Int J Earth Sci (Geol Rundsch) (2007) 96:541–566 DOI 10.1007/s00531-006-0111-9
ORIGINAL PAPER
Chemical evolution of saline waters in the Jordan-Dead Sea transform and in adjoining areas Peter Mo¨ller Æ Eliyahu Rosenthal Æ Stefan Geyer Æ Akiva Flexer
Received: 8 April 2005 / Accepted: 1 May 2006 / Published online: 11 August 2006 Springer Verlag 2006
Abstract The Ca–Mg relationship in groundwaters strongly points to the overall dolomitization and local albitization. The Mg/Ca ratios reveal two trends by which saline waters develop: increase of Mg/Ca ratio by evaporation and decreasing Mg/Ca ratios due to dolomitization and albitization. Br/Cl vs. Na/Cl ratios demonstrate that albitization does not play a major role which leaves dolomitization to be the main source for decreasing Mg/Ca ratios in saline waters. In the eastern and southern Region of Lake Kinneret, salinization occurs by mixing with a Ca/Mg molar ratio <1 brine (Ha’On type). Along the western shoreline of the Lake, a Ca/Mg > 1 dominates, which developed by the albitization of plagioclase in abundant mafic volcanics and the dolomitization of limestones. The most saline groundwater of the Tabgha-, Fuliya-, and Tiberias clusters could be regional derivatives of at least two mother brines: in diluted form one is represented by Ha’On water, the other is a Na-rich brine of the Zemah type. Additionally, a deep-seated Ca-dominant brine P. Mo¨ller (&) Geoforschungszentrum Potsdam, Telegrafenberg, 14473 Potsdam, Germany e-mail:
[email protected] E. Rosenthal Æ A. Flexer Department of Geophysics and Planetary Sciences, Tel Aviv University, 69978 Tel Aviv, Israel e-mail:
[email protected] A. Flexer e-mail:
[email protected] S. Geyer Centre for Environmental Research, Leipzig-Halle, Theodor Lieser St. 4, 06120 Halle/Saale, Germany e-mail:
[email protected]
may ascend along the fractures on the western side of Lake Kinneret, which is absent on the eastern side. Groundwaters of the Lower Jordan Valley are chemically different on both sides of the Jordan River, indicating that the exchange of water is insignificant. All saline waters from the Dead Sea and its surroundings represent a complex mixture of brines, and precipitation and local dissolution of halite and gypsum. Many wells of the Arava/Araba Valley pump groundwater from the Upper Cretaceous limestone aquifer, the origin of the water is actually from the Lower Cretaceous Kurnub Group sandstones. Groundwater drawn from the Quaternary alluvial fill either originates from Kurnub Group sandstones (Eilat 108, Yaalon 117) or from altered limestones of the Judea Group. The origin of these waters is from floods flowing through wadis incised into calcareous formations of the Judea Group. On the other hand, as a result of step-faulting, hydraulic contact is locally established between the Kurnub- and the Judea Groups aquifers facilitating the inter-aquifer flow of the confined Kurnub paleowater into the karstic formations of the Judea Group. Two periods of Neogene brine formation are considered: the post-Messinan inland lagoon resulting in drying up of the Sdom Sea and the evaporation of the Pleistocene Samra Lake, which went further through the stage of Lake Lisan to the present Dead Sea. For the first period, major element hydrochemistry suggests that the saline waters and brines in the Jordan-Dead Sea–Arava Valley transform evolved from the gradual evaporation of an accumulating mixture of sea-, ground-, and surface water. Due to the precipitation of carbonates, gypsum, and halite, such an evaporating primary water body was strongly enriched in Mg, Br, and B and shows high
123
542
molar ratios of Br/Cl, B/Cl, and Mg/Ca but low Na/Cl ratios. The development of the Br/Cl ratio is chemically modelled, showing that indeed brine development is explicable that way. Along with the evaporation brine, evaporites formed which are leached by infiltrating fresh water yielding secondary brines with Na/Cl ratios of 1. When primary brines infiltrated the sub-surface, they were subjected to Mg–Ca exchange in limestones (dolomitization) and to chloritization and albitization in basic igneous rocks turning them into Ca-Cl brines. These tertiary brines are omnipresent in the Rift. The brines of the late Lisan and Dead Sea were generated by evaporating drainage waters, which leached halite, gypsum, and carbonates from the soil and from the sub-surface. All these brines are still being flushed out by meteoric water, resulting in saline groundwaters. This flushing is regionally enhanced by intensive groundwater exploitation. In variable proportions, the Neogene and late Lisan Lake and Recent Dead Sea brines have to be considered as the most serious sources of salinization of groundwaters in the Rift. Deep-seated pre-Sdom brines cannot strictly be excluded, but if active they play a negligible role only.
in adjoining areas, i.e. east of the emerging water divide between the Rift and the Mediterranean Sea, the dominant brine was the final product of the evaporation of an inland marine lagoon (the Sdom Sea, Fig. 1), which infiltrated into an environment prevalently built of previously formed carbonate rocks and particularly of clastic beds filling the nascent Rift. Evaporites precipitated in this lagoon and the subsequent dissolution of the resultant salts also contributed to the generation of brines. During the Pleistocene, fluvio-lacustrine conditions prevailed in the Jordan Valley and the fresh water Samra Lake developed. This lake became progressively saline, probably as the result of the dissolution and flushing of salts from the previous hypersaline Sdom Sea (Picard 1943). This evolved into the saline Lisan Lake, which represents the last step in the hydrochemical evolution in the Rift. From the Pleistocene onwards, Lake Lisan became more and more saline by evaporation and the subsequent flushing of evaporites. At present, the remnant of this period is the Dead Sea. Thus, during four major periods, the rock sequences were flushed by previously formed brines
36˚ E
35˚
N
Legend:
34˚
Major rivers Volcanics Maximal extension of lakes Emergent land Maximal extension of post-Messinian ingression
N
n
34˚
Leb
ano
Introduction
Be
q a‘ Õaa Z
ah
le
Keywords Jordan-Dead Sea–Red Sea Rift Æ Formation water Æ Brine Æ Paleohydrogeology Æ Water/rock interaction
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Hula G o l a n
GOL AN
Lake Kinneret
Hula
Lake Kinneret
Yiz re
Lisan Peninsula
Beer Sheva
31˚
Negev
Sdom
Se
a
31˚
32˚
Lisan Lake
De a d Se a
Jericho
Smara Lake a
a Zarq W.
Plateau
32˚
el
Bet Shean
Transjordanean
y
Central Jordan Valley
el
Va lle
30˚
A Elat
33˚
Lake Kinneret
re
123
Hula
33˚
Yi z
Brines and evaporites were formed during major phases of the geological history of the Jordan-Dead Sea transform. Between these phases, the rock sequences were flushed of previously formed brines and evaporites and were ‘‘made ready’’ for new generations of liquids (Rosenthal et al. 2006). During the Paleozoic to Early Mesozoic, brines were generated by the dissolution of evaporates (Stoecklin 1968; Husseini and Husseini 1990). Brines were also formed in areas on both sides of the Rift during the Upper Triassic evaporative period, in which thick beds of gypsum (the Mohilla and the Abu Ruweis Formations) were deposited in Israel and in Jordan (Zak 1963; Bandel and Khoury 1981). During the Mio-Pliocene, brines were mainly generated by the post-Messinian ingression of seawater, which dissolved evaporites previously deposited on the dried-up Mediterranean Sea bed and in the erosive channels incised into the adjoining coastal areas (Rosenthal et al. 1999; Hirsch 2005a, b). In addition, brines reacted with the invaded rock sequence and with contemporaneous basic volcanics. In the Rift and
B
30˚
C 0
50
100 Km
Gulf of Aqaba
35˚
36˚
Fig. 1 Schematic map of the geographic outlines of the Sdom Sea (a), Samra Lake (b), Lisan Lake (c) (after Horowitz 2001). The present-day Dead Sea is superimposed
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
and dissolving evaporites and were made ready for the following generations of liquids. The contemporary phase in the Rift is characterized by an ongoing process of flushing out of residual brines and the dissolution of evaporites by currently recharged fresh water. Lake Kinneret, which serves at present as a reservoir of fresh water, is constantly endangered by saline springs emerging along its shores and on the bottom of the Lake. The Cl– concentrations of these waters vary within a wide range of 400–18,000 mg/L. Considering Mg/Ca ratios, two hydrochemical types of saline waters were identified (Goldschmidt et al. 1967; Rosenthal 1988): (a) mostly phreatic and non-thermal waters characterized by Mg/Ca > 1, which occur along the eastern shore of Lake Kinneret and in the north-eastern parts of the Jordan Valley (Newe Ur type), and (b) confined, thermal, and saline groundwaters, with Mg/Ca < 1, occurring mostly along the western margins of the Rift, in the area of Lake Kinneret, in the Jordan Valley, and further south along the western shores of the Dead Sea and the Arava (Devora type). The phreatic Newe Ur waters always overlie the confined Devora type waters. The following major scenarios of brine formation and of salinization of groundwaters were hitherto discussed in literature: •
•
Seawater penetrating from the Mediterranean Sea into the Rift being subsequently confined (Mazor and Mero 1969; Kafri and Arad 1979; Gat et al. 1969; Magaritz and Nadler 1980). The upflow of brines is assumed to be controlled by tectonic processes. Bergelson et al. (1999) outlined a schematic hydrochemical evolution of the saline springs, according to which seawater evaporated to generate an ‘‘original brine,’’ which was suggested to be represented by the thermosaline springs of Tiberias. The groundwaters emerging in the springs of Fuliya and Tabgha were assumed to be dilution products of this brine. According to Kolodny et al. (1999) the water of the saline springs was the result of leaching brine pockets trapped in the sub-surface by freshly recharged pristine waters. Mg-rich brines are relics of the inland sea that existed in the Rift and preceded the Dead Sea (Rosenthal 1988; Starinsky 1974). By subsequent dolomitization, the Mg-chloride composition of these brines turned into Ca-chloride brines and became (in an unexplained manner) confined and pressurized. The Mg-rich brines, emerging from the springs along the western shores of the Dead Sea, were thought to represent residual brines derived from evaporated Pliocene seawater, which was
543
•
•
trapped in the primordial Sdom depression within the Rift. Ca-chloride brines are known from a thick sequence of evaporates in wildcat well Zemah 1 (Simon and Mero 1992). Based on d11B studies, Vengosh et al. (1994b) considered these Rift brines as the main source of the contemporary Dead Sea water. Brines formed in the Rift by dissolution of sulphates, halite, and of K–Mg-rich and Na-poor post-halite evaporites (Flexer et al. 2000) identified in wildcat borehole Zemah 1 located in the immediate south of Lake Kinneret. Further, buried Neogene salt structures are found at Zaharat Qurein (Belitzky and Mimran 1996) and in the sub-surface between Jericho and Fazael (Shulman et al. 2003) in the Lower Jordan Valley. With bittern salts being present (Bender 1974; Raab 1998), ablation of such bodies creates Mg-rich brines such as encountered along the eastern and southern shore of Lake Kinneret. The Ca-rich brines encountered along the western littoral of Lake Kinneret and elsewhere in the Rift could be the result of massive Mg-removal by dolomitization. Klein-BenDavid et al. (2004) suggested a salinization mechanism involving the mixing of two defined end-members with distinct salinities. The brine evaporated slightly beyond the onset of halite being controlled by the rates of inflow and outflow of seawater. In the same basin, more concentrated brine was generated, when loss by evaporation was greater than the inflow of seawater. This brine represents the extreme composition of evaporated seawater well within the field of bischofite.
The chemical evolution of groundwater in different aquifers in Jordan was attributed by Salameh and Bannayan (1993) mostly to water–rock interaction by which precipitation water recharging the upper aquifers changes gradually mostly by interaction with calcareous and argillaceous rocks. The major changes in water quality are due to the dissolution of evaporites. Over-exploitation and depletion of water resources facilitate the salinization of fresh groundwater by these saline waters (Salameh 1996). Exploratory boreholes drilled on the Jordanian side in the Central Jordan Valley into the Plio-Pleistocene lacustrine strata revealed the occurrence of brackish waters (8.4 g Cl–/L) and of brines (38.5 and 48 g Cl–/L), which have the characteristic features of Ca-chloride brines occurring mostly along the western margins of the Rift (Awad et al. 1997). Zak (1963), Belitzky and Mimran (1996), Flexer et al. (2000), Al-Zoubi and ten Brink (2001), and Shulman et al. (2003) described the
123
544
occurrence of salt diapirs in the Dead Sea basin and in the Central Rift Valley and related salinization of groundwater to these salt bodies. According to Gvirtzman et al. (1997), the outflow of saline waters is nowadays controlled by the head and the flow of fresh groundwater and possibly by thermal convective flow. However, the high hydraulic pressures of several hundred bar characterizing deep-seated Ca-chloride brines around Lake Kinneret such as the boreholes Devora 2A, Jordan 1, Rosh Pina-1 (Rosenthal 1988), and Zemah-1 clearly indicate confinement of these brines and contradict the model of Starinsky (1974), who suggested percolation of the evaporated and Ca-enriched brine into rock-formations adjacent to the Rift. During the Pliocene, there were no geological processes that could have generated deep confinement of brines and their pressurization (Rosenthal 1988; Ilani et al. 1988). The divergence between the high pressures and the history of brine emplacement in the Rift (Rosenthal et al. 2006) may indicate the possible existence of several generations of Ca-chloride brines, which were formed and emplaced at different times and under different conditions. The reported divergence of cited opinions emphasizes the essential questions on the origin of the brines, their present distribution in the Rift, and on the processes of their emplacement. Hence, it is of outmost importance to elucidate whether the brines generated in the Rift are (i) primary (i.e. evolved as the result of seawater evaporation), (ii) secondary (resulting from the leaching of evaporate minerals), or tertiary (i.e. altered primary or secondary brines). A further scope of the present paper is to examine the chemical processes generating brines of different compositions in the study area. The presence of these brines endangers the quality of groundwater by salinization. However, there is no reason to assume an uniform process of salinization all over the Rift.
Data base New chemical analyses of fresh and saline groundwaters are complemented by those from Guttman (2000), Salameh (2001), Mo¨ller et al. (2003), Klein-BenDavid et al. (2004), and the unpublished records of the Israel Hydrological Service. In Table 1, only analyses are compiled for which the balance 100(cation - anion)/ (cation + anion) is less than ±7%. Many of the deep wells are wildcat drillings with only poor documentation, but the reliability of the water analyses was critically checked by Bentor (1969), Starinsky (1974), Fleischer et al. (1977), Greitzer (1980), Menashe
123
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
(1991), Fleischer and Varshavsky (2002). In Table 1, the analyses of groundwaters and their hosting lithostratigraphic units are grouped regionally. It is known from previous studies that along the western margins of the Rift the inter-aquifer flow is significant (Vengosh and Rosenthal 1994; Rosenthal et al. 1998; Mo¨ller et al. 2003). Therefore, some of the waters given in Table 1 represent mixtures of waters of various origins, which are now hosted in the denoted lithostratigraphic units. The spread of analytical data from the same locality is due to sampling the wells at different depths, in different years, and different times of the years. Figure 2 shows the locations of springs and wells in the Rift and adjoining areas. The brines are characterized by their chemical composition and by their Cl-excess (Eq. 1) and Q value (Eq. 2), which yield a qualitative information on the presence of Ca–Mg–Cl brine when exceeding 1. Cl-excess ¼ Cl Naþ Kþ
ð1Þ
2 Q ¼ Ca2þ =ðHCO 3 þ SO4 Þ
ð2Þ
All concentrations are in mequiv./L
Chemical characterization of saline groundwaters in the Rift Ca–Mg relationship In the plot of 2(Ca + Mg) vs. Cl-excess, samples align along a 1:1 relationship (Fig. 3a). This ratio corresponds to dominant dissolution of a hypothetical (Ca, Mg)Cl2 component. However, Fig. 3b shows that this is only the case when Cl-excess exceeds 500 mg/L. Below a Cl-excess of 300 mg/L, the (Ca + Mg)/Cl-excess increases inversely with the Cl-excess. In both plots, many brines plot slightly to the Ca + Mg side, indicating an additional source of alkaline earths elements. It is noteworthy that the Ca + Mg content of one brine exceeds that of the Dead Sea. At first sight, most of the saline waters can be described as mixtures in varying proportions of fresh water with NaCl- and (Ca, Mg)Cl2 brines. The cross plots of Ca vs. Mg (Fig. 4a) reveal several types of brines in the study area. Two types of brines with Mg > Ca follow the indicated trend lines, representing mixing of highly saline end-members with fresh water. The relationships between Ca and Mg either scatter widely or plot along a bundle of mixing lines. In limestone, Ca and Mg exchange by dolomitization. Conversion of Ca < Mg into Ca > Mg brines depends
S W W S W W W W W S S S W W S S W S W
Fuliya cluster Fuliya-spring Fuliya-well D 906 Fuliya-well A Fuliya-well B Fuliya-well Kinneret 5 Fuliya-6 Fuliya-well 6/2 Fuliya-well Kinneret 10 B
Tiberias Hot Spring cluster Tiberias, Kikar well Tiberias Hot Springs-Main Spring Tiberias Hot Springs-Main Spring Tiberias Hot Springs-Main Spring Jordan-2 Jordan-1 Tiberias Hot Springs-Roman Spring Tiberias Hot Springs-Roman Spring
Eastern Shore of Lake Kinneret Bet Zaida, well Gofra spring Ha’On beach, well
S
S S S S S S S S W W W W+L
Tabgha cluster Tabgha-Druzi Spring Tabgha-Druzi Spring Tabgha-En Sheva Tabgha-En Sheva Tabgha-Sartan spring (saline) Tabgha-Sartan spring (saline) Tabgha-Sartan spring (fresh) Tabgha-Sartan spring (fresh) Tabgha-well Kinneret 7 Tabgha-well Kinneret 7 Tabgha-well Kinneret 8 Barbutim well
Yarmuq Gorge Northern side Yarmuq Gorge-En Reah
1 1
W W
1
3 1 1
1 1 3 3 2 2 1 3
1 1 1 1 1 3 3 1
1 3 1 3 1 3 1 3 1 3 1 1
2 2
Data source
W W
Sample origin
Lake Kinneret area Rosh Pina 1 Rosh Pina 1 Rosh Pina 1 Kalanit 2 Huqoq well
Location
Cret. Cret. Cret. Cret. Cret.
Tert.
Quat. Quat. Quat.
U. Cret. U. Cret. U. Cret. U. Cret. L. Cret. L. Cret. U. Cret. U. Cret.
U. Cret.
U. U. U. U. U.
U. Cret. U. Cret. Tert. Tert. U. Cret. U. Cret. Tert. Tert. U. Cret. U. Cret. U. Cret. Quat.
Jur. Jur. Jur. U. Cret.
Hosting formations
6.9 6.8 7.1 7.0 6.9
28.8 27.8 23.2 27.8 23
38
6.9
7.6 6.6
6.4
57.5
25.5 24.2
7.3 6.0
8.9
40.8 59
27
7.0 7.2 8.7 7.0 8.0
7.1
23.9
28.4 25 28 28.1 25
7.6
7.2 7.2
pH
18.3
25.5 18.6
T (C)
143
3,580 1,187 5,212
10,540 7,197 6,940 6,800 1,436 6,080 6,335 7,050
374 186 1,111 475 1,869 352 450 8,044
670 579 293 493 1,274 1,420 812 741 139 317 7,296 1,103
33,415 8,004 33,400 17.6 42.5
Na (mg/L)
Table 1 Chemical and isotopic composition of regionally grouped formation waters
12.2
35.3 71.5 442
491 333 325 316 19.9 257 336 326
12.1 4 40 20.3 67.4 11.2 15 306
29.1 21.7 12.1 19.1 51.4 51.8 31.2 29.2 6.6 12.7 260 73.9
1,157 358 1,160 1.8 2.7
K (mg/L)
40
857 213 1,657
1,093 655 662 651 24 377 639 666
65 46 86 81 181 64 73 748
85 70 42 62 149 146 92 85 24 42 466 132
2,918 724 2,920 33 29
Mg (mg/L)
142
724 230 1,078
5,157 3,418 3,400 3,330 40 1,407 3,403 3,480
167 111 129 212 426 151 178 1,856
305 249 170 249 518 501 330 316 123 189 1,742 448
26,946 4,794 27,000 79 77
Ca (mg/L)
309
9,680 2,522 15,800
29,746 18,810 18,300 18,000 1,584 11,413 19,018 18,400
755 432 2,169 987 4,271 720 939 17,213
1,437 1,280 690 1,110 3,000 3,200 1,930 1,770 320 741 15,490 2,662
105,349 22,819 105,000 30 90
Cl (mg/L)
4.2
123 36.4 202
252 240
484 241 238 230
4.49 2.6 15.2 6.3 28.1 3.91 5.71 124
15.3 12.5 7.24 10.6 28.1 26.5 17.6 15.9 2.55 6.93 120 22.3
1,740 0.1 0.2
1,742
Br (mg/L)
115
11 98 60
474 796 812 788 227 2,188 755 808
98 58 4 191 295 93 121 1,606
140 107 56 87 245 259 143 140 29 55 706 203
347 1,161 700 8 13
SO4 (mg/L)
362
185 436 504
40 197 143 140 912 250 135 138
313 312 48 326 35 314 310 21
219 269 259 291 325 304 287 266 265 286 17 346
171 370 171 464 357
HCO3 (mg/L)
20
33 53
35
12 28
21 15 0 22 2
0 13
16
15
13
15
14
13 22
SiO2 (mg/L)
–5.8
–4.3 –2
2.64 –3.2
–1.4
–5 –4.9 –4 –4.9 –4.1
–2.3
–5.9
–5.5
–5.4
–5.9
–3.1
–5.2 –4.8
d18O&
–30
–21 –12
26.6 –18
–5.1
–22 –23 –11 –22 –21
–11
–31
–26
–27
–30
–17
–25 –22
d2H(&)
786
15,010 4,391 24,504
47,997 31,478 30,677 30,115 3,330 21,721 30,773 30,970
1,497 854 3,554 1,994 7,140 1,395 1,782 29,896
2,695 2,319 1,286 2,031 5,278 5,604 3,371 3,097 660 1,364 26,080 4,658
171,873 37,859 171,920 182 275
TDS (mg/L)
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566 545
123
123 1 4 4
S W W
3 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 4 4 1 4 1 4 1 4 4 4
W W W W W W S S S S W W W W W W W W W W
Central Jordan Valley Western side Zemah 1 Devora 1 Devora-2A Bira 2 Newe Ur Bet Yosef En Huga En Amal En Moda Hammam el Malih 1 Beqaot 2 Beqaot 1 20-17/11 Argaman 14 Argaman 29 Fazael 2 Fazael 6 Fazael 8 Jericho 2 (salty) Mitspe Jericho 2
1 4 1
1 1 1
Data source
W W S W W W W W W W
W W S
Southern side Yarmuq Gorge-Bir Mukheiba 8 Yarmuq Gorge-Mukheiba Himma
Eastern side Adasiya N-Shuna Shuna, north 1,010 m JICA well 3 Manshiya 2 Manshiya well Waqqas Waqqas well Wadi Ziglab Teibeh Rajib Abu Zigan Muallaha sp. Bassat el-Faras
S W W
Sample origin
Yarmuq Gorge-En Makla Yarmuq Gorge-Meizar well 2 Yarmuq Gorge-Meizar well 3
Location
Table 1 continued
L. Cret. Quat. Quat.
Quat. U. Cret. U. Cret. Quat. L. Cret. L. Cret. U. Cret. U. Cret. Quat. Quat.
Quat. U. Cret. Tr. Neog. Quat. Neog. Tert. U. Cret. U. Cret. Jur. U. Cret. U. Cret. Cret.–Tert. Cret.–Tert. Cret.-Tert. U. Cret. U. Cret. U. Cret. L. Cret. U. Cret.
U. Cret. U. Cret. U. Cret.
Tert. U. Cret. U. Cret.
Hosting formations
34.8 23.3 28.7
28 23.8 54.8 27.5 48.8 52 49.4 52 24.3
25.2 29.2 31.1 25 28.2 26.3 35.1 31.1 28.9 29.2 30 29.7 25.3 26.1 28.5 26.0 24.3
29.1 44.3 41.5
47.3 60 41.8
T (C)
6.1
6.9
6.8
7.2
7.8 8.0 7.7 7.6 7.1 7.1 7.0 7.3 7.3 7.0 7.2 7.2 7.3 7.3 7.3
7.1
7.0
6.6 6.6 7.1
pH
271 127 84 1,187 29 30 32 41 51 41 102 1,054 4,600 23
79,274 759 40,204 100 258 569 764 498 98 755 85 55 495 156 225 26 57 89 5,560 39
39 43 124
212 178 49
Na (mg/L)
9.44 7.2 4.2 117 1.9 1.6 3.1 3.12 3.9 3.9 8.97 76.9 362 2.5
402 2 14 51.8 24.8 11.8 3.2 44.5 4 3.1 36.7 13.4 16.4 1.9 4.6 6.8 490 2.8
140
3.1 5.3 13.9
18 21.8 4.7
K (mg/L)
58 69 50 158 37 41 37 39 50 41 62 140 332 30
63 146 1,944 31 22 64 169 94 43 103 34 33 171 46 71 26 38 50 641 31
30 46 35
44 35 21
Mg (mg/L)
102 152 70 670 64 76 64 88 88 56 54 546 2,408 42
1,200 216 34,820 37 21 48 256 176 102 380 59 59 202 73 97 57 68 79 1,115 75
91 86 108
178 142 65
Ca (mg/L)
380 181 120 1,928 35 47 44 55 99 78 108 1,318 10,877 51
119,813 1,633 129,540 111 179 665 1,716 1,076 250 1,730 135 103 1,248 304 472 49 154 238 12,199 72
57 62 208
488 317 62
Cl (mg/L)
0.31
4.5
0.10
0.60
2,102 0.4 0.8 4.4 9.4 6.5 1 23.8 1.8 1.2 21.8 4.6 8.4 0.2 3.7 6.6 220 0.4
54
2.3
0.30
6.2 3.7 0.30
Br (mg/L)
229 258 70 1,822 40 46 27 60 74 115 143 1,309 3,479 28
7,978 211 192 36 95 484 303 136 45 343 37 31 313 61 108 14 16 22 868 31
54 78 148
158 278 4
SO4 (mg/L)
217 400 406 1,003 348 372 363 396 393 192 309 1,223 397 248
110 354 427 233 364 260 316 310 361 257 245 267 249 245 259 267 228 230 91 264
357 400 337
340 244 315
HCO3 (mg/L)
18
18
16
20
13
40 32 28 25 20 20 20 20 19 27 19 21 16 15 15
25
16
26 38 23
SiO2 (mg/L)
–6.3
–5.8
–5.7
–5.7
–40
–28
–27
–28
–24
–21 –22 –23 –21 –21 –24 –23 –23
–5.1 –5.2 –4.8 –4.9 –4.8 –5.5 –5.6 –5.6 –5.6
–10 –14 –11 –12 –19 –21
–32
–27
–32 –33 –29
d2H(&)
–3.9 –3.6 –3.2 –3 –4.9 –5.2
–6.1
–5.7
–6 –6.8 –5.9
d18O&
1,050 795 418 5,882 223 242 224 285 367 336 477 4,465 22,058 177
264
208,521 2,965 209,203 357 620 1,914 3,268 2,017 561 3,399 376 306 2,514 677 1,019 190 357 505
291 320 664
1,131 1,013 228
TDS (mg/L)
546 Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
1 3 3 2 3 3 2 2 2 2 1 1 1 2 2 2 2 2
W W W S S S W W W S W W W W W W W W W W W W W
Eastern side Rama JICA well 5 Hisban artesian deep well Hisban deep JICA well 1 1 4 1 4 1
3 3 3 3 3 1 1 3 3 2 1 1 1 1 3 1 1
Data source
S S S S S S S S S W S S S W W S L
Sample origin
Dead Sea area Western side Yam-7C Yam 7N Yam-1 Yam-2 Yam-5 En Feshkha creek En Feshkha spring En Feshcha spring Feshcha Well Radon 1, 60 m Kane A (spring) Kane B (north) Kane C (south) Mitspe Shalem-Mineral beach Mitspe Shalem-Mineral beach Ein Gedi spring Dead Sea, 3128.664¢N; 3524.980¢E; depth: –60 m En Gedi Spa En Gedi-3 En Gedi-2 Yesha Springs En Kedem En Zeruya Massada 1 Massada 1 Massada 1 Zohar Springs En Boqeq, Nirvana Htl. Gofrit well En Boqeq, PB 26 Hammei Zohar 15 Lot-1 Lot-1 Sdom-1 Sdom-2 Sdom Deep-1
Location
Table 1 continued
Paleoz-L.Cret. Quat. L. Cret. Quat. Paleoz.–L.Cret.
Quat. Quat. Quat. Quat. Quat. Quat. Jur. Jur. L. Cret. Quat. Quat. Quat. Quat. Tr. Tr. Quat. Quat. Tr.
Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. Quat. U. Cret.
Hosting formations
6.2
6.4
6.3
5.1 5.8 7.0
27.3 31.1 34
33.5 35.3 32 32 33.7
6.5
7.5 6.2
pH
39.4
#### 25.8 26.9 26.5 43.6 43.6 27.9 25.8
28.2 26.6
T (C)
515 1,328 347 689 582
23,990 25,800 54,500 27,824 27,000 26,900 2,990 8,395 1,357 20,518 24,223 6,830 902 9,890 15,111 23,110 33,200 32,904
1,230 954 462 414 451 446 436 5,130 8,560 12,870 138 178 193 25,501 26,200 47 33,800
Na (mg/L)
63.2 176 46.2 94.0 73.5
0 4,551 304 51 799.5 1,677 30,023 999 4,778
2,380 2,520 4,780 0 4,110 2,870 316 924.3
155 117 45.9 41.4 45.2 45.1 41.4 815 1,350 1,839 17.9 21.4 25.9 3,769 4,020 4.96 8,050
K (mg/L)
86 146 77 83 122
11,705 12,700 21,800 18,685 21,500 14,500 1,081 2,742 106 22,481 34,844 1,229 246 9,185 13,936 21,083 8,510 5,456
782 579 216 193 207 250 215 4,000 6,900 9,480 90 114 130 20,169 21,000 28 46,012
Mg (mg/L)
240 459 197 266 284
9,136 8,980 15,000 10,290 12,500 9,730 1,840 6,360 844 11,389 15,113 203 235 42,840 34,600 85,030 17,300 77,345
410 342 141 139 138 187 182 2,030 3,550 4,954 85 112 124 10,790 11,500 55 20,720
Ca (mg/L)
771 2,272 560 1,015 814
88,161 92,600 176,000 117,663 126,000 102,000 9,017 33,974 1,916 120,035 159,474 13,720 2,048 119,209 131,137 275,110 107,000 216,444
4,840 3,620 1,490 1,310 1,440 1,383 1,415 23,800 40,200 57,092 400 597 697 120,048 123,000 82 222,677
Cl (mg/L)
2.9
1.8
2.5
2,954 4,102 292 24.6 2,468 3,050 3,103 1,790 4,305
1,673 1,770 3,100 2,209 2,770 2,000
98.6 72.2 25.7 22.5 24.9 28.9 28.7 541 886 1,200 12 11.6 14.3 3,148 2,690 1.07 5,700
Br (mg/L)
966 1,264 767 873 1,134
121 150
905 42 319 536 251 349 450
61
163 150 230 142 86 160 61 122 317 130 63 199 188
320 325 270 290 290 262 276 227 298 168 302 300 284 177 100 208 220
HCO3 (mg/L)
466 1,580 400 1,060 1,240 1,730 2,640 2,496 2,436 1,551 2,312 295 395 816 1,008
103 93 59 57 57 76 61 147 225 700 50 66 73 1,061 1,110 34 343
SO4 (mg/L)
14
11
15
4 13 14
16
13 5
15 15 14 12
14 14
SiO2 (mg/L)
–5.7 –5.7 –5.6
–5.3
–1.8 –2
–2.1
–5 –4.2
–2
4.24
–5.6 –5.5 –5.5 0.96
–5.3 –5.4
d18O&
–36 –36 –42
–36
–23
–39
22.3
–23 –24 –23 –3.4
–22 –23
d2H(&)
2,341
2,011 6,180 1,491
137,527 145,950 275,580 177,731 195,120 159,730 17,884 54,891 6,659 178,928 244,622 22,886 3,915 185,207 200,519 437,460 169,704 341,274
809 1,115 1,271 184,498 189,520 266 335,331
7,619 5,778 2,440 2,177 2,363 2,430 2,392 36,463 61,671
TDS (mg/L)
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566 547
123
123 2 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
W
W W W W W W W W W W W W W W W W W W
W W
W S
Eastern side Wadi Araba 1 Wadi Araba 5
Southern Jordan Disi 1 Wadi Yutum Paleoz. Quat.
Jur. Jur.
Quat. Quat. L. Cret. L. Cret. U. Cret. L. Cret. U. Cret. L. Cret. L. Cret. U. Cret. L. Cret. L. Cret. Quat. U. Cret. U. Cret. U. Cret. U. Cret. U. Cret. Quat. Quat.
Paleoz.–L.Cret. Low. Cret. Low. Cret.
Hosting formations
Q values, Cl-excess, and various ionic ratios are given
1 1
1 1
1 1 1
S S S
Zarqa Ma’in Waterfall Zara Princess Basma Farm Ibn Hammat 2 Arava/Araba Valley Western side Arava 1 Arava 1 En Ofarim 5 En Yahav 6 Tzofar 220 Tzofar 20 Tzofar 4a Paran 20 Paran 18 Yaalon 116 Yaalon 6A Yaalon 7 Yaalon 117 Yotvata 13 Yotvata 12 Grofit 4 Samar 4 Samar 2 Eilat 103 Eilat 108
Data source
Sample origin
Location
Table 1 continued
28.5 29.7
29.9 27.3
25 44.3 39.4 35.3 26.5 59.2 36.3 32.6 42.2 41.6 39.6 29.1 29 33.3 31.8 28.3 32 31.8
57 51.7 38.8
T (C)
7.9 7.7
7.2 8.4
6.8 6.7 6.8 7.2 7.1 6.7 6.7 6.7 7.0 7.1 6.9 7.1 6.7 6.9 7.2 6.8 7.2 7.2
6.2 6.2 7.2
pH
26 114
79 783
50,510 50,500 367 313 179 1,176 196 323 200 255 106 168 215 285 476 312 337 681 1,139 431
452 268 79
Na (mg/L)
1.5 2
2.2 9.6
1,030 1,030 19.8 19.8 26.7 64.8 8.7 28 29.3 12.1 6.6 5.7 16.2 10 14.2 15.8 10.9 45.9 22.4 12.8
49.1 31.4 6.1
K (mg/L)
7.0 20
32 111
14,100 12,000 77 100 96 191 80 95 81 151 51 74 120 141 221 110 106 248 312 160
34 21 18
Mg (mg/L)
34 79
129 676
50,600 50,600 193 217 177 895 212 286 208 335 81 100 211 203 326 272 175 619 679 233
177 122 53
Ca (mg/L)
34 194
202 2,723
202,500 202,000 663 607 312 3,636 421 748 408 746 257 340 368 479 873 765 631 1,346 3,151 967
727 494 106
Cl (mg/L)
0.1 1.5
2.4 43.6
3,674 3,670 1.4 1.9 0.7 35.6 0.9 3.8 0.9 1.5 0.8 1 1.1 1.5 2.1 3.9 3.3 9.4 30.2 7.7
3.2 2.4 0.5
Br (mg/L)
25 93
125 66
144 122 454 545 570 729 448 609 627 653 165 295 656 738 1,113 574 678 1,832 967 636
258 155 65
SO4 (mg/L)
97 109
235 8
61 61 243 265 364 14 204 207 307 322 234 264 264 213 241 220 131 195 66 140
371 212 206
HCO3 (mg/L)
14 21
12 7
13 17 18 21 17 25 22 15 17 17 17 19 20 14 19 15 18 20
23 20 13
SiO2 (mg/L)
–6.5 –5.8
–6.2 –7
–7 –7.2 –6.8 –8.1 –4.5 –8 –7.5 –5.9 –5.9 –5.9 –6.9 –6.5 –6.4 –7.6 –6.8 –6.7 –7.2 –6.8
2.76
–4.1 –4.6 –7.2
d18O&
–40 –31
–33 –35
–41 –46 –42 –55 –23 –53 –50 –35 –28 –29 –40 –41 –38 –47 –39 –46 –46 –42
–32 –37 –42
d2H(&)
142 525
584 4,419
322,558 319,922 1,789 1,820 1,379 6,749 1,383 2,117 1,575 2,169 684 1,001 1,604 1,876 3,046 2,067 1,960 4,796 6,319 2,467
1,722 1,115 341
TDS (mg/L)
548 Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
A
Fuliya
R. Jordan
88
24 27-28 32 25-26 29-31
67
89
84 85
63-64 22-23
65 66
106 107
61-62
18-21Tiberias
A
83
90-92
93
Yarmuq R.
94 68
Borde rline
11-17
10
Jordan R .
ITER
4-9 Tabgh a
3
87
80-82
Dead Sea
2
Lake Kinner et
SYRIA
ANE SEA AN
LE BA NO
N
1
33˚
MED
D
C Jord an R .
36˚
35˚
E
549
69-70 71
B 32˚
B
77-79 50 34-35 52 51 36 53 54 55 38 37 39
C
99-101
102-103
56
31˚
D
95-98
72-75 76
33
ISRAEL
86
40
104 105
43-45
46-47 48
JORDAN
Jordan R.
41-42
108 109
57 58
59
30˚ 60
EGYPT
49
RED SEA N Fig. 2 Sampling map. Figures refer to identification numbers in Table 1
either on the progress of dolomitization of limestones or on the albitization of basaltic rocks (Table 2). Albitization and dissolution of sulphates and carbonates increase the (Ca + Mg)/Cl-excess ratios. Figure 4b shows that the Mg-rich brines follow strict trends, whereas Ca-dominated brines scatter widely. Mg/Ca vs. Na/Cl The Mg/Ca vs. Na/Cl molar ratios yields a complex distribution of data points (Fig. 5). Modern seawater plots outside the displayed area. The Na/Cl ratio of seawater is shown by the broken line. Left to this line, the trend of Mg/Ca ratios increasing with decreasing Na/Cl ratios represents the evolution of brines produced by evaporation. Increasing Mg/Ca ratios correlate with younger brines. This trend largely represents the development of the brines of the Lisan Lake period with samples from the Dead Sea plotting at the outmost point. Another trend in Fig. 5 is more diffuse and is directed towards both low Mg/Ca and Na/Cl ratios,
finally meeting many of the deep brines. The low Mg/Ca ratios are explicable when brines react with limestone. The Mg–Ca exchange process such as dolomitization shifts any saline water towards low Mg/ Ca ratios. Concomitant precipitation of calcite and of gypsum buffers this process to some degree. Due to active Plio-Miocene volcanism, Na is consumed by the albitization of the anorthite component of plagioclase (reaction b in Table 2) and, when including pyroxenes, by spilitization (reaction c and d in Table 2). ‘‘Brucite’’ is precipitated as part of chlorite (reactions c–e in Table 2), which lowers the Mg/Ca ratios vertically downwards in Fig. 5. Adularization (reaction e in Table 2) increases the Na/Cl ratio. All these reactions take place under sub-surface conditions. Oxidation of sulphides supports precipitation of gypsum, and increases the Mg/Ca ratio (reaction f in Table 2). The trends of Mg/Ca brines in Fig. 5 cannot be explained by the dissolution of carnallite (KMgCl3Æ 6H2O), bischofite (MgCl2Æ 6H2O), kainite (MgSO4Æ KClÆ 3H2O), and kieserite (MgSO4Æ H2O), because these salts yield sub-vertical to vertical upward shifts of
123
550
a
6000 Lake Kinneret area Yarmuq Gorge
5000
Central Jordan Valley
2(Ca+Mg) mmol/L
Fig. 3 Cross plots of 2(Ca + Mg) and (Ca + Mg)/ Cl-excess vs. Cl-excess. Both plots visualize that at Clexcess exceeding 500 mg/L the saline waters consist largely of dissolved halite and a hypothetical (Ca, Mg)Cl2 compound. Symbols with a dot inside denote deep brines
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Dead Sea area
4000
Arava/Araba Valley Dead Sea, northern basin 3000
2000
1000
0 0
1000
2000
3000
4000
5000
6000
Cl excess mmol/L
b
2 Lake Kinneret area
(Ca+Mg)/Cl excess, molar ratio
Yarmuq Gorge Central Jordan Valley Dead Sea area Araba/Arava Valley
2
Dead Sea, northern basin
1
1 0
1000
2000
3000
4000
5000
6000
7000
Cl excess mmol/L
sample points. All waters with Na/Cl > 1 result from the weathering of basalts. Br/Cl vs. Na/Cl The data set in Table 1 yields a diffuse trend of increasing Br/Cl ratio with decreasing Na/Cl ratios (Fig. 6), contrasting the sharp Dead Sea Rift trend line as defined by Klein-BenDavid et al. (2004). Evaporation beyond the beginning of halite precipitation decreases Na/Cl but increases Br/Cl. The evaporation of seawater does not yield a straight line because of the precipitation of diverse salts (Fontes and Matray 1993). Both trends border the data cloud towards the left. Few of the deep brines plot near the seawater evaporation curve, only Sdom 1 plots below. The distribution of samples suggests that various processes are involved
123
in generating these brines. The Dead Sea and Sdom brines are characterized by high Br/Cl ratios and represent residual brines generated at different times. The Zemah brine with very low 1,000 Br/Cl molar ratio of 0.2 and Na/Cl of about 1 clearly indicates dominant dissolution of halite. Any horizontal spread of samples is only explicable by increase of Na not being associated with Cl. This could be NaHCO3 and/or Na2SO4 possibly resulting from the oxidative weathering of basaltic rocks. Albitization shifts the Na/Cl line horizontally to the left. Dissolution of MgCl2 salts such as bischofite (MgCl2Æ 6H2O) and carnallite (KMgCl3Æ 6H2O) decreases Na/Cl and Br/Cl ratios. Facit: The Ca–Mg relationship strongly points to overall dolomitization and local albitization. Mg/Ca ratios reveal two trends by which saline waters develop: increase of Mg/Ca ratio by evaporation and
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566 Fig. 4 Cross plots of Mg vs. Ca and Mg vs. Cl-excess. Symbols with a dot inside denote deep brines. Bold trend line indicate Mg > Ca saline waters. Symbols with a dot inside denote deep brines. The factor of 2 is introduced because Mg and Ca are actually given in mequiv./L
551
a
4000 Lake Kinneret area
3500
Yarmuq Gorge Central Jordan Valley
3000
2Mg mmol/L
Dead Sea area Arava/Araba Valley
2500
Dead Sea, northern basin
1:1 1:1
2000 Ca
Ca>Mg
1500 1000 500 0 0
500
1000
1500
2000
2500
3000
3500
4000
4500
2Ca mmol/L
b
4000 3500
Lake Kinneret area Yarmuq Gorge Central Jordan Valley Dead Sea area Arava/Araba Valley Dead Sea, northern basin
2Mg mmol/L
3000 2500 2000 1500 1000 500 0 0
1000
2000
3000
4000
5000
6000
7000
Cl excess mmol/L
Table 2 Compilation of alteration reactions being considered in this study
Alteration reactions a
Dolomitization of limestones 2CaCO3 þ Mg2þ , CaMgðCO3 Þ2 þ Ca2þ
b
Albitization of mafic rocks CaAl2 Si2 O8 þ 2Naþ þ 4H4 SiO4 , 2NaAlSi3 O8 þ Ca2þ þ 8H2 O
c
Chloritization of Neogene gabbroic intrusions 2KðMgÞ3 ½AlSi3 O10 ðOHÞ2 þ Mg2 Si2 O6 þ 2Mg2þ þ 2CaAl2 Si2 O8 þ 6H2 O þ 2Naþ ,
calcite
dolomite
anorthite
albite
anorthite
pyroxene
biotite
2ðMg2 AlÞ½AlSi3 O10 ðOHÞ2 Mg3 ðOHÞ6 þ 2NaAlSi3 O8 þ 2Ca2þ þ 2Kþ albite
clinochlore
d
or CaAl2 Si2 O8 þ 2Mg2 Si2 O6 þ 10H2 O þ Mg anorthite
2þ
,
pyroxene
ðMg2 AlÞ½AlSi3 O10 ðOHÞ2 Mg3 ðOHÞ6 þCa2þ þ 3H4 SiO4 clinochlore
e
Adularization of albite NaAlSi3 O8 4þKþ , NaAlSi3 O8 þNaþ
f
Sulphate reduction by organic matter 2 SO2 4 þ CH4 , CO3 þ H2 S þ H2 O
albite
adularia
123
552 4
Lake Kinneret area Yarmuq Gorge Central Jordan Valley 3
Dead Sea area
Ha'On
Arava/Araba Valley
Mg/Ca mol/mol
Fig. 5 Cross plots of Mg/Ca ratio vs. Na/Cl ratio. Broke line denotes the Na/Cl of modern seawater. The solid line represents the Dead Sea Rift trend line (adapted from Klein-BenDavid et al. 2004), which is identical with the estimated ‘‘optimal fit’’ in Fig. 18a. The insert shows the probable changes due to mineral/brine interactions. Symbols with a dot inside denote deep brines
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Dead Sea, northern basin 2 +MgCl2
dedolomitisation
adularisation
1 + (Ca,Mg)Cl2 albitisation
dolomitisation
0 0.0
0.5
1.0
1.5
2.0
2.5
Na/Cl mol/mol
15
1000Br/Cl mol/mol
Fig. 6 Cross plot of 1,000 Br/ Cl vs. Na/Cl. The insert shows the probable changes due to mineral /brine reactions. Symbols with a dot inside denote deep brines. Dashed line represents the Dead Sea Rift trend line (adapted from Klein-BenDavid et al. 2004)
Lake Kinneret area Yarmuq Gorge Central Jordan Valley Dead Sea Area Arava/Araba Valley Seawater evaporation Modern seawater Dead Sea Rift trend line seawater ratio Dead Sea
10
+Br
Ha'On 5
albitization adularization
Sdom 1
+(Mg,Ca)Cl 2
0 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
+ halite
1.8
2.0
Na/Cl mol/mol
decreasing Mg/Ca ratios due to dolomitization and albitization. Br/Cl vs. Na/Cl ratios demonstrate that albitization does not play a major role which leaves dolomitization to be the main source for decreasing Mg/Ca ratios in saline waters.
Identification of saline water bodies As outlined in Figs. 3, 4, and 5, several processes may be implied in the salinization of groundwater in the Rift. The presence of diversely composed saline waters is not given by the chemical distinctions between Ca > Mg and Ca < Mg brines but by the individual plots of Ca + Mg vs. Na + K and Ca + Mg vs. Cl-excess.
123
The origin of saline waters in the Rift is discussed regionally. The Region of Lake Kinneret Close to the water line of Lake Kinneret, four large clusters of springs are discerned (Fig. 2). Along the western shore, three such clusters are positioned on or close to the western faults of the Rift and mostly with diagonal, NW–SE directed intersections. The Tiberias Hot Springs(Hammei Tveria) emerge along a zone of step-faults that cut through the Upper Cretaceous limestone and dolomite beds. Similar thermosaline waters emerge from the bottom of the Lake as sub-marine springs with temperatures of up to
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
a
180 Tiberias Kikar well
160
Tiberias Hot Springs Fuliya cluster
140
Tabgha cluster
Ca+Mg mmol/L
Fig. 7 Lake Kinneret. a Cross plots of Ca + Mg vs. Na in waters from the four spring clusters at Lake Kinneret. Note that the eastern and Tiberias clusters plot along the same line. b Ca vs. Mg plot shows distinct trends for all four clusters along Lake Tiberias
553
120
Eastern cluster
100 80 60 40 20 0 0
50
100
150
200
250
300
350
400
450
500
1600
1800
Na mmol/L
b
6000
5000
Ca mg/L
4000 Tiberias cluster 3000 Fuliya cluster
2000 Tabgha cluster
Eastern cluster
1000
0 0
200
400
600
800
1000
1200
1400
Mg mg/L
60C. The gross chemical composition of these highchlorinity waters (up to 18 g/L) originates from a slightly altered limestone by which they are also naturally recharged (Mo¨ller et al. 2003). In Fig. 7a, the saline waters from the Tiberias Hot Springs and the eastern cluster (Ha’On, Gofra) plot along a common line with a molar ratio of (Ca + Mg)/(Na + K) = 0.4, but their Ca vs. Mg trends are different (Fig. 7b). The hot waters from Tiberias (Fig. 7b) show an intercept on the ordinate. This quantity of Ca is independent of the Ca–Mg exchange and is explicable by the albitization of basalts that occur in great thickness south of Tiberias and along the adjacent faults (Shaliv et al. 1991). High d18O and d2H values (Fig. 8) vary over a wide range. The water from well Kikar represents a least diluted brine. The tritium content of 2.8 TU compared to 5.6 TU in the lake indicates a high fraction of recent precipitation. The Fuliya springs (Fig. 2) emerge from a tilted fault-block built of Upper Cretaceous carbonates. The gross chemical composition indicates that the
thermosaline waters originate from limestones. Figure 7a suggests that the Fuliya cluster has at least two saline sources. Kinneret 10b is the most saline water occurring in this group with the lowest ratio of (Ca + Mg)/(Na + K) = 0.2. On the other hand, the lowsalinity waters plotting together with Tabgha waters are characterized by (Ca + Mg)/(Na + K) = 0.3. The salinity of the emergent Fuliya springs is lower (<17 g/L Cl) than that of the Tiberias springs and so are their temperatures (<28C). Sub-marine seepages also emerge around the eastern margins of this faultblock. Irrespective of mixing, the Ca/Mg ratio is uniform (Fig. 7b), i.e. it is locally controlled. Measurable tritium indicates young ages (Table 1) and stable isotope ratios of H2O cover a wide range. Springs near Tabgha produce the largest quantities of saline water. The Nur and Sheva springs flow out from fault-blocks built of Upper Cretaceous and Tertiary limestone beds. The gross chemical composition indicates that the limestone is slightly hydrothermally altered. The recharge of these springs occurs
123
554
over Tertiary (Eocene) chalks (Mo¨ller et al. 2003). Further to the west, the water from wells Kalanit 2 and Huqoq (also replenished from Upper Cretaceous limestone) shows low chlorinities (<100 mg/L), whereas the Tabgha wells and springs range up to 16 g/L. Kinneret 8 yields the least diluted brine of this cluster. The Ca vs. Mg trend line of Tabgha waters (Fig. 7b) is sub-parallel to that of the Tiberias water, indicating that the Ca–Mg exchange is regionally comparable but Tiberias is affected by additional albitization. Compositionally, Kinneret 8 belongs to the Fuliya cluster because it is affected by a Na-rich but Ca-poor brine like most of the waters of the Fuliya group (Fig. 7a). The uniformity of Ca/Mg ratios of the Tabgha cluster includes Kinneret 8, but different Mg–Ca trends of the Tabgha and Fuliya cluster in Fig. 7a proves local dolomitization. Stable isotopes cluster in Fig. 8 indicate that the fractions of mother brines mixed with fresh water are low. Along the eastern and southern shores of the Lake, saline waters in boreholes and in springs reach Cl concentrations up to 20 g/L. Their chemical compositions differ from that of waters encountered along the western shore of the Lake mainly by Mg/Ca ratios >1 (Fig. 7b). These saline waters emerge mostly from clastic beds of the Neogene age. The water from Gofra is recharged over Cretaceous limestone, but Ha’On gains its water mainly from the basalts of the Golan Heights (Mo¨ller et al. 2003). The chemical development of the Mg-dominated saline water of Ha’On type does not truly resemble the composition expected for evaporating seawater (Figs. 5, 6). It was generated by drying up the Sdom Sea, a mixture of seawater and drainage water. During lowering of the water level,
Fig. 8 d18O vs. d2H cross plot showing slightly different trends for the water from the Tabgha and Fuliya clusters compared with waters from Tiberias, eastern clusters, and the Yarmuq Gorge. EMWL eastern meteoric water line (adapted from Gat and Carmi 1970). Precision of measurements is ±0.1 and ±0.8% for d18O and d2H, respectively
123
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
brine pockets were trapped in the sub-surface and are now leached by pristine waters (Kolodny et al. 1999). With 2(Ca + Mg) equalling Cl-excess (Fig. 3a), all saline waters from the region may carry fractions of an original brine mixed with locally different fractions of a Na brine of the Zemah type with (Ca + Mg)/Na ~ 0.01 (Fig. 7) and regionally varying Mg–Ca exchange by dolomitization and Na–Ca exchange by albitization (Fig. 7b). Thus, the most saline groundwater of the Tabgha-, Fuliya-, and Tiberias clusters could be regional derivatives of at least two mother brines: in diluted form one is represented by Ha’On water, the other is a Na-rich brine of the Zemah type. Additionally, a deep-seated Ca-dominant brine may ascend along the fractures on the western side of Lake Kinneret, which is absent on the eastern side. This Ca > Mg brine is not explicable as an ablation brine from the Zemah evaporite body as suggested by Simon and Mero (1992), which would have to be low in Ca because of the high sulphate content, and any sulphate reduction would have caused calcite precipitation. Flexer et al. (2000) suggested that Ca-chloride thermosaline waters could be dilution products of either deep-seated, ancient Ca-chloride brines, or the ‘‘western’’ Ca-chloride brine could have geochemically evolved from the ‘‘eastern’’ Mg-rich brine by water/ rock interaction. In the nearby gorge of the Yarmuq River, four springs of the Hammat Gader group emerge, which differ by their temperatures (25–50C) and by Cl– contents of 80–530 mg/L. All springs flow out of Eocene chalks outcropping along the gorge and on the northern slopes of the Ajlun. As a rule, in these springs salinity increases in direct relation to temperature. The
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
555
7 Northern side
6
Southern side
Ca+Mg mmol/L
5 4
Mukheiba
3 2 1 0
0
2
4
6
8
10
Na mmol/L
Fig. 9 Cross plots of elements in groundwaters from the Yarmuq Gorge indicating mixing
water from wells Meizar 2 and 3 are chemically controlled by altered limestones (Mo¨ller et al. 2003). Unlike the waters of En Makla and En Reah, the low d18O and d2H values of the Meizar waters indicate recharge at high topographic elevations of the Golan Heights (Arad and Bein 1986). On the Jordanian side of the gorge, the groundwater flows westward in the Upper Cretaceous aquifer. The marly aquicludal layers overlying this aquifer create confining conditions with piezometric heads tens of metres above ground level (E. Salameh, personal communication). These waters are exploited by the artesian wells of Mukheiba and Wadi el Arab. The wells at Mukheiba are extracting water from the Upper Cretaceous limestone, which is recharged over basaltic areas extending to the north and east. At Himma, the water is extracted from the altered Upper Cretaceous limestone, but it is recharged over unaltered Upper Cretaceous limestones. On the Jordanian side, the confined parts of the aquifer show enhanced temperatures and pressures and correlated dissolution of minerals and water salinization (Salameh 2001). The few analysed samples from both sides of the Yarmuq Gorge show similar trends in cross plots of Ca + Mg vs. Na (Fig. 9). The data from Mukheiba always plots aside the general trend, which is explicable because the recharge occurs over basalts, whereas the other groundwaters are recharged over limestone areas. The Lower Jordan Valley The Lower Jordan Valley is the regional recipient and mixing zone for groundwater originating from adjacent calcareous and basaltic aquifers. It hosts two rhombic
Fig. 10 d18O vs. d2H cross plot showing the grouping of groundwaters from the Lower Jordan Valley. EMWL eastern meteoric water line (adapted from Gat and Carmi 1970)
pull-apart basins—the Bet Shean Valley and the Fazael depression. Because of structural conditions specific to the western part of the Valley, the Judea-, Avedat-, and two basaltic groundwater bodies drain to the Valley through interconnected outlets, which are common to these aquifers (Rosenthal 1988). Upon reaching the interconnecting zones, created by the marginal faults of the Rift, fresh groundwater deriving from different aquifers are mixed. West of the Jordan River, most groundwaters are hosted in the Upper Cretaceous aquifer limestone, and recharge occurs over the Upper Cretaceous and Eocene limestone areas. The groundwater at En Huga is recharged over a basaltic area (Mo¨ller et al. 2003). The groundwaters from Neogene and Quaternary aquifers (Bira, Newe Ur, and Bet Yosef wells) are chemically controlled by leaching of basalts. Well Bira 2 exploits water from a basaltic aquifer, which is recharged from Eocene limestone by lateral flow (Mo¨ller et al. 2003). East of the Jordan River, groundwater flows from the foothills westwards towards the River through alluvial fans and other young clastic sediments interfingering with the beds of the Lisan Formation, which in these areas contain gypsum and halite. Recharge takes place through lateral flow from aquifers extending east of the foothills. Deep groundwater in the Lower Cretaceous aquifer is saline and under artesian conditions. It seeps upwards through young sediments and forms saline springs. The salinity of these springs reaches up to 1,000 mg/L (Salameh 1996). Locally, alluvial fans contain water of excellent quality. However, the extensions of these fans towards the River contain increasingly brackish and salty waters, which
123
556
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Fig. 11 Cross plots of Na vs. d18O in groundwaters from the Lower Jordan Valley. With increasing d18O values salinities decrease in Jordan but increase in the western part of the Rift
a
30 Western Jordan Valley
Ca+Mg mmol/L
Bet Shean area Eastern Jordan Valley 20
10
0 0
20
40
60
Na+K mmol/L
b
30 Western Lower Jordan Valley Bet Shean area
Ca mmol/L
Eastern Lower Jordan Valley 20
originate from irrigation return flow and mix with salty formation waters. The deep well of North Shuna produces water from an Upper Cretaceous limestone aquifer, but REE indicate dominant water from a sandstone environment, which indicates ascending flow from Lower Cretaceous beds (Mo¨ller et al. 2006). At Manshiya, major ion chemistry and rare earth element patterns characterize the water as originating from the Upper Cretaceous, but it is drawn from Lower Cretaceous sandstones. Here, young water descends into deeper rock horizons. At Abu Zigan, the spring water originates from the Lower Cretaceous sandstone. In the d18O vs. d2H cross plot (Fig. 10), the data from the western side of the Jordan River split into two groups. The upper group denotes waters from basaltic aquifers in the Bet Shean area, whereas the lower one comprises waters from the limestones of the Judea Group. The isotopic difference between the two is caused mainly by an altitude effect and both groups strongly differ in Na vs. d18O (Fig. 11). The waters from Beqaot and Argaman are isotopically different from Fazael and Jericho well waters due to the different altitude of the recharge areas. The Jordanian waters d2H and d18O show the lowest isotope ratios (Fig. 10). In Fig. 11, d18O shows an opposite trend with Na and Br compared to water from the western side, indicating sweetening by fresh water. The correlations of Ca + Mg vs. Na and Ca vs. Mg are similar with those in the west (Fig. 12) and are characterized by (Ca + Mg)/(Na + K) molar ratios of about 0.3. Only waters from the basaltic areas (Bet Shean) show significantly lower ratios. Considering the prevalently clastic character of the beds filling the centre of the Valley, hydraulic contact between east and west of the River Jordan cannot be excluded a priori. However, the chemical compositions of water from lithologically comparable aquifers differ on both sides of the Jordan River (Table 1). This denies significant exchange of water beneath the river bed of the Jordan River. The surroundings of the Dead Sea
10
1:1
0 0
2
4
6
8
10
Mg mmol/L
Fig. 12 Lower Jordan Valley. Cross plots of earth alkaline vs. alkaline elements (a) and Ca vs. Mg (b). There is no significant difference between groundwaters from both sides of the Jordan River
123
In areas west of the Dead Sea mostly calcareous rock formations of Upper Cretaceous age outcrop. The Neogene salt diapir of Mt Sdom protrudes along the south-western margins of the Dead Sea. Other buried salt structures are identified further northwards along the western shores of the Sea and also beneath the Lisan Peninsula (Bender 1974; Shulman et al. 2003). The central part of the Rift contains thousands of metres of clastic fill covered by the Pleistocene Lisan Formation, the thickness of which does not exceed
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Fig. 13 Cross plots of d18O vs. d2H for saline waters from the surroundings of the Dead Sea area showing a diffuse trend. Precision of measurements is ±0.1 and ±0.8% for d18O and d2H, respectively
100 m. The Dead Sea is imbedded into this sequence and its bottom is covered by very thick layers of rocksalt (Neev and Emery 1966). From the west three regional aquifers drain to the Dead Sea: •
•
•
The Lower Cretaceous Kurnub Group sandstones, which prevalently host paleowater (Rosenthal et al. 1998). Along their flow paths towards the Dead Sea, these waters become progressively saline due to mixing with brines in the fault zones outlining the Rift (Arad 1964). Rosenthal et al. (1991) found that water from the Kurnub Group flows laterally into the down-faulted blocks made up of Judea Group beds. Rosenthal et al. (1998) discussed the various geochemical processes that shaped the chemical composition of the Kurnub Group water. Upper Cretaceous limestone and dolomite of the Judea Group aquifer that hosts groundwater mainly recharged by flash floods over the eastern and north-eastern Negev. The Arava Graben Fill aquifer consists of coarse gravel deposited as terraces or as alluvial fans and hosts water mainly recharged by flash floods in the wadis (Yechieli et al. 1992). Springs and seepages emerge along the Dead Sea shore. The high salinity is related mainly to mixing with remnants of the Dead Sea when it was at higher stands (Rosenthal et al. 1998).
Along the western shore of the Dead Sea, water is drawn from Quaternary clastic sediments, which are predominantly composed of Upper Cretaceous
557
limestone and dolomite. Only the fresh waters of the En Gedi and En Boqeq springs discharge directly from the Upper Cretaceous limestone aquifer, which is recharged by current rainfall over the eastern flanks of the Judea Mts. Close to the fault zone, these waters mix with brines circulating mainly in the fault zone and with the remnants of Dead Sea brines from the times it was at higher stands (Yechieli et al. 1992). The waters of En Gedi 2 and of En Boqeq Gofrit well are high in d18O and show limestone signatures (Mo¨ller et al. 2003). Contrarily, the water from well Zohar 5 is extremely low in d2H (–42&), indicating a contribution of Pleistocene water from Kurnub sandstones. The thermal waters from the western shoreline plot along a mixing line of a very saline component like the Dead Sea brine and Kurnub–Judea water (Fig. 13) represented here by the intersection of the two lines (Mo¨ller et al. 2003). Na + K (Fig. 14a) disprove simple mixing between only two end-members. The En Gedi 2–fresh water trend shows the lowest (Ca + Mg)/(Na + K) molar ratio of 0.5. All brines below the Dead Sea–fresh water line are explicable by assuming ablation of halite, which is proved by enhanced Na (Fig. 15a), which is not associated with corresponding increase in Br (Fig. 15b). Most of the saline waters are highly enriched in Ca (Fig. 14b) and are plotting away from the Dead Sea brine–fresh water mixing line. The saline waters plotting along the line departing from En Boqeq (Gofrit well) represent a mixing line with Ca-rich, low-salinity fluid. Ca vs. SO4 shows several trends indicating variable ablation of gypsum in the sub-surface (Fig. 16). The lowest limit is given by Ca/SO4 molar ratio of 1, which is met by Massada 1 samples. Note that the water of the En Boqeq (Gofrit well) has a much higher SO4 concentration (2,200 mg/L) than the Dead Sea (120 mg/L). The Feshkha trend points to En Gedi 2 as a possible end-member. The high-salinity waters in Fig. 16 represent late Lisan- or proto-Dead Sea brines. All brines seem to be a complex mixture of contributions from different sources and local dissolution of halite and gypsum. The springs of En Feshkha and Kane emerge at the north-western end of the Dead Sea, at the fault contact between Upper Cretaceous limestones and the fill of the Rift. d2H of –23& is typical for water from the Judea Group (Fig. 13). These non-thermal waters are mixing products between fresh water draining from the west and north-west and probably of Dead Sea water when it was at higher stands. Along the western shore of the Dead Sea, there are also three major clusters of thermosaline springs emerging at Hammei Zohar, Hammei Yesha, and
123
558
a 3000
Sdom 1
2500
Ca+Mg mmol/l
Fig. 14 Dead Sea Area. a Cross plots of Ca + Mg vs. Na + K. Most of the brines plot below the Dead Sea– fresh water line indicating considerable ablation of halite. b Cross plots of Ca vs. Mg. Dead Sea saline waters plot in the Mg > Ca field, whereas the deep brines assemble in the Ca > Mg field. Many saline waters align not along a trend with fresh water as an end-member
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
Dead Sea
2000
Sdom Deep 1
En Boqeq Lot 1
1500
Zohar Springs
1000
En Gedi 2 Sdom 2
500
0 0
500
1000
1500
2000
2500
3000
Na+K mmol/L
b
90000 Sdom 1
80000 Sdom Deep 1
70000
Ca mg/L
60000 50000 Lot 1
40000 30000
Ca>Mg
Mg>Ca
Lot 1
Dead Sea
En Gedi Spa
20000
En Boqeq
10000 0 0
10000
20000
30000
Mg mg/L
Fig. 15 Cross plots of Na and Br vs. d18O for waters from the Dead Sea area
123
40000
50000
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
a
600
b
Dead Sea
Sdom 1
2000
Sdom Deep 1 Sdom 2
400
En Boqeq
En Gedi
Ca mmol/L
Ca mmol/L
Fig. 16 Cross plot of Ca vs. SO4 for groundwaters from the Dead Sea area
559
En Mitspe En Gedi
200
Yesha Springs
Zohar Springs En Zeruya En Gedi
Lot 1
1000
Lot 1 Dead Sea
En Feshkha
Sdom 2
Massada
0
0 0
5
10
15
20
25
0
30
5
10
Hammei Shalem. The temperatures of these waters reach 42C and their Cl– content ranges from 30 to 130 g/L. The waters from Hammei Zohar (Gofrit well), Hammei Yesha, En Gedi 2 well, and Mitspe Shalem seem to represent mixtures of recent Dead Sea water with fresh water from adjoining aquifers in the west and from flash floods in the wadis flowing to the Dead Sea. Bentor (1969), Arad (1964), and Mazor and Mero (1969), however, suggested that these saline waters are the result of mixing fresh water derived from adjacent aquifers with upflowing brines. The Dead Sea water does not plot on the Ca + Mg vs. Cl-excess line (Fig. 3a), which indicates the presence of (Ca, Mg)Cl2 brine. Since this line confirms distinct residual brine from the Sdom Sea period, the Dead Sea water must be the result of a different evolution path, i.e. the evaporation of the Lisan Lake—a continental lake nourished by drainage water and rain flushing residual brines and dissolving evaporites from the Sdom Sea period. Seawater infiltration as suggested by Mazor and Mero (1969) can neither be proved nor is it necessary.
a
20
25
30
Starinsky (1974) suggested that the salts dissolved in the Dead Sea were derived from the springs discharging during geological times. Considering the variations of B contents and of its isotopes, Vengosh et al. (1991) confirmed Starinsky’s view and excluded the possibility that the waters of the thermosaline springs represent diluted Dead Sea water, which is in contrast to our results. East of the Dead Sea, groundwater occurs in two different aquifer complexes, i.e. in the upper limestone complex and in the lower sandstone complex. Along the eastern side, Palaeozoic to Lower Cretaceous sandstones are the main aquifers. The upper limestone aquifer receives replenishment by infiltrating precipitation. Large volumes of groundwater discharge through springs along the upper reaches of the wadis of Zerqa Ma’in, Walla, Mujib, Karak, Shaqiq, and Ibn Hamad. The lower aquifer (sandstone complex) does not receive any appreciable amounts of direct recharge by precipitation. Hisban deep well water represents water from the Upper Cretaceous aquifer infiltrating the
1000
b
35 Eilat 103, Tzofar 20
900
30
Tzofar 20
800
Ca+Mg mmol/L
W. Araba 5
700
Ca mg/L
15
SO4 mmol/L
SO4 mmol/L
Fig. 17 Cross plots of elements in groundwaters from the Arava/Araba Valley. Tzofar 20 and Wadi Araba 5 but also Eilat 103 behaves different from the indicated trend lines
En Boqeq
Elat 103
600 500 400 300
Wadi Araba 5
20 15 10
Western side 200
25
Eastern side
Western side Eastern side
5
100 0
0 0
100
200
Mg mg/L
300
400
0
10
20
30
40
50
60
Na mmol/L
123
560
Triassic/Jurassic sandstones, where the gross chemical composition is achieved. The Arava Valley Structurally, the Arava Valley is a graben bordered to the east by the Mountains of Edom built of Precambrian igneous rocks overlain by sandstones, and to the west by calcareous beds of the Judea Group overlying the sandstones of the Kurnub Group. The Arava Valley constitutes a drainage base for the large wadis that drain eastern Sinai, the Negev, and the Mountains of Edom. On both sides of the Valley, a dense swarm of faults created fault–blocks, which are buried under very thick (thousands of metres) sequences of clastics of the Neogene Hazeva Formation overlain by Quaternary to Recent alluvium. In the western Arava Valley, the Lower Cretaceous aquifer contains water recharged over sandstone areas (Mo¨ller et al. 2003). This is typical for En Ofarim, En Yahav 6, Paran 20 Yaalon 6A and 7, and Grofit 4. The water from well Paran 18 derived from the Lower Cretaceous sandstones is recharged over Upper Cretaceous limestones. Well Tzofar 20 located near a fault, derives its water from Lower Cretaceous sandstones. Although many wells pump the water from the Upper Cretaceous limestone aquifer, the origin of the water is actually from the Lower Cretaceous Kurnub Group sandstones (Tzofar 4a, Yaalon 116, Yotvata 13 and 12, and Samar 2 and 4). Groundwater drawn from the Quaternary alluvial fill either originates from Kurnub Group sandstones (Eilat 108, Yaalon 117) or from altered limestones of the Judea Group. The origin of these waters is from floods flowing through wadis incised into calcareous formations of the Judea Group. On the other hand, as a result of step-faulting, hydraulic contact is locally established between the Kurnub- and the Judea Groups aquifers, facilitating inter-aquifer flow of the confined Kurnub paleowater into the karstic formations of the Judea Group. The waters from well Tzofar 20 and Eilat 103 do not resemble the majority of waters occurring in the Arava (Fig. 17). The water of Eilat 103 is the result of evaporation of seawater in a playa-like environment (Yechieli et al. 1992). Tzofar 20 is compositionally comparable with Tiberias Hot Spring waters (Table 1) and indicates that the salinizing Ca > Mg brine is present throughout the Rift. In the central parts of the Arava, groundwater is extracted from aquiferous horizons in the Neogene Hazeva Formation and in the young alluvial fill. These waters are recharged mainly by flash floods in the
123
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
wadis draining the mountainous areas east and west of the Rift and also by groundwater originating in the Kurnub Group leaking into the buried fault-block. The Cl– concentrations of these waters vary and range from 200 to 4,700 mg/L. The main source of salinization is the upflow of brines along the various systems of faults (Yechieli et al. 1992). In the southern Arava, the groundwaters exploited from wells Samar 2, Yotvata 12 and 13, and Yaalon 116 derive from the alluvial fill of the Rift, which is chiefly composed of limestone debris and sand. However, the rare earth element patterns of these waters are similar to those of waters of the Hot Spring of Tiberias and of Bet Yosef 5 derived from a basalt–limestone contact zone (Mo¨ller et al. 2003). These waters are probably recharged by flood events occurring in the Wadis Daraba and Turhan in Jordan, the drainage network of which is prevalently along weathered basaltic dykes (Rosenthal et al. 1990). In the Jordanian part of the study area, i.e. in the northern Wadi Araba, the sub-surface of the Valley is also built of a very thick (thousands of metres) sequence of alluvial sediments. At great depth, the water is saline due to the effects of the Dead Sea interface and to the relics of its predecessors. Groundwater in the area occurs in alluvial deposits, talus, and in alluvial fans, the thickness of which attains 250 m. Groundwater flows from the mountains in the east, west- and north-westwards to the Dead Sea. Generally, water salinity increases in the direction of flow. In the confined parts of the aquifer, higher temperatures and pressures enhance the dissolution of minerals and water salinization. The fresh and the brackish groundwater are found in the uppermost portions of the aquifer. Here, groundwater flow is directed southwards to the Red Sea. Recharge is from precipitation falling on the surrounding mountains in the east, where it infiltrates the barren rocks and flows laterally into the fluviatile old alluvial deposits covering the floor of the Valley. Some recharge takes place along the side wadis and on Wadi Araba itself. Salinity increases in the direction of groundwater flow, i.e. from north to south.
Brine formation in a shrinking lake in a closed basin The model and its parameters Assume a prismatic lake with an initial surface Fo and an average depth ho resulting in Vo, i.e. the volume of the lake located in the centre of the drainage basin of an area FD. For the sake of simplicity, an average
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
561
precipitation rate N is assumed all over the area. The chemical mass balance of the lake (including the pore water in the sediments of the lake) for each chemical element i is derived from the differential equation 3
co,i concentration of element i in lake water at the beginning ct,I concentration of element i in lake water at time t in mmol/l
NcN;i dðFtÞ þ NcD;i dðtðFD FÞÞ ¼ dni ¼ dðVci Þ
Besides the various concentrations that have to be selected, Eq. 8 contains as variables time of evaporation t, shrinkage constant e, the ratio of precipitation rate and initial depth of the lake N/ho. In order to dry up the lake, precipitation must have been low and due to evaporation and evapotranspiration not all precipitation on land results in drainage water.
ð3Þ
F time-bound area of the lake V time-bound volume of the lake cN,i concentration of element i in precipitation in mmol/L cD,i concentration of element i in drainage water in mmol/L t time of evaporation n number of moles of element i c time-bound concentration The changes of the volume V and the surface F of the lake are described by the shrinkage constant e of both the initial depth ho and width ao of the lake according to Eqs. 4 and 5. From geometric considerations, Eqs. 6 and 7 are derived for the changes of volume and surface area of the prismatic lake, which are appropriate for mathematical treatment. h ¼ ho expðetÞ
ð4Þ
a ¼ ao expðetÞ
ð5Þ
Vt ¼ Vo expð2etÞ
ð6Þ
Ft ¼ Fo expðetÞ
ð7Þ
e= shrinkage constant of the lake’s both initial depth ho and width ao Indices o and t = at the ‘‘beginning’’ and time ‘‘t’’ The actual volume of the lake Vt is not only changed by evaporation, but also by filling with clastic sediments and chemical precipitates. These sediments have high porosities (>50%) and are considered to exchange their pore water with that of the lake, i.e. the sediments always contain water with the time-bound salinity. At the very end, when the lake is dried up, the sediments contain the residual brines. Therefore, in the above calculations, the pore water of the sediment is considered as part of the water volume. Substituting h, V, and F in Eq. 3 by Eqs. 4, 5, 6, and 7 and integration, Eq. 8 is obtained.
NFD 2N cD; i t þ tðcD; i cN; i Þ ho Vo ðexpðetÞ 1Þ expð2etÞ ¼ ct; i
co; i þ
ð8Þ
Scenario of the late Sdom Sea The Pliocene seawater, which deeply penetrated inland and created the Sdom Sea, was probably more saline than the modern Mediterranean seawater because of massive dissolution of Messinian evaporites and mixing with brines that accumulated on the dried-up sea floor of the proto-Mediterranean (Vengosh and Starinsky 1994; Vengosh et al. 1994a). Salts and brines also accumulated in the canyon-like channels of the Kishon–Yizreel–Bet Shean Valleys in the north and the Afik-Beer Sheva channel in the south through which seawater penetrated into the Rift creating the Sdom Sea (Rosenthal et al. 1999). At present, remnants of highly saline brines are still present in deep sub-basins of the Mediterranean Sea (Vengosh et al. 1994a; Bau et al. 1997). The active Plio-Miocene volcanism at that time interacted with the invading seawater. Spilitization, oxidation of sulphides, precipitation of carbonates and sulphates, albitization and chloritization of basaltic rocks might have changed the chemical composition of the invading seawater. The cyclic disconnection of the Sdom Sea from the Mediterranean Sea in the west generated thick sequences of evaporites in the deeper parts of the Rift and accumulated Mg-chloride-dominated residual brines in sediments and adjacent sedimentary rocks. In order to start the calculations, modern seawater is concentrated to an unknown degree by an evaporation factor f, which has to be adjusted by trial and error. The shrinkage constant of the depth of the lake e is adjusted to yield an evaporation period of about 3 Mio years. The average precipitation rate N of 0.2 m.a–1 is assumed to be constant all over the drainage area and over time. The initial depth of the lake is assumed to be 200 m at average. The composition of the drainage water is assumed to be a mixture of 95% Jordan River water and 5% Feshkha spring water representing brackish groundwater draining from the Judea Group
123
562
carbonate aquifer (Table 3). The fractions of drainage and rainwater are proportional to the time-bound exposed areas of land and lake surface. Applying Eq. 8, one has also to consider halite precipitation, which is introduced into the calculation by setting Cl– concentrations constant after having reached 5,900 mmol/L. This value is derived from data reported by Fontes and Matray (1993). Starting the calculations with assumed values of f = 1 and e = 0.000001, the ‘‘starting estimate’’ line in Fig. 18a is obtained, which shows both too low initial ratios and a too high negative slope. Increasing f towards 10 shifts the lowest Br/Cl and highest Na/Cl ratio towards that of the Dead Sea Rift brine trend line as defined by Klein-BenDavid (2004). Increasing e decreases the negative slope and increases the duration of evaporation. By trial and error the best fit for the assumed composition of drainage water is found. The ‘‘optimal fit’’ in Fig. 18a is slightly curved. Trying differently composed drainage waters, it turns out that only waters with Na/Cl < 1 yield reasonable results. Water with Na/Cl > 1 leads to positive slopes because Na increases and Cl decreases with time. Such trends markedly differ from that of evaporating seawater. Although rainwater is low in salinity, it has an impact on the Br/Cl ratio of the lake water with time, which is easily proved by changing the Br content in precipitation in Table 3. Assuming Br/Cl ratio in rainwater to be the same as in seawater, a Br content of only 0.045 mg/L results which is much less than the average Br content of rainwater with 1.25 mg/L. Applying the seawater, adjusted Br content alters the ‘‘optimal fit’’ into the ‘‘assumed Br/Cl ratio of seawater in precipitation’’ in Fig. 18a. This scenario is based on geological evidence, on known phases of seawater transgressions, and on the leaching of salts from soils, sediments, and sedimentary rocks by natural replenishment water in the drainage basin. It starts after the last closure of the Rift from the Mediterranean Sea. Thereafter, only flood-, drainage-, and rainwater are continuously added. Contrarily, the
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
dual-mode evaporation model of Klein-BenDavid et al. (2004) only considers the period of repeated inflow and evaporation of the closed basin without considering the drainage water with its characteristic chemical composition. Scenario of the late Lisan Lake Starting from a low-salinity mixture of drainage and rainwater and a contribution of saline water, the change of a fresh water lake (e.g. Lake Samra) into a saline lake (like the Lisan Lake) is simulated by Eq. 8. Evaporation is expected to have occurred during the last 6,000 to 30,000 years. Correspondingly, the shrinkage constant e of the depth and width of the lake was found by trial and error to be about 0.00005. During the Pleistocene and its enhanced rainfalls, huge amounts of salts were flushed from rocks, soils, and sediments and transported into the Rift Lake. During later stages of the Lisan period, the lowering of the lake level led to the mixing of groundwater with earlier saline waters entrapped in the sediments. Thus, the groundwater could have been of highly variable salinities. In Fig. 18b, the trends are shown for an assumed starting mixture of 10% rainwater, 86% Jordan River water, and 4% of Feshkha spring water (Table 3). The lowering of the lake level started after 10, 100, and 500 times salt concentration. The fractions of drainage and rainwater are proportional to the exposed areas. The resulting various V-shaped paths are caused by the different salinities at which the shrinkage of the lake started (positive slopes) and by precipitation of halite (negative slopes). The higher the pre-concentration was before the shrinkage started, the earlier the precipitation of halite started. The increasing negative slopes visualize the impact of Br/Cl ratios in precipitation with time. As long as halite precipitation does not occur, Na/Cl ratios are nearly constant, but Br/Cl decreases because the lake’s surface decreases and the contribution of the high Br/Cl in rain over the lake diminishes. Conse-
Table 3 Composition of waters used in estimating the composition of brines during drying up of the Sdom Sea and Lake Lisan
Seawater Rainwater average Jordan River Feshkha Drainage water (A) = 0.95 Jordan + 0.05 Feshkha Drainage water (B) = 0.1 rain + 0.86 Jordan + 0.04 Feshkha
Na (mg/L)
Cl (mg/L)
Br (mg/L)
1,000 Br/Cl (molar ratio)
Na/Cl (molar ratio)
10,764 7.2 16.2 436 37.2 34.2
19,383 13.1 15.1 1,420 85.3 78.1
67.1 1.23 0.1 27 1.45 1.42
1.54 42 2.94 8.45 7.52 8.10
0.86 0.85 1.66 0.47 0.67 0.68
Rainwater composition is presented as averages estimated from data given by Salameh and Rimawi (1988) and Herut (1992)
123
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
a
12
10
1000Br/Cl molar ratio
Fig. 18 Evaporation of a closed inland lake. Results of model estimates according to Eq. 8. a Shows the results of three runs. For details see text. b Displays an example of evaporation under the conditions of the late Lisan Lake. Necessary data are compiled in Fig. 3
563
8
6 Klein-Ben-David (2004) Starting estimate
4
Optimal fit Assumed Br/Cl ratio of seawater in precipitation
2
0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.7
0.8
0.9
1.0
Na/Cl molar ratio
b
12
1000Br/Cl molar ratio
10
8
6 Klein-BenDavid (2004)
4
Drainage water concentrated by factor 10 drainage water concentrated by factor 100 drainage water concentrated by factor 500
2
0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
Na/Cl molar ratio
quently, the drainage water with low Br/Cl ratio becomes more influential. However, when halite precipitation starts, Br/Cl ratios increase and Na/Cl ratios decrease because Br is only insignificantly coprecipitated. Depending on the starting conditions, various trend lines can be given. Thus, depending on the fresh water and brine components, a whole bundle of evaporation lines covers the domain right to the trend line of Rift brines.
Origin of saline water bodies in the Rift The chemical evolution of the late brines of the Sdom Sea period comprises evaporation of seawater loaded with dissolved Messinian salts and that of drainage
water. This created the Sdom Sea brine with Mg > Ca composition. Referring to Fig. 7a, the waters of the eastern side of Lake Tiberias (Ha’On, Gofra) plot together with the Hot Spring waters from Tiberias. This strongly points to a common source. They correspond in Br/Cl, B/Cl, and Na/Cl. They differ in Mg/Ca and (Ca + Mg)/(Na + K) because of alteration processes such as dolomitization and albitization, which could have taken place in the deep structural low of the eastern Lower Galilee. At Tzofar 20, water resembling that of Tiberias is identified. This implies that the Tiberias type of saline water is omnipresent throughout the Rift covering the area of the former Sdom Sea (Fig. 1). Considering the geological evolution of the area, these saline waters originate from the same mother brine. It was formed
123
564
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
during the latest stage of evaporation of the Neogene Sdom Sea. As part of this evaporative process, salt bodies formed in the Rift (Zemah, Zaharat-el Qurein, Jericho, Lisan Peninsula, Mt Sdom). The sub-surface ablation of the Zemah salts produced a secondary brine characterized by Na/Cl ~ 1 and low Br/Cl ratios (Table 4). Knowing that the Tiberias type of saline water is present throughout the Rift raises the question whether there are even older brines that take part in the salinization of groundwaters. Referring to Table 4, the deep brines are higher in Br/Cl ratio and are pressurized (partial pressure release test in Devora 2A and Rosh Pina 1 up to 480 bar). Any dilution of these brines by less saline waters enabling the ablation of halite would result in lower Br/Cl ratios. Such a process, however, can hardly explain the uniformity of the Tiberias type of brine throughout the Rift. Nevertheless, it cannot be excluded that deep-seated brines from the Triassic/Jurassic strata may ascend along deep reaching faults and penetrate into active aquifers, causing their salinization. These prevalently Ca-Cl brines are not of uniform composition and may reflect
Table 4 Compilation of ionic ratios of identified primary, secondary, and tertiary brines in the Jordan-Dead Sea–Arava Rift that denote mother brines Na/ Cl
Mg/ Ca
(Ca + Mg)/ 1,000 1,000 (Na + K) Br/Cl B/Cl
Pre-Sdom Sea periods, probably Triassic–Jurassic: all tertiary brines Devora 2A 0.48 0.09 0.54 Rosh Pina 0.53 0.2 0.45 Sdom 2 0.3 0.79 0.9 Arava 1 0.85 0.39 1 Sdom Deep 1 0.25 0.12 1.4 Lot 1 0.16 0.5 2.6 Arava Valley 0.77 0.85
7.2 7.4 7.5 8.1 8.8 9.7 1.4
Sdom Sea period Representatives of primary brines Ha’On 0.51 2.5 Eilat 103 0.56 0.8
5.7 3.6
0.4 0.58
Representative of secondary brine Zemah 1 0.09 0.01 Representative of tertiary brines Tiberias-Main Spring 0.51 0.32 2.8 Tzofar 20 0.5 0.4 1.7 Late Lisan Lake–Dead Dead Sea Jericho 2 Radon 1 Araba 5 Dead Sea Hot Springs
Sea period 0.3 3.8 0.7 0.95 0.3 3.1 0.4 0.3 0.4 2.7
1.5 0.21 0.84 0.6 0.6
0.36 0.95
Conclusions
0.2 5.7 5.8
0.5 0.38
26.5 8.0 9.3 16 9
In italics additional examples are given that are mentioned in the text
123
different genetic processes in the past. The influence of Dead Sea water is best discussed by its unique Gd anomaly. Any brine of this type under sub-surface conditions looses the Gd anomaly by ion exchange with REY bearing minerals. Thus, thermal mineral waters of the springs (En Boqeq-Gofrit, Hammei Yesha, and Hammei Shalem) with their strong Gd anomalies indicate considerable contribution of Dead Sea water flushed out from its previous higher stands with high-temperature saline water from deep horizons. Going further back into the development of the previous Dead Sea, the saline water from the well Radon 1 in the Jericho area seems to be of a similar composition as the Hot Springs at the Dead Sea (Table 4). The contribution of saline water of well Arava 1 to the salinization of groundwater bodies in the Arava does not seem to be significant. Comparing the Br/Cl ratios of groundwater in the Arava area with the saline Arava 1 brine, it occurs that the contribution of the latter is negligible (Table 4). As previously indicated by Rosenthal et al. (1998), salinization is mainly due to water–rock interaction. By comparing Br/Cl ratios of Radon 1 well with those of the Dead Sea springs, it occurs that the salinization of groundwater in the Jericho area is mainly by flushing out of remnants of the late Lisan Lake and of the previous Dead Sea (Table 4). Summing up, one may derive that deep-seated brines of Devora type do not take part in the groundwater salinization process in the study area. Similarly, the brine of Arava 1 is not reflected in the chemical composition of the waters in the Arava Valley in agreement with conclusions by Rosenthal et al. (1998). Groundwater salinization is mainly influenced by the omnipresent Rift brines as encountered at Haon, Tiberias Hot Springs, and Tzofar 20. In the Lower Jordan Valley and the Dead Sea area, groundwaters are also influenced by flushing out of the remnants of late Lisan Lake and the previous Dead Sea.
By considering chemical and isotope data, it occurs that the saline water bodies contributing to saline groundwater in the Rift are the following: • • •
Recent Dead Sea and late Lisan Lake waters, which are primary evaporation brine with Mg > Ca; Neogene Sdom Sea primary evaporation brines exemplified by Ha’On brines with Mg > Ca; Neogene to Recent secondary ablation brines of the Zemah type Na/Cl~1;
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566
• •
Neogene tertiary brines with Ca > Mg occurring in Tiberias Hot Spring and Tzofar 20; Deep-seated brines of the Devora type with variable chemical compositions.
In variable proportions, the Neogene and late Lisan Lake and Recent Dead Sea brines have to be considered as the most serious sources of salinization of groundwaters in the Rift. Deep-seated pre-Sdom brines cannot strictly be excluded, but if active they play a negligible role only.
References Al-Zoubi A, ten Brink US (2001) Salt diapers in the Dead Sea basin and their relationship to Quaternary extensional tectonics. Mar Petrol Geol 18:779–797 Arad A (1964) The geology and hydrogeology of the Lower Cretaceous of the Northern Negev and Judean Mts. Ph.D. thesis in Hebrew with English abstract, Hebrew University of Jerusalem, pp 1–154 Arad A, Bein A (1986) Saline versus fresh water contribution to the thermal water of the Northern Jordan Rift Valley. Isr J Hydrol 83:49–66 Awad AR, Taqieddin SA, Al-Homoud AS, Barakat EA (1997) Hydrochemical assessment of a deep Jordanian aquifer. Environ Geol 30:198–208 Bandel K, Khoury HN (1981) Lithostratigraphy of the Triassic in Jordan. Facies 4:1–26 Bau M, Mo¨ller P, Dulski P (1997) Yttrium and lanthanides in eastern Mediterranean seawater and their fractionation during redox-cycling. Mar Chem 56:123–131 Belitzky S, Mimran Y (1996) Active salt diapirism at the Zahrat el Qurein Dome, Lower Jordan Valley. Isr J Earth Sci 45:11–18 Bender F (1974) Geology of Jordan. Gebru¨der Borntra¨ger. Stuttgart, Berlin, pp 1–196 Bentor YK (1969) On the evolution of subsurface Brines in Israel. Chem Geol 4:83–110 Bergelson G, Nativ R, Bein A (1999) Salinization and dilution history of ground water discharging into the Sea of Galilee, the Dead Sea Transform, Israel. Appl Geochem 14:91–118 Fleischer L, Varshavsky A (2002) A lithostratigraphic data base of oil and gas wells drilled in Israel. Earth Science Research Administration, Oil and Gas Unit Rep. OG/9/02, Jerusalem, pp 1–280 Fleischer E, Goldberg M, Gat JR, Magaritz M (1977) Isotopic composition of formation waters from deep drillings in southern Israel. Geochim Cosmochim Acta 41:511–525 Flexer A, Yellin-Dror A, Kronfeld E, Rosenthal E, Ben Avraham Z, Artzstein PP, Davidson L (2000) A Neogene salt body as the primary source of the salinity in Lake Kinneret. Arch Hydrobiol Spec Issue Adv Limnol 55:69–85 Fontes JC, Matray JM (1993) Geochemistry and origin of formation brines from the Paris Basin, France. 1. Brines associated with Triassic salts. Chem Geol 109:149–175 Gat JR, Carmi I (1970) Evolution of the isotopic composition of atmospheric waters in the Mediterranean Sea area. J Geophys Res 75:3039–3048 Gat JR, Mazor E, Tsur Y (1969) The stable isotope composition of mineral waters in the Jordan Rift Valley. J Hydrol 7:334– 352
565 Goldschmidt MJ, Arad A, Neev D (1967) The mechanism of saline springs in the Lake Kinneret depression. Geol Surv Isr Bull 45:1–19 Greitzer Y (1980) Hydrodynamic investigation of geological formations in Israel to the purpose of oil exploration—stage 1, vols. 1–2. Rep Isr Nat Oil Co Ltd, Tel Aviv, pp. 1–210 Guttman J (2000) Hydrogeology of the eastern aquifer in the Judea Hills and Jordan Valley. Rep. 468. Mekorot Co., Tel Aviv pp 1–47 Gvirtzman H, Garven G, Gvirtzman G (1997) Hydrogeological modelling of the saline hot springs in the Sea of Galilee, Israel. Water Resour Res 33:913–926 Herut B (1992) The chemical composition and sources of dissolved salts in rainwater in Israel. Ph.D. thesis (in Hebrew, English abstract), Hebrew University of Jerusalem pp 1–150 Hirsch F (2005a) The Oligocene–Pliocene of Israel, chapter 18H. In: Hall JK, Krasheninnikov VA, Hirsch F, Benjamini C, Flexer A (eds) Geological framework of the Levant: the Levantine Basin and Israel, vol. II. Tel Aviv University and Ben Gurion University, Special Publication, pp 459–488 Hirsch F (2005b) The Late Pliocene to Quaternary of Israel, chapter 18H. In: Hall JK, Krasheninnikov VA, Hirsch F, Benjamini C, Flexer A (eds) Geological framework of the Levant, vol. II. The Levantine Basin and Israel. Tel Aviv University and Ben Gurion University, Special Publication, pp. 489–514 Horowitz A (2001) The Jordan Rift Valley. Balkema, Rotterdam, pp 1–730 Husseini MI, Husseini SI (1990) Origin of the infra-Cambrian salt basins of the Middle East. In: Brooks J (ed) Classic petroleum provinces. Geological Society, Special Publication, 50:279–292 Ilani S, Rosenthal E, Kronfeld J, Flexer A (1988) Epigenetic dolomitization and iron mineralization along faults and their possible relation to the paleohydrology of southern Israel. Appl Geochem 3:487–498 Kafri U, Arad A (1979) Current subsurface intrusion of Mediterranean Sea water, possible source of ground water salinity in the Rift Valley system, Israel. J Hydrol 44:267–287 Klein-BenDavid O, Sass E, Katz A (2004) The evolution of marine evaporitic brines in inland basins: the Jordan-Dead Sea Rift Valley. Geochim Cosmochim Acta 68:1763–1775 Kolodny Y, Katz A, Starinsky A, Moise T (1999) Chemical tracing of salinity in lake Kinneret (Sea of Galilee). Isr Limnol Oceanogr 44:1033–1044 Magaritz M, Nadler A (1980) Re-interpretation of 18O and D isotopic composition of the Tiberias subgroup hot waters. In: Proc Geol Isr Annu Meet Southern Sinai and the Gulf of Elat, Ophira, Sinai, Feb. 4th, 1980, pp 1–26 Mazor E, Mero F (1969) The origin of the Kinneret-Noit water association in the Kinneret–Dead Sea Valley, Israel. J Hydrol 7:318–333 Menashe S (1991) The deep aquifers in the northern part of Israel. M.Sc. thesis, Center for Soil and Water Sciences, The Hebrew University, Rehovot, Israel, pp 1–63 Mo¨ller P, Rosenthal E, Dulski P, Geyer S, Guttman J (2003) Rare earths and yttrium hydrostratigraphy along the Lake Kinneret–Dead Sea–Arava transform fault, Israel and adjoining territories. Appl Geochem 18:1613–1623 Mo¨ller P, Dulski P, Salameh E, Geyer S (2006) Sources of salination of groundwater along the Jordan-Dead Sea–Red Sea transform: new results for the area around Lake Kinneret, the western Jordan- and Arava Valley. Acta Hydrochim Hydrobiol 34:86–100 Neev D, Emery KO (1966) The Dead Sea. Sci J 2:50–55
123
566 Picard L (1943) Structure and evolution of Palestine with comparative notes on neighbouring countries. Bull Geol Dept Hebrew Univ Jerusalem 4:1–187 Raab M (1998) The origin of evaporites in the Jordan–Arava valley in view of evolution of brines and evaporites during seawater evaporation. Geol Soc Israel Rep. GSI 1/98, pp 1– 138 Rosenthal E (1988) Occurrence of Ca-chloride brines at common outlets of the Bet Shean multiple aquifer system, Israel. J Hydrol 97:107–128 Rosenthal E, Adar E, Issar A, Baatelan O (1990) Definition of groundwater-flow pattern by environmental tracers in the multiple aquifer system of southern Arava Valley, Israel. J Hydrol 117:339–369 Rosenthal E, Weinberger G, Berkovicz B, Flexer A, Kronfeld J (1991) The Nubian sandstone aquifer in the western Negev, Israel: delineation of the hydrgeological model underconditions of scarce data. J Hydrol 132:107–135 Rosenthal E, Jones BF, Weinberger G (1998) The chemical evolution of Kurnub Group paleowater in the Sinai–Negev province—a mass balance approach. Appl Geochem 27:3–17 Rosenthal E, Weinberger G, Kronfeld J (1999) Ground water salinization caused by residual Neogene and Pliocene sea water, an example from the Judea Group aquifer, southern Israel. Ground Water 37:261–270 Rosenthal E, Flexer A, Mo¨ller P (2006) The paleoenvironment and the evolution of brines in the Jordan-Dead Sea transform and in adjoining areas. Int J Earth Sci 95:725–740 Salameh E (1996) Water quality degradation in Jordan. Friedrich Ebert Stiftung and Royal Society for the Conservation of Nature. Printed by The National Library, Amman, Deposit No. 1134/8/1996, pp 1–178 Salameh E (2001) Sources of salinities in the Jordan Valley Area/Jordan. Acta Hydrochim Hydrobiol 29:329–362 Salameh E, Bannayan H (1993) Water Resources of Jordan—present status and future potentials. Friedrich Ebert Stiftung, Amman, 183 pp Salameh E, Rimawi O (1988) Hydrochemistry of precipitation of northern Jordan. Int J Environ Stud 32:203–216
123
Int J Earth Sci (Geol Rundsch) (2007) 96:541–566 Shaliv G, Mimran Y, Hatzor Y (1991) The sedimentary and structural history of the Bet She’an area and its regional implications. Isr J Earth Sci 40:161–179 Shulman H, Flexer A, Guttman J, Yellin-Dror A, Rosenthal E, Anker Y (2003) The hide-and-seek game between geological factors and groundwater flow in the Jericho-Fazael Region. In: Proceedings, annual meeting of the Israel Geological Society, Ein Boqeq, p 118 Simon E, Mero F (1992) The salinization mechanism of Lake Kinneret. J Hydrol 138:327–343 Starinsky A (1974) Relationships between Ca-chloride brines and sedimentary rocks in Israel. Ph.D. thesis, Department of Geology, Hebrew University, Jerusalem, Israel, pp 1–176 Stoecklin J (1968) Salt deposits of the Middle East. In: RB Mattox (ed) International conference of saline deposits, 1962. Geol Soc Am Spec Pap, no. 88, pp 157–181 Vengosh A, Rosenthal E (1994) Saline groundwater in Israel: its bearing on the water crisis in the country. J Hydrol 155:103– 124 Vengosh A, Starinsky A (1994) Relics of evaporated sea water in deep basins of the Eastern Mediterranean. Mar Geol 115:15–19 Vengosh A, Starinski A, Kolodny Y, Chivas AR (1991) Boron isotope geochemistry as a tracer for the evolution of brines and associated hot springs from the Dead Sea area, Israel. Geochim Cosmochim Acta 55:1689–1695 Vengosh A, Starinsky A, Anati D (1994a) The origin of Mediterranean interstitial waters—relics of ancient Miocene brines; a re-evaluation. Earth Planet Sci Lett 121:613–627 Vengosh A, Starinsky A, Kolodny Y, Chivas AR (1994b) Boron isotope geochemistry of thermal springs from the northern Rift Valley, Israel. J Hydrol 162:155–169 Yechieli Y, Starinsky A, Rosenthal E (1992) Evolution of brackish groundwater in a typical arid region: Northern Arava Rift Valley, southern Israel. Appl Geochem 7:361– 374 Zak I (1963) Remarks on the stratigraphy and Tectonics of the Middle Triassic Makkhtesh Ramon. Isr J Earth Sci 12:87–89