Aquifer Characterization Using Shallow Geophysics Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Effective management of groundwater resources is a well known problem in several areas around the world. It is especially important to areas suffering from an intrinsic lack of fresh water such as islands. Detailed study of available aquifers has great interest in the high touristic greek island of Crete. The growing water demand makes water resources management extremely important for sustainable development. This is the case in the prefecture of Chania in Western Crete, where there has been lack of success of many different groundwater management plans that have been produced over the years and submitted for application to water management authority. The inefficiency of the management plans is verified by the continuous water shortage reported every year which forces the public authorities to transfer water from far away sources and wells. Until today there are only geological data as well as information from the borehole logs, so the current work is focused to study in detail the tectonic and hydrogeological characteristics of the Keritis watershed in order to make a correlation of substantially information with the geophysical data of the study aquifer. Transient electromagnetic soundings were conducted in order to obtain detailed information about: the tectonic, hydrogeological, hydrolithological and geometrical characteristics of the aquifer under investigation. Its limits as well as the level and direction of groundwater movement are clearly identified. In addition, optimum areas of water well drilling were identified in order to minimize the uncertainty and the total cost (economical, managerial, etc) of future groundwater surveys.

Keywords: water management, hydrogeophysics, TEM, aquifer characterization

1. Introduction

Groundwater management has led to the need for accurate investigation and description of aquifers. A hydrogeological characterization of a specific area can be achieved by estimating a set of quantitative parameters of the aquifer, such as aquifer thickness and extent, hydraulic conductivity, hydrolithological units and tectonics. We should have always in mind that the active tectonic is a main reason for dividing the region into different hydraulic units (Keller and Pinter 1996; Aryamanesh et al. 2009). The aforementioned parameters are most important in areas where groundwater management plans prove insufficient or inaccurate.

One of the two main watersheds (aquifer systems) in the prefecture of Chania and the nearest to the capital city is the Keritis basin. Until today, there is no (published) large scale detailed geophysical study in this area since the main exploratory tools for the Water Management Authority of Chania (WMACh - Dr. K. Vozinakis, personal communication, 2008) are boreholes and local geophysical investigations. Previous geoenvironmental studies on the area (Lionis and Perleros 2001; Soupios et al. 2007; Kouli et al. 2008; Nikolaidis et al. 2010) involved the application of geological, geophysical, geochemical and other methods in the study area.

The NW Crete suffers from the lack of a central water management plan. The main reason for this is the absence of the complete knowledge of the hydrogeological status of the area. Several areas are not sufficiently irrigated during the critical periods due to insufficient aquifer delineation information. In addition due to a non-centralized management organization (there are four independent involving authorities) there are limited financial potentials. This status leads to the use of groundwater investigation methods that are less expensive and time consuming than boreholes. Processing of satellite and aerial images can be helpful for the identification, using indirect methods, of invisible in-situ fracture zones and to confirm already defined tectonic features. River networks may also be correlated with the tectonic regime of the area. After all this has been done, groundwater geophysics can be applied to map the groundwater system for both domestic and irrigation purposes. All the above are crucial for the assessment and management of the shallow (phreatic) aquifer and, if it is possible, also the deeper, karstic aquifer of Keritis Basin.

The theoretical and practical background of the application of geophysics for groundwater investigations has been extensively presented and reviewed by researchers (Grant and West 1965; Telford et al. 1976; Dobrin 1976; Parasnis 1979; Kearey and Brooks 1991; Sharma 1997; Milsom 2003; Reynolds 2010). Near-surface geophysics for groundwater applications include mapping the depth and thickness of aquifers (Mazac 1985; Huntley 1986; Goldman et al. 1988; Robain et al. 1996; Godio and Bottino 2001; Albouy et al. 2001; Krivochieva and Chouteau 2003; Danielsen et al. 2007; Sultan and Santos 2008; Wattanasen and Elming 2008), locating geological formations (such as fractures and fault zones) that could act as preferential fluid paths (Christiensen and Sorensen 1998; Jongmans et al. 2000; Young et al. 2004; Batte et al. 2008) and mapping contamination to the groundwater such as that from saltwater intrusion (Custodio and Bruggeman 1987; Mills et al. 1988; Goldman and Kafri 2006; Hamzah et al. 2006; Kafri and Goldman 2005; Kafri et al. 2007; Cimino et al. 2007; Duque et al. 2007; Adepelumi et al. 2008).

Methods for applying geophysical techniques in groundwater exploration have been presented (Van Dongen and Woodhouse 1994) but as MacDonald et al (2001) point out, areas with complex geology and hydrogeology are not covered by the general approach and require specific methods for particular problems. Most geophysical techniques have been used for groundwater characterisation but the electrical and electromagnetic methods have the greatest success and can be directly used for mapping and monitoring fresh or contaminated groundwater areas around the world (Meju et al. 1999; Peavy and Valentino 1999; Sørensen and Søndergaard 1999; Gwaze et al. 2000; Corriols et al. 2000; Mendoza et al. 2000; Miele et al. 2000; Maillol et al. 2000; Wynn et al. 2000; Farrell et al. 2000; Paine et al. 2000; Soupios et al. 2010).

In this paper the current hydrogeological status of the Keritis Basin of NW Crete Island in Greece, by means of geoenvironmental approaches is presented. More specific, electromagnetic measurements in the mode of transient (time domain) electromagnetic soundings were conducted in the Keritis basin, in order to obtain detailed information about: the tectonic regime, the geometry of different hydrogeological/lithological units and the geometrical characteristics of the aquifer of the area under investigation. A total of 919 geophysical measurements (sounding) were acquired in 314 different locations, in correlation with other related available information (such as geology, borehole logs and tectonic) and finally, the 2D and 3D model reconstruction of the subsurface of the study area was achieved. The results confirm a structure of 5 main different geological formations, as well as the aquifer geometry and the optimum areas for hydrowell construction were identified.

2. Study Area

2.1 General

The study area is the Keritis basin which is situated in the northwestern part of Chania Prefecture, at Crete Island in Greece. It is located between latitudes 35o2450N and 35o3000N and longitudes 23o4959 and 23o5800E (Fig.1).

The area is an agricultural district where many villages surround the basin (Fig.1). The Keritis basin covers an area of approximately 136km2. The hydrological basin of Keritis represents one of the most important basins in the municipality of Chania and belongs in the province of Kydonia. The main municipal districts which belong in the hydrological basin are: Agyia, Platanias, Agia Marina, Galatas, Barypetro, Geranion, Alikianos, Vatolakos, Koufos, Fournes, Skines, Brysai, Modion, Orthounion and Meskla.

The west and central region of the Municipality of Chania appears strong geomorphologic characteristics which alter from north to south. High variation of the elevation and the slope is observed from south (the area of White Mountains) to north where the relief became soft. This continues until the coastline areas with lowland cultivate areas.

2.2 Geological and Tectonic Settings

The major part of Keritis basin is covered (Fig. 1) by Quaternary deposits (Q1, Q2) and Miocene sediments (N1) expand to the northeast, southeast and southwest. In addition, Pliocene sediments (N2) in the northwestern part of the study area, and the Tripolis nappe (K2) also appear in the northeastern and northwestern part of the basin. Dissected hills of phyllites and quartzites (P-T3), a Late Carboniferous to Late Triassic package of sedimentary rocks composed mostly of quartz-rich siliciclastic sediments, with minor limestone, gypsum, and volcanic rocks (Krahl et al. 1983) are observed mainly in the south-western and south-eastern part of the study area. The Trypalion nappe (T3-J1) is exposed in the central-eastern part and finally the Plattenakalk Limestone (J2-E2) is exposed in the south-east (Fig.1). The lithostratigraphy and tectonic units that make up the geological structure of Keritis basin is shown in figure (1) (from newer to older) and are listed in Table (1).

The study area consists of a complex tectonic system surrounded by normal faults. The majority of the formations are calcareous and karstified. The current tectonic regime is dominated by north-south and east-west extensions. There are two main groups of currently identified faults: visible (which are depicted with solid black lines in Fig. 1) and concealed (which are depicted with red dashed lines in Fig.1) (Lionis and Perleros 2001). The main strike of visible faults is northwest-southeast and east-west (Fig. 1), which defines the boundaries between the existing hydrolithological units as well as the groundwater flow direction, since some of the faults bound and direct the groundwater flow.

2.3 Hydrogeological Settings

From the hydrogeological point of view, the Keritis basin rocks can be divided into four main categories according to the permeability of geological formations, as shown in figure (2),

High permeability rocks (Category I), inside of which the karstic limestones of Tripolis and Trypalion are comprised. High potential aquifers are expected to be hosted in these units.

Medium permeability rocks (Category II), which include Quaternary deposits belonging to the Holocene as well as to the Pleistocene. Medium potential aquifers are expected to be observed in these units.

Low permeability rocks (Category III), which include Neogene sediments of Pliocene to middle-upper Miocene age. The upper members consist of marly thick-bedded, organogenic limestones, locally brecciated, and the lower members consist of white-grey clastic, usually biogenic, marls-marly limestones.

Impermeable rocks (Category IV) consist of phyllites-quartzites. In some cases, due to the tectonism of the bedrock, low to medium potential aquifers are observed in the otherwise generally impermeable bedrock, (Lionis and Perleros 2001).

Specifically, two main hydrogeological systems and one secondary surface system are presented, as described below: The first is the underground hydrogeological system of the permeable carbonate formations that appear in the southeast part of the basin near the villages of Barypetron and Meskla. The main groundwater supply comes from the southern limestones formations recharged mainly on the massive carbonate limestone of the White Mountains (Lefka Ori). On the north part, this carbonate formation is cut off tectonically (Agyia) along an East-West fault, which causes the region to be flanked with impermeable phyllites-quartzites formations (more northern) and the creation of the springs in the Agyia area with an elevation of around 40m (Platanos-Kolumpa-Kalamiwnas). Based on the aforementioned geological and hydrogeological conditions of the study area, the water supply for Meskla springs (fig. 2) presents an average annual supply of 30.5 Mm3/year with maximum and minimum of 59.3 Mm3/year and 7.7 Mm3/year, respectively. For Agyia springs (fig. 2) the average annual supply is 60 Mm3/year with maximum and minimum of 98.3 Mm3/year and 60.3 Mm3/year correspondingly.

The aforementioned karstic springs supply this high amount of water (one of the biggest water supplies in Greece) but we should keep always in mind that a permanent karstic underground reservoir with unknown (unexplored) geometric and hydraulic characteristics exists in the study area. However, the existence of a high potential aquifer, inferred from the annual drainage of the springs, is expected (WMACh - Dr. K. Vozinakis, personal communication, 2008). It is mentioned that in the basin that feeds the local springs, three wells are in use from the Organization for Development of West Crete (ODWC): one in the Barypetron (Mylwniana) area with a supply of 700 m3/h and two wells in the Fournes area with a supply of 260 m3/h. Without those hydrowells the spring flows would be correspondingly higher.

The second hydrogeological system of the Quaternary deposits is exposed in the Phyllites nappe on the central part of the Keritis Basin and around the villages of Skines, Alikianos, Fournes, Vatolakkos and Koufos. Its supply comes from Keritis, directly from the surface supply river drainage system of the phyllites-quartzites and expands to the south, as well as from the subsurface flanking regions from the upper carbonate formations. The groundwater quality at the area around Barypetron (Myloniana) in the eastern part differs from the area of Koufos (, north-western part, a case that lead to the assumption of two sub-basins in the area (WMACh, Dr. K. Vozinakis, personal communication, 2008). Koufos' sampling point exists on the catchment of Phyllites-Quartzites which is a totally different rock-type from Plattenkalk limestones of Trypalion nappe (where the Myloniana sampling point exists). Additionally, there is utilization in the Quaternary deposits inside the hydrogeological basin, using boreholes with remarkable results (Table 2), such as high water discharges followed by only small changes of the water level.

The surface hydrogeological system of the phyllite-quartzites in the north part of the basin, which is generally an impermeable formation, collects the rainfalls and delivers into the streams of the study area. On the south-eastern part of the study area, the phyllites-quartzites (near Meskla village) are in tectonic contact with the Trypali limestones. The springs of Meskla (Panagia-Kefalobrysia), exist at the elevation of 210m, and their supply comes from the limestones and from the shallow drainage/leakages of the phyllites-quartzites of the Keritis Basin (WMACh - Dr. K. Vozinakis, personal communication, 2008).

3. Time Domain Electromagnetic Method (TDEM) geophysical method

3.1 General

The TDEM method has been used in environmental and hydrogeological studies over the last 25 years. The method is described in detail in several textbooks (Kaufman and Keller 1983; Fitterman and Stewart 1986; Stewart and Gay 1986; Goldman et al. 1988; Mills et al. 1988; Nabighian and Macnae 1991; McNeill 1994; Sharma 1997; Reynolds 2010). Therefore, only a short description is given below, in order to briefly discuss the essential features of the method.

The TDEM method belongs to the category of "controlled source" EM methods. A typical TDEM system consists of a transmitter (Tx) loop and a receiver (Rx) loop with equal or less dimensions. The size of Tx loop varies according to exploration depth but without being the parameter that directly controls the exploration depth as happened in other methods (e.g Vertical Electrical Sounding method). An increase in Tx loop initially affects (increases) the signal to noise ratio resulting in an increase in exploration depth (Barsukov et al. 2007).

A current flowing through the Tx loop generates a primary, stationary field. By using an abrupt on-off switching sequence, currents are induced into the ground according to Faraday's law, due to Ohmic resistance of the subsurface; the current system will decay and further induce a secondary magnetic field. This field, which is the transient response, is measured by induction in the coil (the Rx loop). The decay rate of the electromagnetic field depends on the distribution of the resistivity in the subsurface. In a conductive medium, the field decays slower comparing to a resistive medium. Based on this principle, the measured voltage on Rx coil can provide information about geoelectrical structures at several depths. In order to provide easier interpretation the measured voltage is usually converted to apparent resisitivity, ρa, according to following formula (Barsukov et al. 2007):


where μ0 is the magnetic permeability, R= (L/π1/2) is the effective radius of the Tx loop (L x L), r= (l/π1/2) is the radius of Rx loop (l x l), E is the voltage and I is the current.

The exploration depth is mainly affected by the time interval between subsequent turn-off and next turn-on. In order to explore deeper, a bigger time interval is required. Because of skin-effect, at early times, the induced currents are concentrated on upper layers leading to measurements that are sensitive to shallow structures only. As the time interval increases, the current intensity migrates to greater depths and the measured secondary field will depend more on the properties of deeper layers. In addition, the current density in shallow structures decreases, relaxing in this way the influence of them within the measured secondary field. This elimination of near-surface resistivity variations is a unique feature of TDEM method resulting in high quality data where other geoelectric methods failed.

3.2 Data Acquisition

The acquisition of the TDEM soundings took place using the TEM-FAST 48 (AEMR Company, Netherlands). The TDEM surveys were carried out by using a single square loop configuration of dimensions 50x50m, allowing an interpretation of the data in terms of the subsurface resistivity structure down to a depth of a maximum of 200m. The system was set to transmit current up to 4 Amp with 32 active time gates from 4 μs to 1024 μs and the stacking time about 3 minutes. To define and avoid aliasing effects (high frequency - HF noise) the measurements were repeated several times at each sounding location. GPS, (Garmin - 60CSx) was used for the accurate (less than 4m) positioning of the TDEM soundings. To be able to construct an image of the subsurface, the survey was conducted along profiles in different directions adopting an average distance between the sounding of about 200m. The preferential directions were NS based on the tectonic and geological characteristics of the study area (Fig.1).

The preparation and implementation of each sounding took 15-20 minutes, and for a 2-person crew, was able to take 10-15 soundings per day, depending on the weather conditions. In total, 917 electromagnetic soundings were acquired in 312 different locations, over a period of 4 months (from May 2008 until August 2008), which were initially calibrated using the available borehole-logs. The root mean square (rms) error of the final data set was less than 6%.

3.3 Data Analysis

The TEM-RES software package was used for processing the raw data solving the inverse problem in time domain electromagnetic soundings. The Receiver's transient voltages of each sounding are transformed into apparent resistivity to aid in the qualitative interpretation and inversion (Raiche and Spies 1981). The interpretation from Apparent Resistivity variation to 1D profile is achieved by the use of inversion. Initially, for each sounding, the best apparent resistivity versus (vs) time curve is selected in order to produce an inverted 1D, horizontal layering model. Processing of the raw TDEM data yields a vertical profile of apparent formation's apparent resistivity vs depth. Low resistivity values (smaller than 10-30 Ohm-m) usually are a result gained from the presence of fresh water in the subsurface. Resistivity values substantially less than 10 ohm-meters provide a strong indicator for the presence of salty or contaminated (by other sources) groundwater at that depth.

Reconstruction of the section from apparent resistivity curves (which is the common form of field data representation) is carried out by either the transformation (in this case a pseudosection based on gradient media, is calculated), or by solving the inverse problem (in this case a piecewise-homogeneous section is calculated) in a class of layered media. Although both approaches have their own inherent significance and can be used independently, in the current study a joint form of these two approaches is used: initially the transformation results are calculated and then, these results used to construct the initial model for the inversion process. Under this approach, the initial model for the inversion process is defined in respect of the study area characteristics. Thus, the first essential step of data analysis (after preliminary procedures) is data smoothing. Then the apparent resistivity is calculated.

These variations affect the shape of the curve ρh = ρ(h) and thus the resolving capacity of transformation. The reconstruction of ρ(h) is done automatically without the need of additional information since it is not necessary to specify the number of layers. Thus, the reconstructed sections are being pseudo-sections and there are times that they represent the resistivity distribution more satisfactory than inversion process for layered media (Barsukov et al. 2007).

4. Results

4.1 2D projections

The results from inverted 1D soundings are initially calculated. Generally, the 1D modeling is inadequate to reconstruct and describe the subsurface, so 2D imaging is required due to its ability to construct pseudosections. In the forthcoming geoelectrical sections, red colors represent high resistivity formations (e.g. limestone), green and yellow colors demonstrate the medium resistivity formations (clay, sand, phyllites, quartzites), while the blue colors depict the low resistivity values (e.g. fresh- saline water). 1D model after the inverse processing can produce a 2D image that allows the user to get more detailed information of the study area. Results for each one of the 3 sections (as depicted in Fig. 1) are presented in figure (3). All of the sections are correlated with the geological, hydrogeological and borehole logs available information of the investigated area, in order to have the best fit reconstruction model for the Keritis Basin.

In the geological sections, four main categories of rocks dominated (fig. 3): a) quaternary deposits (A), b) neogene sediments (B), c) phyllites-quartzites (C) and d) trypali carbonates (D). More specifically, the formations composed of Quaternary deposits are alluvial deposits, terra rossa, clays, and weathered material from the erosion of the bedrock. Neogene deposits are composed mainly of marly limestones, marls, sandstones and clays. The phyllites - quartzites nappe is composed of phyllites, quartzites, shales, schists and meta-sandstones, while the trypalion carbonates are composed of cohesive limestones and dolomitic limestones.

Regarding fault recognition in a geoelectrical section, a first indication is the high resistance contrast (Vanneste et al. 2008). When these types of results were produced they were treated as preliminary indicators which were then evaluated by information from geological maps, borehole logs (discontinuities in geological layers, etc) and field trips. The resulted tectonic structures were additionally correlated with all the known faults. New faults (as indicated by geoelectrical section) were evaluated by field trips since all the new faults (except the depiction of the buried fracture zone near Agyia Lake) were on the surface and some were extensions of previously known ones. About concealed faults, since the geoelectrical section was evaluated for surface ones, the indications were checked and evaluated using the most updated geological maps, the available borehole logs and by other sub parallel geoelectrical sections. Since all the well known faults were greatly correlated with the geological maps it was safe to assume that new faults indications (from geolectrical sections) dictate unidentified concealed faults.

Regarding the current tectonic regime, the derived (existing as well as for new) tectonic lines as resulted from borehole logs evaluation, application of geophysical method and geological mapping/fieldwork, are presented in Figure (4).

More specifically, 21 faults were pointed out in the study area (Fig. 4). From these, 15 were already identified and 6 new faults were identified from fieldtrips and geophysical methods. From those, 2 were totally new (NF_2 and NF_6). The remaining 4 were extensions to already known faults. More specifically, NF_1 extends 21, NF_3 extends 7, NF_4 extends 16 and NF_5 extends 13. It is important to mention that in many fault the rupture zone is defined between conductive and/or resistive formations. Another important finding is that the fault with the ID 17 (the main ENE-WSW fault cross the Keritis valley) was referred as concealed and now is verified from the geophysical measurements. This fault plays vital role for the water management of the study area since, a) it stops the water flow to the North (there are no observed sub seas water discharge by processing satellite images) and b) it acts as pathway for the groundwater to the ENE (towards to Souda gulf).

Overall, the 2D sections show that the aquifer system is complex and show locations with a high potential for groundwater. With the 2D sections, the general scientific objective is the characterization of the physical properties and distribution of the thicknesses of sediments (Godio and Bottino 2001). With the help of the 2D geophysical interpretation, all of the main tectonic features of the area were able to be identified.

4.2 3D imaging - depth slices

By grouping and projecting at various depths all the 2D sections, a set of horizontal slices for 3D imaging for resistivity distribution in the study area is constructed and presented in Figure (5). A set of five independent slices is produced according to depths of 0-25m, 25m-50m, 50m-100m, 100m-150m and 150m-200m.

The first result is that all the formations depicted in depth slices are well correlated with the geology as depicted in Figure (1). The first two shallow depth slices (Fig. 5A, B, 0-50m) demonstrates the highly and intermediate resistive zones on the SE and SW of the investigated area, respectively. The high resistivity values in the SE direction can be correlated with the Trypalion nappe (T3-J1) (more specifically the limestone-dolomite formations) as shown in Figure (5B), whilst the intermediate resistivity values in the SW direction and north central part (bounded by thick dashed lines) indicate the Phyllites-Quartzites nappe (P-T3). Areas with low resistivity, depicted in blue-like colors, are composed of quaternary deposits and neogene sediments which in places are saturated with water. This assumption is verified in several ways: by the geological mapping, from the borehole logs and shafts' water level and from personal communications with local experts. This shallow aquifer area (till the depth of 50 meters), orientated E-W, covered almost all the central part of the basin. The contact between the calcareous rocks and the Neogene formations in the SE part of the basin and between the phyllites-quartzites and the Neogene formations in the northern part of the basin is sharp, and is considered to be fault zones striking approximately ENE-WSW and NNW-SSE, respectively. The red lines delineate the faults and illustrate the fracture zones between the Trypallion nappe and the phyllite-quartzite nappe. Another important point is that these red lines are in very good agreement with the regional geological field map (Lionis and Perleros 2001) and fault depiction by satellite imaginery methods (Kouli et al. 2008).

At the third and fourth depth slices (Fig. 5C, D, 50-150m), high/medium and low resistivities are depicted in the investigated area. Specifically, a conductive zone (dotted ellipse) on the northern part of the basin and along the coastline can clearly identified as well as a high resistance zones depicted in the SE. The conductive area in the northern part of the basin (Neogene sediments) is interpreted as sea water intrusion from the NE side, since the coastline is only a few hundred meters away. The separate conductive zone in the central area is evidence of fresh water and verified by boreholes' logs. The fourth depth (100m-150m) slice (Fig. 5D), has intermediate and low resistivities which are detected in the northern part of the basin. As mentioned above, the Trypallion nappe is the bedrock of the investigated area (Douwe and Meulenkamp 2006; Papanikolaou and Vassilakis 2008) and is shown in the 100-150m depth slice as a cohesive and unfractured formation. It is important to note that both during the survey, and after the interpretation, a superparamagnetic effect (noisy data) was observed, when the measurements were acquired over the Trypalion unit (due to the rich in iron composition). Two more areas were also depicted (see the purple rhombs) which shows the Fournes spring (at the southern part) and Agyia lake (and the nearby springs) at the central part of the Keritis basin.

At the last fifth depth (150m-200m) slice (Fig. 5E), dashed arrows show the directions of the water movements in correlation with the hydrolithological map (Lionis and Perleros 2001). Overall, the aquifer system is recharged from the karstic system of White Mountains (2600m height) with direction from South to North and from the Plattenkalk formation (J2-E2, Fig. 1) to the Trypalion carbonates (T3-J1) which has the highest transmissivity in the study area (Soupios et al. 2007). The fracture zones play a vital role in water movement through the Keritis Basin, as mentioned above.

5. Discussion

5.1 Tectonic implications

The tectonic features (fracture zones) of the study area act either as underground barriers or as pathways for the groundwater movement as already mentioned. This could happen because the main elongated WSW-ENE fault with id number 17 and the existence of the impermeable phyllites-quartzites nappes and Neogene sediments at the N and NW part respectively, both act as a barrier for the groundwater movement to the North and consequently there are no significant subsea groundwater leakages/discharges.

The Trypallion nappe, as a carbonate formation, includes all the features related to karst development, tectonism and fragmentation. These procedures make it an important hydrogeological formation. Keeping in mind the large area covered by the Plattenkalk limestones of the Trypallion nappe in the broader area, there is a considerable amount of water fed in due to precipitation. This massive formation of the Plattenkalk limestone gives valuable supplies of water through infiltration to the groundwater. The maximum precipitation from White Mountains that penetrates the surface of Plattenkalk limestones produces water reservoirs which then decant their content to neighboring permeable formations. The produced aquifers are then identified by the uprush of karstic springs due to tectonic conditions of each area. Typical examples of this type of spring are the surficial lake of Agyia and the springs of Meskla and Agyia. Except from the measured supplies, the above situation is verified by an experiment that took place (ODWC 1990, 2007) where a colored substance (tracer analysis) introduced at Omalos plateau (at 1050m altitude) which was then traced at the Meskla springs and Agyia and Mylwniana boreholes.

Based on the knowledge described above, two possible scenarios for the groundwater can be suggested,

1) the groundwater deposited in the area of Agyia Lake, which can be evaluated by the surficial appearance of the groundwater,

2) since we did not observe any groundwater leakage/discharge to the north, and the observed water supplies (from boreholes and springs near Agyia) are less than the expected (as estimated from precipitation and infiltration in recharge zone - White Mountains and presented in Perleros et al. 2004) we can assume that the main WSW-ENE fault (id:17) acts as a pathway for the groundwater discharge to Souda Bay in the NE of our study area. This assumption can be also supported by the fact that the boreholes/shafts along the axis of fault 17 produce good water yields that is not supported by the shallow-porous aquifer.

5.2 Saltwater intrusion

Analysis of the results highlights the presence of some extremely low values of resistivity along the coast, (<1 ohm-m) for at about a 1 km width. Moreover, there were substantial agreement between the geophysical (TDEM) and the hydrogeological results.

Overall, the possibility of contamination (salt water intrusion) of the aquifer in the study area due to the proximity to the coastline does not exist since it was found from the geophysical interpretation that generally, the N-S fracture zones (id_1, 21, 9) are impermeable and they don't allow the groundwater movement (discharge into the sea and/or salt-water intrusion in land). Moreover, at the soundings 307-312, 001, 305-302 the groundwater seems contaminated from the seawater as depicted in the 2D and 3D interpretation. The formations of the phyllites-quartzites and the Neogene sediments in the north of the basin act as barrier to the water movement, so that the saline water cannot pass through these formations.

The aquifer in the study area is hosted into hard (fractured) rocks (calcitic) with secondary porosity (voids, fractures, etc). The water hosting bedrock (dolomitic limestone) appears with low resistivities whereas the dry dolomitic limestone presents high resistivities. The resistivity contrast between these two situations can be about 500-1200 Ohm.m. Due to the complex geological formations that compose the Keritis basin, the post-Alpine rocks, and especially the neogene sediments (Pliocene and Miocene), are marine deposits indicating deposits from fresh/brackish water (Lionis and Perleros 2001). Regarding classification from the geophysical methods, these formations have low resistivity values because of the composition of the formations and not from saline water. The existence of saltwater intrusion cannot be verified only by low resistance values but must also come in conjunction with real situation (e.g how close is the coastline and interconnection with tectonic features). Thus, a low resistance value 1500m away from coastline is not a proof of saline water intrusion. In addition, in order to classify the fresh, brackish and saline water from the geoelectrical sections, always a correlation is made using additional data from the geological map or from borehole information (stratigraphy, water table, water quality information, etc). In a recent study (Kouli et al. 2008) shows that the presence of SO42- ions from dissolution of gypsum and other mineral deposits increase the salinity of groundwater without having any evidence for seawater intrusion. This dictates that chemical analysis must be done in order to acquire the complete knowledge of saline water origination.

5.3 Fresh water

Based on this geological, hydrological, hydrolithological, tectonic and geophysical study of the Keritis basin, there are two optimum positions located for the construction of the boreholes in the area presented in the TDEM depth slices. The positions are the zone between the Phyllites-Quartzite and the Limestone Dolomite of the Trypallion nappe. The boreholes that are close to these positions have large water supply (700m3/h). More specifically, the locations below the soundings 142 (Agyia area) and 270 (Fournes area) seem the best positions for drilling. In addition, there is a candidate area for successful drilling around the circumference of the Trypallion formation which is depicted in Figure (5). The above situations are depicted in Figure (5D) (100-150m) where purple diamonds demonstrate the optimum locations for drilling. Based on these sections, soundings 142 and 270 depict that there is likely to be a valuable amount of fresh water.

6. Conclusions

The purpose of the current study was to identify the current hydrogeological status of the Keritis Basin by means of multidisciplinary geoenvironmetal approaches. Initially, results from TDEM measurements were evaluated and calibrated using borehole data at numerous locations. Then, by means of 2D pseudosections, formations and known faults were evaluated and corrected and new faults were identified. By applying the 3D imaging, one can calculate the groundwater depth and the thickness of the aquifer. Supplementary, the depth slices are helpful in identifying the tectonic feature of the study area in order to have better understanding if the formations/faults act as a groundwater barrier, e.g. phyllites -quartzite formation or flow pathways, e.g. fault id_17. The results of the interpretation of the depth slices allow the identification of the 4 geological units (Quaternary deposits, Neogene sediments, phyllites-quartzites, Trypallion Limestone). Finally, from the above slices, optimum areas of drilling can be identified in order to minimize the uncertainty and the total cost for groundwater survey and management. This can be achieved because the new knowledge that produced from the current study could guide future (probably more detailed) investigations to specific areas where the presence of groundwater is already evaluated.