HYDROGEOCHEMISTRY OF COASTAL CARBONATE AQUIFER IN LUCIJA-PORTOROž (GULF OF TRIESTE, NORTHERN ADRIATIC SEA, SLOVENIA)
HIDROGEOKEMIJA PRIOBALNEGA KARBONATNEGA VODONOSNIKA V LUCIJI PRI PORTOROžU (TRžAŠKI ZALIV, SEVERNO JADRANSKO MORJE, SLOVENIJA)
Mihael BRENČIČ
Abstract	UDC 556.34.06(497.4)
Mihael Brenčič: Hydrogeochemistry of coastal carbonate aquifer in Lucija-Portorož (Gulf of Trieste, northern Adriatic Sea, Slovenia)
The 801 m deep artesian borehole Lu-1 in Lucija near Portorož (SW Slovenia) was drilled in 1994 on the coast. The borehole is constructed in flysch in the upper 257 m and continues in limestone below that depth. The groundwater has a very distinct chemistry. High concentrations of iron (10.4 mg/l), sulphide (15 mg/l), and chloride (6700 mg/l) are representative showing that mixing of seawater (31%) and groundwater (69%) under highly reducing conditions is present in the confined carbonate aquifer. Groundwater flow is extensive and deep, fed by higher elevation precipitation to the northeast.
Keywords: salt-water intrusion, groundwater-seawater interactions, groundwater discharge, confined limestone aquifer, Lucija, Slovenia.
Povzetek	UDK 556.34.06(497.4)
Mihael Brenčič: Hidrogeokemija priobalnega karbonatnega vodonosnika v Luciji pri Portorožu (Tržaški zaliv, severno Jadransko morje, Slovenija)
V letu 1994 je bila v Luciji pri Portorožu (JZ Slovenija) izvrtana 801 m globoka vrtina Lu-1, ki leži v neposredni bližini morja. Vrtina je v zgornjem delu do globine 257 m izvedena v terciarnih flišnih kamninah, v spodnjem delu pa v paleocenskih in krednih apnencih. Podzemna voda v vrtini ima zelo pester kemizem. Prisotne so visoke koncentracije železa (10,4 mg/l), sulfida (15 mg/l) in klorida (6700 mg/l), kar kaže na mešanje morske vode (31%) in podzemne vode (69%) pod vplivom redukcijskih pogojev v hidrodinamsko zaprtem karbonatnem vodonosniku. Opraviti imamo z obsežnim in globokim tokom podzemne vode, ki se napaja s padavinami na območju severovzhodnega sredogorja na razdalji več kot 10 km. Ključne besede: vdor morske vode, mešanje morske in podzemne vode, iztok podzemne vode, zaprt karbonatni vodonosnik, Lucija, Slovenija.
INTRODUCTION
Coastal areas are usually very densely populated and industrial activities, services, and tourist facilities are very well developed. For these reasons, demand for drinking water and other water usages is high. Demand is also present for geothermal water that can be used for heating as well as for balneological uses.
In coastal areas groundwater usage and seawater-freshwater interaction are among the key hydrogeologi-cal problems. A short overview of the problem is given by Todd and Mays (2005), and hydrodynamical calcu-lations regarding the saline-freshwater interface can be found in Bear (1979). For coastal karstic areas, a recent
Department of Geology, Faculty of Natural Sciences and Engineering, University of Ljubljana, Aškerčeva ulica 12, Ljubljana, Slovenia,
e-mail: mihael.brencic@ntf.uni.lj.si Department of Hydrogeology, Geological Survey of Slovenia, Dimičeva ulica 14, Ljubljana, Slovenia Received/Prejeto: 16.04.2009
ACTA CARSOLOGICA 38/2-3, 179-196, POSTOJNA 2009
overview is given by Fleury et al. (2007); however severa! review works about karstic groundwater in coastal areas can be found elsewhere (e.g. Breznik 1998; Cala-forra 2004; Tulipano et al. 2005; Breznik & Steinman 2008). Karstification of carbonate rocks from the Dinaric carbonate platform along the Adriatic coast is relatively deep and karstic features, especially submerged springs (e.g. Kuščer 1950; Kuščer & Kuščer 1964; Bonacci 1995; Fistanic 2006) and caves (e.g. Vrhovec et al. 2001; Suric et al. 2007) can be found below the present sea level. Today they can be found from the Gulf of Trieste in North Italy to the coasts of Albania in the south (Breznik 1973). Groundwater discharges into the Slovene littoral zone are not well known. Carbonates outcrop only in the area of Izola; other parts are covered by flysch rocks (Pleničar et al. 1973). There are several indications that ground-water flow is significant (Lapanje & Prestor 2003; Petrič et al. 2004) and that submerged springs are present (Ur-
banc et al. 2004; Žumer 2004, 2008; Faganeli et al. 2005); however there is no systematic knowledge about the hy-drogeology of the wider Slovene coastal area.
Drinking water supply problems are profound in the Slovene coastal area. For these reasons, many practi-cal hydrogeological investigations have been performed in the past; they were focused on drinking water supply and into low enthalpy geothermal groundwater usage. Many boreholes and wells were constructed and hydro-geology data obtained. Unfortunately, they are mainly not systematized or published. The aim of this article is to present hydrogeochemical characteristics and hydrogeological interpretations of the groundwater flow cap-tured in the borehole Lu-1 positioned in the coastal carbonate aquifer of the Slovene coast. Data from borehole Lu-1 are compared with those for groundwater sampled in the wider Portorož (SW Slovenia) environs.
SITE DESCRIPTION
Borehole Lu-1 is positioned south of Lucija, on the western part of the land that belongs to marine Marina Lucija (Fig. 1). The shortest distance between the borehole and the sea is 40 m and the top of the borehole tubing is at 2 m a.s.l. In the near vicinity of the borehole to its north, natural streams conducts surface water from the coastal hinterland into the sea. Water in the channel is brack-ish with chemistry depending on the tidal conditions. South of the borehole, abandoned saltpans at Fazan are present.
Understanding the geological structure of the area is very important for the interpretation of regional ground-water flow. The southwest area of Slovenia representing the maritime province is composed of Eocene flysch and transitional marl overlying the rocks of the Adriatic Di-naric carbonate platform which spans the Upper Trias-sic to the alveoline nummulite limestone of the Middle Eocene age. The area is located on the border between Istria belonging to the Dinaric foreland and Čičarija and Kras that belong to the External Dinarides. Thrust faults are present in the area that were activated post Miocene due to underthrusting of Istria beneath the External Dinarides. The thrust belt strikes generally NW-SE dipping toward NE. Thrust blocks consist of limestone and flysch rocks that alternate in the direction of dip (Placer 1981, 2005, 2007).
The borehole was completed to the depth of 801 m. To the depth of 23 m, the borehole is drilled in Qua-ternary organic clay sediments. In the interval between
23 m and 128 m in depth, interbeded marl and sandstone within flysch of Eocene age is present. In the interval between the depths of 128 and 257 m marls are inter-cepted; their upper part is Eocene and their lower part is Palaeocene age. These beds are followed by Palaeocene alveoline-nummulite limestone extending to a depth of 378 m. In its upper part pyrite lenses several millimetres in thickness are present. In the interval between 378 and 381 m clay with coal is found (Fig. 2). In the same beds, Sečovlje coal mine operated in the vicinity (Breznik 1973; Plenčar et al. 1973). From 381 m to the final depth of 801 m massive upper Cretaceous limestone with dark nearly black bituminous beds intercalated with white sparitic layers is found. Limestone joints are filled with calcite and thin films of iron minerals. Below 500 m, no open fissures conducting water were observed.
From the surface to a depth of 244.5 m the borehole is cased with concrete. From 244.5 m to the final depth the borehole is open without any casing. The borehole was drilled with reverse borehole fluid circulation. Water in the aquifer was pushed out together with drilling cuttings through the tubing with compressed air blown into the annulus space between tubing and casing. After the drilling was completed, the temperature at the casing mouth was 23°C and the temperature at the bottom under static no flow conditions was 28°C.
Groundwater hydrogeochemistry from borehole Lu-1 has been interpreted considering chemical analyses from other boreholes, springs, and surface water in
Fig. 1: Geographie position of borehole Lu-1 and other sampling points from the study (geology modified after Pleničar et al.1973).
the wider Portorož environs (Fig. 1). Details about the sampling methodology are given below. Sampling was performed in borehole D-1 near Orešje on the left bank of Drnica creek near the village of Dragonja. The borehole was drilled to a depth of 450 m where in the interval between 254 m and 450 m limestone is present. Samples were also taken from boreholes Al-0, Al-1, and Al-2, which are positioned southwest of Krkavče. The boreholes are drilled to depths of 140 m, 183 m, and 240 m respectively and are positioned close to each other. Bore-holes Al-0 and Al-1 were completed in flysch rocks; in borehole Al-2 limestone starts at the depth of 158 m. On the foothill below the village of Parecag near the aban-doned inn on the southern side of the Portorož-Sečovlje road, sample PA-1 was taken from borehole MM-1; this borehole was completed at the depth of 380 m and flysch ends at 210 m. In the area of Krkavče, samples were col-
lected from borehole KA-1. In this borehole flysch ends at 36 m and limestone is present until the final depth of 404 m. For the interpretation, sample PAL-1 from borehole HV-1 near the Palace hotel in Portorož was avail-able. The borehole was drilled to a depth of 705 m in similar lithological conditions to borehole Lu-1; however fissures are very seldom. Samples were also taken from springs in the vicinity (Fazan, Koblar) and from surface waters (river Dragonja and creek Rokava).
Fig. 2: Schematic geological profile with borehole Lu-1 position.
METHODS
SAMPLING AND CHEMICAL ANALYSES
Samples of groundwater were collected during the course of drilling of borehole Lu-1 (marked as P1 to P10). They were taken at the mouth of the tubing at the place where water overflowed. The sampling campaign was part of logging procedures intended to identify any changes in the borehole column. Samples taken during drilling operations were stored in plastic PVC bottles. In the lab they were filtered through a membrane filter with a 0.45 ^m pore diameter. The only exceptions are samples P9 and P10 (Tab. 1), which were filtered and fixed in the field. These samples were acidified to pH<2 for Fe2+ and Fe3+ determination.
Groundwater and surface water samples that were used for the comparative analyses of groundwater from borehole Lu-1 were taken during investigations per-formed in 1993 and 1994. The results are mainly ob-tained from drilling reports from the Archive of Geo-logical Survey of Slovenia. Collected samples were not treated in the field.
The majority of samples were analysed in the Labo-ratory for Water Chemistry, Biology, and Technology at the Chemistry Institute of Ljubljana, Slovenia. The only exceptions are analyses of sample KA 1/22 from 12
February 1993, which were performed in Public Health Survey of Novo mesto, Slovenia. The analyses performed are presented in Tabs. 1 to 4. During the pumping test in borehole Lu-1 between 4 and 5 May 1994, samples were also collected for salinity determination at the marine bi-ology station at Piran.
Seawater samples from Piranski zaliv were taken at two sampling points usually used by the Marine Biology Station at Piran. They were collected from the sea bottom and from the surface (MA samples: longitude 13.57, latitude 45.50333; F samples: longitude 13.55, latitude 45.53833). Samples were stored in the refrigerator at -20°C. They were analysed by suppressor ion chroma-tography at the Department of Chemistry and Chemical Technology, Faculty for Natural Sciences and Technology, University of Ljubljana.
The results of stable 618O isotope determination are expressed in standard delta notation (6) as per mil (%%) deviation from the standard as:
ô180(°/oo) =
( R
sample
R
-l
std
xlOOO
where R and R are ratios of the stable isotopes of
sample	std
18O/16O of sample and standard (V SMOW), respectively.
Tab. 1: Chemical analyses from borehole Lu-1 obtained during drilling.
Sample		P1	P2	P3	P4	P5	P6	P7	P8	P9	P10
Depth	m	43.0	65.0	91.0	113.0	190.0	500.0	611.5	704.0	800.0	800.0
Date		09.03.94	11.03.94	13.03.94	14.03.94	16.03.94	12.04.94	19.04.94	25.04.94	04.05.94	05.05.94
K+	mg/l	4.0	3.3	4.5	5.5	7.3	140	125	130	130	135
Na+	mg/l	290	295	647	613	524	3700	3500	3600	3500	3600
nh4-n	mg/l	<0.04	-	-	-	0.61	2.2	2.0	1.9	1.6	1.6
Ca2+	mg/l	405	505	231	251	327	428	434	416	435	417
Mg2+	mg/l	45.3	59.8	38.7	41.1	48.4	257	258	258	258	258
Cl-	mg/l	1210	1390	1315	1365	1365	6700	6700	6500	6700	6700
hco3	mg/l	300	282	343	325	309	306	232	235	300	262
no2-n	mg/l	0.002	0.002	0.001	0.001	0.002	<0.002	<0.002	<0.002	<0.002	<0.002
no3-n	mg/l	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
SO4-	mg/l	45.9	34.8	16.9	20.4	29.6	800	854	809	800	789
SiO2	mg/l	15.2	13.8	12.6	12.6	13.1	17.4	-	-	16.2	15.6
Fetot	mg/l	<0.01	-	-	-	-	5.2	0.13	<0.01	0.2	10.4
Fe2+	mg/l		-	-	-	-	-	-	-	0.2	8.3
Mntt tot	mg/l	<0.01	-	-	-	-	<0.01	-	-	-	-
pH		7.0	7.0	7.3	7.3	7.2	7.7	7.3	7.6	7.4	8.0
Dry 105°C	mg/l	2380	2625	2695	2550	2715	12100	-	-	-	-
KMnO4	mg/l	6.0	-	-	-	-	-	-	-	136	101
Sr2+	mg/l	-	-	-	-	-	-	-	-	18	18
Br-	mg/l	-	-	-	-	-	-	-	-	28.6	29.4
h2s	mg/l	-	-	-	-	-	-	-	-	33	19
S2	mg/l	-	-	-	-	-	-	-	-	15	0.6
TC	mg/l	-	-	-	-	-	-	-	-	58.6	51.7
IC	mg/l	-	-	-	-	-	-	-	-	55.6	49.3
TOC	mg/l	-	-	-	-	-	-	-	-	3.0	2.4
518O	%%	-	-6.31	-7.21	-7.15	-	-5.79	-5.73	-	-5.70	-5.73
Positive S18O values show relative enrichment of the material in the heavy isotope with respect to the standard and negative S18O values indicate a lighter composition than the standard. The S18O analyses were performed on a dual inlet VARIAN MAT 250 stable isotope mass spec-trometer.
Gas (CH4, N2, Ar, CO2, C2H6, and C3H8) analyses were performed with gas chromatography where volu-metric parts of gases were determined. Stable isotope and gas analyses were performed at Jožef Stefan Institute.
INTERPRETATION METHODOLOGY
For the interpretation of chemical and isotope analyses the following methods were used: ion pattern analyses, hydrogeochemical facies, comparative analyses between dissolved species, grapho-analytical methods, mixing models, and geochemical modelling.
Ion pattern analyses define samples according to the concentration of dominant dissolved species. For the hydrogeochemical facies determination, definitions according to Back (1961) based on the Piper diagram (Piper 1944) were followed (Domenico & Schwartz 1990).
The interprétation of water chemical analyses was based on molar ratios between dissolved species. This is a classical hydrogeochemical method based on the analogy and comparison of empirical values obtained in comparable aquifers. Usually molar ratios are used for the groundwater source determination (Schoeller 1962). We used mainly ratios based on Cl- because absorption of this species on the mineral surfaces from the aquifers is minor and can be used as a conservative tracer (Hem 1985). Absorption of Cl- is possible only in aquifers with very old groundwater; however this is not the case in the carbonate coastal aquifer of Lucija-Portorož. Further molar ratios of Ca2+/Cl- and SO42-/Cl- and Cl-/HCO3- indicate mixing of seawater and groundwater. Ratios where SO42- and HCO3- are included indicate redox conditions in the aquifer. Differentiation between waters from dolomite and limestone can be done with the help of the Ca2+/Mg2+ ratio. Parameter IEB1 is defined as the molar quotient (Cl- - Na+ + K+)/Cl-, which shows whether part of Na+ and K+ has been replaced with Ca2+ and Mg2+ (Schoeller 1962).
In the grapho-analytical methods, Piper (1944) diagram and indication diagrams based on D'Amore et al. (1983) were used. A Piper diagram is represented as a projection of cation and anion ternary diagrams into rhombic space. Diagrams according to D'Amore et al. (1983) are based on the interpretation of the Piper diagram. In these diagrams indication parameters denoted by the letters A to F are used. Parameters are presented as line diagrams where the whole diagram and not just one particular parameter is used for the chemical determination of the sample. Parameters are defined according to the following equations:
A =
B =
C =
D =
E =
'[HCO-]-[SO2A] { II-)
'toH Mi 2C-) 2(+)J
' [nci+] [er]N
100
100
U<+) 2(-)J
([Na+]-[Mg2+])
100
100
2« . f [Ca2+]+[Mg2+] [hCO-]
2(+)
100
F =
M-M-Ml
2«
100
Parameters define the ratio between dissolved spe-cies where square brackets represent concentrations in meq/l. Sum Z represents the sum of the anion (-) and cation (+) concentrations. Parameter A discerns water flowing through carbonate aquifers from water that flow through rocks enriched with gypsiferous evaporites. Parameter B distinguishes waters related with sulphates (e.g. gypsum) from waters that originate from marls and clay rocks rich with Na+. Parameter C distinguishes waters from flysch rocks and waters from evaporitic rocks. Parameter D defines waters from dolomitic rocks. Parameter E distinguishes waters that originate from carbonate rock from waters rich in sulphate. Parameter F shows enriched K+ concentrations (D'Amore et al. 1983).
Due to the very low yield until the depth of 190 m the groundwater temperature in borehole Lu-1 was in-terpreted as 14°C, representing the annual average air temperature in the area. From this depth onward the temperature of flowing water was considered to be 23°C as it was measured at the casing mouth.
MIXING MODEL
S18O and Cl- are considered to be conservative tracers. They can be used in the two component end member mixing models for the interpretation of seawater-groundwa-ter mixing. The model is based on the equations:
x1 + x2 = 1
where
x1 is the volumetric proportion in the first compo-nent,
x2 is the volumetric proportion in the second com-ponent,
c1 is the concentration of the conservative tracer in
the first component, , is the concentration i the second component, and j is the conce hole Lu-1).
THERMODYNAMICAL CALCULATIONS
c2 is the concentration of the conservative tracer in
c3 is the concentration in the sample mixture (bore-
Thermodynamical calculations were performed with the help of the geochemical code WATEQ4F (Plummer et al. 1976). Saturation indexes (SI . ,) of minerals available
mineral
in the aquifer, partial pressures of CO2 defined as pCO2, and redox potential Eh were calculated.
RESULTS
Results of chemical analyses are given in Tabs. 1-4. The results of grapho analytical and thermodynamic calcula-tions are given in Tabs. 5-8 and Figs. 2 and 4.
Comparison between cation and anion sums shows that relative error in chemical analyses is in the range between 1% and 5%. Higher concentrations of Cl- presented in the majority of samples, causing problems in determination of concentrations of other dissolved spe-cies. Therefore some determinations are subject to higher analytical errors than usual. During the interpretation all efforts were taken to obtain reliable data from the archive records.
During the pumping test in borehole Lu-1, the sa-linity of water flowing out from the borehole was deter-mined as 13.71%o. Together with stable isotope results of S18O from Lu-1, which are given in Tab. 1, samples for ö18O were also taken at the depth of 278 m where the value of S18O of -5.75% was obtained and at the depth of 405 m where the value of S18O of -5.73% was obtained.
In the seawater near Punta in Piran the ö18O value of +0.42% was obtained. In the spring at Fazan the ö18O value was -6.44% and in the spring behind the petrol station near the abandoned Koblar restaurant the ö18O value was -6.37%. Both samples were taken on 5 May 1994.
In the borehole D-1 near Orešje (Fig. 1) more sam-ples were taken over a longer period. In December 1981 the Ö18O value was -8.19%, on 16 July 1982 the Ö18O value was -8.57%, and on 19 April 1993 the ö18O value was -8.69% (Marinko 1994).
More detailed analyses were performed on the sample taken from borehole Al-0 (Fig. 1) on 7 July 1993. The pH of the water was 7.13 and the isotopic compositions of oxygen and deuterium were -6.19% and -41.5% re-spectively. The ö13C value in DIC was -1.96%. The gas emanating from the borehole has the following volu-metric composition: 97.58% CH4, 1.83% N2, 0.02% Ar, 0.13% CO2, 0.34% C2H6 and 0.09% C3H8. The Ö13C isoto-pic composition of the methane was -71.2%.
DISCUSSION AND
CONCEPTUAL MODEL
Borehole Lu-1 was drilled next to the sea. It was drilled in the upper part through the Palaeocene and Eocene flysch and in the lower part through the Palaeogene and Cretaceous limestone. According to the relatively high transmissivity of T = 6.3 x 10-4 m2/s determined during the pumping test it can be concluded that below the flysch cover of low permeability, a high yield and permeable confined carbonate aquifer exists (Marinko 1994). Based on the pumping test we can conclude that the aquifer is very permeable and that existent fissures and cracks con-duct water efficiently; however, based on drilling results and later logging the existence of caverns inside the limestone was not proved. Due to the near vicinity of sea and limestones' high organic content, the hydrogeochem-istry of the groundwater is complex and not typical for Cretaceous and Palaeogene limestone aquifers from the Dinaric carbonate platform. At the same time the results for borehole Lu-1 indicate that similar hydrogeochemical conditions exist in the whole coastal aquifer of the Slovene part of Istra; the results of chemical analyses from submarine springs next to Izola (Žumer 2004, 2008; Fa-ganeli et al. 2005) show some resemblance.
The hydrogeochemical conditions of aquifer systems that were drilled through by borehole Lu-1 were
INTERPRETATION
interpreted using the conceptual model shown in Fig. 4. This conceptual model is constructed based on a master log of the borehole (Marinko 1994) and on the geological and structural interpretations of the wider area (Placer 1981, 2005, 2007). The mixing of seawater from Piranski zaliv and groundwater originating from the hinterland karstified aquifer appeared in the coastal carbonate aquifer. Limestone's high organic matter content induces a chain of redox reactions.
HYDROGEOCHEMICAL FACIES ANALYSES
Comparison of ion pattern and hydrogeochemical facies shows that water in the aquifer captured with borehole Lu-1 is chemically stratified (Tabs. 5 and 6). Until the depth of 65 m, Ca2+-Na+-Cl--SO42--HCO3- facies is present, followed by Na+-Ca2+-Cl--SO42--HCO3- until the depth of 113 m, and from the depth of 500 m until the end of the borehole Na+-Ca2+-Cl--SO42- facies is present. The latter facies is also present in boreholes HV-1/94 (sample PAL-1) and Al-0 and in seawater of Punta Piran, showing that hydrogeochemical stratification of the aquifer is present in the whole coastal aquifer around Portorož. Facies stratification is common in coastal aquifers in intergranular aquifers (Appelo & Postma 1993) as well as in carbonate coastal aquifers where local stratig-
raphy does not consist only of carbonate rocks (Wicks & Herman 1996).
Molar ratios of the samples from borehole Lu-1 represent the autonomous group, in which resemble in some instances the samples from the Palace borehole PAL-1 and borehole Al-0 (Fig. 1 and Tab. 7). Three distinctive groups of molar SO42-/Cl- ratios can be ob-served. The ratio from the Punta in Piran is 0.05. Values for samples from deeper than 500 m from borehole Lu-1 are somewhat lower, between 0.043 and 0.046. In sample PAL-1 the actual ratio is 0.06, higher than samples from Lu-1. Relatively low values below 0.01 can be found in samples from P2 to P5 and Al-0. In all other samples the ratio is much higher, showing the predominant influence of groundwa-ter from the continent. Three groups can also be observed in the molar Ca2+/Cl- ratio. Again samples P6 to P10, Al-0, and PAL-1 represent the independent group. Slightly higher ratios can be found in samples P1 to P5. Samples Al-1 and AL-2 can be included in this group. In all other samples, higher values are observed.
A similar grouping can be expected in the molar Cl/HCO3- ratio; however this is not the case. Samples Al-1 and Al-2 cannot be in-cluded in the group of P1 to P5. The molar Ca2+/Mg2+ ratio has great discrimination potential. Samples P1 to P5 represent independent groups with values between 3.6 and 5.4. Samples P6 to P10 have very comparable
Fig. 3: Indicator diagrams according to D'Amore et al. (1983).
Fig. 4: Hydrogeochemical conceptual model of the coastal carbonate aquifer from Lucija portorož (borehole Lu-1).
values near to 1. Similar values between 0.6 and 1.3 can be found in samples Al-0, Al-1, Al-2, KA-1/1, PA-1, and PAL-1, which originate from flysch rocks. All other samples show a predominance of Ca2+.
Indicator diagrams by D'Amore et al. (1983) show the very distinctive origin of sampled waters (Fig. 3). Diagrams of samples P6, P7, P8, P9, P10, PAL-1, and Al-0 have nearly identical shapes to the diagram for the sea near Punta in Piran. Very similar shapes can be seen for samples P1 to P5; however, the range of parameters is smaller, especially for parameter B, which represents the influence of Na+, and parameter E, which discerns waters rich in sulphates. All these patterns show that the concentrations of Na+ and SO42- are lower compared to the first group. Other diagrams are completely différent. A diagram of sample Al 1 shows high similarity with dia-grams from the second group. Surface waters and waters
Tab. 2: Chemical analyses from the borehole groundwater samples in the wider environs of Lucija Portorož.
Sample		Al-0	Al-1	Al-2	KA-1/1	KA 1/21	KA 1/22	PA-1	PAL-1
Date		03.07.93	31.08.93	22.05.94	20.01.93	12.02.93	15.02.93	07.07.93	03.06.94
Depth	m	-	-	-	100-400	-	-	-	-
K+	mg/l	40	7.5	5.4	3.7	2.0	1.1	5.0	108
Na+	mg/l	3450	150	50	22.0	29	20.6	115	2640
nh4-n	mg/l	<0.04	1.1		0.06	0.87	2.3	3.3	0.9
Ca2+	mg/l	405	62.8	66.7	44.3	136	123	11.8	369
Mg2+	mg/l	185	37.4	32.3	43.7	12	12.4	5.9	179
Cl-	mg/l	6500	215	110	12.8	9.9	8.0	30	4750
hco3-	mg/l	228	357	287	355	509	508	295	233
NO-- N	mg/l	<0.002	<0.002	-	<0.002	<0.01	<0.002	<0.002	-
NO--N	mg/l	1.1	<0.4	-	<0.01	<0.5	<0.01	<0.4	-
SO4-	mg/l	3.7	44.4	44.2	31.7	3.6	1.6	76.1	800
SO	mg/l	-	-	24.0	-	13.5	-	0.6	24.4
FeM	mg/l	0.8	0.79	-	0.02	5.3	4.9	17.3	-
Mntot	mg/l	<0.01	<0.01	-	<0.01	0.28	0.27	<0.01	-
pH		6.9	8.4	-	8.1	7.0	7.2	8.9	7.6
Dry 105°C	mg/l	12990	780	-	367	-	456	397	-
Dry 180°C	mg/l	12300	-	-	-	433	-	-	-
KMnO4	mg/l	180	120	-	12.0	2.3	10.4	9.0	-
Sr2+	mg/l	-	-	5.7	-	-	-	-	11
Br-	mg/l	-	-	0.2	-	<0.1	-	-	13.5
H2S	mg/l	-	15.6	-	-	-	-	-	-
S-	mg/l	-	-	-	7.3	-	<0.1	-	-
TC	mg/l	-	-	-	-	-	-	-	46.5
IC	mg/l	-	-	-	-	-	-	-	45.2
TOC	mg/l	-	-	-	-	-	-	-	1.3
SO H ÍQ 20	20 JQ 0) 53
Ca	Na*K hc03	cl
Fig. 5: piper diagram.
Fig. 6: Vertical distribution of Cl- and S18O in groundwater outflowing during driling from borehole Lu-1.
from the springs show patterns that are representative for aquifers under the predominant influence of carbonates with low Mg2+ concentrations. Similar to the latter group is the diagram of samples KA-1/21 and KA-1/22, with a
Tab. 3: Chemical analyses from the spring water, surface waters, and sea of piranski zaliv.
Watercourse		Fazan	Koblar	Rokava	Dragonja 1	Dragonja 2	Punta - sea
Date		05.05.94	05.05.94	03.06.94	20.01.94	03.06.94	13.04.94
K+	mg/l	1.5	1.5	2	0.7	2	480
Na+	mg/l	13	9	11	7.7	11	11800
nh4-n	mg/l	<0.04	<0.04	<0.04	<0.04	<0.04	<0.1
Ca2+	mg/l	129	130	80.0	86.2	71.4	450
Mg2+	mg/l	20.2	10.8	10.2	9.6	10.8	1380
Cl-	mg/l	46	21	15.2	9.7	10.9	21300
hco3-	mg/l	402	396	249	265	226	160
no2 - n	mg/l	-	-	<0.009	<0.002	0.27	0.003
no3-n	mg/l	7.6	7.4	0.21	1.4	0.02	<0.1
SO42-	mg/l	-	-	30.4	44.2	34.4	2930
s¡°2	mg/l	-	-	-	-	-	0.13
FeM	mg/l	0.03	0.02	-	0.09	-	0.01
Mntot	mg/l	-	-	-	<0.01	-	<0.01
pH		7.2	7.4	8.0	7.9	8.2	7.8
Dry 105°C	mg/l	-	-	-	324	-	39200
KMn°4	mg/l	9.5	6.3	-	6.2	-	-
more negative parameter D, showing a greater influence of Mg2+ concentrations, prob-ably from the dolomite rocks. The diagram of Al-2 shows similarity with the diagram of Al-1. The diagram of sample PA-1 has no resemblance with other diagrams. It is one of the diagrams typical of sandstone aquifers.
MIXING MODEL
From the results of chemical analyses of Cl- (Tab. 1) and from distribution along the drilling length (Fig. 6), it can be seen that two ranges of Cl- concentrations are present in borehole Lu-1. In the depth interval between 45 m and 190 m, concentrations are in the range 1210 mg/l to 1390 mg/l, and in the interval from 500 m to 801 m, measured concentrations are 6700 mg/l, ex-cept in the case of the depth 704 m, where concentration is 6500 mg/l. As one of the end members in the
mixing model we have used Cl- concentrations from sample KA-1/22 (Tab. 2) and as another end-mem-ber in the mixing model, seawater from Punta Piran was used (Tab. 3). Based on the application of two end-members, the mixing model proportion of seawater in the depth interval between 0 m and 190 m is between 6% and 6.5% and in the interval between 500 m and 801 m the proportion of seawater is between 30.5% and 31.0%. The mixing model was also calculated based on salin-ity data from the end of the pumping test. The results for the proportion of seawater in the groundwater are slightly higher than for the Cl- cal-culation: the proportion is between 36% and 38%.
Tab. 4: Results of seawater chemical analyses and water from borehole Lu-1, June 1994.
		Borehole Lu-1	Sea F - bottom	Sea F - surface	Sea MA - bottom	Sea MA - surface
Cl-	mg/l	6461	13487	15309	16170	15897
SO42-	mg/l	847	2849	3034	2825	2659
Br-	mg/l	22.1	46.8	50.1	50.4	53.7
hco3-	mg/l	210	112	128	115	138
pH		7.8	8.1	7.9	7.8	7.8
T-i—riTni]-1-1—rn-ni]-1-1—n-rrrr]-1-1—t-ttttt]-1-1—i i i m i|-1
0,1 1 10 100	1000
Cl [mmol/l]
Fig. 7: Mixing model.
The mixing model was also calculated with Br-concentrations as a conservative tracer (e.g. Davis et al. 1998; Kim et al. 2003; Freeman 2007). In the groundwater end member the concentration of Br- is assumed to be much lower than groundwater Cl- concentration and in calculations a concentration of 0 mg/l was used. The share of seawater based on Br- concentrations is in the range of 41% to 57%. Discrepancies between calculations of Br- and Cl- are rather large. The reason for this cannot be explained exactly. One possible cause may be inaccuracy of the analytical procedures; the other may be that the geochemistry of the Br- ion differs from that of Cl-, whether as an additional source in the aquifer or as a non conservative tracer. This question must remain open.
ION EXCHANGE PROCESSES
Seawater-groundwater mixing effects can also be ob-served in the diagram of the mixing ratio of the whole set of samples (Fig. 7). The line represents the dilution of the original seawater solution and particular samples are represented by dots. Following from the two end member tracer mixing model, Na+ and Cl- concentrations at different mixing ratios are distributed along the line. For the samples where this presumption is valid, dots are distributed along the line. If the sample is shifted from the line and if we suppose that seawater is present then this deviation is the consequence of some processes; usually ion exchange appears on the ex-changeable sites in clay minerals of the rock or sediment penetrated with seawater (Appelo & Postma 1993). If the sample is positioned above the line then enrich-ment of Na+ is present; if it is positioned below the line, concentrations of Na+ have been depleted. Ion exchange reactions can be presented by the equations: CaX + 2Na+ « Na2X + Ca2+ CaX + 2K+ « K2X + Ca2+ MgX + 2Na+ « Na2X + Mg2+ MgX + 2K+ « K2X + Mg2+
where X represents ion exchange sites in the aqui-fer minerals. These processes are usually profound when concentrations of cations are higher as is the case in sea-water.
Based on the mixing model, samples from borehole Lu-1 are positioned along the line. Shallower samples from the upper part of the borehole (P3, P4 and P5) are slightly shifted down. Samples Al-0, Al-1, Al-2, and PAL-1 are positioned along the line showing that their shares of seawater are 30%, 1%, 0.5%, and 22% respectively.
Tab. 5: Distribution of the ion pattern.
Tab. 6: Hydrogeochemical facies according to Back (1961).
In samples KA-1, PA-1, P1, and P2, déviation from the dilution line is very distinctive. According to the position of borehole KA-1 it can be concluded that groundwater captured by this borehole is not under the influence of the sea and the positions of these samples in the diagram do not reflect ion exchange processes. Sample PA-1 re-flects an additional source of Na+, probably clay minerals in the flysch (Tab. 7). This sample has the lo west IBE1 parameter showing high release of Na+ and high absorption of Ca2+ and Mg2+. This is also in accordance with the D'Amore diagram showing totally diffèrent characteris-tics of sample PA-1.
In the mixing diagram (Fig. 7), the position of Lu-1 samples is the consequence of seawater-ground-water mixing. Differences among particular samples are also observed. Concentrations of Na+ in samples P1 and P2 are lower than expected from the diagram, meaning that some Na+ was replaced with Ca2+ or Mg2+ released from the flysch into the solution. From the diagram (Fig. 7), a more intensive ion exchange reaction can be deduced than for samples P3, P4, and P5. The presence of ion exchange reactions can also be
deduced from values of the IEB1 parameter; the values for P1 (IEB1=0.6) and P2 (IEB1=0.7) are the highest among analysed samples, showing a high exchange of Na+ in seawater with Ca2+ and Mg2+ from the rock. The same can be seen from the molar ratios of Na+/Cl-, Ca2+/Cl-, and Cl-/HCO3-and the sample position in the Piper diagram, where samples from the upper part of borehole Lu-1 above 190 m are positioned nearly along the line.
Differences in ion exchange parameters can be the outcome of the low per-meability of rocks in the fly-sch cover and consequently longer retention times inside the rock column. In the sam-ples originating from depths greater than 500 m, ion exchange processes cannot be detected. A possible reason for this is that in limestone where ion exchange reactions are possible only at very long residence times and elevated temperatures. In the present aquifer this is not the case.
THERMODYNAMICAL CALCULATIONS
The results of thermodynamical calculations are given in Tab. 8. It can be clearly seen that samples from the upper part differ from samples from below 500 m. The sharp boundary between the two groups can be clearly seen from TDS calculations. Both groups can also be separated based on the SI and SI values. The
r	anhydrite	gypsum
water in the upper parts is more under saturated in gypsum and anhydrite than water from the lower parts. The differences between SI and SI are less profound.
calcite	dolomite	r
Over-saturation occurred in all samples with respect to calcite and in all samples except P1 and P2 with respect to dolomite. Values of SI, show that groundwater
strontianite
in the deeper parts is in equilibrium with the strontium mineral strontianite.
Eh calculation based on the Fe2+/Fe3+ ratio proved that strong redox conditions are present in the deeper part of the aquifer. Water is highly over-saturated with
Cation pattern		
Sequence	Distribution	Samples
1	Na+ > Ca2+ > K+	P3, P4, P5, P6, P7, P8, P9, P10, Al-0, Al-1, PA-1, PAL-1
2	Ca2+ > Na+ > Mg2+ > K+	P1, P2, Al-2, KA-1/21, KA-1/22, Rokava, Dragonja-2
3	Ca2+ > Mg2+ > Na+ > K+	KA-1/1, Fazan, Koblar, Dragonja-1
4	Na+ > Mg2+ > K+ > Ca2+	sea
Anion pattern		
1	Cl- > so42- > hco3-	P6, P7, P8, P9, P10, PAL-1, sea
2	Cl- > hco3- > SO42-	P1, P2, P3, P4, P5, Al-0, Al-1, Al-2
3	HCO3- > SO42- > Cl-	PA-1, Dragonja-1, Dragonja-2
4	hco3- > Cl- > SO42-	KA-1/21, KA-1/22
	Facies	Samples
1	Ca2+ - Na+ - Cl- - SO42- - HCO3-	P1, P2
2	Na+ - Ca2+ - Cl- - SO42- - HCO3-	P3, P4, P5, Al-1
3	Na+ - Ca2+ - Cl- SO42-	P6, P7, P8, P9, P10, PAL-1, AL-0, sea
4	Na+ - Ca2+ - HCO3- - Cl- - SO42-	PA-1
5	Ca2+ - Na+ - HCO3-	KA-1/21, KA-1/22
6	Ca2+ - Mg2+ - HCO3-	Koblar
7	Ca2+ - Mg2+ - HCO3- - Cl- - SO42-	Dragonja-1, Dragonja-2, Fazan
8	Ca2+ - Na+ - HCO3- - Cl- - SO42-	Al-2, KA-1/1
respect to pyrite. Sample P9 shows slightly higher Eh (-0.24 V) and lower saturation with respect to pyrite and siderite than sample P10 (Eh = -0.28 V), which was tak-en at the end of the pumping test.
For the silica minerais, over saturation (quartz) or under saturation (amorphous silica) can be observed. It is important to note that saturation indexes for SI ,
r	quartz
and SI... , are similar along the depth of the whole bore-
silicagel	o	r
hole. This can be interpreted as a consequence of an SiO2 source in the vicinity of the borehole.
REDOX CONDITIONS IN THE AQUIFER
Water flowing out from the borehole mouth during the drilling operation has a very strong smell of rotten eggs indicative of H2S gas; sometimes even traces of ammonia can be smelled. Extensive quantities of black amorphous FeS precipitated in the drilling fluid basin. All these phe-nomena indicated highly reducing conditions in the aq-uifer, and this was also confirmed by thermodynamical
Tab. 7: Important molar ratios for samples from borehole Lu-1 and Portorož environs.
Sample	Na+/Cl-	Ca2+/Cl-	Cl/HCO3-	SO/VCl-	Ca2+/Mg2+	IEB1
P1	0.36	0.29	7.03	0.014	5.36	0.63
P2	0.32	0.32	8.59	0.009	5.07	0.67
P3	0.75	0.15	6.68	0.005	3.58	0.25
P4	0.68	0.16	7.32	0.005	3.66	0.31
P5	0.58	0.21	7.70	0.008	4.05	0.41
P6	0.84	0.06	38.16	0.044	1.00	0.14
P7	0.79	0.06	50.33	0.046	1.01	0.19
P8	0.84	0.06	48.21	0.045	0.97	0.14
P9	0.79	0.06	38.92	0.044	1.01	0.19
P10	0.82	0.05	44.57	0.043	0.97	0.16
Al-0	0.81	0.05	49.69	0.000	1.31	0.19
Al-1	1.06	0.26	1.05	0.075	1.01	-0.09
Al-2	0.69	0.53	0.67	0.146	1.24	0.26
KA-1/1	2.62	3.03	0.06	0.903	0.61	-1.87
KA-1/21	4.46	12.02	0.03	0.133	6.80	-3.64
KA-1/22	3.92	13.45	0.03	0.073	5.95	-3.04
PA-1	5.83	0.34	0.18	0.925	1.20	-4.98
PAL-1	0.85	0.07	35.53	0.061	1.24	0.13
Fazan	0.43	2.45	0.20	-	3.83	0.54
Koblar	0.65	5.42	0.09	-	7.22	0.28
Rokava	1.10	4.61	0.11	0.729	4.71	-0.22
Dragonja-1	1.21	7.78	0.06	1.661	5.39	-0.27
Dragonja-2	1.54	5.73	0.08	1.151	3.97	-0.70
Punta	0.84	0.02	232.02	0.050	0.20	0.14
calculations (Tab. 8). Due to the presence of organic matter, represented by convention in calculations as formaldehyde, CH2O, the following redox reactions can explain the processes in the aquifer: Fe3+ reduction:
CH2O + 8H+ + 4Fe(OH)3 - 4Fe2+ + 11H2O + CO2 Sulphate reduction:
2CH2O + SO42- + H+ - 2CO2 + HS- + 2H2O
or
SO42- + 2CH2O - 2CO2 + S2- + 2H2O Precipitation of FeS on the surface:
Fe2+ + HS- - FeS + H+ Nitrate reduction:
NO3- + 2CH2O + 2H+ - NH4+ + 2CO2 + H2O As cast iron was used for casing, a possible reaction is also:
8Fe + SO42- + 8H+ - 8Fe2+ + S2- + 4H2O We have not proved that these reactions were mediated by reducing bacteria. However, according to reports from a similar reducing environment (Snyder et al. 2004)
and the relatively slow kinet-ics of inorganic reduction processes, the influence of bacteria is very likely (Kon-hauser 2007).
ALTITUDE OF THE RECHARGE AREA
Two groups of waters and big différences in the vertical distribution of 618O values can also be observed in Lu-1 (Fig. 6). They can be used for the determination of the mean altitude of the recharge area. Values of 618O are posi-tively related to air temperature during the precipitation event. According to the well known altitude effect, air temperatures decrease with altitude, and consequently ö18O also decreases (Gonfian-tini et al. 2001). If the relation between S18O values and altitude is known, the recharge altitude of groundwater can be calculated (Gat 1996).
The information on S18O values from Piranski zaliv is important for the interpretation of S18O values in
Tab. 8: Equilibria calculations: samples from Lu-1 (performed by WATEQ4F).
	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10
depth	43.0	65.0	91.0	113.0	190.0	500.0	611.5	704.0	800.0	800.0
pCO2	-1.66	-1.64	-1.89	-1.92	-1.83	-2.39	-2.10	-2.40	-2.09	-2.76
TDS	2315	2583	2609	2634	2623	12355	12105	11949	12220	12255
Carb.Alk.	4.93	4.63	5.64	5.34	5.08	5.08	3.85	3.90	4.98	4.35
Eh	-	-	-	-	-	-	-	-	-0.24	-0.28
Anhidrite	-1.89	-1.97	-2.51	-2.40	-2.15	-0.98	-0.94	-0.97	-0.94	-0.99
Calcite	0.24	0.28	0.38	0.38	0.38	0.81	0.31	0.59	0.52	1.02
Silicagel	-0.74	-0.78	-0.83	-0.83	-0.83	-0.78	-	-	-0.81	-0.83
Dolomite	-0.27	-0.16	0.19	0.20	0.17	1.75	0.75	1.33	1.17	2.18
Gypsum	-1.54	-1.62	-2.18	-2.07	-1.83	-0.75	-0.70	-0.74	-0.73	-0.76
Ilite	-0.48	-0.66	-0.66	-0.60	-0.58	-0.53	-	-	-	-
Kaolinite	1.33	1.24	0.96	0.95	0.97	-0.41	-	-	-	-
Quartz	0.59	0.55	0.49	0.49	0.49	0.53	-	-	0.48	0.48
Pyrite	-	-	-	-	-	-	-	-	12.80	14.06
Siderite	-	-	-	-	-	-	-	-	-0.87	1.36
Strontianite	-	-	-	-	-	-	-	-	-0.06	0.06
groundwater from Lu-1. In the sample from Punta Piran, the S18O value of +0.42%o is comparable to average S18O values of +0.60% for seawater in Piranski zaliv reported by Herlec (1998). It is also reported that the 618O value for seawater in the Gulf of Trieste is +0.8% (Dolenec et al. 1996). In Sečovlje saltpans, S18O values in seawater fluctuate in the range of -0.8% to +5.9%, which is the consequence of evaporation at higher air temperatures and the influence of precipitation (Pezdič et al. 1998). The world average S18O value for seawater is 0.0% but for closed seas, ö18O values are usually slightly positive. For the central part of the Adriatic Sea near Lovrečica, Pezdič (1999) reports a S18O value of + 1.49% and for the southern part of Adriatic Sea near Bari, +1.07%. Both values are considerably higher than values measured in the vicinity of Piran. For our end member mixing model calculations of the 618O value reported by Herlec (1998) were taken. However, based on previously mentioned data it is questionable which S18O value is representative for the Piranski zaliv. From the available isotope data and from the geographical and geological position of Piranski zaliv it can be inferred that influences from freshwa-ter on the seawater are considerable. The river Dragonja discharges a substantial amount of surface water to the bay and it is possible that there is an influence of groundwater seeping through the bottom of the sea.
In the end-member mixing model calculations, sea-water-groundwater mixing was proposed as reasonable and end-member ratios are the result of pure physical
mixing without absorption of the tracer Cl-. In the interpretation of the S18O evolution, the presence of 6% sea-water in the upper part and 31% in the lower part of the borehole Lu-1 was consid-ered. In the interval between 65 m and 113 m, two S18O values were used: -6.31% which represents the shal-lower sample and -7.18% which represents the middle of samples in the interval 91 m to 113 m. For the interval of 500 m to 801 m the S18O value of -7.45% was used. The results of mixing model calculation for S18O values of groundwater from the hinterland from the shal-lower part to the deeper part of the borehole are -6.75%, -7.67%, and -8.59%. These are S18O values that can be observed in the groundwater of the hinterland (Urbanc et al. 1992). The calculated value is also similar to ö18O values that were observed in borehole D-1 near Orešje.
The isotopic composition of rainfall during the drilling of Lu-1 is not known. During the course of the years 2001 and 2003, investigation of stable isotope composition in the precipitation in the area was per-formed (Vreča et al. 2007), showing that for Portorož airport, positioned 2 m above sea level, the monthly average weighted S18O value is -5.3%, ranging between -0.9% and -10.9%. For the evaluation of the altitude effect in the hinterland, data from the same period from the precipitation station Kozina positioned at 497 m can be used. In this station the monthly average weighted S18O value is -7.6% with the range between 12.3% and -2.1%. The calculated mean altitude effect between Porotorož and Kozina precipitation station for the period between 2002 and 2003 is 0.46%/100m, which is very high. Values reported in the literature range be-tween 0.16%/100 m and 0.4%/100 m and are only ex-ceptionally higher (e.g. Sigenthaler & Oeschger 1980; Doctor 2008). For the same area and for the earlier and more sporadic observations between 2001 and 2003, altitude effects are reported between 0.27%/100 m and 0.33%/100 m (Vreča et al. 2005). High altitude effects are also reported for the Alpine area of Northern Italy (Longinelli & Selmo 2003) and for the Dinaric mountain range from Slovenia to Croatia (Vreča et al. 2006),
with values around 0.30%o/100 m. The altitude effect is not dependent only on the relief but also depends on the prevailing weather situation and especially on the amount of precipitation. However, it can be seen that due to high relief alteration over short distances and due to the mixing of various air masses (Vreča et al. 2007), altitude effects in the hinterland region of Portorož are higher than those reported elsewhere.
To determine the highest possible altitudes of the recharge area, the highest altitude effect of 0.46%o/100 m was used. Based on this gradient and on the altitude S18O value equation reported by Vreča et al. (2007) the mean altitudes of the recharge areas are 314 m, 512 m, and 710 m respectively. The question of whether the es-timated altitudes are physically possible can be raised. According to the high relief of the hinterland and based on the geological structure (Placer 1981, 2005, 2007), the calculated recharge areas are physically possible. In the near vicinity to the north, flysch hills with altitudes of up to 400 m are present. To the northeast in the area of Podgorje kras the altitudes range between 400 m and 600 m and more to the north on the line between Kozina and Slavnik, the altitude reaches over 1000 m.
The problem of the hydrogeochemical interpretation of waters from borehole Lu-1 is also reflected by the fact that the age of groundwater is not known. Relatively negative values of S18O can be the result of older ground-water ages when isotopic composition of precipitated water was more negative. Based on available data this cannot be proved. Transmissivity values of the carbonate aquifer are high and groundwater is well homogenised, especially in deeper parts below 500 m. With much
older ages of the groundwater with more negative ö18O values, a more pronounced ion exchange reaction could be observed in the deeper part of the aquifer; however we think that this is not the case. Interpretation is more complicated for shallower groundwater where exchange reactions are more explicit. The sample from the depth of 43 m is probably the result of recharge from the near vicinity of the borehole and the more negative S18O value is the result of longer residence time due to low rock per-meability. For samples between 91 m and 113 m, higher recharge areas are more probable because in this interval more permeable layers may appear and there may be better homogenisation of groundwater.
REGIONAL HYDROGEOLOGICAL INTERPRETATION
Based on the interpretation of chemical and isotope analyses of groundwater originating from borehole Lu-1 and groundwater from the Portorož environs and with the help of the mixing model, we propose the following regional hydrogeological interpretation. In the area of south-western Slovenia, deep and extensive groundwater flow in the direction from the northeast to the southeast is present. This flow is recharged in the wider area of Ma-tarsko podolje, Slavnik ridge, and Kraški rob and flows to the eastern part of Trieste Bay (Fig. 1). In a similar manner, submarine springs of Izola could be interpreted as well as Rižana spring, which is recharged in Matarsko polje (Krivic et al. 1989), representing high groundwater overflow from the deep aquifer structures.
CONCLUSIONS
Artesian borehole Lu-1 is 801 m in depth and sited very close to the sea in Lucija near Portorož (Piranski zaliv), and was drilled in 1994. It was constructed through the 257 m deep sequence of flysch rocks after which it enters limestone of Palaeogene and Cretaceous age. The aqui-fer where the borehole is constructed is classified as high yield and highly permeable confined carbonate aquifer with fissured porosity.
Groundwater samples were collected during drilling and final pumping tests after borehole completion. The results of these analyses were compared with groundwa-ter samples from the wider Portorož environs. It was es-tablished that groundwater in the well Lu-1 is a result of seawater-groundwater mixing and due to the presence
of high organic matter content in the aquifer is subject to highly reducing conditions. Groundwater from the carbonate aquifer is over saturated with respect to calcite, dolomite, iron sulphides, and iron carbonate, with the presence of H2S and NH4+. In the carbonate aquifer, 33% seawater is estimated by the mixing model, and in the flysch part, 6% seawater is represented.
In the shallower part of the borehole Ca2+-Na+-Cl--SO42--HCO3- hydrogeochemical facies is present followed by Na+-Ca2+-Cl--SO42--HCO3- facies and in the deepest carbonate part of the borehole Na+-Ca2+-Cl--SO42- facies is present. Based on the comparison with other boreholes a similar stratification is interpreted to exist in the whole coastal aquifer around Portorož. The
transition from groundwater to seawater is gradual and stratified. Hydrogeochemical stratification is related to the geological structure with différences in physical and chemical properties between flysch and limestone rocks.
It is shown by S18O values that fresh groundwater is recharged from the hinterland. We conclude that in the area of south-western Slovenia deep and extensive groundwater flow in the direction from the northeast to the southeast exists and discharges as submarine seepage into the eastern part of the Gulf of Trieste. Above this regional groundwater flow, many local perched aquifers exists with a different hydrogeochemistry that is mainly controlled by a simple Ca2+-Mg2+-CO2-H2O open system.
The chemical analyses presented in this paper are not part of a systematic sampling plan. They are the re-sult of applied hydrogeological research in various bore-holes in the wider Portorož environs during the years 1993 and 1994. Some determinations of dissolved spe-cies are not directly comparable; they differ according to the analytical methods, sampling procedures, and detection limits. An additional problem for the analytical procedures is the high contrast in dissolved species concen-
trations, especially high concentrations of Cl-, which can greatly influence other species determinations. In such cases, the results of chemical analyses must be treated as semi-quantitative. However, in spite of all the drawbacks connected with the inconsistent sampling plan and vari-ous levels of analytical accuracy, the hydrogeochemical results together with stable isotope measurements of ö18O show a consistent picture of the hydrogeological conditions in the carbonate coastal aquifer of the Slovene maritime province.
In the carbonate aquifer of Portorož and its environs many research questions are remain open. In the future more systematic sampling based on the consistent plan should be performed. This will present firm basis for the investigation of hydrogeochemical processes in the transition zone of seawater groundwater mixing inside of limestone aquifer rich with organic mater. More detailed investigations are needed to determine the age of groundwater at the different depths and parts of the aquifer, as well as recharge zone and its altitude. For this purpose several different investigation methods must be used, including stable and radioactive methods in the combination with organic and inorganic geochemistry.
ACKNOWLEDGEMENTS
The paper is dedicated to the late Miha Marinko, leader of the borehole Lu-1 project. The help of Andrej Lapajne, Polona Vreča and Simon Mozetič during the preparation of the article is greatly appreciated. Part of the research in 1994 was financed by Metropol hoteli d.d. The article was
prepared through the research programme "Groundwater and Geochemistry" of the Geological Survey of Slovenia and financially supported by the Slovene Research Agency, ARRS.
REFERENCES
Appelo, C.A.J. & D. Postma, 1993: Geochemistry, Groundwater and Pollution.- A.A. Balkema, pp. 536, Rotterdam.
Back, W., 1961: Techniques for mapping of hydrochemi-cal facies.- U.S. Geological Survey Professional Paper, 424-D, 380-382.
Bear, J., 1979: Hydraulics of Groundwater.- McGraw-Hill, pp. 569, New York.
Bonacci, O., 1995: Brackish karst spring Pantan (Croa-tia).- Acta Carsologica, 24, 99-107.
Breznik, M., 1973: Nastanek zaslanjenih kraških izvirov in njihova sanacija.- Geologija, 16, 83-186.
Breznik, M., 1998: Storage Reservoirs and Deep Wells in Karst Regions.- A.A. Balkema, pp. 251, Rotterdam.
Breznik, M. & F. Steinman, 2008: Hydromechanism and desalination of coastal karst aquifers: theory and cases.- Acta Carsologica, 37, 197-212.
Calaforra, J.M. (ed.), 2004: The Main Coastal Karstic Aquifers of Southern Europe.- European Comission, Directorate General for Research, pp. 123, Brussels.
D'Amore, F., Scandiffio, G. & C. Panichi, 1983: Some observation on the chemical classification of ground waters.- Geothermics, 12, 141-148.
Davis, S.N., Whittemore, D.O. & J. Fabryka-Martin, 1998: Uses of cloride/bromide ratios in studies of potable water.- Ground Water, 36, 338-350.
Doctor, D.H., 2008: Hydrologic connections and dynamics of water movement in the classical Karst (Kras) aquifer: evidence from frequent chemical and stable isotope sampling.- Acta Carsologica, 37, 101-123.
Dolenec, T., Pezdič, J. & U. Herlec, 1996: Stable isotope study of the Adriatic Sea.- Acta Geologica Hungari-ca, 39, Suppl. 35-38.
Domenico, P.A. & F.W. Schwartz, 1990: Physical and Chemical Hydrogeology.- John Wiley & Sons, Inc., pp. 824, New York.
Faganeli, J., Ogrinc, N., Walter, L.N. & J. žumer, 2005: Geochemical characterization of the submarine spring of Izola (Gulf of Trieste, N Adriatic Sea).-RMZ - Materials and Geoenvironment, 52, 35-39.
Fistanic, I., 2006: Sustainable management of brackish karst spring Pantan (Croatia).- Acta Carsologica, 35, 65-72.
Fleury, P., Bakalowicz, M. & G. de Marsily, 2007: Submarine springs and coastal karst aquifers: A review.-Journal of Hydrology, 339, 79-92.
Freeman, J.T., 2007: The use of bromide and chloride mass ratios to differentiate salt-dissolution and formation brines in shallow groundwaters of the Western Canadian Sedimentary Basin.- Hydrogeology Journal, 15, 1377-1385.
Gat, J.R., 1996: Oxygen and hydrogen isotopes in the hydrologic cycle.- Annual Review of Earth and Plan-etary Sciences, 24, 225-262.
Gonfiantini, R., Roche, M.A., Olivry, J.C., Fontes, J.C. & G.M. Zuppi, 2001: The altitude effect on the isoto-pic composition of tropical rains.- Chemical Geol-ogy, 181, 147-167.
Hem, J.D., 1985: Study and interpretation of the chemical characteristics of natural waters.- U.S. Geologi-cal Survey Water Supply Paper, 2254, pp. 256.
Herlec, U., 1998: What can we learn from the stable oxygen isotope composition of recent molluscs from the Adriatic sea: the problem of mean annual marine water temperature.- RMZ - Materials and Geoenvironment, 45, 51-55.
Kim, Y., Lee, K.S., Koh, D.C., Lee, D.H., Lee, S.G., Park, W.B., Koh, G.W. & N.C. Woo, 2003: Hydrogeo-chemical and isotopic evidence of groundwater sa-linization in a coastal aquifer: a case study in Jeju volcanic island, Korea.- Journal of Hydrology, 270, 282-294.
Konhauser, K., 2007: Introduction to Geomicrobiology.-Blackwell Publishing, pp. 425, Malden.
Krivic, P., Bricelj, M. & M. Zupan, 1989: Podzemne vodne zveze na področju Čičarije in osrednjega dela Istre (Slovenija, Hrvatska, NW Jugoslavija).-Acta Carsologica, 18, 265-295.
Kuščer, I., 1950: Kraški izviri ob morski obali.- Razprave - Razred za matematične, fizikalne in tehnične vede, 1, 99-147.
Kuščer, I. & D. Kuščer, 1964: Observations on brackish karst sources and sea swallow - holes on the Yugo-slav coast.- In: Haralambos, D., (ed.) Association Internationale des Hydrogeologues Memoires Tome v: Reunion d'Athenes 1962. Greek Institute for Geology and Subsurface Research, pp. 344-353, Athenes.
Lapanje, A. & J. Prestor, 2003: Leau thermale dans les calcaires Paléogènes et Crétacés de la côte slovene.- In: Circulations Hydrothermales en Terrains Calcaires, pp. 31-38, Carcassonne.
Longinelli, A. & E. Selmo, 2003: Isotopic composition of precipitation in Italy: a first overall map.- Journal of Hydrology, 270, 75-88.
Marinko, M., 1994: poročilo o opravljenih delih na vrtini Lu-1 v portorožu. Inštitut za geologijo, geotehniko in geofiziko, Report.
Petrič, M., Kogovšek, J. & L. Ballarin, 2004: Coast of Trieste and Kras (Italy - Slovenia).- In: Calaforra, J.M. (ed.): The Main Coastal Karstic Aquifers of Southern Europe. European Comission, Directorate General for Research, pp. 46, Brussels.
Pezdič, J., 1999: Izotopi in geokemijski procesi.- Naravo-slovnotehniška fakulteta, Oddelek za geologijo, pp. 269, Ljubljana.
Pezdič, J., Dolenec, T., Faganeli, J., Čermelj, B., Vukovič, A. & N. Ogrinc, 1998: Water evaporite effects on stable isotopes in the salt pans of Sečovlje (Slovenia).- RMZ - Materials and Geoenvironment, 45, 145-148.
Piper, A.M., 1944: A graphic procedure in the geochemi-cal interpretation of water analysis.- Transaction of American Geophysical Union, 25, 914-923.
Placer, L., 1981: Geološka zgradba jugozahodne Slovenije.- Geologija, 24, 27-60.
Placer, L., 2005: Structural curiosity of the northern Istria.- Geologija, 48, 245-251.
Placer, L., 2007: Kraški rob (landscape term), Geologic section along the motorway Kozina - Koper (Capo-distria).- Geologija, 50, 20-44.
Pleničar, M., Polšak, A. & D. Šikic, 1973: Osnovna geološka karta SFRJ 1:100.000. Tolmač za list Trst.-Zvezni geološki zavod, pp. 68, Beograd.
Plummer, L.N., Jones, B.F. & A.H. Truesdell, 1976: WA-TEQF - A Fortran version of WATEQ, A computer program for calculating chemical equilibria of natural waters.- U.S. Geological Survey Water Resources Investigations Report, 76-13, pp. 70.
Schoeller, H., 1962: Les eaux souterraines.- Masson & Cie, pp. 642, Paris.
Siegenthaler, U. & H. Oeschger, 1980: Corrélation of 18O in precipitation with temperature and altitude.- Nature, 285, 314-317.
Snyder M., Taillefert, M. & C. Ruppel, 2004: Redox zo-nation at the saline-influenced boundaries of a permeable surfical aquifer: effects of physical forcing on the biogeochemical cycling of iron and manganese.- Journal of Hydrology, 296, 164-178.
Suric, M., Jalžic, B. & D. Petricioli, 2007: Submerged speleothems - expected the unexpected. Examples from the eastern Adriatic coast (Croatia).- Acta Carsologica, 36, 389-396.
Todd, D.K. & L.W. Mays, 2005: Groundwater Hydrology.-John Wiley & Sons, Inc., pp. 636, New York.
Tulipano, L. Fidelibus, M.D. & A. Panagopoulos, (eds.), 2005: Groundwater Management of Coastal Karstic Aquifers.- COST Office, pp. 363, Brussels.
Urbanc, J., Pezdič, J., Juren, A. & J. Prestor, 1992: Isoto-pic study of relation between the Rižana spring and waters from its vicinity.- In: Hötzl, H. & A. Werner, (eds.): Tracer Hydrology: proceedings of the 6th International Symposium on Water Tracing, Karlsruhe (Germany) A.A. Balkema, 295-298, Rotterdam.
Urbanc, J., Marinko, M., Kogovšek, J. & M. Petrič, 2004: Izola.- In: Calaforra, J.M. (ed.): The Main Coastal Karstic Aquifers of Southern Europe. European Co-mission, Directorate General for Research, 47, Brussels.
Vreča, P., Kanduč, T., Žigon, S. & Z. Trkov, 2005: Iso-topic composition of precipitation in Slovenia. In: Isotopic Composition in precipitation in the Mediter-ranean Basin in Relation to Air Circulation patterns and Climate, TECDOC Series No. l. 1453, IAEA, pp. 157-172, Vienna.
Vreča, P., Krajcar Bronic, I., Horvatinčic, N. & J. Barešic, 2006: Isotopic characteristics of precipitation in Slovenia and Croatia: Comparison of continental and maritime stations.- Journal of Hydrology, 330, 457-469.
Vreča, P., Brenčič, M. & A. Leis, 2007: Comparison of monthly and daily isotopic composition of precipitation in coastal area of Slovenia.- Isotopes in Envi-ronmental and Health Studies, 43, 307-321.
Vrhovec, T., Mihevc, A., Lauritzen, S.E. & J. Lundberg, 2001: O starosti potopljenih stalaktitov v potopljeni jami pri otočku Galiola, Dalmacija, Hrvatska.- Naše jame, 43, 31-36.
Wicks, C.M. & J.S. Herman, 1996: Regional hydrogeo-chemistry of a modern coastal mixing zone.- Water Resources Research, 32, 401-407.
Žumer, J., 2004: Odkritje podmorskih termalnih izvirov pred Izolo.- Geografski obzornik, 51, 11-17.
Žumer, J., 2008: Podmorski kraški izviri pred Izolo.-Naše jame, 47, 39-44.