Study and Modelling of Saltwater Intrusion into Aquifers. Proceedings 12th Saltwater Intrusion Meeting, Barcelona, Nov. 1992. CIHS. © CIMNE. Barcelona, 1993: 245-266.
SEA WATER INTRUSION HYDROGEOCHEMISTRY OF THE GARRAF CARBONATE AQUIFER (BARCELONA-TARRAGONA)
PASCUAL, J. Manuel. Dr. Ap. Sc. (1) (3) CUSTODIO, Emilio. Dr. In. Eng (2) (3) GALOFRE, Andreu. Geologist (1) (3)
(1) Water Authority of Catalonia, Barcelona and Lleida ( 2) Dept. of Ground Engineering·, Polytechnical University of Catalonia, Barcelona (3) Foundation International Center for Groundwater Hydrology, Barcelona
ABSTRACT
The Garraf massif is a thick sequence of limestone and dolomite formations of Cretaceous age with minor Miocene calcarenites in direct contact with the sea. Intensive exploitation of groundwater for urban and residential areas causes a continuosly increasing salinization by sea-water intrusion.
In the Calafell-Bellvei area, two PVC-cased boreholes were selected for study, one with only fresh water and the other penetr'ating the upper part of the mixing zone. Systematic sampling was done at three depths in the brackish borehole, and at only one depth in the fresh water one.
According to the chemical data, the possible equilibrium chemical reactions are considered. Mass balance studies have been carried out in order to check and complete first assumptions obtained after comparing actual and theoretical (closed system) ion concentration values of water mixtures. This allows the formulation of a conceptual model of the hydrodynamic behaviour, of the Garraf aquifer consistent with hydrogeological and hydrochemical data.
1. INTRODUCTION
During the last 20 years studies of diverse aquifers along the Catalan coastal areas have been carried out and a better undertanding of sea water intrusion mechanisms under different circumstances has been gained.
The Garraf carbonate massif has been studied in detail and has been selected as a case study. The present paper tries to explain the general hydrogeological characteristics affecting sea water intrusion and considers, in some detail, the freshwater-saltwater mixing zone behaviour at an experimental site near calafell (Tarragona). 246 CHEMICAL ASPECTS
2. LOCATION AND GEOLOGY
The Massif begins 15 km to the SW of Barcelona and extends southwestward: (fig. 1 ). An important tourist infrastructure for summer and weeken visitors has developed along the coast line. Groundwater was the sole source of water supply to the ·area until recently. Surface water collects in a reservoir on the river Foix (fig. ·1) as does much of the domest~~ sewage from Vilafranca del Penedes. Significant infiltration occurs througl the reservoir floor. The water from the reservoir is used for loca irrigation since it is not fit for human consumption.
A boom in urban and tourist development over the past 20 years has b~ought about the present situation, characterised by high density of populat~on.~n summer (over 500,000 inhabitants) and peak water demands in an area w~th limited resources. Serious deficiencies in water supply and water qual~ ty exist. The majority of wells exploit brackish and saline water· In sofme formerly freshwater wells, up to 16 g/1 of Cl- ion were reached a ew years later.
The Garraf Massif is part of the southern end of the Catalan Littoral Range ( Catalanides) and separates the sea from the Penedes sedimentary basin: It is a Jurassic (dolsparite and dolsmicrosparite) and Cretaceous (micr~~~c limestone, marly limestone and marly layers) isocline carbonate format~on that overlies a Triassic series which outcrops at the northern edge of the Llobregat valley (figure 1).
3. GENERAL HYDROGEOLOGICAL CHARACTERISTICS
The Garraf carbonate formations are affected by intense brittle tectonics with NW-SE structural directions and other orthogonal, in the NE-SW direction. A layer of Miocene calcarenites covers the Mesozoic materials and fills small secondary depressions inland and along the coast. A karstic morphology has developed along the fracture lines with numerous caves' sinkholes and surface karstic carvings. Karstic solution paths develo~' particularly at higher levels. They only attain the watertable at spec~f~c sites, where underground channels exist below sea level, forming some large submarine springs (La Falconera, Aiguad~l9).
After a heavy rain season, it's common to observe high increases of submarine spring flow, creating a brown stain spreading over the sea surface. An important part of the Garraf massif mean anual recharge is lost by groundwater flow, through a series of submarine springs. J.M. Pascual et al. 247
GARRAF
N. ~ km
MIOCENii a 10 fM z-1 Marla §>T.::I Triassic 0 Poloozolc !IEill Calcarenites ~ uurosoic- Crotocooua 1----· ---·
J-C
I '+rl-\ B-B'
~z _Mar -,___. -·,~~
Figure 1. Geographical and geological situation 248 CHEMICAL ASPECTS
From a regional hydrogeological point of view, several areas can be differenciated in the massif (fig. 2). There is a northern area of low permeability and high watertable slope (Cervell6-Vallirana) where the development of deep karstic features is hindered by the presence of the Triassic bedrock at high elevation. There is a central coastal area in which groundwater flow goes directly to the sea, towards the SE, and presents potentiometric depressions along the coast that may be linked to the existence of transversal fracture lines. There is a central area in which ground1vater flow is directed towards the SW. This is due to the NE-SW oriented structural lines and the barrier effect of the marl and clay formations that fill the Vilanova-Sant Pere de Ribes depression (1). There is also an eastern area (around the river Foix) where groundwater flow goes toward the sea, following the transversal fracture system.
Along the central and southern coastline, extensive areas exist where piezometric levels are just above or· below sea level due to the effect of pumping, even when the wells are several kilometres inland. A large number of wells have water heads higher than the regional piezometric level due to the presence of marls and marly limestone layers of Aptian age found in the upper Cretaceous carbonate series. These layers produce numerous perched aquifers, partially recharg~d from the Foix reservoir.
The study of data on specific lvell discharge and penetration of wells below sea level shows a probable progressive transition from the upper part of the rock, fractured and karstified to some extent, to the lower part, little fractured and of low permeability. This transition zone is below sea level near the coast, and close to zero altitude or above it inland, where the wells yield little water.
Table shows that wells drilled far from the coast generally have low yield _(f~~· 3). Only 20% of them have a specific discharge above 300-350 m3.day m , independently from penetration below sea level. These values are sensibly more frequent near the coast. If, during the well drilling enough yield is not found when it penetrates a few metres below sea level, it is little probable that it will improve with a greater penetration. 110-----
• /;Canyelles \\:
...... \':-jf~.~ o---- -·-. ?p·-- ,, ___ .. /_,~ l ·.. • C3Ja.fell 'JY (~,0-- \ ~--o, ;.··*·>'' ~·7Vilanova MErnTERRANEAN SEA
Figure 2. Piezometric levels in the Garraf aquifer (1984).
N .1:> 'D 250 CHEMICAL ASPECTS , § Q
Vilanova Sitges MEDITERRANEAN SEA Q km
Rgure 3. Map of specific yields. (a) Area with shortly penetrating wells below sea level. When penetration is higher than 20 m specific )'ield is generally low.
l ··GARRAF/ 7 / l
- • ~ ~ ·i:;; ·.~····· ••••••••• •••• ·,.··.· •• ----·~_·.·.:-.·"-- I iII- II . .. I:~~ ...... I
E Calcarenite Miocen~ deposits of the Penedlls depression
A iJ <1-- PERCHED P1EZOMETRIC LEVEL (eg. Mojo nquifer). 0 F lmpermenble formntions (Albion mnrl) 0
A • • • • Well d!Qined. kDtstified ~rnp. Goqd speleofoqjcn.l del(elopmenl Numeous • •. '2 ; • perChed nquifers of smlll dimensions. Vert1cdl flows domrnnte. e f;;:::k ~ -i :<-P1EZOMETRIC LEVEL IN WET PERIODS. :· .;:i. ·n\'f? ,Zone only snturnted in wet periods. Horizonllllflow domtnnte. ..,._. P1EZOME1RIC LEVEL IN DRY PERIODS c Uttor~ zone of high permenbilily. Secondruy development Sen wnter § . imnunon zona. ~~r-l'.C:.C'r"-0./~~ ._ KARSTIFICAllON BASE LEVEL Non kMBtilied or obliternted zono. Closed frncturntion. Sumo preferential flow pnttems mny be found. although they ore not lrequenl
Rgure 4. Hydrogeological conceptual model. Schema1ic cross-section J.M. Pascual et al. 251
Table 1 . - Specific yield of wells according to distance to the coast and penetration
Well location Penetration below Probability (%) to exceed Distance from sea level (m) a discharge given3 sp:;lj'i~fc the coast (m.day m )
20% 50% 80%
20 300 13 0.6 '>3 km 20 350 25 1.5
20 250 30 4.0 <3 km 20 3700 350 40.0
The permeable baselevel in the Garraf massif is neither a sharp surface nor appears at a fixed altitude, but it is a zone in which the probability to found fractures and joints open to groundwater flow progressively decreases downwards. The same is true for the probability of finding productive sections during well drilling.
Figure 4 shows schematically the probability model after the study of well yield, potentiometric fluctuations and water balances.
Zone A, between ground level and the regional water table, contains inner karstic features developed in vertical direction following existing features in the rock. There are other less frequent horizontal karstic features that follow small marly and/or clayish seams. They may sustain small aquifers, perched above the regional water table. Wells open in these perched aquifers show fast piezometric changes. Generally zone A is unsaturated or has a thin aquifer in the lower part. In wet periods, with high water table, the water thickness in the wells may rise suddenly (zone B). Horizontal as well as vertical groundwater flow is fast, with short water residence time. Zone C only exists near the coast and shows features of secondary permeability development. This development is associated to sea level changes during the recent Quaternary. It is the zone that presents the best condition for drilling high yielding wells, but it is also the zone less favourable in obtaining good quality water since it is easily affected by sea water intrusion. Generally, permeability is higher, presents smoother head gradients and water level fluctuations are small, except during and shortly after intense rainfall. Groundwater flow is slower and dominantly horizontal, with longer residence times, and greater reserves. 252 CHEMICAL ASPECTS
Below zones A + c there is zone D, where permeability is very low. The probability of drilling a good yielding well is very low. Some preferential groundwater flow paths may exist, although they are rare. They correspond mainly to the fault system, sub-normal to the coastline, that may affect the whole Mesozoic formations.
In practice, the transition zone beetween zones A + c and zone D represent the base of the Garraf aquifer. Landwards it goes up and may attain altitudes above sea level. Near the coast it is at least a few tens of metres below sea level. Groundwater in zone D flows sluggishly and has a long residence time.
Zones E and F in figure 4 represent the Moja perched aquifer system, one of the most important, that behave· independently from the regional aquifer of the Garraf Masif.
4. SEA I~ATER INTRUSION
Due to continuous groundwater abstraction most of the time there is a landward flow of groundwater from the coast, except for a short time after the autumn intense rainstorms. This produces a trend of increasing salinity and a thick mixing zone between locally recharged freshwater and sea\~ater. The mixing zone extends practically up to the water table. Figure 5 shows schematically the recharge and groundwater flow process and the isosalinity contours. Progresive urbanization of the coastal area and the development of touristic and week-end residential areas produce a high groundwater demand in an area traditionally known as water scarce. As a consequence, many exploitation wells have been constructed, which gradually have been contaminated by sea water. Many of them have continued to operate in spite of the high salinity since, until recently, there has been no other water supply alternative. Figure 6 shows the chloride content in the upper part of the water table.
Although existing information is very incomplete, it is well known that there is a continuous trend of increased salinization, with seasonal fluctuations.
Figure 7 shows the recent evolution of chloride ion in several wells from the Garraf Massif. Figure 7a represents daily variations; sharp decreases appear as the consequence of rainy periods. The magnitude of change depends on the medium permeability (1) and from the fast and discontinuous contributions from upper levels and perched aquifers. There is a clear seasonal variability as well (fig. 7b). J.M. Pascual et a!. 253
_JZ_ Wate-r table --- t.it Figure 5. Schematic representation of groundwater flow conditions and mixing with sea water in the study area. Local freshwater recharge has a more or less homogeneous chemical composition. Some recharge follows preferential paths. The bedrock is formed by non-karstified limestones and dolostones. with a transitional passage to the permeable. upper layers. MEDITERRANEAN 0 km Figure 6. Chloride in abstracted groundwater (gll a-) in the Cretaceous carbonate formations of the study area. Values represent the mean value for the uppper layer penetrated by wells. 254 CHEMICAL ASPECTS a/ 60 E 40 E I 20 II I • II L. .I I I I II .... ~, 4000 t /Iv~ s I v I I I r.l ~ 3000 ·- ;;-- .... 1- D --· I v - ~./ "'E 2000 "0" 1---- 0 h) :E u r; i A 1000 ~I\ I r v A- Collado B -Collada v C -Coladors '-..I D-Mas Ric.ord 4 6 8 10 MONTH 5 bt 60~------~•• ~----~------~----~ I ',., ,' ',, 0. I , 1 -e 4o. ....r-----::-.t-'"-· /. --~'__.+.:----~'-----1r-~ ...... ____ ~ ~ 1 2 3 4 S 6 7 II 9 l:> 11 12 1 2 3 4 5 6 1 8 9 10 11 12 19 86 198 7 YEARS Figure 7. Evolution of the a- ion concentrations. aJDaily variation in several wells ofVilanova i La Gettru (1 ). b/ Bimensual variation in the S.Ol calafell borehole. J.M. Pascual eta!. 255 Sea water intrusion largely affects the southern half of the Garraf massif. It determines the chemical characteristics of recharge water, changing it from a calcium or magnesium bicarbonate type to a sodium chloride type (figure 8) 1 except in the areas lvhere infiltration from the Foix reservoir occur. Downstream from the reservoir, more or less following the same path as the permanently dry creek, the cation content of the groundwater seems to show the influence of modifying factors. The theoretical (closed system) amount of sodium that should be present in the mixing zone (figure Sa) is decreased. Considering all the chemical data on groundwater, the behaviour of the major ions relative to the concentration of Cl- is variable, although a first approximation shows2the following tendencies: Na+ and Mg 2+, (i) marked excess of Ca + with a parallel deficit in mainly in those areas where salinization is most advanced, (ii) sligther or negligible exc~rs of ~o 4 = in groundwater with a concentration less than 2000 mgl of Cl 1 with a subsequent swing to a marked deficit in water most affected by salinization, (iii) slight excess of Hco concentration throughout the whole process 3 of salinization, (iv) temporary reduction of the ratio rNa/rCl (r = meq/1) 1 except in the groundwater from the inland strip of Miocene calcarenites, where salinity is low (2). Simple preliminary treatment of the samples as a whole shows the fossible existence of modifying factors such as cation exchange between Na in the 2 water and Ca + in the rock where salinization is more advanced. 5, HYDROGEOCHEMICAL PROCESSES IN THE FRESHWATER-SALTWATER WATER MIXING ZONE IN CALAFELL AREA Mixing of fresh water with marine salt water in coastal aquifers produces changes in calcium carbonate saturation as a consequence of the mixing itself and as the result of water-rock interactions (3). Field studies made in Florida and Yucatan ( 4) ( 5) and elsewere involve pure limestone under given geological and climatological conditions. This leads to calcite dissolution for a given interval of salinities inside the mixing zone, and enhances coastal karst development. This has been modelled recently (6). In order to check if previous results can be considered common hydrogeological behaviour in coastal aquifers, other areas have been studied. These include some in the Mediterranean Sea, such as the Greek islands, southern Turkey and Israel, in the Eastern area, Apulia (Italy) in the central one, and Catalonia and the Balearic Islands (Mallorca Island) (7) in the western area (Spain) . N Ul 0\ @ VILAFRANCA ... ® so4 DEL PENEDES (') :I: ~ ~ >C/J. ~ ~ figure 8. Hydrochemical factes in the Garraf Massif. Zonation derived from trilinear diagrams. AJ cations. 81 Anions. J.M. Pascual et a!. 257 Since there is a fair hydrogeological and hydrogeochemical knowledge of.the Garraf massif, it was selected to carry out detailed hydrogeochemical studies, to know proceses in other carbonate rock circumstances, especially under sustained exploitation. Different papers (2)(8)(9)(10)(11)(12)(13) deals with the results obtained. Existing bore-holes (figure 9) have been selected for repetitive sampling during a period close to two years. \'later samples have been obtained with a depth sampling bottle. Samples were filtered in the field, with in the field measurement of pH, electrical conductivity and temperature. The other major chemical constituents were determined in the laboratory in Barcelona as soon as possible, by transporting the samples in tightly closed bottles inside a cooled box. Two existing PVC-cased boreholes were selected in Calafell-Bellvei area for study, one with only fresh water and the other penetrating the upper part of the mixing zone (figure 10). Due to dominant landward groundwater flow, at the saline water borehole, no freshwater layer is present, and the mixing zone inside the tube extends up to the water table. The vertical salinity gradient is maintained by local recharge, but it is also controlled by the dis·tribution of heterogeneities along the bore (9). Systematic sampling in the brackish water borehole was done at three depths. However only data from the shallowest depth, 23 m deep or about 2 m below the water table, will be considered here, since there chemical changes are greatest. Chemical equilibria relative to environmental minerals have been studied with the help of the 1'/ATEQF programme (14). After determining ion composition deviations relative to theoretical (closed) mixing of fresh and seawater, ma~s balance studies have been carried out with the BALANCE programme (15) and the chemical behaviour has been studied by means of the PHREEQE programme (16). SI refers to the saturation index of a given mineralogical species and is defined as the logarithm of the ratio of the ion activity product to the solubility product. Differences of 0.3 are not significant. Actual groundwater composition is compared to the calculated results of closed system mixing of two end waters (fresh and marine, or bracksish water above and below), refered to as theoretical mixing. The mixing proportion is deduced from the chloride content, since it is assumed conservative. Salinity profiles have been determined by means of calibrated electrical conductivity and temperature probes. In some instances loss of signal and electrode polarization problems have distorted the readings, but the shape of the profile is conserved. \'later and carbon isotope analyses of dissolved inorganic carbon have been carried out, but they are not considered here. 258 CHEMICAL ASPECTS 0 MEDITERRANEAN SEA km Rgure 9. Situation of monitored boreholes A 8 20 lr':l::':rrl 1r I II I. I \ I \ I I I ,~~~ • 1ow permeab i 1 i ty moderate permeabi 1 ity JO I ~· ~ vertical flow ~ ' '-< Q) ~ • ...'- •O ~ horizontal flow ~ ~ "d c: "'~ fractured stretch - '""'(") .c... 10 c. "'e. Q) 0 g. e. 50 ·--· ·-··· 70 Rgure 10. Characteristics of borehole S.O 1. AJ lithology: karstified and highly fissured parts are indicated as indentations of the column and change of the symbol of carbonates (brick-like); below 61 m the borehole is uncased =dinfilled with debris. Bl Gamma-ray log ; high radioactivity produce deflections towards the right. Cl Radioactive tracer (1-131) dilution tests resuts (the one made in July 1986) after tracing the whole water column inside the PVC casing; figures show the elapsed time after the first log (0). inmediately after the extraction of the tracer injection hose. Dl The two extreme water salinity (electric conductivity) logs from a series of ten: the shape is conserved in spite of large seasonal salinity variations. EIExplanation. N v. \0 260 CHEMICAL ASPECTS The time evolution of ionic strengh is show in figure 11 • A seasonal variation of calcite and dolomite saturation indices exist in the mixing water and fresh water (figure 12). Assuming a closed system with respect to , the comparison of the theoretical and actual evolution of calcite and co 2 dolomite saturation indices shows no clear correlation. Oversaturation in winter and undersaturation during the summer, not only are explained by the variation of the activity coefficients due to the mixing of waters with different ionic strengh, but also to other processes such as cation exchange. It is assumed that clays from the impure carbonates provide the ion exchange capacity. The diference between theoretical and actual values of co partial pressure (pco ) and total 2 2 dissolved inorganic carbon are assumea those of a system that is really open with respect to co at this shallow depth below the water table. In 2 summer there is an lnflow of co from the vadose zone, and in 2 winter, an outflow of co zone. As a result, waters are undersaturated in 2 summer and oversaturatea in winter with respect to b'ltf ;1':(-ci te and dolomite. The vadose zone has not been moni tared, but C/ c changes in the water agree with the gas exchange process. The parameter evolution in fresh water is assumed to have driven the evolution of the same parameters ~n the mixing water. ANa+ deficit is observed that does not·equal the Ca +excess ~figure 13). The Na+ deficit is greater in winter than in summer, and the Ca + excess does not match the winter Na+ deficit. This is ~ssumed the result of Ca/Na exchange and to calcite precipitation during winter. In the mixing zone conditions are favourable for dolomitizacion of calcite to take place. The ratio Sid /SI shows variable values in the mixing 1 zone. In some instances it a~Eain~athe right values for the formation of dolomite: calcite un2ersaturation and dolomite oversaturatfon. In fact, there is always a Mg + depletion in the mixing zone water. The processes mentioned above are also supported by mass balance calculations (11) (figure 14). They show that calcite precipitation takes place in autumn, winter and spring, and that dissolution takes place in summer. Dolomite formation is more likely ·between summer and autumn. The Ca/Na ion exchange takes place all year round, although it is less important in the summer. Similar studies have been done in more saline, deeper parts of the same water column. Conversely to what has been said with respect to the higher part of the mixing column, calcite dissolution and dolomite formation processes are clearly more important than those of precipitation and dedolomitization at the 23 m depth. The results show that kinetics play an essential role since short undersaturation periods intercalated between longer oversaturation periods are enough to produce a net dissolution effect ( 11 ) . Figure 1 5 is a conceptual model of processes inside the groundwater of the Garraf massif showing \~hen and where they may occur. J.M. Pascual et al. 261 0.1 u c 0 b 23m. +-' c Ql 0,.> u .... Ql 0 u ~ 0.4~ +-' > +-' u ...: o.•o I 2 l 4 ' 6 7 I 9 10 II 12 I 2 l • ' 6 1 I 9 10 II ll 1966 1987 Figure 11. Time evolution of chemical characteristics in Garraf boreholes. aJ ionic strength in the freshwater borehole and at 23 m depth in the saline water borehole. b/ activity coefficient of Mg. Ca and COg ions at 23m depth in the saline water borehole. N Ri \ pH pH \ !Jij J( i pHM I 7,5 ------...... ---______..J I 7,5 :~<------~~~/ ' \ -\--· .. \ ... / 1,0 1,0 ' \ / " \ PHI1 \ ' ... , : \ / \ / '-·--, \ ,:1 '\ ~~ \ ' \ '\.1''/ \, / Q, '/ -;. \ / '~ / 7,0 7,6 ~\ '-..,. \\/ / j·''\"/ \ l I l'(:s!(j 'v 6,5 --\ 6,5 0,5 N o,s N 0 0 \ \\~>< /1 \ u c.. /""; u \ I"" "'-o/ - n \ I \v· ..,_of - ::r: 0 H 0 'K.-'"'-\ \ A'h \ V) w \ SIC~ N ij N E E 0 ~ o E 0 E u = u a. \ a. - 1 ,_., =0 >- ~ s'c-r~ \1 ,cAti> \rn..\ u u t-< . I 2"' ~ \ 2"' > ·101 ~ [10 (/) o I : \II\ I ;:::::;r I 'I ~: I Ill ~'1/ ~-1,o 0 -1,0 til n..., c (/) X c1 C1 ----...... __ ------~ ' / ~ ,. -1,5 -1,5 5 ··--zc····'g,_/\1' \ .-t: T Figure 12. Time evolution of calcite and dolomite saturation index (logarithmic). log Pco2,total dissolved inorganic carbon and pH in Garraf boreholes. N in saline water at 23m depth. 8/ in the freshwater borehole. M indicates measured; C calculated from analytical data and X the theoretical mixture of salt and freshwater according with sample c1· content. J.M. Pascual et a!. 263 23 m Ul Ul OJ u • >< OJ + + n:l ~ 0 0" ---OJ E c 0 +' -2 OJ ~ Q_ OJ u -G + n:l ;z: 1986 1897 Figure 13. Time evolution of Ca*"' excess and Na• depletion (actual values less theoretical values from dosed system mixtures of local freshwater with local seawater) in mixed water from Garraf saline water borehole. at 23m depth (about 2m below the water table). -I -2 0 N :::c Ol ..><: _, ---~ 0 -2 E E _, p-' precipitation DOLOMITE d- dissolution 1986 1987 Figure 14. Time evolution of mass balance results at 23m depth in mixed water in the Garraf saline borehole. The assumed processes are calcite and dolomite precipitation/dissolution. and Ca/Na ion exchange. 264 CHEMICAL ASPECTS SUMMER AND PRE-SUMMER PERIOD LATE SUMMER PERIOD SENSU STRICTU KARSTIFIED AND NON SATURATED ZONE HAL~ DIFFUSED FLOW AND SATURATED ZONE LOW PERMEABILITY AND NON KARSTIAED ZONE Roots, brushwood and forest Biological activity. CO2 production lnl~tration of Ire~ wal81s. High C02content. Calc~e and dolomite equilibrium. \\1\\\\\\\ Mixed waters; Y , Y , Y , SIC' Sid decreasing. PC0 inc!easing. lnhibilion of the Na/Ca exchange. Ca Mg C02 2 · Mixed waters; Sic and Sid <0. High PC0 . C~e dissolution. Important dedolomitization Na/Ca 2 exchange. Mixed waters; Sic and Sid <0. VeJy high PCO 2-~e dissolution. Dedolomitization Na/Ca ..... exchange. Roots, brushwood and forest No biological activzy. No CO2 production. Infiltration oi fresh wale~. Low PC02 .Sic and Sid )0. Mixed wat81s; Yea ,Y Mg' Y~,SIC' Sid increasing. Low PC02'1oss of gas. Na/Ca exchange. Mixed waters; Sic and Sid> 0. lnterrnedate PC02 dec!easing. Important dolomitization Na/Ca exchange. Mixed waters; Sic and SI~O. High PC02 decreasing. Dolomitization. Na/Ca exchange. Figure 15. Hydrogeochemical conceptual model for the Garraf's Southern littoral area.· J.M. Pascual et a!. 265 6. REFERENCES (1) ALOM, A. and SOLER, X. (1989). Sintesis de la circulaci6n y dinamica del acuifero carstico de Garraf (Barcelona). Tecnol~a del Agua. 55: 35-41. (2) CUSTODIO, E., PASCUAL, J.M.; BOSCH, X., ·and BAYO, A (1986). sea water intrusion in coastal carbonate formations in catalonia, Spain· Proceedings 9th Salt Water Intrusion Meeting. Delft: 147-164. (3) PLUMMER, N.L. (1975). Mixing of sea water with calcium carbonate groundwater. Geological Society of America. Mem. 142: 219-235. (4) BACK, W, HANSHAW, B.B. PYLE, T.E.; PLUMER, L.N. and WEIDIE, A. E. (1979). Geochemical significance of groundwater discharge and carbonate solution to the formation of Caleta Xel Ha, Quintana Roo. Mexico. Water Resources Research. 15 (6): 1521-1535. (5) BACK, W, HANSHAW, B.B.; HERMAN, J.S. and VANDRIEL, I.N. (1986) · Differencial disolution of a Pleistocene reef in the ground-water mixing zone of coastal Yucatan, Mexico. Geology. 14: 137-140. ( 6) SANFORD, W. D. and KONIKOW, L. F. ( 1 989) . Simulation of calcite dissolution and porosity changes in saltwater mixing zones in coastal aquifers. Water Resources Research. 25 (4): 655-667. ( 7) PRICE, R.M. ( 1988). Geochemical investigation of salt water intrusion along the coast of Mallorca, Spain. Departament of Environmental Sciencies. University of Virginia. M.S. Thesis: 1-186. (8) PASCUAL, J.M. and CUSTODIO, E. (1987). Procesos hidrogeoquimicos en la zona. de mezcla agua dulce-agua salada en el li toral del extrema meridional del macizo carbonatado de Garraf ( Tarragona) • Hidrogeologia y Recurs as Hidraulicos. Madrid· XI: 477-492. (9) CUSTODIO, E. , PASCUAL, J .M.: BAYO, A. and BOSCH X. (1989), processes in the m1x1ng zone in carbonate formations: central and Southern Catalonia. Natuurwet Tijdschr. Ghent. 70: 263-277 • (10) BOSCH, X. CUSTODIO, E, and PASCUAL, J.M. (1990). Geochemical reactions in carbonate coastal aquifers, Catalonia, Spain. Selected Papers in Hydrogeology. IAH.1: 147-160. (11) PASCUAL, J.M. and CUSTODIO, E. (1990). Geochemical observations in a continuously seawater intruded area: Garraf, Catalonia (Spain)· Procedings 11th Salt Water Intrusion Meeting. Gdansk: 308-330. 266 CHEMICAL ASPECTS (12) CUSTODIO, E., BAYO, A., PASCUAL, J.M. and BOSCH, X. (1990). Results from studies in several karst formations in southern Catalonia (Spain). International Symposium and Field Seminar on Hydrogeologic Processes in karst terranes. AIH. Antalya. Turkey (in press). (13) PASCUAL, J.M. (1990). Hidrogeoquirnica del macizo carbonatado de Garraf: Analisis de los procesos relacionados con la mezcla de aguas subteraneas dulces y saladas en el literal de Calafell-Bellvei (Tarragona). Doctoral Thesis. Dept. of Ground Engineering. Polytechnic University of catalonia: 1-255. (14) PLUMMER, N.L.; JONES, B.F. and TRUESDELL, A.H. (1984). WATEQF-A FORTRAN IV verslon of WATEQ, a computer program for calculating chemical equilibrium 'of natural waters. U.s. Geological Survey, Water Resources Investigations. 76-13: 1-70. (15) PARKHURST, .D.L.; PLUMMER, L.N. and THORSTENSON, D.C. (1982). BALANCE-a computer program for calculating mass transfer for geochemical , reactions in groundwater. U.s. Geological Survey, Water Resources Investigations. 82-14: 1-29. (16) PARKHURST, D.L.; THORSTENSON, D.C. and PLUMMER, L.N; (1985). PHREEQE - A computer program for geochemical reactions. U.S. Geological Survey, Water Resources Investigations. 80-96: 1-210.