiles,creating a more accurate geological model.

As a second step in this study, it was necessary to invert theapparent resistivity values obtained during the field survey of the realresistivity of the subsoil. For this purpose, the RES2DINV softwarepackage (Geotomo Software) was used to create a 2D resistivitymodelusing the data obtained from electrical imaging surveys (Griffiths andBarker, 1993). The inversion routine is based on the smoothness of theconstrained least-squares inversion (L2-norm). It attempts to mini-mise the square of the difference between the observed and calculated

w the fresh water–salt water relationships in the confined aquifer. In spite of the criticaled a significant improvement in the groundwater quality and a retreat of the salt water

Fig. 4. Evolution of the electrical conductivity and the piezometric level at piezometer IGME-2. The groundwater quality improvement in the confined deltaic aquifer is clearly relatedto the rise in the piezometric levels.

165J. Martínez et al. / Engineering Geology 108 (2009) 161–168

apparent resistivity values (DeGroot-Hedlin and Constable, 1990;Sasaki, 1992) implemented by a quasi-Newton optimisation technique(Loke and Barker, 1996; Loke and Dahlin, 2002). Then, a mesh gridwith four nodes between adjacent electrodes is established to moreaccurately calculate the apparent resistivity values. Finally, model cellsof refinement with widths of half the unit spacing are applied. Thisapplication is considered to improve the results when there are verylarge resistivity variations near the ground surface.

In the geophysical prospecting campaigns, four profiles wereselected (Fig. 1). Profile 1 (NW–SE oriented) was performed parallel tothe Vélez river on its left bank. There was a space of 5 m between eachelectrode and a total length of 475 m. Therefore, 96 electrodes wererequired. A total of 2021 measurements of apparent resistivity weretaken with the equipment, with a model of 1117 blocks, reaching amaximum study depth of 90.3 m. The obtained RMS was 8.4% (Fig. 6A).

Profile 2 was oriented in a SE–NW direction and located on theright bank of the Vélez river. There was a space of 5 m between eachelectrode and a total length of 555 m. Therefore, 112 electrodes were

Fig. 5. Geological cross sections. Longitudinal (A) and transverse (B) profiles were obtaincomplexity of the aquifer system. A water table oscillation from a severe drought period (1

required. In this case, 2809 measurements were taken. The blockmodel consisted of 1417 cells and a maximum study depth of 105.7 mwas attained. The obtained RMS was 25.2% (Fig. 6B).

Profile 3was oriented in aNE–SWdirection and aligned inparallel tothe coastline until it crossed the river. The electrode setup was identicalto that of Profile 1, and so it had the same length and reached the samedepth. The obtained value for RMS was 15.6% (Fig. 6C). The field worktook place during themonth of July, when the Veléz riverwas dry, and itwas possible to put the electrodes in a profile crossing the stream.

Profile 4 was oriented in a NE–SW direction and was also alignedparallel to the coastline. It started to the west of Profile 3 in the alluvialbasin of the Vélez river. As with the other profiles, the electrodes werespaced 5m apart. The total lengthwas equal to that of Profile 2 (555m).Therefore, 112 electrodes were required. 2809 measurements weretaken and 1417 blocks were obtained, with a maximum study depth of105.7 m. The obtained RMS was 54%. This value is higher than thoseobtained from the inland profiles, but it is a normal figure when largecontrasts in resistivity are registered. Profile 4 included the coastline,

ed from borehole log correlations. The profiles show the geometrical and structural995) through a recent date (2007) is also included. Locations are shown in Fig. 1.

Fig. 6. Obtained Electrical Resistivity Tomographic profiles. The position of each profile is shown in Fig. 1.

166 J. Martínez et al. / Engineering Geology 108 (2009) 161–168

where resistive dry sands and gravel saturated with saline water werepresent. Nevertheless, the information from the boreholes allowed us toconfirm and validate the obtained geoelectrical model (Fig. 6D).

The profiles reached depths of 90 to 105m, close to the impervioussubstratum of the basin. Since the resistivities measured at thosedepths have a certain degree of uncertainty, direct information from

167J. Martínez et al. / Engineering Geology 108 (2009) 161–168

the mechanical boreholes was used to help with the geological inter-pretation of the indirect measurements obtained from the resistivityprofiles.

4. Results and discussion

Table 1 shows the depth of the Palaeozoic basement and the totaldepth of the 35 mechanical research boreholes. The boreholes weredrilled in rotation while obtaining continuous samples from the corebarrels under direct circulation (location shown in Fig.1). Fig. 2 is basedon the data obtained from these boreholes, showing 19 stratigraphiccolumns correlated in threeprofiles. The columnswere nearly parallel tothe tomography profile, including the two in the NW–SE direction andthe one in the NE–SW (Fig. 1). From this information, importantvariations in the thickness of sediments, which ranged from 10 to 80 min sectors that were close to one another, could be observed. Moreover,the correlation revealed that the basin had a highly irregular geometry,as a result of tectonic faults (Fig. 5). Thefillingbasically consistedof lutitefacies, sands and alternating rudites, which are typical of point bars andare laterally wedged.

Water table levels were measured at all the boreholes andsubsequently compared with the resistivity data from the electricaltomography profiles. In the example discussed in this paper, the watertable level corresponded to the decrease in resistivity between thesuperficial level and the intermediate one. Also, in sectors close to thecoastline, marine intrusion was characterised by low resistivityadvancing inland from the coast.

The geoelectrical profiles (Fig. 6) showed resistivity values thatvaried laterally and in depth. In the vertical aspect, three levels couldbe distinguished in accordance with the resistivity values obtained. Inthe upper part, there was a superficial level 3 to 15 m thick with high

Table 1Basement depths measured in the mechanical boreholes and total lengths of drilling inthe Velez river aquifer.

Borehole Basement depth (m) Total depth (m)

P1-S1 12 15.5P1-S2 27 29.5P1-S3 36 40.5P1-S4 36 37P1-S5 34 36.5P1-S6 18 20S-6 25 43.4S-7 18 22.5S-8 24 35S-9 27 47.5S-10 9 56S-10bis 13 13.5S-11 13 19.8S-12 40 47.4S-13bis 80 86.3S-13 80 86.3S-14 25 44.5S-15 24 28.2S-16 25 50.2S-18bis 30 35S-20 15 26S-21 16 29.4S-22 25 30.6S-23 24 34.5S-151bis 47 58.1S-152 47 53.2S-153 48 55S-154 48 59.3S-155 N41.9 41.9S-156 N50 50S-157 N50 50S-158 N50 50S-159 (IGME3) 48 50IGME1 47 50IGME2 47 50

resistivities ranging from 20Ω·m to 80Ω·m. There was a second levelwith very low levels of resistivity, below 15 Ω·m. and even as low as2 Ω·m. Finally, there was a third level characterised by a gradualincrease in resistivity values from 25 to more than 80Ω·m. In the firstand the second levels, lensmorphologies could be identified that wererelated to channel filling. As can be seen in Fig. 6, the depth of thelowest level was highly variable, ranging from 20 to 80 m at pointsthat were very close to each other.

ERT displayed the existence of sectors with coarser detrital facies,mainly in the right bank of the delta. These coarse deposits are areas ofhigher hydraulic conductivity and they favoured the advance of marineintrusion. Previous hydrochemical analysis and vertical profiles ofelectrical conductivity of thewater showed areas in the superficial levelwith 20 mS/cm. ERT allowed relating the spatial variation in ground-water salinity with lateral facies variations that control the permeabilityof the subsoil. These zones of higher permeability explain the presenceof salinewater in certain areas inwhich thepumping rateswerenot highenough to justify an important advance of marine intrusion.

The upper level had high resistivity values, which correlated withgravels and sands above the phreatic level. Equivalent patterns havebeen described by García-Aróstegui 1998, Auken et al., 2003, andSumanovac 2006. The intermediate level, whichwasmore conductive,corresponded to the water-saturated detrital facies. At this level,channelled morphologies could be identified. Moreover, at this levelthere were important lateral variations in the resistivity values. Profile1 in Fig. 6A reveals a contour of low resistivity near the coastline(extreme SE), which, as in the other sectors, could be related tomarineintrusion. Similar patterns and interpretation have been considered inother coastal aquifer environments (Nielsen et al., 2007). The ERTprofiles detected a salt water wedge located near the cost line. Thissituation was considered as a consequence of fresh water seepage tothe sea. This interpretation was in agreement with the Velez riveracting as a gaining stream in the last stretch before the mouth.

The saline wedge was constant in Profiles 3 and 4, which wereobtained close to andparallel to the shoreline. Thedeepest levelwas thatof the Palaeozoic substrate. The resistivity values here were notparticularly high due to the alteration that had taken place at the top.Abrupt changes in the altitude of this levelwere causedby the faults thataffect the basement. These faults could be equivalent to those previouslydescribed in other regions (Gallipoli et al., 2000; Suzuki et al., 2000;Caputo et al., 2003). The geometry of the impervious basement deducedfrom the ERT profiles indicated an average thickness in the easternsector of the delta higher than in the western sector.

5. Conclusions

The analysis of the information obtained from the mechanicalboreholes drilled in this region revealed the existence of importantvariations in the thickness of the sedimentary coverage, which rangedfrom 10 to 80 m in nearby areas. These stratigraphic colums correlated,leading to the conclusion that the geometry of the river basin is highlyirregular and determined by tectonic processes. The filling is comprised,fundamentally, of lutite facies and sands. In addition, there are ruditefacies typical of point bars, which are laterally wedged.

The correlationbetween thedata from themechanical boreholes andthe data from the electrical tomography shows that the lattergeophysical prospecting technique is a suitable tool for in-depth analysisof sedimentary bodies. It provides greater definition in geologicalcorrelations as well as uninterrupted monitoring of the thickness of fillzones. Electrical tomographymakes it possible to determine the nature,depth and morphology of the basement. In addition, faults can bedetected, which is of special interest in this case because they cannot bemapped at the surface.

The four obtained geoelectric profiles had variable values ofresistivity, both laterally and in depth. In the vertical dimension, threelevels could be distinguished. The superficial level had high resistivity

168 J. Martínez et al. / Engineering Geology 108 (2009) 161–168

values, which correlated with gravels and sands above the phreaticlevel. The second level had very low resistivity values, correspondingto waterlogged detrital facies. At this level, there were importantlateral variations in the resistivity values. These variations reflectlateral changes in the facies, which are usually bar channels andwedges. The thickness of this intermediate area is highly variable,ranging from 10 to 80 m at nearby points. The variation is related tothe synsedimentary tectonics that were deduced from the correlationsobtained from the mechanical boreholes. Finally, the lowest levelcorresponds to the Palaeozoic substrate and was characterised by aprogressive increase in resistivity values.

From a hydrogeological standpoint, electrical tomography is a veryuseful tool for determining the position of the water table in alluvialaquifers. In the example discussed in this paper, the phreatic levelcorresponds to the decrease in resistivity between the superficial leveland the intermediate one.Moreover, in sectors close to the coastline, thistechnique highlights possible marine intrusion, which is characterisedby a front of low resistivity that advances inland from the coast.

References

Auken, E., Jørgensen, F., Sørensen, K.I., 2003. Large-scale TEM investigation for ground-water. Exploration Geophysics 34, 188–194.

Caputo, R., Piscitelli, S., Oliveto, A., Rizzo, E., Lapenna, V., 2003. High-resolutionresistivity tomographies in active tectonic studies. Examples from the TyrnavosBasin, Greece. Journal of Geodynamics 36, 19–35.

Colella, A., Lapenna, V., Rizzo, E., 2004. High-resolution imaging of the High Agri ValleyBasin (Southern Italy) with electrical resistivity tomography. Tectonophysics 386,29–40.

DeGroot-Hedlin, C., Constable, S., 1990. Occam's inversion to generate smooth, two-dimensional models from magneto telluric data. Geophysics 55, 1613–1624.

Elorza, J.J., García Dueñas, V., Comas, M.C., González-Donoso, J.M., 1981. Mapa geológicoy memoria explica de la hoja 1.054 (Vélez Málaga), a escala 5:000. InstitutoGeológico y Minero de España (IGME), Madrid.

Gallipoli, M., Lapenna, V., Lorenzo, P., Mucciarelli, M., Perrone, A., Piscitelli, S., Sdao, F.,2000. Comparison of geological and geophysical prospecting techniques in the

study of a landslide in southern Italy. European Journal of Environmental andEngineering Geophysics 4, 117–128.

García-Aróstegui, J.L., 1998. Estudio hidrogeológico y modelización del acuífero de losríos Vélez y Benamargosa (Málaga). PhD Thesis, Univ. de Granada, 377 pp.

Gourry, J.Ch., Vermeersch, F., Garcin, M., Giot, D., 2003. Contribution of geophysics to thestudy of alluvial deposits: a case study in the Val d'Avaray area of the River Loire,France. Journal of Applied Geophysics 54, 35–49.

Griffiths, D.H., Barker, R.D., 1993. Two-dimensional resistivity imaging and modelling inareas of complex geology. Journal of Applied Geophysics 29, 211–226.

Lario, J., Zarzo, C., Dabrio, C.J., Somoza, L., Goy, J.L., Bardají, T., Silva, P.G., 1995. Records ofRecent Holocene sediments inputs on spit bar and deltas of south Spain. Holocenecycles: climates, sea levels and sedimentation. Journal of Coastal Research 17,241–245 (Special Issue).

Loke, M.H., Barker, R.D., 1996. Rapid least-squares inversion of apparent resistivitypseudosections by a quasi-Newton method. Geophysical Prospecting 44, 131–152.

Loke, M.H., Dahlin, T., 2002. A comparison of the Gauss–Newton and quasi-Newtonmethods in resistivity imaging inversion. Journal of Applied Geophysics 49, 149–162.

Maillet, G.M., Rizzo, E., Revil, A., Vella, C., 2005. High resolution electrical resistivitytomography (ERT) in a transition zone environment: application for detailedinternal architecture and infilling processes study of a Rhône River paleo-channel.Marine Geophysical Researches 26, 317–328.

Massey, A.C., Taylor, G.K., 2007. Coastal evolution in south-west England, UnitedKingdom: an enhanced reconstruction using geophysical surveys. Marine Geology245, 123–140.

Naudet, V., Lazzari, M., Perrone, A., Loperte, A., Piscitelli, S., Lapenna, V., 2008. Integratedgeophysical and geomorphological approach to investigate the snowmelt-triggeredlandslide of Bosco Piccolo village (Basilicata, southern Italy). Engineering Geology98, 156–157.

Nielsen, L., Jørgensen, N.O., Gelting, P., 2007. Mapping of the freshwater lens in a coastalaquifer on the Keta Barrier (Ghana) by transient electromagnetic soundings.Journal of Applied Geophysics 62, 1–15.

Sasaki, Y., 1992. Resolution of resistivity tomography inferred from numericalsimulation. Geophysical Prospecting 40, 453–464.

Store, H., Storz, W., Jacobs, F., 2000. Electrical resistivity tomography to investigategeological structures of earth's upper crust. Geophysical Prospecting 48, 455–471.

Sumanovac, F., 2006. Mapping of thin sandy aquifers by using high resolution reflectionseismics and 2-D electrical tomography. Journal of Applied Geophysics 59, 345–346.

Suzuki, K., Toda, S., Kusunoki, K., Fujimitsu, Y., Mogi, T., Jomori, A., 2000. Case studies ofelectrical and electromagnetic methods applied to mapping active faults beneaththe thick quaternary. Engineering Geology 56, 29–45.

Telford,W.M., Geldart, L.P., Sheriff, R.E., 1990. Applied Geophysics. Cambridge UniversityPress. 770 pp.