Content uploaded by Joel Kronfeld
Author content
All content in this area was uploaded by Joel Kronfeld on Apr 28, 2016
Content may be subject to copyright.
Arch. F{ydrobiol. Spec. Issues Advanc. Limnoi. 55, p. 59-85, February 2000
Limnoiogy and Lake Management 2000+
A SJeogeffie saEt body as the primary s@ue'ce Gf saHFmnty
an Lakc KEmmcret
Akiva Flexer, AnnatYellin-Dror, Joel Kronfeld, Eliyahu Rosenthal,
Zvi Ben-Avraham, Philippe Pascal Artsztein and Lea Davidson
with 6 figures
Abstract: A mineraiogicai study of sporadic samples from the thick salt sequence found within wildcat
Zemah-!, revealed the presence of evaporite minerals beyond the level of halite precipitation. The pres-
ence of post halite minerals extending up to, and including, bischofite cannot be excluded. The finding of
the more soluble K and Mg salts shows evidence of an advanced degree of seawater evaporation that
acted, presumably cyclically, on seawater that intruded during Miocene-Pliocene times into the Rift
Valley. The wildcat Zemah-l borehole is located immediateiy south of Lake Kinneret. The region is
dissected by regional faults that extend longirudinally across the lake. These faults may act as conduits
for groundwater that flushes the salt body at various levels as it rises to the surface. The presence of such
an extensive, mineralogically differentiated salt body (similar to other such structures in the Rift) may
explain the high Cl, Br and Mg contents and the deficiency of Na in groundwater and spnngs emereing
along the eastern shore of the lake and in the nearby Rift extension. The Zemah-l salt body as weil as
similar structures in the Rift, may be the sources of the brines discharging along the eastern, southern and
southwestem shores of the lake and pcssibiy may be diffusing upwards throu-eh the sediments alon-q the
entire lake floor.
lntroduetEosi
Lake Kinneret (Fig. 1) is the major water reservoir of Israel. The preservation of its water
quality is of national importance. The source of saiinity of Lake Kinneret has been a topic of
investigation for decades. As yet, no consensus of opinion has been reached regarding either
the sources of salt nor the processes of salinization. These are basic elements that need to be
detennined before practical methods can be applied to decrease the sait inputs. All investiga-
tors agree that within the catchment area there must be internal sources of salinity. The upper
Jordan River, which contributes 2/3 of the recharge to the lake, brings in only 1l-20 mg l-i Cl
whereas the chloride content of lake water is in the 220-390 mg l-t Cl range (Snron & Mgno
Au"fh.ors' addresses: Akiva Flexer, Annat Yellin-Dror, Joel Kronfeld, Eliyahu Rosenthal, Zvi Ben-
Avraham, Philippe Pascal Artsztein and Lea Davidson, Faculty of Exact Sciences, Department of Geo-
physics and Planetary Sciences, Tel-Aviv University, Tel Aviv 69978,Israel.
007r -1 128/00/0055-069 $ 4.25
A. Flexer et al.
'\d!1
iirr;::rJi.i
:':itt!.:1jr*
n4i::'i .i
;;li;r:i$
!n$8;;Q.
:*:.ii*:la$
r. ;,.:. L.:
-* . .'. -l
: 1:.i
.:t!.::::lillir
:!i:*$
,:i:::tii--..ia.r:
aa!1!rrli:n
:' r":,: I
..".",i-":r
i;ir:Li$i
i:,;lr.iir;i,ii
,,r:*i;,-ir-
j:;iii,ii:ilt
iil:'{,$i
:!$ili.!.$
Fi'*uecr'.€.8i{ *arn'es
Iniene,J Sait ilerno*-q
iS-*i=;itrc- * i
r ' ..1^
lLlU IJt-r L*v
Fiig" 3.. Location map showing Lake Kinneret and the salt bodies along the R'ift Valley'
1987, GopgEN 1991). Additional recharge comes from rainfall on ihe surrounding hitrls. Along
the shoreline and subrner_eed are many -saiine springs which are considered to be major con-
tributors of sait to the lake (snaoN & MERo !g87, slltrs et al. 1989). In 1964, a salt water
carrier was built which diverts the saline water, from the Tiberias, Fuliya, and rabgha springs
(Fig. 2), to the iower Jordan River. Recently, Srn-I-Bn (1994) demonstrated that there is a net
chloride flux through lake bottom sediments to the watel body of the lake' she proposed that a
hypersalinebrineunderliestheSedimentsandpercolatesthroughthem,
The presenr study was carried out to invesiigate the feasibility that a solid salt body (or
bodies) that was discovered in the area can gen-*erate the brines which are the main salinitv*
inputs to the lake. previous srudies discountld the possibility of a solid body of salt as the
contributor of salinity because sait was regarded aS synonymous with halite' Moreover' neither
the Na/Cl nor the CVBr ratlos, found in iater emerging from the saline springs' conform to
+a j,i
i'
J.
i
-I
A Neogene sait bodv V7
32050'
2E 0 anl
,tr8
?50
?E ran'
32'50'
245
240
Fig" 2" i-ocation map of the Zem*'l borehoie and the saline sprin,ss alon-e the slioreline of l-ake Kinneret.
tirat expected from the dissoiution of halite. Accordingly, MeRcADo & Meno (1984), pointed
oui that the cieficiency of Na. SOu and the enrichment of Ca, N{g and Br reiative to a sea water
precursor, cannot be explained by dissolution of a salt body composed of haiite. Nlencus &
SI-aceR (1985), without carrying out any systemati.c minerai identification, reported that the
drilied section in the Zemah-l borehoie containeci soiely halite and anhycirite.
The present study investigated the mineraiogy of pari of the evaporite section encountereC
in the Zeman-l borehole. Our preliminary' findings presented here reveal the occurrence of
minerals of the more soluble salts of potassium and magnesium in addition to halite. This
evidence suggests that the presence of a solid body (or numerous bodies) of evaporites in the
catchment area may be the primary source of a hypersaline brine that contributes soiutes to
groundwater emerging in the saiine springs and percolating through the bottom sediments.
GeoicgteaE as^dd gc@pF?yeEeaE baekEround
i-ake Kinneret (Sea of Galiiee) is one of two lakes (the other being the Dead Sea ro the south)
that fill morphotectonic depressions alon-e the Dead Sea-Jordan R.ift Valley (Figs. 1, 3). Lake
72 A. Flexer et al'
G)
$
$
T
3
.J
t- ! & *\.*}' + 4 |
sJ4Jre41+
! -..R!
Kinries"et
gL4+J
!&+z:
4;4J49
4LJ&J
+iLlg
+3
4+4+4
lF-r + + 4 !
-
+ -l lst+Til;d votcanics
+ | l?ti .',:'iI]
I
: +t
-
(-
$
€
o
a
NVolcanics
Carbonates
Fault
0 5 10km
| ----------.t-----
: _j
.:.. :
.':: L
iI
,:l
_,11
,,.::. j
::-.
i.t i
i'!
i3
'. fl
'.1
"i;ii '
.jii
t:t
j'] l
::il
:il
'.;':
F,ig. 3. The structural and surface geology of the Lake Kinneret area (Ivlodified after BEN Avn'q'HAM et al'
1996). A. F. -Alm"g;;i;;,1' r"-- lo'-dun fault' S'A' - Sheikh Aii faull
Kinneret is located at the northern portion of the Kinneret-Bet shean basin (BEN A'RAHAI* et
a'. 1996). The lake i, zo tm 1ong, 12 km at its widest, ancl has a maximum depth of 46 m' The
A Neogene salt body Ta
lake surface is situated at approximately -210 m MSL. This low elevation makes it a re_eional
drainage base of rainfail recharge failing in the surrounding hills and transported through the
stepwise downfaulted blocks (Fig. a) to beneath the lake floor.
The Dead Sea- Jordan Vailey Rift is a piate boundary of the transform type which connects
the Red Sea, where sea floor spreading occurs, with the Zagtos zone of continental collision.
This rift is one of the most prominent tectonic features in the Middle East. The movement
along the transform has caused a ieft-lateral relative displacement of 105 km since the time of
its formation (GanruxKEL et al. i98 l,). Several fault systems occur in this area, the main ones
being the N-S transform system (Fig" 3) forming a longitudinal graben containing the buried
salt body of Zemah. The main eastern border fault of the lake has a general N-S strike (Fig. 3).
The western side of Lake Kinneret and its southern extension to the Bet Shean Valley is distin-
guished by a series of tilted blocks separated by transversal faults (SalrzivteN 1964). Seismic
studies show that the lake area is tectonically active. The average heat flow measured in the
Lake Kinneret is quite high (75 mW m-2, BEN.\vnernu et ai" 1986).
Sarrznaex (1954) and MicHBLSoN et al. (1987) investigated the compiex geology of the
areas surrounding the lake. Previous studies of the sub-bottom structure and stratigraphy of
Lake Kinneret included seismic refraction and reflection measurements (Beu AvnaHeM et a1.,
1981, 1986), magnetic measurements (FolrutAN & Yuvel- 1916, BEN AvneHAM et al. 1980,
GNzeuRc & BsN AvReHnvr 1986, EppsLBAuM & BsN AvRasalt 7991), heat flow measure-
ments (tsEN AvneHAM et al. 1978) and gravity measurements (BEI.I AvnaHeiv et al. 1996).
These studies clearly indicate that, like the surface geology, the sub-bottom structure of the
lake is also quite complex and is being affected by active faulting. The sedimentation rate in
Lake Kinneret is quite high; various estimates range from2toT mrn/year (Sriu-eR 1974).
This causes smoothenin,e of the floor and, as a result, the morphology of the floor does not
reflect the subbottom structure. (Bex AvRRnavt et a1. 1990).
Thc Zemafx=tr saflt body
The wiiCca tZemah-1 borehoie was drilled immediaiely to the south of Lake Kinaeret (Fig. 2),
This is anan'ow se-gmentof theJorrlan- DeadSeaRJftValleyborieredby nonnalfauits andfclds.
Two extensive faults that transect the sedimentary cclumn of the lake from the north to the south
are situated on either side cf the well (Fi-e. 4). A 2.8 km thick unit of evapoites, magniatic,
carbonates andmaris was encountered betweenl344-4X42maeptb. Evaporites constitute 3570,
or fii1iy 980 m of the unit (Mencus & SLAGpn 1985). Much further to the south within the R.ift
Valiey occur ihe hypersaline brines of the Dead Sea and the adjacent Mount S'dom salt body.
Henass (pers. comrn., 1998) has noted 27 layers of post-halite evaporite mineralizatlonin the
S'dom salt plug. Oil exploration on the Lisan Peninsula in the Kingdom of Jordan encountered,
beneath the Upper Pieistocene marls, halite as well as beds of sylvite and carnalite. The salt body
found beneath the Lisan Peninsula is believed to be correlative to the Mount S'dom deposit
(Bevonn lg74).Though the chronological controls are weak, the upper portion of these salt
bodies have been tentatively placed as being from Late Miocene to Pliocene (BeNnps.l9"7 4,ZeK
1974). This would make them approximately coeval with the Zemah-l evaporites (Miocene-
Pliocene age) (Sner-iv 1991). Within Lake Kinneret proper, seismic profiling has not as yet
identified salt diapirs. However, gravity data from the northern and western portions of the lake,
indicate several negative anomalies that could be interpreted as salt bodies (L. Eernenuu,
personal comm., 1998). If these are indeed diapirs, then horizontal beds of salt may be also
I
V4 A. Flexer et al.
Kiss:er€t Ei*-{3ev R=n'wt-i-hgstst
lti
ttt
?+*
Its'*kzen
-ffi& I
385,/2S +
Iel-fr(€ta{
I
I
I
B-€-
?.ffiiP+z
B NryEl€ Csver Brsati
Distence {km}
Fig. 4. NW- SE geological cross -section tirough Lake Kinneret and the Zemah-1 borehole.
present in the sediment succession of the iake. Important concentraiions of gypsum deposited
along the mar-qins of the Miocene lake were found atMenahamia andWadi Issachar (ScuurlieN
1959, Silailv 1991). Though Mencus & Slacnn (i985) described the evaporite in wiidcat
Ze1;1ah-I as being soieiy halite anrl anhydrite, Raae (i998) noted the possibiiity that in weil
post-haiite ininerals were also deposited Zemah-!. Tlierefore mineral icent'rficaiicn of a se-
quence of approximateiy 400 m section in the evapoiite section was cariied out to check, ii
indeec, ihe evaporite bcciy is iimited mineraiogicaiiy to halite and anhydrite.
ftilethods and resulHts
The cuttings frorn the wiidcat Zemah-l borehole have not been systeniaticaily anaiyzedprevi-
ously for their chemicai and mineralogicai compositions. Therefore neutron, bulk ciensity and
acoustic logs obtained from the Zemah-1 drilling were cross-plotted to identify the iithology
and mineralogy (Scrn-EIN & Spwer i986, Yei.i-IN-DRoR 1984).
A common practice in log interpretation is to cross-plot various iog readings in order to
deterrnine formation lithology and compute porosity accurateiy. Cross plots of sonic versus
density logs are wicieiy used in the interpretation of shaiy sands. For carbonates, the density
versus neutron cross-plot is commonly empioyed. These piots and the caiculations based on
them are extremeiy useful, but when the lithoiogy is a complex mixture of minerals, interpre-
tation of data often becomes ambiguous.
The "litho-porosity" cross-plot was introduced for interpretation in formations of compiex
iithoiogy (Bunre et al. 1963). It presents simultaneously data from a1i three standard porosity
A Neogene salt body
eYE7,!i:i3=)L*=
=:lg+iEi:=E;i\?=
:lr';ag€EEU;rli
;d,'€f : FE:*;E:: =:
EE=Ee:;*=s$E?E:
:€g;*IEiES:€€EE
r&;odUEgE-?.==i==2
a
x
o
C
c
o
!
Z
-J
Z
IJ
fl E fr q E q I 6 R 3 sd
-rFoEec:a hi
crsiE4tq€o
-q-'nqccq
'j;'j;.irioir.i.; q
6
z
X
o
.=
c
o
I
J
0)
z
Z
U
lN
li
1*
IL:
l=
l.
i"
t
la
I
I
lo
f@
I
l*
lF-
'5
o'6 "d"
o!tr
orr'r-u6 Alrsuap ritng b
q1
r
c
6
N
6
N
r
a
o
q:
c
o
n
o
o
o
F
.=
,6
C
c
L
,9
:
UJ ;\
E-QNA€
o--Fqo
ll/ras 7t aujt] ltsuerl lluos lF
crlu-rO Allsuap 4.te b
a
A. Flexer et al.
tools, the sidewall neutron porosity 1og, the formation density compensateci iog and the
borehole compensated sonic lo-e. Frorn the reading of these 1ogs, two porosity inciependeni
parameters, "M" and "N", are derived - M from the sonic and density logs, and I"[ from the
neuffon and density logs. in the piot of M versus N, a unique point regardless of porosity
represents each pure rock mineral. For a formation of complex lithology, the position of 1og
data-points on the M-N plot reiative to the pure mineral points is of great assisrance in identi-
flring various minerals in the formation.
The sonic, neutron and density 1og data, from the depth interval -1345 m (top of sait body)
io -1731 m (base of upper salt body layer), from the Zemah-l borehole, was digitized. Cross-
plots of sonic versus density, density versus neutron, sonic versus neutron and M versus N
were produced (Fig. 5). The salt evaporite mineraiogy was thus detennined by the use of the
cross-plots. The three companion plots were used for verifying the interpretation of the litho-
porosity plot.
Xn the 2100 m of section studied, two different lithological point-groups were clearly recog-
nized on the cross plots (Fi-e. 5): (i) a magmatic group, and (2) a halite and post-halite salr
group. Within this group there appear to be various evaporite minerals made up of the sulfate
and chioride salts containing K and Mg, not all of which have yet been identified exactly. No
anhydrite was detected in this sequence.
Figs. 5 a,b,c, show that the iithology is well described by the three cross-piots. However the
final rnineralogic identification is determined by the iitho-porosity plot (Fig. 5d) which shows
that this 400 m section contains appreciable amounts of halite anC post-haiite evaporite miner-
als. A trend of points to the right and upwards on the M versus N cross ptrot is indicative of the
salt zone. The continuation of this trend upward indicates the existence of a variety of salt
minerals that may well include carnalite or even bischofite. It should be recalled that the se-
quence of evaporite minerais deposited from seawater found in the classic Zechsiein section
from Gerrnany (citeC in Kaeus<oPF i967) is roughiy as follows: carbonares, suifates (gypsuin
anhydrite), halite and afterwards kieserite, pol,vhalite, cal-nalite, syivite and bischofite. A lovr
Fia/Cl ratic and an enrichment in bromine distinguishes the post-halite. depositionai sequence .
The present research should stili be ccnsidered preiiminary. nevertheiess, it shouli be ein-
phasized that ourin-e terminal evaporation, the pcrtion of post-haiite minelals can becorne a
si-enificant fraction of the evaporite succession.
Hsthento proposed s^nodcls fon hnlnc fonrnatlon
The results of the mineralogical analysis of the Zeman-l sait beds are very pertinent to undei-
standing the chernistry and origin of the thermal springs and have bearing on the salinization
process of the iake as a whole. The water chemistry of the thermal springs has been extensiveiy
studied in the past and many theories have been put forth as to the origin of the brine. Most
models that have been proposed to explain the salinization of the lake water assumed mixing
of a primary brine with fresh water. GoloscsulDT et aL. (1967) found that the Rift brines are
of the Ca-chloride type and are characterized by Na-deficiency. They discerned two hydroche-
mical subgroups: groundwater with Mg/Ca > 1 found along the eastern and southern shores cf
the lake and groundwater with Mg/Ca < 1 characterizing sources along its western and north-
western shores. Mezon & Mgno (1969 a, b) suggested that the saline end member was essen-
tia11y seawater that intruded into the Rift Valley and evaporated there. According to these au-
thors, the hydrochemical characteristics of groundwater along the entire Rift Valley from the
A Neogene salt body ZV
Kinneret to the Dead Sea, are of sufficient constancy to warrant the existence of a single sea-
water-derived saline en<i mernber. SteRwsrY (197 4) also proposed thai the source of the 111ft
brines was from liquicis derived from Miocene sea water intrusions, that have further evapo-
rated beyond the halite stage. This was recently,supported by boron isotopic data (VexcosH et
al. 1994). After their partial evaporation, the residual Mg-rich brines interacted with the car-
bonates of the Judea Group exposed along the margins of the Rift, thus acquiring a Ca-chlonde
composition"
Gvnrzveu and co-workers (1997) presented a hydrological model of the Kinneret basin
that takes into consideration the thermal regime and the fauits surroundin-e the basin. In its essen-
tial form, the model shows that freshwater recharged by rainfall on the mountain ranges sur-
rounding the lake flows through aquifers that drain to the lake. The flow of the fresh water com-
ponent is controlled by the hydraulic head, which is determined by the topographic differences
(approx. 650 m) between the mountain recharge area and the lake. The initially cool recharge
water flowing towards the lake through deep-seated aquiferous strata is heated. In the vicinity of
the lake, groundwater driven both by its hydraulic head as well as by thermai convection rises
along fauits. mixes with proposed brines and emerges as brackish or saline springs.
Based on chemical and isotopic observations, BERGELSoN et aL. (1999) proposed that the
salt source is an ancient, intensively evaporated brine (21 ta 33 foid sea water) which perco-
lated throu,eh the valley forrnations from a lake thai formed in the Rift Valley foilowing
seawater intrusion during the Late Miocene.
The new suggcstcd rnodcl and ats negEenaE rmpEaeateons
Any hydrochemical models intended to explain the formation and distribution of brackish and
saline waters arounC and in Lake Kinneret have to consider tne following points:
a. The sftarp differentiation between the Mg-chloride saline waters along the eastern and
southem shores and the Ca-chloride waters along the wesiem anci northwestern marEins of
ihe lake.
b. Both waters are Na-deficieni, with NarCl <A.7, and are consiiierabiy enriched in Bi (Ci,lBr
weight ratios as low as 70).
c. High-resolution sei,smic profiles indicate ihat the lake bottoin and ileai'by areas aro dorni-
nated by major fault- and graben-structuies that extend across the la,ke from Degania, to-
wards and near to the location of wiidcat Zemah-L.
d. The mineralogicai data demonstrates that within theZemah-1 salt bodv there occur layers
containing K-Mg-salts, which are of the post-halite evaporite group.
In view of these constraints, we propose a new modei for the formation of a brine causing,
in the first place, salinization of groundwater along the eastern, southern and southwestern
shores of Lake Kinneret. This model involves dissolution of sulfates, of halite and of K-Mg
rich and Na poor, post halite evaporites from the studied section of Zemah-1. The creation of a
Ca-chloride brine with Mg/C a> l, the result of the previously suggested dissolution process,
was tested and confirmed by experimental modeling using NETPATH code (Pr-utvrueR et al.
1994).This was also confirmed by applying the SNORM code (BoorNE & JoNEs 1986, JoNes
& Boonrie 1937) which showed that saline groundwater along the eastern, southern and south-
western shores of the iake evolved as the result of dissolution of sulfates, halite, bishofite,
carnalite and polyhalite.
5
.:g
78 A. Flexer ei al.
Along the western shores of the lake such groundwaters were encountered as far northward
as 1 km south of the Roman spring of Tiberias. It is further suggested that dissolution of the
salts is by fresh groundwater flowing into the Rift and reaching the buried salt body of Zemah
(and other such bodies, which are known tb exislin the Rift). The model of GvrnrzueN et al.
(L997) allows for the descent of the recharge water to depth, great enough to encounter the salt
bodies such as encountered in the Zemah-I borehole. Leaching of the sait and the subsequent
dilution of the newly formed brine with fresh water can occur during ascent. This convection
model can be used to explain the different salinities and temperatures observed in the springs
emerging around the lake. Rising convection cells under the lake as a whoie would also be in
agreement with the findings of increasing chlorinify in the sediment pore water with increas-
ing depth of the sediment (Sm-i-rn 1994).
Leaching by the recharge water of soluble, late precipitated Mg-salts would obviously cre-
ate a Mg-rich brine such as has been encountered aiong the eastern shores of Lake Kinneret.
The chemical composition is uniike that of the known Rift brine (GolnscHMIDT et al. 1967)
which is characterized by Mg/Ca < 0.6 and low CllBr ratios. However, if the lithoiogy of the
rock-sequence characterizing the Rift fiii surrounding the Zemah salt piug is considered, and if
plausible waterlrock reactions are envisaged, then the hydrochemical link between the Mg-
rich brines (generated by dissolution of Zemah salts) and the Mg/Ca < 0.6 Rift brines, may be
established. The lithological section surrounding the plug inciudes hundreds of meters of clay-
rich clastics and thick sequences of gabbro. The clays could easily act as exchange reactors
removing Na from the brine and enriching it in additional Mg (originating from the volcanic
rock-masses)" The Ca-rich brines encountered along the western littoral of the iake could be
the result of massive Mg-removal by dolomitization (see aiso BgncelsoN et al. 1999) of the
surrounding carbonate sr.rata. All these assumptions were tested by experimental modeling by
appiyin,e the NETFATH code (Pt-ulturn et a1. 1994). Numerous runs confirmed that Ca-chio-
ride brines emerging along the western littoral of the iake could eventually cievelop from the
eastern, Mg-rich brines, as the resuit of the previously suggested hydrochemicai processes.
Leachin-e of haiite would yield hi-eh CllBr ratios, whiie very 1ow CVBr ratio vaiues are
chaiacteristic of the brines in Lake Kinneret. Hence, the high bromide concentrations found
ubiquitously in the Kinneret basin, have to be expiained in order tc validate the whole rnodel.
Ocean water has a C1/Br ratio of approxrmately 286. The Jordan River water inputs have iower
raiios, approximateiy 252. The Cl/Br ratio for the Lake Kinneret water deciined from i20-i40
before the Salt Water Carrier diverted the salty springs, to a present vaiue of approximateiy
100-110. The most saiine springs alon_e the western littoral of the lake have Cl,{Br ratios of
about 70-80. Lower ratio values were observed in water seeping into bottom sed"iments. Otirer
springs (Fuliya, tsarbootim and Tabgha) exhibit ratios that fall in the ran-ee from 100-i60 (a11
Br data frorn SIILLER & NISSENBALIM 1996). Therefore, mixing of seawater with fresh water
cannot yield the CVBI that is observed.
High concentrations of Br in the Rift (generating low Cl/Br ratio vaiues) couid be attributed
to severai sources. According to MnzoR & Mnno (1969 a, b) and SrRRiNsrv (1974), Br was
enriched during the progressive stages of salt precipitation from a seawater bocly. This sea
penetrated during the Miocene into the Rift and is purported to have become the precursor of
the Ca-chioride Rift and Dead Sea brines (SreixHonN 1983). These brines which are the re-
sidual fluids after the massive precipitation of halite, will thus be enriched in tsr (tsneITICH &
F{nRuaxN 1963, R.ees 1998), as the distribution coeffrcient between halite and soiution is iess
than 1 (Ztx 1974). According to SrurR & NISSENBAUM (1996). the pore water from sedi-
A Necgene sali body Zg
ment cores taken from boreholes drilled into the bottom of the 1ake, are characterized by low
CllBr ratics. These ratios decrease down the sediment cores. The lowest ratios encountered are
approximately 55, which are significantly lower than values noted in the Tibeias Springs
(70-80)"
There are, however, other processes that can be invoked to explain the high Br-content of
the brines. It is also possibie that Br enrichment can come from the degradation of organic
matter (Ntssexreul,t & Mencenirz I99I) as weil as leaching of Br from basic igneous and
volcanic rocks and metamorphics (Leurnvo & LeltpEN 1987, NuRwrr et ai. 19SS). Thus, the
gabbro and the basalt, which are widely distributed in the subsurface as well as throughout the
whoie area aiong the margins of the rift, may also supply considerable amounts of Br (Raen
1998). F{ence, the dissolution of salt minerais, beyond the stage of initiai halite deposition, can
produce the highly concentrated Mg-Ci brines found along'the western margin of the lake as
well as the ubiquitously iow Cl/Br ratios. The high Br concentration may be additionaliy en-
hanced by gaining Br from the volcanic rocks and/or by degradation of organic matter. Inde-
pendent support fcr the proposed hydrological model is presented in a recent geo-electrical
anaiysis (Golouex et al. 1997) carried out to locate saline water in the bottom sediments of
Lake Kinneret. The salinity models, deveioped to explain the resistivity profiles, show that in
the southern part of the lake there are brine halos with the salinity increasing inwards towards
the direction of the Zemah solid salt body.
Dise usslon
It is entirely plausibie that within the area extending between the Dead Sea in the south and the
surroundings of Lake Kinneret in the north, the Late Miocene sea that intruded the Rift Va11ey,
treft behind it mineralogically differentiated evaporite deposits. The presence of VIg and Br-
rich brines are of potential economic interest, both for the extraction of chemicals and for
balneotheraphy. From a hydrologicai perspective, these brines are of considerabie significance
in that they affect the quality of the Rift Val1ey water sources" SranilisKy \1914) already
proposed that the bines found in the subsurface of the Rift, represent a resiciual product of
evaporated Pliocene seawater, which precipitate<i halite duing the course of its evolution. The
Ca-chloride composition of these brines was attained by doiornitization. This part of the model
appears to be geochemicaiiy plausible. However, StaRtNSKY (1914) su,egesteci that these
brines migrated from their source of origin westwards, up to 100 km from the margins of the
Rift. .Aiong their migration paths, these brines were supposedly diiuted chemically b,v fresh
water. F{owever, the very high hydraulic pressures (hundreds of atmospheres) characterizing
the Ca-chloride brines in the deep boreholes around Lake Kinneret (Devora 2A., Jordan 1 and
Rosh Pina 1; RoseNTsai- 19BB), ciearly indicate the deep confi.nement of these brines. This
evidence wouid contradict the model of SreRINsKy (1914) for brine formation. His model
implied that, in the Rift, the residual evaporated seawater, percoiated into adjacent permeable
formations and accumulated there. No explanations were given for the process of confinement
of these fluids. I4oreover, since the Pliocene, no geological processes have occured., which
could have generated such deep confinement (Rosnxrriar 1988, 1989, ILANr et a1. 1988).
At the time of Sranntsrv'S publication (1974), the existence of the Zemah salt plug was
unkncwn. This relativeiy new evidence simplifies the regional salinization model as it points
to 1ocal brine-forming processes making iong distance migration unnecessary (RosrxrHAL
1988). Moreover, it could possibly offer a different view on the issue of confinement. The new
J
:l
n
,{i
4
'dl
u
I
i:!
rl
,3
;j
:li
fi
;.la
:i.t
8CI A. Flexer et al,
modei impries occurrence of Mg-rich residual fluids in the vicinity of the sait piug. Fresh
water recharged by rainfall on the surrounding areas and driven by high hydrauiic heads,
slowly percolates through rhe ciastic fi1l of the Rift reaching the saiine fluids, diluting them
and piston-pushing them into preferentially penneable strata or along the existing fault sys-
tems. The piston action courd be enhanced both by diapir flow and/or by the ongoing block
movements in the Rift, which would squeeze the liquids and push them into permeable strata'
The existence of intensively evaporated marine brine, beyond the onset of halite precipi-
tation, as suggested by BrncELSoN et ai. (1999), is plausible' However' the downward per-
colation of such concentrated residual liquors percolating through hundreds of meters of
undisturbed, impermeable, previously deposited sequences (some of them containing thick
formations of evaporites) is, geoiogically infeasiblt' 9n the other hand' the 637C1 and 63aS
data of BERGELS.N et ar. (1999) are consistent with the extreme evaporative stages of our
model.
Theproposedgeochemicalmodelcanbeverifiedifitcanalsoexplaintheconcomitant
isotopic changes to both 2H and rsO, that occur during each step in the evolving water/brine'
The model presented here would initially have the descending groundwater dissolve the salt
from the diapir,and then rise. The most saline and hottest spring (the Tiberias Roman Spring;
Ger et al. i969) should be the isotopicaliy, closest surface expression of this process. Ideally,
in the next step, the isotopic composition of other saiine springs or saline groundwater should
be estabiished by mixing of the lepresentative Roman Spring water or the primary brine with
locai fresher water (Lake Kinneret watef or loca1 ground water)' Thus' all of the relevant re-
gional waters strouri falr aiong isotopic mixing rines if the solid sait body can be considered as
a primary saiinity source. The isotopic compoiition of the freshwater end members-precipita-
tion,localgroundwatersandLakeKinneretareknown'asarethevaluesofthemoresaline
springs and groundwater (Ger et aI. L969, Br,nCBr-SON 1999), which wouid constitute the
intermediate group or evolved mixtures. A't first glance' defining the isotopic influence of the
solid sait *uy upplu, problematic. Dissoiution of sa1t, again oniy as a first consideration' may
bethoughttochangethesaiinitybutnottheisotopiccomposi.tionofthedissolving
groundwater. This, hJwru"r, would hold true only if the diapir were composed of halite' The
entireprecedingchemicaiargumentwasmadetoestabiishthepresenceofthemostsolubie
sulfate and chloride mineral saits of K and Mg' These salts began to precipitate when the
original sea water had been reduced by evaporation to gnlV t '54Vo af its original voiume (Ma-
soN 1964). What characterizes these .uupotut" minerals is that most of them do contain rem-
nants of the water from which they prr.ipitut.d. This evaporative signature shouid be pre-
served in the water of hydration that.ryrtuigr.t with the common evaporate minerals' such as
epsomite iNngSOnZH,O;, h""uhydrite (MgSOo6HrO)' kieserite (MgSO4'H'O)' bischofite
(MgClr6Hro) andcarnaiite 6UgCL 6HrO)' t1',11G highly saline evaporative environments'
the isotopic signature should U" ,rniqo"i The 6,rsO values should be enriched' though only
moderateiy positive. Values not unlft those of the Dead Sea (6 ']Ou-o*, =+4'5Voo) may be
expected; but, the 62H values, which started out being enriched in the evaporating brine' un-
dergo a reversal in the later stages and can end up becoming quite depleted, especially com-
pared to the 6rso vaiues. The isotopic reversals are caused by a decreased activity of the water
due to the high solute content. This leads to reducing the humidity contrast between the bound-
arylayerandtheatmosphere(e.g.Sornn&GAT1g15),Additionalisotopicfractionationbe-
tween the saline solution and the water of hydration can accentuate the separation between r8O
and 2l{ that is preserved in the water of hyiration' For example, the equiiibrium fracti'onation
A Neogene sait body Bf
factor (u) between brine and the water of hydration of gypsum has been determined (Sonen
I9l5) both for deuterium (cro=0.980) and for ttO (s o_,r= 1.004). Water of ,,anomalous" iso-
topic composition, but very similar to that of the expected values have indeed been encoun-
tered in the RiftValley, at the Mount S'doin diapir. Moreover, these brines are in direct contact
with the sait body, which has been recognized (as reported above) to contain evaporative min-
erals beyond the stage of initiation of halite precipitation. The brine from the deep S'dom-1
well exhibits 6"O,,-o*f+4.4%o and 6tH1,*o*f -28%o, while the Maayan Hareach seepage ex-
hibits 6t8o1.*o*1=+2.I%o and E'H1.*o *t= -29%o (data from GAT et al. 1969, unfortunately the
salinities of the brines were not reported). These values are obviousiy not pure water of hydra-
tion; but probabiy represent difrerent degrees of dilution with meteoric water.
A plot of these samples, along with ail of the relevant water types in the area. on a 62H
versus 6180 diagram (Fig. 6) enables the mixing relationships to become apparent. In this
diagram (modified from Get et al. 1969) the brine from Mount S'dom is extrapolated to repre-
sent the expected isotopic values derived from the leaching of the Zemah-l salt body. This is
oniy an approximation, because although the isotopic values of precipitation in the l{egev also
lie along the Eastern Mediterranean Water Line (62H =8 618O+22%o) the precipitation is less
depleted compared to that of the Hills of Galiiee. By drawing a iine that connects the S'dom-1
water to the fresh eastern Galilean aquifer, the isotopic relationship of the prirnary brine can be
Locol meteoric
woter line
Ic stern
oquifer Rom cn
spring,^@-
s-A
-t?
Lo ke K in nerei
Jordon
River M oayc nHarecch
-.--j- 1
I
n
l
S'dom-i
ern
fer
rt
^r
ui
We
oq -4-20
8l8c (%")
Fig. 6. l8O-Deuterium relationships of the saline and fresh water sources in the Lake Kinneret area. The
isotopic values for Lake Kinneret represent an average year, not of drought nor especially rainy. The
S'dom-i brine is taken to represent an approximation of the isotopic values to be expected from the
dissolution of the Zemah- I salt body by ground water.
A mixing line connecting the S'dom-1 values to the local Meteoric line (Eastern Galilean aquifer) is
denoied by an arrow and the number 1. This line defines the primary brine. An arrow and the number 2
Cencte a second mixing line connectin-e Lake Kinneret to the point of intersection with primary brines.
The saline \i/ater sources fail along this line. Other symbols are: small open circles for sweet water
springs; large open circles for aquifers; triangles for the mineral springs of ihe Fulya area; filled circles
for the Tiberias mineral springs; fllled squares for the minerai springi along eastlrn shore region and
stars for the springs around the Mt. S'dom area.
-o
^\
O
(i.)
40
20
U
-lu
-40
-ou 4
-A
-8
----r
A. Flexer et al.
established. This line is an isotopic mixing line connecting two end members. trt also repre-
sents a mixing of salinity between these same two end rnembers, the rain and the Zemah-1 salt.
Thus, not only will dissolution of these evaporative minerals change the chemicai compcsition
of the water but they will aiso impart to it a highly evaporative signature. In Fig. 5 this line (i)
falls beneath the point of the Tiberias Roman Spring. However, if a second line (2) is drawn
from Lake Kinneret to intersect the primary brine (1), a second mixing line is established, both
isotopically and of salinity. This iine passes through the Tiberias Roman Spring sample. This
means that after the brine has been created by the leaching of the evaporite bcdy by
groundwater, it mixes with the water of Lake Kinneret as it moves upwards. This causes an
isotopic shift and a dilution of the brine. If the second mixing line (2) is drawn, the saline water
sources (from Gnt et a1. 1969) also fall along or close to ihis line. Indeed, ail of the saiinity and
isotopic compositions of the water sources of the Kinneret basin can be explained by the mix-
ing of meteoric water, Lake Kinneret water and the primary brine. Using the present model,
both the chemical distribution of the springs is explained as well as the distribution of the
stable isotopes in the water sources.
e ene Husions
The new evidence presented in this paper implies that salinization processes invoiving both
groundwater and lake water, are caused by two different brines. The saline Vlg-rich
groundwater emerging along the eastern, southern and southwestern shores of the lake were
formed by fresh water percolating into the Rift, leaching the salt piug and rising as the water
becomes heated. Along the western and southwestern shores of the lake thennomineral waters
emerge which could be dilution products of a deep-seated Ca-chioride brine. By ccnsidering
possible water/rock interaction processes, it is not excluded that the "westerit" Ca-chiorioe
brine could have geochemically evolved from the "eastefit" M-e-rich brine.
The new evidence indicates the existence of two brines differing in their chemical cctiipcsi-
tion and controlleii by dlfierent hydrauiic mechanisrns. F{ence, the manageinent of giounowar"er
and lake water salinities needs tc consider differeni methods and rneasures for the speciflc
hydrogeolo-eicai regimes of the twc, possibly -eeochemicaily interconnected, brines.
The data presented here would also suppcrt the hypothesis that difiusion of briaes across
the sediment-water interface tarkes place over the entire lake bottoin and is the major source cf
chloide input today (Srnt-nn 1994, SrrlLER & NisssNe,\uM 1995).
The Zemah satrt deposits may offer the key to understanding the salinizaiion processes af-
fecting the whole Nationai Reservoir. This identification of the salt body as being the possibie
primary brine source suggests a method for preserving the water quality of the 1ake. If the
leaching of this salt by inflowing fresh recharge water can be diminished, the resultant brine
formation will be reduced. Therefore, it is suggested to control the amount of deep draina_ee of
fresh recharge water from the surrounding areas, to prevent its reaching and leaching of the
salt body and the subsequent upward movement of the newly formed brine.
ffiefcrecrces
BEN AvRAHAvt, 2., HaNEL, R. & VnLncnR, F{. (1978): Heat flow through the Dead Sea rift. - Marine
Geol.28: 253-269.
I
f
A Neogene salt body Bg
BeN Avneruu, 2., SHoHAM,.Y., KLED.I, E., MICIfiLSON, H. & SERRUYA, C. (1980): Magnetic suwey of
Lake Kinneret - Central Jordan Valley, Israel. - Marine Geophys. Res. 4: 257-276.
BEN AvRAH ANI,Z., GINzsunc, A. & Yuvat, Z. (198I): Seismic reflection and refraction investigation of
Lake Kinneret - Central Jordan Valley, Isiael. -Tectonophysics 80: 165-181.
BsN AvnaruM, 2., SHALIV, G. & Nun, A. (1986): Acoustic reflectivity and shallow sedimentary struc-
ture in the Sea of Galilee - Jordan Valley, Marine Geol. 70: 175-189.
Bpn Avnarnv, 2., TEN-BRINK, tL, BELL, R. & RszwtKov, M. (1996): Gravity fie1d over the Sea of
Galilee: evidence for a composite basin along a transform fault. -J. Geophys. Res. 10tr: 533-5M-
BEN AvRAHAM,Z., ANnr, G., GouN, A. & Bncnq,Z.B. (1990):The bathymetry of Lake Kinneret and
its structural significance. - Israel J. Earth Sciences 39: 71-84.
BnNorR, F. (1974): The Geology of Jordan. - Contributions to the Regional Geology of the Earth, Vol-
ume 7, Gebriider Borntraeger, Berlin.
BERGELSoN, G., NATrv, R. & BEIN, A. (1999): Salinization and dilution history of ground water dj.s-
charging into the Sea of Galiiee, the Dead SeaTransform, Israei. -Applied Geochemistry 14: 91-118.
BoorNe, M.W. Jr. & Joms, B.F. (1986):A quantitative chemical-mineralogical characterization of natu-
ral waters. - U.S. Geological Suwey, Water Resources Investigations, Rep. 86-4086.
BRArrcH, O. & HennMAI{t{, A.G. (1963): Zur Geochemie des Broms in salinaren Sedimenten, Teil 1:
Experimentelle Bestimmung der Br-Verteilung in verschiedenen nati.irlichen Salzsystemen. -
Geochim. et Cosmochim. Acta 27:36L-39I.
BSRKE, J.A., CeMpeELL, R.L. & Scuumr, A.W. (1963): The Litho-Porosity cross plot. -Tenth Annual
Symposium of the Society of Professional Analysts: 271-301.
EpppleauM, L. & BBtt AvnetrANt, Z. (1997): Detail interpretation of magnetic anomalies in Lake
Kinneret. -Israel Geoiogical Society Annual Meeting (abstracts), Kfar Giladi, Israel, p. 17.
FoLrNreN,Y. & Yuval, Z. (1976):Aeromagnetic map of Israel, 1:250000. -Israel lnstitute for Petroleum
Research and Geophysics, Holon.
GAT, J.R., MAZoR, E. & Tzun,Y. (1969): The stable isotope composition of mineral waters in the Jordan
Rift Valley, Israel. - J. of Hydrology 7: 334-352.
G,qgnNrcr-,Z.,ZIK,I. & FnsuNp, R. (1981): Active faulting in the Dead SeaRift. -In: FRpuND, R. &
GanrLrr'rKEI-, Z. (eds): The Jordan-Dead Sea Rift. Tectonophysics 80: 1-26.
GNzeuRc, A. & BeN AvRAHAM, Z. (1985): Sffucture of the Sea of Gaiilee graben, Israel, from magnetic
measurements. - Tectonophysics 126: 152-164.
G9LDMAN, M., YzERSKT, VI., HuRwnz, S. & GvRTZMAN, H. (1997): Locating the salt water in the
bottom sediments of Lake Kinneret by geo-electrical methods. - Israel Water Commission Rep. 907/
154196 (in Hebrew).
Golosmr,mr, M.J., ARAD, A. & Nggv, D. (1967): The mechanism of the saline springs in the Lake
Tiberias depression. - Geol. Survey of Israel Bulletin 45, 19 pp.
GcplEN, M. (1991): The salinity of Lake Kinneret. -In: GopHsN, M. & ORrnNssRc, Z. (eds): Salinity of
Lake Kinneret: reality, objectives and connections to soil and soil managenent. - Lake Kinneret
Authority, Zemach, Jordan Valley lsI33z,Israel, 18-26 (in Hebrew).
Gvnrzuerq, H., GARVEN, G. & GvnrnnaN, G. (1997):Thermal anomalies associated with forced and
free ground-water convection in the Dead Sea Rift Valley. - Geol. Soc. of America Buil. 109: 1167-
tr16.
irei'u. S., RosgNrriei-, E., KRoNFELD, J. & FLBxER, A. (1988): Epigenetic doiomitization and iron
mineralization along fauits and their possible relation to the paleohydrology of Israel. - Applied
Geochemistry 3: 487 -498.
J+hES, B.F. & BoDNlE, M.W. (1987): Nomative sait characterization of natural water. - In: FnIrz P. &
Fts*tPE, S.K. (eds): Saline water and gases in crystalline rocks. Geological Association of Canada"
Special Paper 33: 5-18.
Kr.cusropr, - (1967): Introduction to Geochemistry" - McGraw-!{ill Book Company, New York,
721 pp.
_I
84 A. Flexer et al.
Lauenuo, P.W. & LAMpEN, P.H. (1987): Brackish and saline ground waters in Finland. - In: FRrrz, P. &
FRApE, S.K. (eds): Saline water and gases in crystalline rocks. Geological Association of Canada
Special Paper, 33: 103-109.
MARCUS, E. & Sracnn, J. (i985): The sedimentary-magmatic sequence of the Zemah-l weil (Jordan-
Dead Sea Rift,Israel) and its emplacement in iime and space. - Israel J. of Earth Sciences 34: 1-10.
MASON, B. (1964): Principles of Geochemistry, second ed. - John Wiley & Sons, New York.
MAZoR, E. & MEno, F. (1969a): Geochemical tracing of mineral and fresh water sources in the Lake
Tiberias basin, Israel. - J. Hydrol. 7:276-317.
- - (1969b): The origin of the Tiberias-Noit water association in the Lake Tiberias Basin, Israei. - J.
Hydrol. 7:318-333.
Msncaoo, A. & Meno, F. (1984): The salty inflows to the Kinneret. - Tahal, Water Planning for Israel
(Ltd.), Report 01/84148,l44pp. (in Hebrew).
MtcneLsoN, A., FLEXER, A. & ERBz , Z. (1987): Comparison of the eastern and western sides of the Sea
of Galilee and its implication on the tectonic nature of the Jordan Valley. - Tectonophysics 141: i 51-
r68.
NEEV, D" & ElaEny, K.O. (1967): The Dead Sea: depositional processes and environment of evaporites.
- Geological Survey of Israel, Bull. 4L, I47 pp.
NrssBNsaulvr, A. & MaRceRtrz, M. (1991): Bromine-rich groundwater in the Hula Valley, Israel. -
Naturwissenschaften 7 8: 217 -2L8.
Nurur, P.A., KurcoNEN, I.T. & Laurnvo, P.W. (1988): Geochemistry and origin of saline
groundwaters in the Fennoscandian Shield. -Applied Geochemistry 3: 185-203"
PlulrtrrpR, N.L., PnrsrEMoN, E"C. & PARKHURST, D.L. (1994):An interactive code (NETPAIH) for
modeling Net geochemical reactions along a flow PATH, version 2-0. - U.S. Geological Survey,
Water Resources Investigations Report, 94-4169, Reston, Virginia.
RAAB, M. (1998):The origin of the evaporites in the Jordan-Dead SeaValley in view of the evolution of
brines and evaporites during seawater evaporation. - Geological Suwey of Israel Report GSV1/98,
Jerusalem, Israel, 138 pp.
RoseNrHal, E. (1988): Ca-chloride brines at common outlets of the Bet Shean-Harod multiple aquifer
system, Israel. - J. Hydrol. 97: 89-106.
(1,997): Thermomineral waters of Ca-chloride composition: review of di_agnostics and of brine evolu-
tion. - Environmental Geol. 32:245-250.
SALTZMAN, U. (1964): The geology of the Tabgha-Huqoq-Migdai area. - Report, Tahal, Water Planning
for Israel Ltd., Tel-Aviv,55 pp. (in Hebrew, English summary).
Scnuur,raN, N. (1959): The Geology of the Central Jordan Valley. - Ph.D. thesis, The llebrew University,
Jerusalem, 103 pp. (in Hebrew, Engiish abstract.)-
Scm-Ew, N. & Sptvar, D. (1986):A log-interpretated mineralogical composition for solution mining of
underground caverns in Mount S'dom. - Israel J. of Earth Sciences 35: 136-i48.
Scuogr-lrR, H. (1956): Geochemie des eaux souterraines. Application aux eaux de gisements de petrole.
- Soc. Ed. "Technip", Paris, 213 pp.
Srnuv, G. (1991): Stages in the tectonic and voicanic history of the Neogene basin in the lower Galilee
and the valleys. - Israei Geoi. Survey Current Research, 1-94.
SrMoN, A. & Meno, F. (1987): The influence of lake level on the salinity concentrations in Lake Kinneret
waters. -In: GAL, Y. & SHtLo, E. (eds): The levels of Lake Kinneret in relation to water problems, 41-
53 (in Hebrew)"
SN.{[H, S.V., SpRnuyA, S., GemvaN, Y. & BswreN, T. (1989): Internal sources and sinks of water, P, N,
Ca, and Cl in lake Kinneret, Israel. - Limnol. and Oceanogr.34 1202'1213.
Soren, Z" (.I975): Isotopic Composition of Hydration Water of Gypsum. - Ph.D. thesis, The Weizmann
Institute of Science, Rehovot, 108 pp. (in Hebrew, with English abstract).
Sorpn, Z. &.GAT, J.R. (i975): The isotope composition of evaporating brines: Effect of the isotopic
activity ratio in saline solutions. - Earth and Planetary Science Letters 26:179-186.
A Neogene salt body gs
STARINSKY, A. (I974): Relationship between Ca-chloride brines and sedimentary rocks in Israel. - ph.D.
thesis, The Hebrew University, Jerusalem, Israel, 176 pp. (in Hebrew English abstract).
SrenruonN,I. (1983): In-situ salt precipitation at the Dead Sea. - Limnol. and Oceanogr.2S:580-593.
StrLtER, M. G97$: Rates of transport and sedimentation in Lake Kinneret. - Ph.D. thesis, Weizmann
Institute of Science, Rehovot, 240 pp. (in Hebreri, English summary).
- 0994): The chloride content in pore water of Lake Kinneret sediments. - Israel J. of Earth Sciences
43: l'79-185.
STILLER, M. & NisseNBAIr4, A. ( 1996): CllBr ratio in pore water from Lake Kinneret (Sea of Galilee). -
Israel J. ofEarth Sciences 45: i53-160"
VENGOSH, A., StaruNsKY, A., KoLoDIvy, Y. & Cnwes, A.R. (1994): Boron isotope geochemistry of
thermal springs from the northern RiftValley. Israel. - J. Hydrology 162l' 155-169.
YBI-lnq-DRoR, A. (1984): The relationship between the Talme-Yafe and Judea Groups-Interpretation
from geophysical logs. - M.Sc. thesis, Tel-Aviv University, Tel-Aviv, 116 pp. (in Hebrew, English
abstract).
ZtK,I. (1974): Sedimentology and bromine geochemistry of marine and continental evaporites in the
Dead Sea area. - ln: Coocax, A.H. (ed): Fourth Symposium on Salt. Houston, Texas. Northern Ohio
Geological Society 13: 349-361.
