Environ Geol (2008) 55:1473–1482 DOI 10.1007/s00254-007-1097-8
ORIGINAL ARTICLE
A geophysical and geochemical approach for seawater intrusion assessment in the Acquedolci coastal aquifer (Northern Sicily) A. Cimino Æ C. Cosentino Æ A. Oieni Æ L. Tranchina
Received: 24 April 2007 / Accepted: 22 October 2007 / Published online: 13 November 2007 Ó Springer-Verlag 2007
Abstract Vertical electrical sounding (VES’) surveys and chemical analyses of groundwater have been executed in the coastal plain of Acquedolci (Northern Sicily), with the aim to circumscribe seawater intrusion phenomena. This urbanized area is representative of a more general problem interesting most of Mediterranean littoral areas, where intensive overdraft favors a heavy seawater intrusion through the coastline. Aquifer resistivity seems decisively to be conditioned by the chlorine and magnesium content in the main aquifer of the region. Schlumberger VES’, together with piezometric and chemical–physical information of groundwater, allowed us to perform hydrogeological and geophysical elaborations, to propose the occurrence of a relatively narrow belt marked by fresh–salt water mixing. In the considered plain, pollution risk studies have been already realized by authors with the proposal to identify—by parametric and synthetic zoning of significant hydrogeological elements—the most vulnerable sectors. In detail, an intrinsic vulnerability mapping has been already performed, applying the well-known SINTACS system.
A. Cimino (&) C. Cosentino A. Oieni L. Tranchina Dipartimento di Fisica e Tecnologie Relative, University of Palermo, Viale delle Scienze, Edificio 18, 90128 Palermo, Italy e-mail:
[email protected] A. Oieni e-mail:
[email protected] L. Tranchina e-mail:
[email protected] C. Cosentino A. Oieni Dipartimento di Geologia e Geodesia, University of Palermo, Via Archirafi 20, 90123 Palermo, Italy e-mail:
[email protected]
This paper intends to give—in this sector of Sicily—an example of integration of different methodologies, showing the role of geophysics to describe the degradation of aquifers on the whole as well as to assess their pollution risk better. Keywords Sicily Acquedolci plain Groundwater Apparent resistivity Seawater intrusion
Introduction Recently, interdisciplinary research programs have permitted the collection and organization of a great number of territorial information, including geophysical and geochemical data in Sicily. The aim was mainly to perform a cartography of pollution vulnerability in this sector of Sicily. In fact, various methodologies of vulnerability assessment have been carefully applied and compared in many countries of the world (Gemitzi et al. 2006; Gogu et al. 2003). At this proposal, the notable role played by vulnerability in the general ambit of pollution risk assessment is well known (Civita and De Maio 1997). This is particularly true in similar contexts as the Acquedolci case, in which this very crowded sector of Sicily is locally subjected to heavy groundwater overdraft: this area has been opportunely interested by the SINTACS method application (Cimino et al. 2006). This point-count system model (PCSM) (Civita 1994; Civita and De Maio 2000) considers seven parameters strictly related with intrinsic hydrogeological features of aquifers, permitting their elaboration in GIS environment. SINTACS has been diffusely applied in other Italian areas (Civita et al. 1995). In the purposely considered area of Acquedolci (Fig. 1), the most vulnerable sectors lie in the alluvial fan of the
123
1474
Fig. 1 a Hydro-structural sketch and b AA’ section of the Acquedolci area. 1 Sandy–gravely–arenaceous complex; 2 sandy–arenaceous quaternary complex; 3 clayey–marly–arenaceous complex; 4 mesozoic calcareous–dolomitic complex; 5 piezometric level and 6 contour lines of piezometric levels (meters above sea level)
Furiano torrent, where the grain size features cause a notable increase of permeability of saturated and non-saturated strata. Here, as in other Sicily sectors, authors have created georeferenced archives of directly collected and analyzed data, also including available records from public bodies, field notes and GIS features. Records were generally organized at different spatial scales: so, georeferenced database structures have met additional difficulties, wholly overcome by the efforts of various researchers involved in this program. Considering the noticeable help offered by geophysical prospecting to delineate groundwater flows and pollution phenomena (Orellana 1982), in this note, authors also explain interpretations of electrical resistivity data. In the last decades, the progressive increase of water requirements and the consequent depauperation of its availability, together with the qualitative deterioration must certainly be included among the most serious environmental problems of Sicily and other world countries. In fact, the wild exploitation causes a notable lowering of water table, in spite of the natural recharge by rainfall and carbonate relieves. In anthropized areas—as the considered one—many sources of potential contamination points of groundwater
123
Environ Geol (2008) 55:1473–1482
occur; among these, authors remember sanitary landfills, municipal wastewater, cemeteries, agricultural fertilizers and accidental gasoline spills. In a certain area, these sources can be classified assigning to them particular danger contamination indexes (DCI) (Cimino and Andolina 2002). An important cause of increasing groundwater contamination in coastal plains as Acquedolci is commonly represented by salt waters. Heavy pumping, if associated with geological factors as grain size, can originate a landward migration of fresh–salt groundwater interface. Human action resulting in marine water entering an aquifer is generally called seawater intrusion. It occurs as ‘‘a result of the diversion of fresh water that previously had discharged from a coastal aquifer’’ (Fetter 1973, 2001). As a result, intense anthropic activity influences coastal hydrologic systems, leading to groundwater pollution by seawater intrusion. Incidence of this problem is noticeably increased in many littoral, urbanized regions of the world (Chachadi et al. 2003). Here the continuous exploitation of aquifers can be frequently observed, justifying a growing attention by the scientific community, as testified by numerous recent experiences (Demirel 2004; Liu and Cheng 1997; Melloul and Goldenberg 1997; Polemio et al. 2006). The evaluation of seawater intrusion has been dealt through different approaches. Some authors used radioactive isotopes in order to explain the increase of salinity, due to seawater intrusion, in the coastal aquifer of Israel (Yechieli et al. 2006). Other authors (i.e. Polemio et al. 2006; Pulido-Leboeuf 2004) employed only geochemical methods based on variations of salinity and, cation and anion concentrations, while others have introduced both geophysical and geochemical approaches to obtain a more comprehensive picture of this phenomenon (Di Sipio et al. 2006; Melloul and Goldenberg 1997; Sodde and Barrocu 2006). The aim of this paper is to display how different approaches (by geophysics and geochemistry) have been integrated and successfully used to identify and circumscribe seawater intrusion near the coastline in the Acquedolci area (Fig. 1). The definition of this aspect is very important for its close relation with the vulnerability assessment of a definite area and, consequently, with the whole hydrogeological risk pollution (Cimino et al. 2006). Indeed, it has to be considered in the studied plain, occurrence of intensive agricultural practices, with a relatively great diffusion of greenhouses and related spreading of chemical fertilizers. A contemporary inhomogeneous lowering of water table, estimated in [5 m during the last 10 years, is also observed. This imposes a correct definition of the hydrogeological problem, suggesting the possible
Environ Geol (2008) 55:1473–1482
ways to monitor and recover, for agricultural and urban needs, the partially compromised aquifer.
Geological and hydrogeological setting The coastal plain of Acquedolci is located in the Northern coast of Sicily (Fig. 1); this plain is bounded by Tyrrhenian Sea on the North, by a group of steep relieves (Pizzo Castellaro and San Fratello Mt.) on the South, by Inganno and Furiano torrents on the East and West, respectively. It is mainly constituted by quaternary alluvial deposits, often terraced, characterized by different permeability degrees, in accordance with the grain size, and a generally medium– high pollution vulnerability. In particular, it is possible to distinguish four principal geostructural complexes, characterized by hydrogeological homogeneity (Abbate et al. 2003; Cimino et al. 2002): (1)
(2)
(3)
(4)
A sandy–gravelly-arenaceous complex, grouping the quaternary alluvions of the torrent fans and the thin coastal belt deposits, with medium to high permeability for porosity. A sandy–arenaceous quaternary complex, covering most of the plain, with medium permeability grade for porosity and in close hydrogeological continuity with the first one; this complex, together with the first one, hosts an unconfined aquifer, intensely exploited by farms and greenhouses. Figure 1 shows contour lines of piezometric levels. A clayey–marly-arenaceous complex, mostly including all the deposits belonging to metamorphic fragments as well as to tertiary flysch units; these ones usually exhibit low or very low permeability values, locally performing a tamponage function with regard to the groundwater circulation; in detail, numidian flysch unit represents the impervious bed to the upper quaternary aquifers (see section AA0 in Fig. 1). A Mesozoic calcareous-dolomitic complex, with medium–high secondary permeability for fractures and karst (Pizzo Castellaro and San Fratello Mt.); it forms a conspicuous aquifer (up to 200 m deep) below the plain, supplying it along the detrital foothills. This carbonate aquifer can be reached in certain wells of the southern sectors of Acquedolci.
Karst relieves of Pizzo Castellaro and San Fratello Mt. are comprised in the Nebrodi East–West chain, mostly constituted by rocks belonging to Mesozoic complex and, subordinately, by metamorphic and flysch units (Cimino et al. 1998). Furthermore, the great permeability of these inland units evidences high vulnerability too, influencing
1475
the protection areas of certain springs supplied by karst groundwater (i.e. the Favara of Acquedolci source, indicated in Fig. 1). Since 10 years, periodical inventories of main hydrogeological and geophysical features have begun, with the aim to elaborate a wide-ranging vulnerability cartography of Acquedolci plain (Cimino et al. 1997).
Seawater intrusion in coastal aquifers The seawater intrusion phenomenon is well known in coastal aquifers. It occurs when the sea, interesting permeable rocks (for porosity or fractures), creates an interface below freshwaters, according to density contrast and to aquifer geometry. In fact, density of saline water (qw) is greater than density of fresh water (q). The interface salt water–fresh water depends very little on marine level, except for low sea fluctuations due to tidal and long-term climatic changes (Fetter 2001). So, the boundary can be considered in a quasi-equilibrium state, any movement caused by seasonal fresh–water discharge as well as by groundwater exploitation. As a matter of fact, this phenomenon can be considered—in a first approximation— essentially stationary, and the equilibrium between the two fluids subjected to the common hydrostatic laws. According to the hydrostatic equilibrium between sea water and fresh water: ðHi þ HÞq ¼ Hi qw
ð1Þ
the Ghyben-Herzberg principle states that: Hi ¼ H q=ðqw qÞ
ð2Þ
where qw saline water density, q fresh water density, Hi depth to the interface below sea level and H elevation of the piezometric surface above sea level. The Ghyben–Herzberg principle states that interface depth depends on the density of liquids—considered as immiscible—and on the distance between piezometric surface and sea level. Its application is possible only if the equilibrium is permanent and the interface is regular; but this equilibrium is difficult to be reached, because the mixing zone is strictly dependent on tidal cycles; while the fresh water zone is also regulated by seasonal conditions (above all rainfall) and human actions, as overdrafts by pumping systems. In fact, Hubbert (1940) verified that the equilibrium is never established because fresh water causes a dynamic balance of interface; so, it must be considered as a surface that can advance landward owing to the abovementioned factors. Besides, due to the different density, fresh groundwater generally grades into saline water with a steady increase in the content of dissolved solids. In some cases, the contact
123
1476
may be quite sharp, producing a very thin mixing zone. In aquifers with tidal hydraulic fluctuations, this layer will be thicker. Where this zone is only few meters thick, the Ghyben–Herzberg principle can be fairly applied. For the purposes of this note, it is essential to express that the mixing layer can likely involve wells and degrade water quality. According to Ghassemi et al. (1993), the intrusion features depend, besides depth, exploitation and recharge rates and well distance from the coast, also on aquifer geometry, porosity, hydraulic conductivity and dispersivity, taking into account local or general anisotropy of these hydrogeological characteristics. So, complex models are needed to quantify these factors. In fact, over the years, different mathematical and numerical models, more or less complex, have been developed with the aim to understand seawater intrusion and to establish the position and thickness of transition zone between fresh and salt water in coastal zones (Bear 1979; Reilly and Goodman 1985; Oude Essink 2001; Narayan et al. 2007). Moreover, in different recent works (Cheng and Ouazar 1999; Cheng et al. 2000; Barlow 2003; Mantoglou 2003; Mantoglou et al. 2004), as well as in this paper, freshwater and saltwater zones are considered to be separated by a sharp boundary, with the approximation based on the Ghyben–Herzberg law and on unconfined aquifers and steadystate flow. To determine groundwater deterioration, seawater intrusion can be detected both directly (in wells by electrical probes or samplings) and indirectly by geophysical methods (Melloul and Goldenberg 1997). This study takes into account direct groundwater samplings and analyses as well as indirect geophysical approaches in order to suggest the fresh–salt water interface geometry.
Environ Geol (2008) 55:1473–1482
analyses as well as piezometric measurements have been executed in the late-fall season, after a rainy period. This high recharge phase of aquifers constitutes the best conditions to opportunely evaluate the depth to water parameter, utilized in the pollution vulnerability assessment of Acquedolci region (Cimino et al. 2006). In particular, wells of Acquedolci plain show piezometric levels up to at least 120 m, with higher values in correspondence with the deep karst aquifer (wells AO-7 and AO-8 in Fig. 2). Wells generally reach in depth the saltwater–freshwater interface, depending on the overdraft intensity as well as on the distance from the coast. Integrated tools represented by hydrochemistry and geophysics define the seawater intrusion entity. The role of tidal effects, generally affecting the already mentioned salt– fresh transition zone, is here retained wholly negligible in all the zones of the plain, in the alluvial fans as well as in the central sectors.
Groundwater chemical analyses
Materials and methods
All chemical analyses of groundwater samples were carried out in the laboratory of the Azienda Municipalizzata Acquedotti of Palermo, according to the following national reference methods: complexometric method by IRSA 2040 for HCO3 measurement; ion chromatographic method UNICHIM and UNICHIM 800 for Ca2+, Mg2+, Na+, K+ and Cl-, SO24 measurements, respectively. Conductivity was estimated on field using a common portable instrument. These values have been automatically compensated to a temperature of 20°C by the instrument during measurements. Figure 2 illustrates the location of wells; in each point, the circle size is proportional to chloride concentration. Table 1 reports the results of geochemical prospecting relevant to the mostly characterizing ions; it
Hydrochemical and geophysical surveys have been carried out in the study area. Inventories, arrays and geochemical analyses have been executed in different times, interesting a total of 41 wells and 41 VES’. Wells involved generally more aquifers. In detail, quaternary complexes are hydraulically connected, both being supplied by the karst aquifer. As quoted above, essential data relevant to water wells have been organized in a georeferenced database. In particular, information on geometry of wells, including piezometric levels and depths, were opportunely recorded. Most of the drilled boreholes present 0.20–0.40 m diameters, while older wells are up to 3 m larger. Sampling method is strictly dependent on well diameter: for small diameter wells (mainly involving the karst aquifer) a private pump was utilized, while for the large ones a direct sampling has been executed. Samplings for chemical
Fig. 2 Map of the investigated area, exhibiting well locations and distribution of electrical conductivity of groundwater. Size of circles is proportional to chloride concentration
123
Environ Geol (2008) 55:1473–1482
1477
Table 1 Chemical–physical data, involved in the encroachment phenomena, relevant to part of groundwater samples Well ID
UTM
T (°C)
Conductivity (20 lS cm-1)
Piezometric level (m)
Cations (mg L-1) Ca2+
Mg2+
Anions (mg L-1) Na+
K+
Cl-
SO24
125.66
73.98
108.76
311.1
115.80
68.84
527.04
45.93
34.93
9.11
484.95
43.29
99.48
3.72
506.3
265.60
172.60
72.23
3.66
509.96
178.97
177.66
38.86 27.11
70.86 62.78
10.14 10.50
184.22 289.14
156.70 75.80
97.24 159.75
86.66
24.14
51.25
3.60
267.18
59.15
159.65
134.00
28.52
76.38
6.70
337.94
107.97
186.77
8
71.15
14.70
30.60
5.93
351.36
40.24
74.70
5,820
0.8
288.30
157.02
768.90
24.33
412.97
1241.00
470.16
19.9
760
10.8
109.20
20.20
42.76
5.72
358.68
42.50
108.35
4213240
18.8
1,587
14.2
178.30
57.97
106.40
6.40
290.36
360.00
176.62
4211360
16.5
1,057
131.59
193.44
18.70
42.70
34.24
440.42
161.04
89.64
463250
4211210
17.6
714
109.7
94.91
21.09
39.22
3.06
342.82
79.45
48.32
OT-4
464000
4212000
17.9
1,125
78.8
151.60
40.85
73.24
3.60
497.76
132.50
77.90
OT-5
461320
4211080
19.9
1,015
34.7
145.70
30.17
59.62
9.00
315.98
59.31
210.02
OT-6
460580
4212200
19.1
1,254
7.6
170.51
32.80
71.88
5.23
357.46
78.30
219.06
OT-7
460430
4212000
21.0
1,512
5.97
230.55
42.83
82.67
5.70
351.36
98.80
274.05
OT-8
461075
4212470
19.0
945
8.46
145.61
26.77
51.74
8.84
323.3
51.92
162.16
OT-9 PM-1
461570 462050
4211275 4212160
21.4 13.9
894 2,800
35.09 7.98
116.52 192.11
28.72 75.37
58.43 386.64
4.32 7.55
287.92 500.2
55.25 709
163.24 244.80
PM-10
465450
4212775
18.6
1,501
23.53
160.40
44.64
138.90
4.62
335.5
204.35
207.55
PM-11
463525
4211960
19.0
2,960
53.23
138.50
70.62
507.44
5.40
524.6
417.52
511.70
PM-12
463450
4212450
17.3
910
10.25
97.42
34.06
80.38
3.20
323.3
96.5
101.27
PM-13
462575
4212610
18.0
1,240
4.33
111.30
33.45
132.00
9.34
341.6
158.65
138.85
PM-2
462440
4211425
18.8
2,080
50.5
173.50
73.36
216.20
6.23
518.5
532
186.95
PM-3
462110
4210925
16.8
893
123.03
108.30
28.60
67.22
7.90
213.5
68.9
165.00
PM-4
462100
4212460
19.0
1,295
8.31
148.00
40.09
83.58
3.06
256.2
360
156.00
PM-5
464050
4211950
15.0
1,637
67.45
165.20
60.70
176.00
1.50
457.5
249
233.50
PM-6
463715
4211605
17.2
900
115.88
151.43
18.30
46.45
2.03
433.1
85.58
54.80
PM-7
463150
4212060
17.9
738
45.08
106.26
15.07
55.98
7.15
341.6
50.48
75.96
PM-8
464710
4212125
15.5
1,426
44.8
175.50
55.86
111.37
4.08
512.4
180.5
165.20
X
Y
AO-7
462540
4210590
19.8
611
181.35
57.17
12.26
51.40
9.83
AO-8
462725
4210725
18.5
852
177.53
84.39
8.65
56.45
12.37
FC-10
462840
4211265
16.2
622
165.4
109.00
3.20
35.00
12.00
FC-7
466050
4212450
17.1
779
113
106.64
22.76
41.61
FC-8
465775
4213190
19.5
1,767
10.8
198.42
58.08
95.30
FC-9
465650
4212260
18.7
1,322
95
72.92
76.64
LD-1 LD-2
461740 461510
4212545 4212460
17.7 18.8
1,070 961
1.68 9.14
112.56 119.31
LD-3
461490
4212200
18.0
946
8.24
LD-4
461660
4212310
18.4
1,076
15.15
LD-6
465050
4212960
19.2
670
LD-7
464440
4212885
19.3
LD-8
465300
4213210
LD-9
465550
OT-2
463475
OT-3
HCO3
PM-9
464225
4212410
18.2
1,014
23.3
126.50
39.80
53.58
3.01
408.7
97.92
95.42
RM-10
464700
4211650
12.0
552
91
65.26
32.86
27.00
1.60
311.1
50.66
24.82
AO-6
466590
4213645
19.1
963
2.37
138.90
28.62
56.43
11.55
418.46
102.55
139.69
FC-1 FC-5
466300 466275
4213790 4213180
19.0 18.3
1,332 1,039
3.6 17.16
155.04 134.71
29.53 20.44
117.40 85.08
6.06 10.25
573.4 459.33
159.8 116.3
154.36 163.90
FC-6
466560
4213290
18.8
2,150
6
139.06
61.03
211.82
5.03
539.24
300.6
341.70
PF-1
466575
4212450
12.1
2,130
83.22
97.80
114.40
274.80
10.74
628.3
532
159.00
must be underlined that other groundwater measurements have been performed, interesting bacteriological analyses (Cimino et al. 2006) and trace elements as well. In this
paper, only ions which may characterize the seawater intrusion phenomenon are summarized. Table 1 also shows data on well temperatures, which are characterized by great
123
1478
variability. In fact, in the studied area at least two different water springs occur (Favara of Acquedolci and Mascarino sources in the North and South of calcareous relieves, respectively) with different temperatures. Their waters belong to different hydrological pathways that contribute to feed the arenaceous aquifer; waters of the northern spring are characterized by temperatures higher than the southern one: this can explain the differences of temperature in certain sampled wells. Since the aquifers—as above quoted—are hydraulically connected, well waters show a gradient of temperature due to the mixing of ground waters relevant to different sources. More in detail, total and fecal coliforms (Entero bacteriacee and E. coli) as well as bacteria belonging to the genus of Streptococcus have been detected near the Inganno torrent course, reaching values up to respectively 55,000 and 63,000 UFC/100 mL. This occurrence in groundwater suggests a well-localized bacteriological risk due to nitrate fertilizers in agriculture or to non-treated urban discharges. Statistical principal component analysis (PCA) of chemical–physical groundwater features was also performed, in order to assess possible correlations among variables, also evaluating possible matching between well locations and measured ion concentrations.
Geophysical survey As universally recognized and described (Kunetz 1966), geoelectric prospecting is the most suitable geophysical method for hydrogeological studies. It differentiates pervious and impervious formations, easily depicting geometry of hydrostructures as well as groundwater contaminations through seawater intrusion. In most of the cases, VES’ (vertical electrical soundings) surveys by Schlumberger arrays are normally executed, using four probes put into the ground. Briefly, a vertical electrical sounding is represented by a discrete sequence of apparent resistivity measures of underground, carried out with a growing spacing between a couple of current electrodes, so interesting deeper and deeper formations. Centre and orientation of VES’ array are maintained fixed. Quantitative interpretation processes permit to investigate geometric and hydrogeological features of aquifers: among these, VES’ allow us to find out salt contaminations of groundwater and to perform a zoning of a considered area on the basis of aquifer resistivity, taking into account a preliminary hydrogeological model (stratigraphic and geochemical information). Among the very numerous application of Schlumberger VES’ to identify seawater intrusion in coastal aquifers, the authors consider the experiences in Mediterranean areas of Shaaban (2001) and Khalil (2006). In both the cases, resistivity relevant to intruded aquifers is\5–10 X m.
123
Environ Geol (2008) 55:1473–1482
In the Acquedolci area, the geophysical survey has been principally carried out in the coastal deposits; in particular, a small group of soundings has been opportunely located in the sandy–gravelly alluvial fan of Furiano torrent, in the western sector of the plain (Fig. 3). The entire studied area has been densely interested by a set of 41 VES’. As it is well known, the applied geophysical method portrays the distribution of ground electric resistivity (Kunetz 1966; Orellana 1982). In this paper, the specific experience carried out in the structurally complex Acquedolci area shows the good integration among hydrogeological, hydrochemical and geophysical methodologies, thanks to the fair correlation between resistivity of aquifer and salinity of groundwater for seawater intrusion. Authors refer to similar VES’ surveys in Egypt (Shaaban 2001; Al-Sayed and El-Qadi 2007), also mentioning the application of other geophysical methods, as TDEM (time domain electromagnetic method). These electromagnetic surveys have been carried out to investigate sea-intrusion in coastal aquifers of Israel (Melloul and Goldenberg 1997). The application of TDEM is recommended in presence of very low resistivity values of the conductive layers as well as of notable shallow lateral heterogeneity, where interpretation of DC measures can be characterized by uncertainty and ambiguity. Easy interpretations of VES’ curves in Acquedolci area, corroborated by the stratigraphic knowledge, widely justified the use of this inexpensive method. This is confirmed by the survey carried out by the Azienda Nazionale Autonoma Strade, during the preliminary geological study of the Furiano torrent delta construction of a new bridge, where electro-stratigraphic cross section of the alluvial fan was derived (Abbate et al. 1994). Geoelectrical prospecting has been accurately planned in order to depict the trend of sea pollution, which follows
Fig. 3 Location of vertical electrical soundings (VES’), with relevant number, in the surveyed area. Topographic contour lines are also exhibited
Environ Geol (2008) 55:1473–1482
the above quoted Ghyben–Hertzberg law taking into close account the irregular permeability features of aquifers and their seasonal exploitation for different uses. For the purpose of this work, authors here present and discuss only VES’ numbers 1, 2 and 3, close to coastline, and VES number 4, far from the shore (Fig. 3). The shown VES’ (Fig. 4) have been executed by means of Schlumberger arrays; resistivity values have been measured using a PASI digital georesistivimeter, model 16 GL. The maximum VES’ spacing was 600 m. The investigation depths were suitable to the aquifer geometry (see piezometric levels in Table 1), reaching the saturated zone and also revealing the eventual occurrence of impervious interbeddings (clays). According to the depths of saturated zone and/or clay top, the depiction of aquifer geometry needed an investigation depth of at least 150 m, assuring the reaching of the polluted sectors. Furthermore, stratigraphic controls aided the inversion of geophysical data. Examples are easily found in the Furiano torrent fan, where well-drillings together with close VES’ occurred, crossing completely the delta sector, as described in the relevant reference (Abbate et al. 1994). Very low resistivity of the saturated zone of aquifer was clearly depicted in certain VES’ and compared with stratigraphic and geochemical outlines. As a matter of fact, values \10 X m were interpreted in the alluvial western fan of Furiano torrent and in the eastern narrow belt of the area, in which high chlorine contents in groundwater were detected ([500 mg L-1). VES’ interpretation allowed us to easily and quickly solve the geoelectrical inverse problem, applying relatively simple geological model based on horizontally layered stratification. In fact, acquired curves exhibited a reliable assessment of the curve asymptotes relevant to the main
1479
hydrostructures also involved in the marine encroachment process as well. As a final result, the distribution knowledge of apparent resistivity parameter, running in GIS environment, permitted to explain local problems, with a comprehensive and up-datable view of the hydrogeological patterns.
Results and discussion Figure 2 shows that—as expected—groundwater electrical conductivity, in the investigated area, agrees with the chloride distribution, as shown in Table 1. This is confirmed by the significant value of their correlation factor, as illustrated in Table 2. Observing this table, it is possible also to evidence a very high correlation value between Na+ and conductivity. This result has to be carefully taken into account, because sodium rate can also be affected by ionic exchange processes between groundwater and clay minerals: Na+ concentration can locally increase in relation to the clay interbeddings in layered sectors of aquifer. In this case, further procedures could be useful to better assess the geostatistical trends of the quoted elements. Furthermore, the concentration of HCO3 , in spite of the apparent absence of any appreciable statistical correlation with almost the remaining variables (Table 2), allows us to recognize the eastern zones of plain where aquifers are supplied by the karst and fracture network of inland relieves (Pizzo Castellaro and San Fratello Mt., see Fig. 1), also through the Inganno torrent. These considerations appear decisive to draw the groundwater pathways in the whole Acquedolci region, including the calcareous-dolomitic southern outcrops, distinguishing sectors with different chemical behaviours, as shown in PCA diagram (Fig. 5).
Fig. 4 VES’ curves with relative interpretation
123
1480
Environ Geol (2008) 55:1473–1482
Table 2 Correlation matrix among chemical–physical parameters measured on the 41 selected well-waters, see also Table 1
Conductivity
Conductivity
Ca2+
Mg2+
Na+
K+
HCO3
Cl-
SO24
1
0.687
0.893
0.966
0.305
0.354
0.949
0.813
1
0.533
0.552
0.319
0.296
0.616
0.573
1
0.841
0.137
0.421
0.849
0.680
Ca2+ Mg2+ +
Na K+
HCO3
1
0.280
0.346
0.917
0.813
1
0.029
0.320
0.110
1
0.332
0.239
Cl-
1
SO24
0.653 1
Significant values (except diagonal) at the level of significance a = 0.050 (two-tailed test) are given in bold numbers
Fig. 5 Principal component analysis (PCA), relevant to the measured parameters (see Tables 1, 2). In the delimited region of the graph, the most HCO3 enriched waters are evidenced
VES’ curves generally exhibit highest values of resistivity along torrent deposits and, primarily, at the foothills of the limestone relieves (debris), where resistivity values overcome 2,000 X m (VES number 4, Fig. 4). Resistivity increases in the inland part of the alluvial deposits of Furiano torrent, sharply decreasing towards Western and Eastern coastal belts. Lower resistivity values, relevant to clayey layers, have been found in central sectors of the investigated area. Geoelectrical interpretation has evaluated the true resistivity of the quaternary aquifer, which ranges between 50 and 200 X m, with higher values up to 500 X m, as tested by shallow measures. The understanding of the geophysical results has evidenced the clear influence of sea water intrusion on apparent resistivity. This confirms, in the investigated
123
plain, the fair relationship among the investigated parameters relevant to geochemical and geophysical prospecting, in spite of the possible occurrence of shallow clayey interbeddings into the sandy–arenaceous overburden. At this proposal, eventual interpretation ambiguities can be easily solved by available stratigraphic information and resistivity gradient. It has to be noted that geoelectrical interpretation of intrusion phenomena—with the consequent salt–fresh water mixing—does not appear anywhere coherent with the shore–well distance. Indeed, referring to the quoted figures, VES 2, relatively far from the coastline, exhibits an asymptotic resistivity value lower than VES 3, closer to the sea. This indicates an inland advance of brackish belt in the Furiano torrent delta: in this sector, high permeability of
Environ Geol (2008) 55:1473–1482
the sandy–gravely formation and a local groundwater overdraft determine an anomalous trend of the seawater intrusion phenomenon, recognized thanks to resistivity data acquired in the alluvial fan. In detail, bore–hole data and indirect permeability evidences, based on well productivity estimations, confirm the noticeable occurrence of gravels and conglomerates in well-localized portions of this alluvial aquifer (Abbate et al. 1994). As a matter of fact, this area—in which an alluvial fan is present—is characterized by higher values of permeability owing to the presence of coarser materials with respect to the central coastal sector of the plain. These considerations permit to hypothesize diverse distances of the saltwater–freshwater interface from the coastline, owing to difference grain sizes among the formations of the plain, as evidenced by the resistivity survey outcomes in Western and Eastern sectors of the plain. In fact, Eastern sectors of Acquedolci plain are characterized by a narrow salty belt, very close to the littoral line, see VES 1 in Fig. 4. Here, aquifer is characterized by more homogeneous grain size, with a thinner or almost missing mixing layer and a sharp salinity gradient towards the sea. Finally, higher resistivity values generally evidence the absence of salt–brackish groundwater or clayey interbeddings.
1481
georeferenced data, in order to typify the aquifers according to chemical–physical features of groundwater. Considering that groundwater could constitute, in future, the main quality resources of the whole Nebrodi region, authors, by monitoring their quoted chemical and physical features, consider essential a protection and a recovery strategy are essential. Further water samplings and geoelectrical measurements, periodically executed, can be inserted in a control program of the pollution trend in the Acquedolci area. By now, the continuous demand of waters for the different needs, also considering the growing demand for tourism, is only partially satisfied by local aqueducts, mainly supplied by the mentioned Favara spring. So, the plain is subjected to a whole hydrogeological risk not only for seawater intrusion and groundwater overexploitation, but also for the intense spreading of fertilizers. The shown updateable representations, pertaining to vulnerability and risk concepts, are essential to consider the problem of the aquifer protection in this crowded sector of Sicily. Acknowledgments Authors would like to thank the staff of the chemical laboratory of the Azienda Municipalizzata Acquedotti di Palermo for the help during the collection and analyses of wellwaters, the Azienda Nazionale Autonoma Strade for the availability of VES’ and boreholes data, and anonymous referee for his critical revision of the manuscript.
Conclusions This article highlights the notable importance of integrated methodologies to delineate groundwater flows and seawater pollution in a coastal area. The considered Sicily sector is a good test-site to study the aquifer contamination problem, very common to urbanized Mediterranean littorals, where severe groundwater exploitations cause seawater intrusion. The simultaneous interpretation of geophysical and hydrogeochemical outcomes, in the frame of interdisciplinary projects, constitutes the first attempt towards the exhaustive pollution risk assessment in Sicily. Relevant mapping will be available to local government as a necessary new tool of territorial planning. The survey validates the expected contamination risk for sea intrusion from the North, especially in the circumscribed sectors of the torrent fans. The shown investigations and outcomes are relevant to a particular zone of Northern Sicily characterized by elevate hydrogeological risk mainly due to seawater intrusion, as vulnerability and quality evidences have testified. This sector of Sicily represents a significant example of serious and uncontrolled exploitation of groundwater: this paper intends to update the contributions and the improvements carried out by researchers in the ambit of interdisciplinary projects, addressed towards a better management of water resources on the whole. A decisive role in this research has also been played by GIS elaboration of
References Abbate R, Cimino A, Emma S, Martorana Tusa A, Orecchio S (1994) Lineamenti geomorfologici e idrogeologici dell’acquifero carbonatico del territorio di Acquedolci, Messina (Geomorphological and hydrogeological outlines of the karst aquifer of Acquedolci area). Boll Acc Gioenia Sci Nat 27:579–597 Abbate R, Cappadona Ignazzitto S, Cimino A, Di Patti C, Orecchio S (2003) Indagini integrate per la valorizzazione delle risorse ambientali nell’area carbonatica di Monte San Fratello—Zona b del Parco dei Nebrodi (integrated surveys for the valorisation of the environmental resources in the karst area of Monte San Fratello—B zone of Nebrodi Park). Thalassia Salentina 26(supp):65–76 Al-Sayed EA, El-Qady G (2007) Evaluation of seawater intrusion using the electrical resistivity and transient electromagnetic survey: case study at fan of Wadi Feiran, Sinai, Egypt. EGM 2007 International Workshop innovation in EM, grav. and mag. methods: a new perspective for exploration, Capri, Italy Barlow PM (2003) Ground water in freshwater–saltwater environments of the Atlantic coast. US geological survey, information services, Denver federal center, Denver, US, Circular 1262, 121 p Bear J (1979) Hydraulics of groundwater. McGraw Hill, New York. 569 pp Chachadi AG, Lobo Ferreira JP, Noronha L, Choudri BS (2003) Assessing the impact of sea-level rise on salt water intrusion in coastal aquifers using GALDIT model. APRH/CEAS, Semina´rio ´ guas Subterraˆneas, Lisbon, 13 p Sobre A Cheng AH-D, Ouazar D (1999) Analytical solutions. In: Bear J, Cheng AHD, Sorek S, Ouazar D, Herrera I (eds) Seawater
123
1482 intrusion in coastal aquifers—concepts, methods and practices. Kluwer, Dordrecht. pp 163–187 Cimino A, Andolina F (2002) The territorial danger in the cartography of groundwater contamination risk in Palermo Plain. Mem Soc Geol It 57:561–568 Cheng AH-D, Halhal D, Naji A, Ouazar D (2000) Pumping optimization in saltwater-intruded coastal aquifers. Water Resour Res 36:2155–2165 Cimino A, Abbate R, Macaluso M, Orecchio S (1997) Karst hydrogeology and vulnerability in a coastal sector of Nebrodi Mts. region (Northern Sicily). Sci Tech Envir, Me´m H S 12:205–208 Cimino A, Abbate R, Martorana Tusa A (1998) The Regional Park of Nebrodi Mts. (Sicily): a contribution to an integrated management of groundwater. Environ Geol 34:320–328 Cimino A, Abbate R, Cappadona Ignazzitto S, Orecchio S, Sambataro S (2002) Protection and monitoring of water resources in North– West Sicily (with particular regard to the S. Fratello-Acquedolci karst area). Speleologia Iblea 10:107–115 Cimino A, Artino C, Oieni A (2006) Rischi idrogeologici e sanitari in aree urbane della Sicilia soggette a contaminazione degli acquiferi: recenti risultati e rappresentazioni cartografiche in ambiente GIS (hydrogeological and sanitary risks in urban area of Sicily subjected to aquifer contamination: recent results and cartographic representation in GIS environment). 24th environment day meeting ‘‘climate and health’’. Accademia Nazionale dei Lincei, Rome, pp47–50 Civita M (1994) Le carte della vulnerabilita` degli acquiferi all’inquinamento: Teoria e pratica (Vulnerability pollution maps of aquifers: theory and practice). Pitagora Ed, Bologna. 325pp Civita M, De Maio M (1997) Assessing groundwater contamination risk using ARC/INFO via GRID function. ESRI International USER Conference, San Diego Civita M, De Maio M (2000) Valutazione e cartografia automatica della vulnerabilita` degli acquiferi all’inquinamento con il sistema parametrico SINTACS R5 (evaluation and automatic pollution vulnerability of aquifer using SINTACS R5 parametric system). Pitagora Ed, Bologna. 226pp Civita M, Gargini A, Manzone L, Pranzini G (1995) Applicazione del sistema parametrico SINTACS alla valutazione della vulnerabilita` intrinseca degli acquiferi nel bacino intrappenninico del Valdarno Medio. Quaderni di Geologia Applicata 3:37–40 Demirel Z (2004) The history and evaluation of saltwater intrusion into a coastal aquifer in Mersin, Turkey. J Environ Manage 70:275–282 Di Sipio E, Galgaro A, Zuppi GM (2006) Salt water contamination on Venice lagoon mainland: new evaluation of origin, extension and dynamics. Proceedings of 1st SWIM-SWICA (19th salt water intrusion meeting-3rd salt water intrusion in coastal aquifers), Cagliari Fetter CW (1973) Water resources management in coastal plain aquifers. Proceedings of the international water resources association, first world congress on water resources. pp322–331 Fetter CW (2001) Applied Hydrogeology. Prentice-Hall. Upper Saddle River. 598 pp Gemitzi A, Petalas C, Tsihrintzis VA, Pisinaras V (2006) Assessment of groundwater vulnerability to pollution: a combination of GIS, fuzzy logic and decision making techniques. Environ Geol 49:653–673
123
Environ Geol (2008) 55:1473–1482 Ghassemi F, Chen TH, Jakeman AJ, Jacobson G (1993) Two- and three-dimensional simulation of seawater intrusion: performances of the ‘‘SUTRA’’ and ‘‘HST3D’’ models. AGSO J Aust Geol Geophys 14:219–226 Gogu RC, Hallet V, Dassargues A (2003) Comparison of aquifer vulnerability assessment techniques. Application to the Neblon river basin (Belgium). Environ Geol 44:881–892 Hubbert MK (1940) The theory of ground-water motion. J Geol 48:785–944 Khalil MH (2006) Geoelectric resistivity sounding for delineating salt water intrusion in the Abu Zenima area, West Sinai, Egypt. J Geophys Eng 3:243–251 Kunetz G (1966) Principles of direct current resistivity prospecting. Geoexploration monographs, Gebruder-Borntraeger, Berlin. 103 pp Liu CW, Cheng LH (1997) Hydrogeological investigation of a groundwater contamination site in southern Taiwan. Environ Geol 29:238–245 Mantoglou A (2003) Pumping management of coastal aquifers using analytical models of saltwater intrusion. Water Resour Res 39:1335 Mantoglou A, Papantoniou M, Giannoulopoulos P (2004) Management of coastal aquifers based on nonlinear optimization and evolutionary algorithms. J Hydrol 297:209–228 Melloul AJ, Goldenberg LC (1997) Monitoring of seawater intrusion in coastal aquifers: basic and local concerns. J Environ Manage 51:73–86 Narayan KA, Schleeberger K, Bristow KL (2007) Modelling seawater intrusion in the Burdekin Delta irrigation area, North Queensland, Australia. Agr Water Manage 89:217–228 Orellana E (1982) Prospeccio´n geoele´ctrica en corriente continua. Paraninfo, Madrid. 578 pp Oude Essink GHP (2001) Salt water intrusion in a three dimensional groundwater system in the Netherlands: a numerical study. Trans Porous Media 43:137–158 Polemio M, Dragone V, Limoni PP (2006) Salt contamination in Apulian aquifer: spatial and time trend. Proceedings of 1st SWIM-SWICA (19th salt water intrusion meeting-3rd salt water intrusion in coastal aquifers), Cagliari Pulido-Leboeuf P (2004) Seawater intrusion and associated processes in a small coastal complex aquifer (Castell de Ferro, Spain). Appl Geochem 19:1517–1527 Reilly TE, Goodman AS (1985) Quantitative analysis of saltwater freshwater relationships in groundwater systems: a historical perspective. J Hydrol 80:125–160 Shaaban FF (2001) Vertical electrical soundings for groundwater investigation in northwestern Egypt: a case study in a coastal area. African Earth Sci 33:673–686 Sodde M, Barrocu G (2006) Seawater intrusion and arsenic contamination in the alluvial plain of the rivers Quirra and Flumini Pisale, south–eastern Sardinia. Proceedings of first SWIMSWICA (19th salt water intrusion meeting-third salt water intrusion in coastal aquifers), Cagliari Yechieli Y, Kafri U, Sivan O (2006) The interrelation between the sea and the coastal aquifer, deduced from analyses of radioactive isotopes. Proceedings of first SWIM-SWICA (19th salt water intrusion meeting-third salt water intrusion in coastal aquifers), Cagliari