Tsur master

Tsur master
Department of Physics and Astronomy
Ruprecht-Karls-University Heidelberg
Master thesis
in Physics
submitted by
Neta Tsur
born in Ramat Gan, Israel
2013
Noble gas isotopic signatures in thermal waters of the
Dead Sea Transform, Israel
This Master thesis has been carried out by Neta Tsur at the
Institute of Environmental Physics
under the supervision of
Prof. Dr. Werner Aeschbach-Hertig
“There is running water and not a seasonal stream,
the Jordan River is proudly flowing”
Somewhere in the Valley, Noah Warschauer
Abstract:
Noble gas isotope composition (especially helium) in thermal waters from different areas
along the Dead Sea Transform in Israel reveals local intrusions of volatile gases and heat
flux from the mantle into the crust. The distinct 3He/ 4He ratios in the atmosphere, crust
and mantle enable one to separate the total helium concentration into mantle, crustal
and atmospheric components. Helium isotope ratios of all sampled waters exceed the
typical crustal ratio, indicating contributions of mantle derived helium to the total helium concentration. A clear trend can be observed in 3He/ 4He ratios from different areas.
Northern samples show higher 3He/ 4He ratios than southern ones. Noble gas temperatures (NGTs) were used to determine the atmospheric helium component. Most of the
samples contain less than 3 % atmospheric helium, whereas the mantle derived helium
component ranges from 2 % to 38 %. In addition to helium, the origin of CO2 in the water is examined. Measurements of δ 13 C indicate no mantle derived CO2 . Furthermore,
stable isotopes data (δ 18 O and δ 2 H) show no evidence of mantle derived water or water
from reservoirs exceeding 100 ◦ C.
Zusammenfassung:
In dieser Arbeit wird die isotopische Zusammensetzung von Edelgasen (vor allem Helium) in Thermalwasser entlang des israelischen Teils des Jordangrabens analysiert. Da
sich das 3He/ 4He-Verhältnis von Atmosphäre, Kruste und Mantel deutlich unterscheidet, lassen sich die Beiträge der einzelnen Edelgasreservoirs voneinander trennen. Die
Untersuchungen zeigen, dass Gase und Wärme aus dem Erdmantel in die Erdkruste eindringen. Alle Wasserproben weisen ein 3He/ 4He-Verhältnis auf, das deutlich höher ist
als jenes der Kruste; folglich stammt zumindest ein Teil des Heliums aus dem Mantel.
Ferner ist in den 3He/ 4He-Verhältnissen ein klarer geographischer Trend erkennbar: Im
Norden ist das Wasser deutlich stärker vom Manteleinfluss geprägt als die Proben aus
den südlicheren Regionen.
Der atmosphärische Heliumanteil wurde mit Hilfe der Edelgastemperaturen (NGT) bestimmt. Die meisten Proben weisen weniger als 3% atmosphärisches Helium auf, wohingegen der Mantelanteil im Bereich von 2% bis 38% liegt.
Zusätzlich wurde die Herkunft des im Wasser gelösten CO2 untersucht. Dabei zeigen die
Messungen von δ 13 C, dass das CO2 nicht aus dem Mantel stammt. Des Weiteren ergibt
die Analyse der stabilen Isotope (δ 18 O und δ 2 H), dass auch das Wasser nicht aus dem
Mantel kommt und nie heißer als 100 ◦ C war.
Contents
1 Introduction
11
2 Theory
2.1 Solubility of noble gases . . . . . . . . . . . . . .
2.2 Components of noble gases in groundwater . . .
2.3 Noble gases as paleoclimate tracers . . . . . . . .
2.3.1 Excess air models . . . . . . . . . . . . . .
2.3.2 Determination of noble gas temperatures
2.4 Mantle derived isotopes in groundwater . . . . .
2.4.1 Helium isotopes: 3He and 4He . . . . . . .
2.4.2 CO2 and δ 13 C . . . . . . . . . . . . . . .
2.4.3 Stable isotopes: δ 18 O and δ 2 H . . . . . .
.
.
.
.
.
.
.
.
.
15
15
17
18
18
22
22
22
29
31
3 The
3.1
3.2
3.3
3.4
Dead Sea Transform
Geographic setting .
Tectonic setting . . .
Hydrogeology . . . .
Sampling sites . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
33
34
36
39
4 Methods
4.1 Water characteristics . . . . . . .
4.2 Noble gases: Mass spectrometry
4.2.1 Sampling . . . . . . . . .
4.2.2 Measurement . . . . . . .
4.2.3 Data evaluation . . . . . .
4.3 Isotopic composition . . . . . . .
4.3.1 Tritium . . . . . . . . . .
4.3.2 δ 13 C . . . . . . . . . . . .
4.3.3 Stable isotopes . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
41
41
41
41
42
43
45
45
45
46
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
mantle derived helium
. . . . . . . . . . . . .
. . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
47
50
54
55
56
56
60
63
67
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5 Results
5.1 Noble gas temperatures . . . . . . . . . . . .
5.2 Helium isotopes . . . . . . . . . . . . . . . . .
5.3 Helium components . . . . . . . . . . . . . .
5.3.1 Tritiogenic helium: . . . . . . . . . . .
5.3.2 Terrigenic versus atmospheric helium:
5.3.3 Separation of atmospheric, crustal and
5.4 Carbon isotopes . . . . . . . . . . . . . . . . .
5.5 Stable isotopes . . . . . . . . . . . . . . . . .
6 Discussion
69
3
4
6.1 Spatial distribution of He/ He . . . . . . . . . . . . . . . . . . . . . . . . . 69
9
Contents
6.2
6.3
6.4
Mantle, crustal and atmospheric components distribution
The geothermal structure of the Dead Sea Transform . . .
Other isotopes as tracers for mantle flux . . . . . . . . . .
6.4.1 Carbon isotopes . . . . . . . . . . . . . . . . . . .
6.4.2 Stable isotopes . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7 Summary and outlook
A Appendix
A.1 Ostwald solubility calculation . . . . . . . . . . . . . . . . . . . . . .
A.2 Mixing ratios of noble gases . . . . . . . . . . . . . . . . . . . . . . .
A.3 Stable isotopes notation . . . . . . . . . . . . . . . . . . . . . . . . .
A.4 Locations in Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A.5 A detailed description of the sampling sites and the sampling process
A.6 Neon and argon isotopic ratios . . . . . . . . . . . . . . . . . . . . .
A.7 Noble gas concentrations of the second and third measurement runs
A.8 Additional figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
70
71
74
74
75
77
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
79
79
79
79
80
82
91
91
94
List of Figures
97
List of Tables
99
Bibliography
103
10
1 Introduction
The Earth was formed approximately four and a half billion years ago from a cloud of
dust and gases that generated a melted sphere. This sphere started to cool down and
different layers were formed. First, the outer thin crust was created, later the upper and
lower mantle and finally the inner core (Figure 1.1). At the time the mantle was formed,
it trapped a great amount of volatile gases. During the first 50 million years from Earth
formation, massive degassing occurred in the mantle and partly continues at the present
time [Allègre et al., 1987]. Some of the escaping volatiles infiltrate into the crust during
the evolution of igneous rocks and can be dissolved by local groundwater. Among them
are CO2 and noble gases, which can be measured and provide information of a regional
crust-mantle interaction.
Crust
Upper mantle
Lower mantle
Outer core
Inner core
Figure 1.1: Interior layers of Earth (figure is not to scale).
There are five stable noble gases in nature: Helium (He), neon (Ne), argon (Ar), krypton
(Kr) and xenon (Xe). They belong to a group of elements that have a full outer shell of valence electrons. Hence they are chemically inert and conservative. In addition, noble gases
are found in rare amounts, which makes them ideal tracers in environmental studies.
Noble gases in groundwater usually originate from three main reservoirs: The atmosphere,
the crust and the mantle. Each reservoir contributes a different fraction to their total
dissolved concentration in groundwater. The fact that each fraction from each reservoir
retains its composition is a powerful tool, which enables to separate the measured noble
gas concentrations to atmospheric, crustal and mantle components. The distribution of
11
1 Introduction
these components provides information about the groundwater recharge conditions and the
crustal environments it was circulating through, since it had infiltrated into the ground.
The composition of the atmospheric component can be used for paleoclimate records. Since
the solubility of noble gases depends on the temperature, the dissolved concentration of
noble gases from atmospheric origin is attributed to the air temperature in the recharge
zone at the recharge time. This effect is called Noble Gas Temperature (NGT) [Mazor,
1972].
Another use of noble gases in groundwater is to identify volatile fluxes from the mantle into
a certain region of the crust, in which the water circulates. Thermal groundwater usually
reaches large depths, where intrusion of volatiles from the mantle is likely to occur. Helium
isotopes are commonly used to trace these intrusions. The isotope 3He mostly originates
from the mantle and does not have a large surface inventory. Moreover, each of the helium
reservoirs (mantle, crust and atmosphere) has a different fractionation ratio of 3He/ 4He,
thus making it possible to distinguish between mantle, crustal and atmospheric derived
helium. Helium isotopes and other volatiles found in groundwater provide important
information about mantle evolution and crust-mantle interactions, in form of geotectonic
activity, volcanism and advective heat transfer [Sano and Wakita, 1988 and Mamyrin
and Tolstikhin, 1984]. The geothermal state of the crust provides information of thermal
energy resources, which are of environmental and economic importance.
Other isotopes can be analysed as well in order to detect other volatiles from the mantle.
For example, 13C is used as a tracer for mantle derived CO2 . The stable isotopes 18O and
2H are often used as tracers of interaction between groundwater and local rocks. Such
interaction occurs in high temperatures, which are sometimes caused by enhanced heat
fluxes from the mantle. However, most of the volatiles have a high chemical and biological
reactivity. Hence, it is more complicated to identify their fluxes from the mantle, compared
to using noble gases.
This thesis analyses noble gases and other isotopes dissolved in thermal waters along the
Israeli part of the Dead Sea Transform. It has been carried out parallel to the work
of Kaudse [2014], who analyses thermal waters along the Jordanian part of the Dead
Sea Transform. There exist former studies that analysed noble gases along the Dead Sea
Transform in Israel. Mazor et al. [1973] measured noble gases in Hammat Gader springs in
the Yarmouk Valley to identify mixing between groundwaters with different temperatures
and salinities. Herzberg and Mazor [1979] measured noble gases in cold springs along the
Dead Sea Transform and other parts of Israel to determine paleotemperatures. Ratios of
3He/ 4He along the Dead Sea Transform were measured by Bergelson et al. [1999] on the
western side of Lake Kinneret and Lange et al. [2008], who measured 3He/ 4He ratios in the
vicinity of Jericho. Both used the helium isotopes as tracers of groundwater salinization.
Very recently, a detailed study of helium isotopes from springs and wells along the Dead
Sea Transform was published by Torfstein et al. [2013]. The data were measured 15 years
ago (1995–1998) and were analysed in order to find mantle derived helium.
In this thesis, thirteen samples were taken from thermal wells and springs in Israel in
December 2012. The sampling sites are located in different regions from the Hula Valley
in the north, to Arava Valley in the south. Concentrations of noble gases and their isotopes,
as well as other isotopes of 13C, 18O and 2H were measured. They were used for identifying
volatiles from the mantle in the groundwater and determine noble gas temperatures. The
main purpose of this work is to analyse the 3He/ 4He ratio regarding to infiltration of
mantle derived helium into the groundwater. Besides, 13C is used to determine the origin
12
of CO2 , which is one of the main volatiles still escaping from the mantle. Furthermore, 18O
and 2H are used to identify temperature anomalies in the crust, which are often caused by
heat transfer from the mantle into the crust.
13
2 Theory
2.1 Solubility of noble gases
The concentration of a gas dissolved in water under equilibrium conditions is proportional
to its concentration in the gas phase as defined by the so called Henry’s law [AeschbachHertig and Solomon, 2013]:
cig = Hi (T, S) · ciw
(2.1)
cig and ciw are the concentrations of species i in [mol/l], in gas and liquid phases, respectively. Hi is called Henry coefficient and is dimensionless. The concentration of a
substance in a gas phase can be converted to partial pressure, thus Henry’s law can also
be written as:
pi = Hi0 · ciw
(2.2)
pi is the partial pressure of gas species i and Hi0 is Henry coefficient in units of [atm·l/mol],
so that Hi0 = Hi · RT . The pressure pi is calculated from the concentration in a gas phase
via the equation of state:
p=
n
RT = cg RT
V
(2.3)
n is number of moles of the gas, V is the volume occupied by the gas, R is the Gas Constant
and T is the temperature.
Solubility is defined as the ratio between the concentration of a substance in a liquid phase
and the concentration of it in a gas phase:
ki =
ciw
1
=
cig
Hi
(2.4)
Higher solubility increases ciw . There are number of definitions of solubility coefficients,
deduced from different parameters of the system. For example, Ostwald solubility is
defined as the ratio between the volume of a substance in a gas phase and the volume of
it in a liquid phase:
Li =
Vig
1
=
Viw
Hi
(2.5)
Another form of Ostwald solubility is:
λi =
ciw
1
= 0
pi
Hi
(2.6)
λi is given in units of [mol/atm·l].
15
2 Theory
Dissolving is a thermodynamic process which depends on temperature, salinity and pressure:
ks = ks (T, S, p)
(2.7)
The relationship between solubility and temperature can be determined by Van’t Hoff
equation. The following relation for Ostwald solubility was obtained by Benson and Krause
Jr. [1976]:
1
1
1
= ln (Li (T, S = 0)) = a0 + a1 · + a2 · 2
(2.8)
ln
Hi
T
T
The parameters a0 , a1 and a2 depend on the gas species and are calculated for temperatures
in [K]. They are given in Appendix A.1 for each noble gas.
Higher water salinity reduces the solubility, which is called “salting-out” effect. This effect
can be neglected for very low salinities, i.e. fresh water. The decrease of the solubility
with the temperature and salinity is demonstrated in Figure 2.1.
(a) Solubility versus temperature.
(b) Solubility versus salinity.
Figure 2.1: Solubility of the most abundant atmospheric gases and noble gases, adapted
from Cook and Herczeg [2000]. Li (T ) and Li (T, S) refer to the solubility of
each gas species i (see legend) in temperature T and salinity S. They are
normalized to the solubility value in T = 0 ◦ C. Adapted from Kaudse [2014].
The dependency of solubility on pressure is indirect, since the partial pressure of a gas,
pi , from Equation 2.2 depends on the temperature and on the water vapour pressure,
which also depends on the temperature. Since the boundary layer between the water and
air, where gas exchange takes place, is saturated in water vapour, its pressure should be
subtracted from the dry air pressure to obtain the partial pressure:
pi (z, T ) = (p(z) − ew (T )) xi
(2.9)
xi is atmospheric mixing ratio of gas i, relative to dry air. The mixing ratios of noble
gases in the atmosphere given in in appendix A.2. The air pressure on the water surface
altitude, p(z), is given by the barometric Equation [Roedel and Wagner, 2010]:
z
p(z) = p0 · exp −
(2.10)
zs
RT
zs =
Mg
16
2.2 Components of noble gases in groundwater
The scale height, zs = zs (T ) depends on the temperature, but is usually approximated to
be zs = 8000 m. M is the molar mass of air. The water vapour pressure, ew , is determined
empirically.
2.2 Components of noble gases in groundwater
Noble gases dissolved in groundwater origin from different sources and their total concentration is composed of contributions of the different sources. Figure 2.2 shows that
all noble gases origin from the atmosphere and that helium has additional contributions
from non-atmospheric sources. Separation of the different components of the noble gases
in a groundwater reservoir may give information about the characteristics of the reservoir,
such as time scales (dating) and paleotemperatures. The latter will be discussed in Section
2.3.2.
concentration normalized to Cieq
2.0
tritiogenic
terrigenic
excess air
equilibrium
1.5
1.0
0.5
0.0
³He
⁴He
Ne
Ar
Kr
Xe
Figure 2.2: Noble gas components in groundwater, normalized to air equilibrated water.
Note that all gases origin from the atmosphere (equilibrium and excess air
components) and helium has additional non atmospheric components. Argon
has a non-atmospheric component too, which is neglected for the purpose of
this work. Adapted from Kipfer et al. [2002].
Equilibrium component: Recharging groundwater water exchanges gases with the atmosphere according to Henry’s law (equations 2.1 and 2.2). Inserting pi from Equation
2.9 into Equation 2.2, the equilibrium component of gas i in water is:
ceq
iw =
(p(z) − ew (T )) xi
RT
(2.11)
If the groundwater is formed from fresh water, such as meteoric water, salting out effect
can be neglected, thus the equilibrium component depends only on the temperature.
Excess air component: Most groundwater samples contain additional dissolved noble
gases with atmospheric composition, i.e. higher than the concentration of the equilibrium
component, attributed to bubbles entrapment in the water [Heaton and Vogel, 1981]. This
17
2 Theory
excess air amount decreases for heavier noble gases, as can be seen in Figure 2.2. There
are number of models that aim to explain and quantify this component, as described in
Section 2.3.1.
Tritiogenic component: Groundwater acts like a closed system for tritium and helium from the moment it looses contact with the atmosphere, thus, their total amount is
conserved with time:
3He
tri (t)
= 3H(0) − 3Htri (t)
(2.12)
Tritium has a relatively short lifetime of T1/2 = 12.43 years [Schlosser et al., 1989], therefore
finding tritium in old water indicates mixing with younger water. The 3H– 3He age, used
for dating, is:
3He (t) 1
τ = ln 1 + 3 tri
(2.13)
λ
Htri (t)
Note that τ is the time since the water was isolated form the atmosphere. The decay
constant is: λ = Tln 2 = 0.056 a−1 .
1/2
Terrigenic component: This terrigenic component refers to noble gases that originate
from the solid Earth and found in helium isotopes and 40Ar. This component is usually
crustal (radiogenic) from nuclear production. Radiogenic components of other noble gas
isotopes are usually very small and can be neglected [Kipfer et al., 2002]. 3He is produced
from Li and 4He is produced by α decay of Th [Mamyrin and Tolstikhin, 1984 and Schlosser
et al., 1989]:
6
Li(n, α) 3H −→ 3He
U, Th −→ X + α
(2.14)
4He
(2.15)
It is important to notice that 3He is also produced from atmospheric tritium, which is
considered to be a different component. In addition, some part of the helium can come
from the mantle. Distinguishing the mantle from the crustal component, one can tell
about the geothermal structure of the environment that the water comes from, as further
explained in Section 2.4. The isotope 40Ar is produced from 40K and may also have
a radiogenic fraction. Further discussion about helium components and mantle derived
helium is found in Section 2.4.
2.3 Noble gases as paleoclimate tracers
2.3.1 Excess air models
There are number of models that are used for separation of the excess air component
from the equilibrium component in groundwater samples. Each model assumes different
formation processes of excess air, as demonstrated in Figure 2.3. The models which are
used in this work and a few similar models are presented in the following text. In all
models helium is usually excluded from the analyse, since it contains non atmospheric
components, making the separation of excess air more complicated. The information in
this section is mostly based on Aeschbach-Hertig et al., 2008 and Aeschbach-Hertig and
Solomon, 2013
18
2.3 Noble gases as paleoclimate tracers
i) Unfractionated excess air (UA) model
AEW & air
complete dissolution
entrapped air
ii) Partial re-equilibration (PR) model
AEW & air
entrapped air
complete dissolution
diffusive outgassing
Multi-step partial re-equilibration (MR)
iii) Closed-system equilibration (CE) model
AEW & air
entrapped air
partial dissolution,
equilibration
water-gas separation
complete dissolution
diffusive outgassing
v) Gas diffusion Relaxation (GR) model
AEW & O2
depleted air
entrapped air
Figure 2.3: Illustrations of the excess air models described in Section 2.3.1. The different
mechanisms of excess air formation in groundwater are shown here. Staring
from the classical model of UA at the top further modified models below.
Adapted from Wieser [2011] and modified by Mayer [2012].
19
2 Theory
Unfractionated excess air model (UA) This model represents the most simplified
case of complete dissolution of entrapped air bubbles. The noble gas concentration in the
groundwater is the sum of an equilibrium component and additional complete dissolution
of air bubbles with the same equilibrium composition, given by Henry’s law (Equation
2.1):
eq
A
cU
iw = ciw (1 + AHi )
(2.16)
The model has one parameter, A = VVwa , which is the volume of dissolved air, Va , relative
to the water volume, Vw . Equation 2.16 can be written for an “effective air fraction” A0 .
Inserting the second term of Equation 2.16 into Henry’s law (Equation 2.2) yields:
eq
A
0
cU
iw = ciw + A xi
p − ew
A0 = A
RT
A0 xi = ceq
AH
=
A
· cia
i
iw
(2.17)
cia denotes the concentration of noble gas i in the gas phase. It is clear from Equation 2.17
noble gases with higher mixing ratios contribute more excess air, therefore argon, which
has the biggest mixing ratio (see Table A.2) has a higher excess air concentration than the
rest of the gases.
Partial re-equilibration model (PR) This model describes a complete dissolution
of the entrapped air, followed by diffusive degassing leading to re-equilibration of the
groundwater with the soil air. The excess air concentration then decreases and changes
its composition due to depletion in light gases.
cPiwR
=
ceq
iw
β D
−FP R D i
N
e
1 + AHi · e
(2.18)
The parameters of this model are: A, which is the volume fraction of dissolved excess
air relative to the water, FP R is the degree of excess air loss after the re-equilibration, Di
denotes the diffusion coefficient of noble gas i in water, β refers to the degree of gas transfer
due to diffusion and can vary between 0.5 to 1 [Aeschbach-Hertig and Solomon, 2013]. Di
is normalized to the diffusion coefficient of neon, which is assumed to origin only from
atmospheric source, as shown in Figure 2.2. Inserting FP R = 0 into Equation 2.18 gives
the limit case of the UA model. As FP R −→ ∞, fractionation of all the noble gases takes
place, so that all the excess air component diffuses from the water and vanishes, therefore
samples that have high FP R values might be close to Air Equilibrated Water (AEW).
A special case of this model is the partial degassing model (PD), where diffusive
degassing acts on the total gas concentration and not only on the excess air component
[Aeschbach-Hertig et al., 2008].
Multi-step partial re-equilibration model (MR) A modified PR model, in which
the re-equilibration takes places in more than one step [Kipfer et al., 2002]. The model is
described by the following equation:
!
n
−nRi
X
eq
eq
MR
−kRi
−Ri 1 − e
(2.19)
ciw = ciw 1 + AHi ·
e
= ciw 1 + AH · e
1 − e−Ri
k=1
Di β
Ri = FM R
DN e
20
2.3 Noble gases as paleoclimate tracers
n is the number of re-equilibration steps, Ri is the exponent defined by the former PR
model. The parameter A is the dissolved excess air volume fraction relative to the water.
The lower limit case of this model is the PR model which is obtained for n = 1.
Closed-system equilibration model (CE) The scenario described by this model is of
air entrapment at the recharge zone, followed by partial dissolution, due to the increasing
hydrostatic pressure, as the groundwater table rises [Aeschbach-Hertig et al., 2008]. The
concentration of the dissolved excess air is:
(1 − FCE ) AHi
eq
eq 1 + AHi
CE
= ciw 1 +
(2.20)
ciw = ciw
1 + BHi
1 + FCE AHi
The concept of the model is illustrated in Figure 2.4. The parameter A = VVwa represents
the total volume of air entrapped in the recharge zone (Va ) relative to the water volume
(Vw ). Note that it is defined here as the fraction of air entrapped at the recharge zone
and is different than the A parameter from the other models, which represents the total
amount of dissolved excess air. B = VVwb is the total volume of the remaining air after
the partial dissolution, relative to the water volume. A and B define the fractionation
Vb
parameter F = B
A = Va . Inserting F < 1 (i.e. B < A) into Equation 2.20 means that
some of the initially entrapped air is dissolved, which corresponds to excess air formation.
Taking F > 1 (i.e. B > A) means there is more air after the dissolution, than entrapped
in the recharge zone, therefore it corresponds to degassing. F = 0 (B = 0) corresponds
to complete dissolution, hence the UA model. CE is the only model presented here,
which assumes a direct partial dissolution without any intermediate step on complete
dissolution.
Figure 2.4: Derivation of the CE model, adapted from Aeschbach-Hertig et al. [2008].
Excess air is first entrapped in groundwater in the recharge zone. The total
volume of the air is Va and the concentration of noble gas i in the air is cia .
The volume of the surrounding water is Vw and the equilibrium concentration
of each dissolved noble gas is ceq
i . As the hydrostatic pressure increases, a
fraction of the air bubbles dissolves until new equilibrium is reached. The
total volume of the remaining air is Vb and the concentration of each noble
gas in the air bubble is cib . The volume of the water changes negligibly. The
concentration of each noble gas, dissolved as excess air component is cw
i and is
given by Equation 2.20.
21
2 Theory
2.3.2 Determination of noble gas temperatures
As explained before, the solubility of noble gases in fresh water (with S=0) depends only
on the temperature. If it is possible to separate the equilibrium component from the
others (Figure 2.2), the corresponding temperature determined from equilibration solubility (Equation 2.8) is the temperature at the groundwater table at the time of recharge,
called Noble Gas Temperature (NGT) [Ballentine and Hall, 1999 and Aeschbach-Hertig
et al., 1999]. Cook and Herczeg [2000] suggest that the NGT corresponds to the ground
temperature of the water or soil in the recharge area.
Helium and argon contain non atmospheric components, therefore their concentrations
must be corrected to atmospheric components for the NGT determination. Since the solubility of helium depends only weakly on the temperature, it is excluded from the determination [Aeschbach-Hertig and Solomon, 2013]. The equilibrium component of argon can
be separated more easily, since most of it comes from 36Ar. Hence, in case the measured
ratio of 36Ar/ 40Ar is smaller than the atmospheric ratio of 0.003378 [Aeschbach-Hertig
and Solomon, 2013], 40Ar is excluded from the NGT determination.
The measured concentration of each noble gas can be written as follows:
ex
cmeas
= ceq
i
i (T, S, p) + ci (A, F )
(2.21)
Here, cmeas
is the measured concentration of species i, ceq
i
i is the concentration of the
equilibrium component, which depends on water temperature, salinity and air pressure
and cex
i is the concentration of the excess air component, which depends on the parameters
of the chosen excess air model, described in Section 2.3.1. A and F in Equation 2.21 are
the parameters of the CE model, which is used for the analysis in this work. Salinity of
S = 0 can be assumed if the groundwater comes from meteoric origin and the pressure can
be calculated from the estimated recharge altitude, with the barometric formula (Equation
2.10). Equation 2.21 can be written for each species i (Ne, Ar, Kr or Xe), making four
equations. Since there are only three unknowns, the system overdetermined.
Ballentine and Hall [1999] and Aeschbach-Hertig et al. [1999] suggest to use inverse modelling instead of Equation 2.21, which cannot be solved. The idea is to find a set of parameters A, T and F , which give the closest modelled concentrations, cmod
, to the measured
i
ones, cmeas
.
The
minimal
difference
between
the
modelled
and
measured
concentrations
i
2
can be found by minimizing χ , defined as:
X cmeas − cmod
i
i
χ2 =
(2.22)
2
σ
i
i
σi is the experimental 1σ error of the species i. The lower χ2 is, the better the fit is.
According to Sun et al. [2010], the same model should be used for the whole data set to
enable comparison of NGTs to present temperature for a paleoclimate record.
2.4 Mantle derived isotopes in groundwater
2.4.1 Helium isotopes: 3He and 4He
Helium has two stable isotopes: 3He and 4He. The second is very abundant and composes
almost 100 % of the total helium inventory, while the first constitutes only ∼10−4 % of
22
2.4 Mantle derived isotopes in groundwater
it (Table 4.1). Helium in groundwater originates from three main reservoirs: The atmosphere, the crust and the mantle. Each of them has a characteristic 3He/ 4He ratio. The
atmospheric ratio 3He/ 4He = (1.384 ± 0.006) · 10−6 [Clarke et al., 1976], usually written
as Ra , is constant due to the long residence time of helium there, which is 105 times
longer than the mixing time of the atmosphere [Lupton, 1983]. Other helium isotope ratios are usually normalized to this constant ratio, for convenience. AEW has a slightly
lower 3He/ 4He ratio of 1.360 · 10−6 , due to slightly different solubility coefficients of 3He
and 4He [Benson and Krause Jr., 1980]. Crustal helium is produced in nuclear reactions
and depends on the concentration of its precursors (see Section 2.2). The 3He/ 4He ratio
of crustal helium ranges from 1 · 10−8 to 3 · 10−8 , corresponding to ∼0.01 Ra [Mamyrin
and Tolstikhin, 1984]. The upper mantle is characterised with the highest 3He/ 4He ratio
of ∼10−5 , corresponding to (8±1) Ra [Graham, 2002]. The fact that these three reservoirs vary widely in their helium isotope composition, enables scientists to differentiate
the origin of fluids.
Helium in the mantle, also called primordial helium, originates from solar nebulae during the early stages of the universe formation. The mantle reservoir of helium has two
main sources: Ocean Island Basalts (OIB) and Mid Ocean Ridge Basalts (MORB), both
contribute to the formation of the oceanic and continental crust. There exist some other
sources of mantle helium, such as diamonds, which their helium signatures can be produced
from dilution with old oceanic crust and sediments [Lupton, 1983]. The generation mechanism of Mid Ocean Ridge Basalts (MORB) and Ocean Island Basalts (OIB) is illustrated
in Figure 2.5.
Figure 2.5: The two main sources of mantle derived helium are: (a) Mid Ocean Ridge
Basalts (MORB) and (b) Ocean Island Basalts (OIB). Adapted from Elliott
[2009].
23
2 Theory
The characteristic 3He/ 4He ratio in MORBs is believed to originate from the upper mantle.
Most of the helium released from there escapes to the atmosphere and a small fraction leaks
into the oceanic and the continental crust. Melting of the oceanic crust undergoing the
continental crust in subduction zones releases helium, which mixes with the local mantle
and reduces the 3He/ 4He ratios in the magma. In areas that are tectonically active, break
up points in the continental crust enable intrusion of magma with a high 3He/ 4He ratio and
formation of igneous rocks that contain high 3He/ 4He ratios as well. Local groundwater,
which circulates through layers of such rocks, dissolve the helium, which increases their
3He/ 4He ratio.
OIB, also known as “hot spot” appears in the ocean and one of the most famous examples
are the islands of Hawaii. The helium in OIB originates from a deeper undepleted part of
the mantle and is more enriched than the MORB. The 3He/ 4He there varies from 15 Ra to
30 Ra [Lupton, 1983; Graham, 2002], due to a mixture with changing amounts of radiogenic
4He in the hot spot zone.
Separation of the helium components
Helium measured in groundwater can be separated to different components, as depicted
in Figure 2.6. Equilibrium and excess air come from the atmosphere, tritiogenic 3He is
produced from atmospheric tritium and the terrigenic component contains helium from
the crust and from the mantle. Two decompositions are discussed here: Separation of
terrigenic from atmospheric components (“air correction”) and further separation of mantle derived helium from crustal helium. Tritiogenic component can be neglected for old
water.
Equilibrium
Atmospheric
Tritiogenic
3He
Helium
Excess air
Crust
Terrigenic
Mantle
Figure 2.6: Helium components in groundwater. Most of the helium consists of 4He. Note
that tritiogenic component consists only of 3He.
Air correction
The separation between the atmospheric component and the terrigenic component is based
on the assumption that all neon in the sample has an atmospheric source. It is called air
correction since it is desired to remove the atmospheric component from the measured
24
2.4 Mantle derived isotopes in groundwater
helium isotope concentration. Three different approaches are presented here for derivation
3He
of the terrigenic isotope ratio Rter = 4 ter , all lead to the same results: A graphic
Heter
approach, isotopes decomposition approach and with the use of the fit results of excess air
and equilibrium component (see Section 2.3.2).
Graphic approach
Mixing between atmospheric and terrigenic endmembers in groundwater is graphically
demonstrated in Figure 2.7. Terrigenic helium consists of a mix of crustal and mantle
derived helium. 3He/ 4He ratios in water of a pure terrigenic origin lie on a mixing line
between crustal and mantle derived helium endmembers, corresponding to the red line
in Figure 2.7. The measured Ne/He ratio is proportional to the amount of helium of
atmospheric origin. In this plot, groundwater starts from point “A” on the graph, with
Figure 2.7: 3He/ 4He in a water sample, normalised to the atmospheric ratio Ra , against
neon to helium ratio in the sample. The helium isotope composition in groundwater can be treated as a mixing between three endmembers from three reservoirs: The atmospheric, the crustal and the mantle, which are marked on
the plot. The lines represent mixing lines between these endmembers. Blue
line: Atmosphere and crust mixing, red line: Crust and mantle mixing. The
black dashed line is a general mixing line of a sample “s”, between the atmospheric and terrigenic (mantle+crust) endmembers. The corrected 3He/ 4He
ratio, without atmospheric contaminations is found where the dashed mixing
line meets the crust mantle mixing line (point “B”).
25
2 Theory
AEW composition, with (N e/He)AEW ≈ 4.2 for water temperature of about 17 ◦ C and
Rs /Ra ≈ 1. After loosing contact with air, mixing with terrigenic helium takes place,
thus additional helium from the crust and mantle is incorporated in the water, moving the
water towards point “B” with Ne/He=0, as it gets older [Sano and Wakita, 1988].
Consider an arbitrary sample, marked as “s” in Figure 2.7. Assuming all neon in the
sample is of atmospheric source, the terrigenic 3He/ 4He component of the sample is the
intercept of the mixing line from the atmosphere “A” to the sample “s”, marked as “B”.
This is found here by calculating the equation of the line A–B. The general equation of
this mixing line is:
y = mx + n
(2.23)
The equation of the line can be found from points “A” and “s”, each has known coordinates:
Ne
[xA , yA ] =
,1
(2.24)
He a
Ne
Rs
[xs , ys ] =
,
(2.25)
He s Ra
The slope of the line is:
m=
Rs
Ra − 1
Ne
Ne
He s − He a
(2.26)
The intercept is obtained from inserting Equation 2.26 into the line Equation 2.23:
yB = n =
Rter
xs yA − xA ys
= ys − mxs =
Ra
xs − xA
(2.27)
Inserting the coordinates of points “A” and “s” into 2.27 yields the non-contaminated
helium component in the sample, normalized to Ra :
R
Ne
Ne
s
Rter
He s − He a · Ra
(2.28)
=
Ne
Ne
Ra
He s − He a
For convenience, an effective He to Ne ratio, x, is defined as:
Ne
He s
Ne
He a
x≡
(2.29)
Inserting x into Equation 2.28 yields:
Rter ≡ Rcorr =
Rs − x · Ra
1−x
(2.30)
Isotopes decomposition approach
If the tritiogenic helium component can be neglected, a sample consists only of atmospheric
and terrigenic components:
3He
4He
26
s
s
= 3Hea + 3Heter
=
4He
a
+
4He
ter
(2.31)
(2.32)
2.4 Mantle derived isotopes in groundwater
The isotopic ratio in the sample is:
3He
Rs =
4He
= Ra ·
s
3He
s
4He
4He
Setting x ≡
=
a
4He
s
a
3He
3
a + Heter
4He
s
4He
+ Rter ·
=
Ra · 4Hea + Rter · 4Heter
=
4He
s
(2.33)
ter
4He
s
yields:
s
Rs = Ra · x + Rter · (1 − x)
(2.34)
Extracting the terrigenic component gives:
Rter =
Rs − Ra · x
1−x
(2.35)
Assuming that all neon in the sample is from an atmospheric source and having in mind
that [He]≈[ 4He ], x can be written as:
4He
Nea
a
=
·4
=
4He
Nea
Hes
s
He
Ne
=
·
=
Ne a
He s
Ne
x≡
=
4He
a
(2.36)
He s
Ne
He a
The parameter x is calculated from the measured
ratio in AEW for a water temperature of 17 ◦ C.
Ne
He s .
Ne
He a
≈ 4.1939 is taken as the
Using the excess air fitting results:
The atmospheric component of helium is fitted for each sample, as described in sections
2.3 and 4.2. This component corresponds to the individual fitted NGT of each sample,
therefore should be more reliable than the theoretic values used in the graphic and the
decomposition approaches.
Similar to the decomposition approach, it is assumed that a groundwater sample contains
atmospheric and terrigenic components, the corrected isotopes concentrations are then:
3He
4He
ter
ter
=
3He
=
4He
s
s
− 3Heatm (CE)
−
4He
(2.37)
atm (CE)
The terrigenic helium component ratio is simply the fractionation ratio Rter =
(2.38)
3He
4He
ter
.
ter
Atmospheric, crustal and mantle derived helium components
The following decomposition continues the air correction, separating the terrigenic helium
component to crustal and mantle derived helium, which are shown in Figure 2.6. The
separation can be derived from a mixing model between atmospheric, crustal and mantle
derived helium endmembers. Aeschbach-Hertig [2005] derived such a decomposition for
27
2 Theory
two components, which is expanded here to three components. The 3He/ 4He ratio of a
sample, Rs , can be written as:
Rs =
3He
s
4He
s
=
3He
a
+ 3Hem + 3Hec
4He
s
(2.39)
The letters “a”, “m” and “c” refer to the atmosphere, mantle and crust, respectively. The
average 3He/ 4He ratios of the three reservoirs from Section 2.4 are summarized in Table
2.1 and inserted into Equation 2.39:
Rs =
4He
a
· Ra + 4Hem · Rm + 4Hec · Rc 4
= ARa + 4 M Rm + 4 CRc
4He
s
(2.40)
With the atmospheric, mantle and crustal 4He components:
4
A=
4He
a
4He
4
;
s
M=
4He
m
4He
s
;
4
C=
4He
4He
c
(2.41)
s
The relative parts of the 3He components can be calculated from 2.41 and using the
3He/ 4He ratio of each reservoir from Table 2.1. For example, the atmospheric component
reads:
3
A=
3He
a
3He
s
=
4He
a
4He
s
· Ra
Ra
= 4A
Rs
· Rs
(2.42)
The other two components can be calculated in the same way. It is clear from Equation
2.40 that the sum of the components is 1:
4
A + 4M + 4C = 1
(2.43)
Equation 2.43 can be written as: 4 C = 1 − 4 M − 4 A. Inserting it into Equation 2.40 and
solving for 4 M yields:
4
M=
(Rs − Rc ) − 4 A · (Ra − Rc )
Rm − Rc
(2.44)
where Rs is the measured 3He/ 4He ratio and Ra , Rm and Rc are the 3He/ 4He ratios
of the atmosphere, mantle and crust, respectively and are given in Table 2.1. 4 C is then
calculated from Equation 2.43 and 3 A, 3 M and 3 C are calculated using Equation 2.42. It is
expected to find high fractions of mantle derived 3He (3 M ) even in groundwater samples
that have small 3He/ 4He ratios, due to the different characteristics of the atmosphere,
crust and mantle, regarding their helium 3He/ 4He ratios [Aeschbach-Hertig, 2005]. It is
generally the case that most of the terrigenic 3He originates from the mantle, therefore it
is useful to compare only the fractions of atmospheric, crustal and mantle 4He, namely 4 A,
4 C and 4 M , which sum up to approximately the total helium concentration in the water
([ 4He ]≈[He]).
This three components mixing model is connected to the air correction by the parameter
4 A, which is exactly x, defined by the air correction (Equation 2.29).
28
2.4 Mantle derived isotopes in groundwater
Table 2.1: Average isotopic fractionations ( 3He/ 4He) of the main three reservoirs of helium
in groundwater from section 2.4.
Ra
Rc
Rm
1.384E-06
2.000E-08
1.000E-05
2.4.2 CO2 and δ 13 C
CO2 is one of the major volatiles that are still escaping from the mantle. According to
Griesshaber and Oxburgh [1992] CO2 in gas phase is the main carrier of mantle derived
3He. A ratio of CO / 3He =2 · 109 was measured for the upper mantle in ocean ridge
2
basalts [Marty and Jambon, 1987]. The ratio of CO2 / 4He in the mantle is 1 · 104 and
for continental gases it depends on the production rate of radiogenic 4He from U and
Th, thus decreasing with time, as the amount of 4He increases. If fluids from the mantle
manage to rise through the crust and pass it without any contamination or fractionation
they should have ratios of CO2 / 3He and CO2 / 4He that are similar to the ratios in the
mantle [Griesshaber and Oxburgh, 1992 and Sano and Marty, 1995]. This however does
not happen in the reality.
Groundwater usually contains a mixing of CO2 from different reservoirs, where different
processes occur and lead to fractionation of the carbon isotopes (see Figure 2.8). Each
source has a typical δ 13 C signature, as shown in Figure 2.8. According to Hoefs [2009]
and Sano and Marty [1995] the main carbon sources on Earth are marine limestone that
contain carbonate, with −2 h < δ 13 C < 2 h and sediments that contain biogenic organic
carbon and have a signature of −40 h < δ 13 C < −20 h. According to Griesshaber and
Oxburgh [1992], δ 13 C between −5 h to −8 h indicates mantle derived CO2 . Other studies
suggest similar ranges, e.g. Barnes et al. [1978] (cited by O’Nions and Oxburgh [1988])
who state a range of δ 13 C between −4 h to −6 h and Sano and Marty [1995] who suggest −5 h < δ 13 C < −8 h. O’Nions and Oxburgh [1988] analysed gases with a mantle
signature of δ 13 C and compared their 3He/ 4He ratios with their CO2 / 3He ratios. Some of
the gases had higher 3He/ 4He, corresponding to a big mantle component, but others had
lower 3He/ 4He ratios, therefore the CO2 in these samples is not likely to come from the
mantle, due to no correlation between the mantle isotope signature of δ 13 C and mantle
derived helium.
It is not simple to attribute the δ 13 C signature to mantle derived carbon due to overlaps
with isotopic signatures of other sources. Groundwater, for instance, has a signature of
−20 h < δ 13 C < 0 h, due to multiple sources of CO2 there. Moreover, carbon in the
crust, which was originally escaping from the mantle, has gone through different fractionation processes, that changed its isotopic composition and lead to the different isotopic
signatures, thus making it difficult to identify CO2 of a mantle origin.
Sano and Marty [1995] introduced a mixing model to determine the nature of CO2 volatiles
in hydrothermal waters, combining the δ 13 C and CO2 / 3He from CO2 in a gas phase.
The idea is a three component mixing with endmembers from the upper mantle, marine
limestone (carbonate) and sediments, marked as M , L and S, respectively. The resulting
29
2 Theory
Figure 2.8: Typical δ 13 C isotopic signatures of different sources and as a result of different
processes, adapted from Clark and Fritz [1997]. Notice the overlap of the
different sources, which makes is difficult to indicate the origin of CO2 in
groundwater.
mixing equations can be written as following:
RO = RM M + RL L + RS S
M
L
S
1
=
+
+
3
3
3
CO2 / He O
CO2 / He M
CO2 / He L
CO2 / 3He S
M +L+S =1
(2.45)
(2.46)
(2.47)
M , L and S are the fractions of the mantle, limestone and sediment derived CO2 in the
sample. RM , RL , RS and RO are the isotopic ratios 13C/ 12C, of the mantle, limestone
and sediment endmembers and the measured (“Observed”) isotope ratio of the sample,
determined from δ 13 C and the PDB standard for each of the above fractionation ratios
(i):
RP DB = 0.0112372
Ri =
δ 13 C
(2.48)
· RP DB
+ RP DB
1000
i
(2.49)
Solving Equations 2.45, 2.46, 2.47 yields the mantle derived CO2 component:
1
M=
30
CO2 / 3He
CO2
−
o
1
/ 3He
m
1
−
CO2 / 3He
s
1
CO2
/ 3He
s
(2.50)
2.4 Mantle derived isotopes in groundwater
Table 2.2: Isotopic signatures of carbon and CO2 / 3He for mantle derived carbon (M),
marine carbonate from limestone (L) and carbon from sediments (S). They are
used for a three component mixing model following Sano and Marty [1995]. See
text for further information.
δ 13 C
[h]
R= 13C/ 12C
CO2
/ 3He
M
L
S
−6.5 ± 2
0±2
−30 ± 10
1.1164E-02
1.1237E-02
1.0900E-02
1.5 ·
109
1·
1013
1 · 1013
2.4.3 Stable isotopes: δ 18 O and δ 2 H
Precipitation from all over the world show a strong correlation in their stable isotopes
composition. The relationship between their δ 18 O and δ 2 H levels is according to the
following Equation [Craig, 1961a]:
δ 2 H = 8 · δ 18 O + 10 h
(2.51)
δ 18 O and δ 2 H are given in [h] and can be derived from the ratios 18O/ 16O and 2H/ 1H
respectively (see appendix A.3). The line in Equation 2.51 is called the Global Meteoric
Water Line (GMWL) and is plotted in Figure 2.9. The isotopic composition of ocean water
is defined as the reference “Standard Mean Ocean Water” (SMOW), so that δ 18 OSM OW =
δ 2 HSM OW = 0h [Craig, 1961b]. The main origin of precipitation is the ocean, where
massive evaporation takes place. However, the GMWL has an intercept of +10 h and
not 0 h. This is called “Deuterium excess”, often abbreviated to “D-excess”. The water
vapour above the ocean is more enriched in deuterium relative to the ocean water due to
non-equilibrium fractionation during the initial evaporation.
The composition of stable isotopes in precipitation varies in different areas and is affected
by the temperature and relative humidity. In warmer areas, evaporation of part of the
precipitation can occur, causing enrichment in deuterium (Fig 2.9). There is also an
elevation effect, in which precipitation falling on mountains in higher altitudes is more
depleted due to enhanced rain-out there, as a result of decrease in the water vapour
saturation pressure. There are other factors that affect on the isotopic composition of
precipitation, but they are beyond the scope of this work.
The local meteoric water line for Israel is taken as the Eastern Mediterranean Meteoric
Water Line (EMWL) from Gat and Carmi [1970] (cited by Gat and Dansgaard [1972]):
δ 2 H = 8 · δ 18 O + 22 h
(2.52)
The EMWL has the same slope as the GMWL but an enhanced D-excess, due to a stronger
evaporation takes place in the Mediterranean Meer, relatively to the world wide average
and due to its almost completely isolation from the Atlantic Ocean and therefore no isotope
exchange. According to Gat et al., 1969 and Gat and Carmi, 1970, the local δ 18 O ranges
from −7.5 h to −4.5 h.
Groundwater which originates from precipitation, also called meteoric water, can be expected to have a stable isotopes composition similar to that of local precipitation. For
31
2 Theory
40
20
GMWL: δ²H = 8 · δ¹⁸O + 10 ‰
δ²H [‰]
0
shift due to kinetic fractionation
during evaporation
deuterium
excess: 10 ‰
-20
-40
-60
equilibrium fractionation:
δ²H = 8 · δ¹⁸O
-80
-14
-12
-10
-8
-6
-4
-2
0
2
4
δ¹⁸O [‰]
Figure 2.9: Relationship between δ 18 O and δ 2 H in global precipitation, given by the
GMWL. Adapted from Bröder [2011].
example, groundwater formed in mountains is depleted in stable isotopes relatively to
groundwater from lower altitudes, due to the altitude effect mentioned before. Changes in
the isotopic composition relative to the composition of local precipitation may be a result
of fractionation processes in the ground, therefore stable isotopes can be used as tracers
for interaction of groundwater with its surroundings.
Thermal waters, in some cases, are enriched in δ 18 O, due to interactions with crustal
rocks at temperatures higher than 100 ◦ C, which cause exchange of 18O [Hoefs, 2009]. The
enrichment is seen only in δ 18 O, since 2H is not contained in minerals. High temperatures,
leading to δ 18 O enrichment in groundwater may be a result of heat fluxes from the mantle.
Therefore, stable isotopes can be used as tracers of heat transfer from the mantle to
the crust, which can sometimes be a potential source of geothermal energy. However,
Giggenbach [1992] analysed volcanic vapour discharges from volcanic systems with high
temperatures, so called andesitic water, where fractionation and a deuterium shift take
place due to evaporation. He indicated that the detection of this shift is not simple in case
of dilution with local groundwater, which is the case for hot springs.
32
3 The Dead Sea Transform
This section reviews the main geographic, tectonic, geological and hydrological settings
of the Dead Sea Transform with emphasis on the Israeli side, where the water samples of
this thesis are taken from. Many of the geological structures in the Dead Sea Transform,
such as basins and rock formations, are named after cities or places that are located in
the surroundings. Hence, it is important for the reader to have a general orientation of
Israel, especially the eastern part of the country, where the Dead Sea Transform lies. The
names of the places mentioned in this section are marked on the map of Israel in Figure
A.1. Many studies have been done about the different aspects of the Dead Sea Transform.
A major part of this section bases on Horowitz, 2001 and Hoetzl, 2009.
3.1 Geographic setting
The Dead Sea Transform was formed during the Late Cenozoic as a result of the break up
of the Arabian–African continent. It is a segment of a bigger rift system, called the Levant
Rift, which is more than 1000 km long. The Levant Rift is divided into three segments:
The Jordan Rift in the south, El Gharb Kara-Su Rift in the north and the Lebanese Fault
in between. The East Anatolian Fault and the Red Sea are the borders of the Levant Rift
to the north and south, respectively (Figure 3.1).
The Jordan Rift Valley is a 350 km long, narrow and north–south oriented valley on the
border between the state of Israel and the kingdom of Jordan. The topography of both
sides of the Jordan Rift Valley is asymmetric, as a result of a slip-strike fault. The Israeli
side extends from the Hula Valley in the north to the northern Red Sea in the south. It is
composed of a series of basins, that are bounded by large normal faults, which cause the
floor of the basins to sink and the margins to reach up to a height of thousands of meters
[Picard, 1987 ; Mart, 1991 and Horowitz, 2001].
The climate in the Jordan Rift Valley is of Mediterranean type in the north and semi arid
to arid in the south, with warm and dry summers and mild winters. The rain time is from
September to April.
The Israeli part of the Dead Sea Transform can be divided into regions of different climate
and geography. Horowitz [2001] suggests a division into four main parts, from north to
south:
1. The northern part extends from the borders between Israel–Lebanon and Israel–
Syria to the southern side of Lake Kinneret. There are two main surface water bodies
in this region: the Hula Lake and Lake Kinneret.
2. The central part lies between Lake Kinneret and the Dead Sea.
3. The Dead Sea area is located in the Judah Desert and is the lowest place on Earth
(today about −427 m, according to data of the Israeli Meteorological Service). The
33
3 The Dead Sea Transform
Figure 3.1: Tectonic map of the Levant Rift and the area, from Mart et al. [2005].
water level keeps going down due to human activity and massive evaporation. Thus,
the area of the surface sea is constantly being reduced.
4. The Arava Valley extends from the Dead Sea to the Gulf of Aqaba, also called the
Gulf of Eilat, named after the Jordanian and Israeli cities on the tip of the gulf. Since
Arava Valley is located on the eastern part of the Negev Desert it is characterized
with high temperatures and low precipitation.
3.2 Tectonic setting
The Dead Sea Transform connects the Red Sea, which is a spreading zone, with the
Taurus–Zagros Mountains in Turkey, where continental subduction takes place [Freund,
1965 and Wilson, 1965]. The tectonic structure of it consists mostly of faults [Quennell,
1956, 1959]. According to Garfunkel et al., 1981 and Horowitz, 2001 it is considered to be
a transform on the border between the African Plate and the Arabian Plate, with a 105 km
long horizontal displacement [Quennell, 1959 ; Freund, 1965 ; R., 1970 and Freund et al.,
1970]. It is called a transform fault because it is formed as a result of the movement of both
plates away from each other, as depicted in Figure 3.1. According to Bayer et al. [1988],
the displacement has began since the Oligocene–Miocene or since Mid–Miocene. From
seismic and heat flow studies it is known that deformation still takes place along the Dead
Sea Transform [Horowitz, 2001]. According to Garfunkel et al. [1981] the present average
slip rate of the lateral motion is 0.7–1.0 cm/a. A more recent slip rate of 0.4–0.5 cm/a is
34
3.2 Tectonic setting
found by Gomez et al. [2007].
The rifting process and the lateral slip have led to formation of a pull-apart basins chain,
as explained by Garfunkel et al. [1981]. A pull apart basin is formed as a result of a lateral
motion of a fault, as illustrated in Figure 3.2. Refraction data and gravity anomalies
indicate that the southern basins in the Dead Sea Transform are deeper than the northern
ones [Horowitz, 2001]. The sediment fill in the southern basins extends up to 14 km before
the basement is reached. The fill depth in the northern basin is 6–7 km. The main basins
or segments along the Israeli side of Dead Sea Transform, described by Horowitz [2001],
are from north to south: The Hula depression, Bet She’an – Lake Kineret depression, the
Dead Sea – Bet She’an segment and the Dead Sea – Arava depression.
Israel
N
Jordan
Figure 3.2: The Dead Sea Transform is a chain of pull apart basins, which are formed as a
result of a left lateral motion slip of the fault, marked with arrows. Israel and
Jordan are indicated for orientation.
The northernmost basin in the Dead Sea Transform is the Hula basin. It is bordered to
the east by the Jordan River gorge and the Golan Heights, to the west by the mountains
of Galilee and to the south by the Korazim block, which makes a topographic separation
of the Hula Basin from the Kinneret basin. The Hula basin is shallow, only 5 km of
sedimental fill, with faults on both east and west margins.
Southern from the Hula basin is the Bet She’an – Lake Kinneret Depression, which is
under sea level and forms the Kinneret basin. Some active faults still exist in the margins
of the lake.
The Dead Sea – Bet She’an segment begins at around the town of Bet She’an, south of
Lake Kinneret and extends southward, down to the Dead Sea. The width of the valley is
about 8 km around Bet She’an in the north and increases to 18 km around Jericho in the
south. The valley is broadening a few kilometres southern of Bet She’an and connects the
Yizrael Valley from the west. The slopes of this segment become more steep to the south,
connecting to the high slopes of the margins of the Dead Sea basin.
The Dead Sea – Arava is the most prominent and seismically active basin in the Dead Sea
Transform. It begins from about 5 km north to the Dead Sea and extends over 130 km to
the south, down to Gav Ha’Arava. Step faults in the eastern and western margins of the
Dead Sea have caused vertical displacements which lead to a depression of the basement.
The depression is 7–18 km wide and has relatively steep slopes on both sides.
The thickness of the crust underlying the Dead Sea Transform varies from north to south,
as illustrated in Figure 3.3. El-Isa et al., 1987 and Ginzburg et al., 1979 indicate that the
crust in the southern Dead Sea Transform is an oceanic crust covered by the continental
35
3 The Dead Sea Transform
Figure 3.3: Schematic cross sections of the crust underlies the Dead Sea Transform. The
thin arrows on the cross sections refer to the process of volatiles upwelling from
the mantle. Adapted from [Torfstein et al., 2013]
crust of the Arabo–Nubian Massif and a thick wedge-like layer of Mesozoic and Tertiary
sediments, which is up to 35 km thick. Horowitz [2001] indicates different densities of the
crust in the north, also supports the existence of two types. The thickness of the crust
decreases towards the north and reaches a thickness of about 25 km and has characteristics
of a continental crust. For comparison, the crust underlies the south of Israel and the Negev
desert is 40 km thick.
3.3 Hydrogeology
Aquifers
The stratigraphic profile of the Jordan Rift Valley comprises of different layers and rock
formations starting from Precambrian time. The groundwater originates from regional
calcareous and basaltic aquifers, as shown in the stratigraphic sequence in Figure 3.4. The
main aquifers described by Hoetzl [2009] are summarized here.
The oldest main aquifer is the Ram Aquifer, consisting of Precambrian sandstone, limestone and siltstone. Above Ram lies the Zarqa aquifer from Permo–Triassic age. This
group aquifer consists of limestone, argillaceous rocks with shales, dolomite and silt stone.
It lies in the vicinity of Zarqa Ma’in and Hisban area in Jordan and has a thickness between
300 m to 350 m. Above Zarka lies the Kurnub sandstone aquifer, which is from lower Cretaceous age and is about 220–320 m thick. The sandstones of Kurnub are connected with
the Zarqa and Ram aquifer. They form an interconnected aquifer complex that is called
”Lower Aquifer Sandstone Complex” [Hoetzl, 2009]. Some parts of the interconnected
36
3.3 Hydrogeology
aquifer consist of marl and silt, separating the different parts from each other.
Above the lower complex aquifer lies a group of aquifers called the Upper Cretaceous
aquifer. It consists of the regional aquifers that lie in the eastern and western sides of the
Jordan Rift Valley. The Judea group lies under the Judea dessert and Samaria area on
the western side of the Jordan Rift Valley. The thickness of the aquifer is 800–850 m and
it comprises of limestone and dolomite.
The youngest group of aquifers is called the Dead Sea Group of the Jordan Valley Deposits.
The stone formations in this group are from the Pleistocene–Holocene age. This group
subdivides to local aquifers on both sides of the Dead Sea Transform and consists of gravel
and sandstones.
Water salinity
The water bodies along the Dead Sea Transform vary from brackish to saline. According
to Torfstein et al. [2013], the brackish groundwater found in the Arava Valley, originates
from the deep Kurnub aquifer and the shallow Arava Formation and the Arava Fill, shown
in Figure 3.4. The salinity of the water is coming from mixing with other brines and from
dissolution of evaporites and other sediment layers [Yechieli et al., 1992].
The saline water of the Jordan Rift Valley contains surface and groundwater in the area
of Lake Kinneret and the Dead Sea. The salinity of the brines varies from brackish to
very high salinities of 400 g/l in the Dead Sea. The Ca-chloride brines formation can be
explained by penetration of evaporated sea water to the Dead Sea Transform, followed
by interaction of the water with limestone layers and dolomitization. Each of the water
bodies has a different Ca-cloride composition, therefore different sedimentation rates of
salt [Starinsky, 1974] (cited by [Stern, 2010]).
37
3 The Dead Sea Transform
Figure 3.4: Stratigraphic column of Cisjordan, adapted from Möller et al. [2007b].
38
3.4 Sampling sites
3.4 Sampling sites
A sampling campaign was carried out in December 2012 in Israel. 13 samples from springs
and wells were taken at 9 sampling sites along the Dead Sea Transform. Locations and
Table 3.1: General information of the sampling sites in the Israel campaign, December
2012. Note that all the samples except of Shamir and Tzofar are below sea
level.
Site*
Shamir
Meitzar
Hamat Gader
Gofra Beach
Hamat Tveria
Mineral Beach
Ein Qedem
Ein Yahav
Tzofar
Region**
Latitude
Longitude
Hula V.
Yarmouk V.
Yarmouk V.
E. Kinneret
W. Kinneret
Dead Sea
Dead Sea
Arava V.
Arava V.
33◦ 10’22’
35◦ 39’50”
32◦ 42’54”
35◦ 41’46”
32◦ 41’03”
32◦ 48’14”
32◦ 46’02”
31◦ 32’45”
31◦ 30’55”
30◦ 40’32”
30◦ 33’13”
35◦ 39’55”
35◦ 38’32”
35◦ 33’00”
35◦ 23’47”
35◦ 23’43”
35◦ 13’05”
35◦ 10’40”
Elv. [m]***
Site type
245
-90
-163
-183
-183
-386
-404
-80
18
Well field
Well field
Spring field
Spring
Spring field
Well
Spring
Well field
Well field
*The owners of the sites are: Mekorot Israel National Water Co.: Meitzar, Ein Yahav
and Tzofar. Mei Golan Water Co.: Shamir. Hamat Gader Resort: Hamat Gader. Gofra
Beach: Gorfa Beach. Nature and Park Authority: Hamat Tveria. Mineral Beach: Mineral
Beach. Ein Qedem has no owner.
**Region abbreviations: V: Valley, E: Eastern and W: Western.
***Elevation above sea level in [m].
Table 3.2: Depth, screen and contributing rock layer of the sampled wells. All the wells
are artesian, including Mineral, for which the owner could not provide borehole
data. Depth corresponds to the depth of the drilling, relative to the surface and
screen corresponds to the depth range from which groundwater actually comes.
Site
Shamir
Meitzar 2
Meitzar 3
Ein Yahav 6
Ein Yahav 16
Ein Yahav 116
Tzofar
Depth [m]
Screen [m]
Lithography
1,420
807
336
904
509
190
1,010
1,162–1,420
448–807
80–336
765–884
255–509
136–171
836–1,007
Limestone
Mostly Dolomite
Bitumenic Chalk
Sandstone
Mostly Lomestone
Flint
Sandstone
site types are shown in Table 3.1 and Figure 3.5. Information about the sampled wells is
given in Table 3.2. The sites differ from each other in the flow rate and their accessibility.
Therefore, the sampling process was different from site to site, as described in appendix
A.5.
39
3 The Dead Sea Transform
Figure 3.5: Locations of the sampling sites of the Israel campaign, 2012.
40
4 Methods
This chapter describes the necessary sampling, measurement and evaluation methods applied in this study. The following parameters and samples were taken on each site: Water
temperature, electrical conductivity, alkalinity, dissolved noble gases, tritium, δ 13 C and
stable isotopes.
4.1 Water characteristics
The following water characteristics were measured in each site:
• Temperature and electrical conductivity: Measurements were taken with a
WTW water probe. The water parameters were noted after their stabilization. According to the manufacturer, the precision of the temperature is ±0.2 ◦ C and the
precision of electrical conductivity is ±1 µS/cm for measured values bellow 2 µS/cm
and ±10 µS/cm for values above 2 µS/cm.
• Alkalinity: An alkalinity test kit manufctured by Salifert, based on titration was
used, with a precision of 0.1 meq/l for alkalinity above 3.9 meq/l and 0.07 meq/l for
a smaller alkalinity.
Photos of the sampling are found in the appendix A.8, Figure A.11.
4.2 Noble gases: Mass spectrometry
4.2.1 Sampling
As mentioned in Section 2, the solubility of noble gases in groundwater depends on the
temperature and salinity at the time of equilibration with air. Therefore the sample should
be taken as close as possible to the point were the water springs from the ground and the
formation of bubbles should be avoided in order to avoid degassing. For the same reason,
contact with atmospheric air should be avoided as well. Noble gases are very volatile,
therefore a sealed copper tube is used as a container and in order to prevent gas escape
and interaction with air. The copper tube is connected from both sides to hoses, as
depicted in Figure A.11. It is recommended to use transparent hoses in order to be able
to detect and remove air bubbles. The hoses also help to prevent bubbles to form at the
sides of the tube. Water should flow freely through the copper tube and fill the hoses.
The sampling process slightly changes from site to site, depending on the site type (well
or spring), the water pressure and the flow rate at the outflow point. After making sure
that no air bubbles are trapped in the water, the copper tube is sealed.
41
4 Methods
4.2.2 Measurement
The noble gas samples were measured using a GV Instruments MM5400 sector field mass
spectrometer, at the Institute of Environmental Physics (IUP). The distinction of different
elements is done by ionization, where different ions move in different curves under magnetic
field, due to Lorenz force. The radius of the curves depends on the mass and charge of
the ions [Hoffmann and Stroobant, 2007]:
m
B 2 r2
=
q
2U
(4.1)
m is the ion mass, q is the ion charge, B is the magnetic field intensity, r is the radius of
the curve and U is the magnitude of the voltage induced on the ion.
The measurement process lasts a few hours and consists of extraction and separation
of the noble gases from the water, followed by detection in the spectrometer itself, as
demonstrated in Figure 4.2. A water sample is first connected to an inlet zone under
vacuum conditions, where degasification takes place, extracting the dissolved noble gases
from the water. According to Henry’s law (Equation 2.2), the solubility depends on the
partial pressure of the gas. The partial pressure of a gas decreases in vacuum, thus reducing
the solubility, which leads to degasification. A zeolite trap is used to remove water vapour
remnants and the rest flows into a cryo trap, where the individual noble gases are separated.
The lighter gases (helium and neon) and the heavier gases (argon, krypton and xenon)
are cooled down until they freeze and are trapped in different segments of the cryo. The
cryo is then heated and desorbs the individual gases based on their melting point, from
the lighter to the heavier, in different time intervals. The gas is then cleaned by getter
pumps, removing remnants of other gases like nitrogen and is ready to be measured in
the spectrometer, where the molecules are ionized, lead through a magnet and detected
(Figure 4.1). Two kinds of detectors are used: An Electron multiplier for small amounts,
measures counts per second and a Faraday cup for bigger amounts, which translates the
ion current to voltage. The plot of a signal versus atomic mass has a shape of a peak,
centered around the atomic mass of the measured isotope, for example, the peak of 4He
has a maximum around about 4 AMU. The measurement is gas consuming, therefore the
signal intensity decreases with time, hence the peak height decreases too.
In order to be able to convert the measured signal to gas amount, gas with a well known
composition is measured and used for two kinds of calibrations. The first, called “fastcal”
(fast calibration), is done before each measurement in order to determine changes of the
spectrometer sensitivity. The second, called “Cal” (air standatd), is an absolute calibration
of the whole system, including the preparation zone and the spectrometer itself. The
measurement process of a “Cal” is similar to a measurement of a sample, but instead of
extraction of gases from a sample, calibration gas is taken from a container and is inserted
directly into the cryo trap (see Figure 4.2). The calibrations have to be regularly performed
in order to adapt them to changes of background effects, such as temperature changes in
the laboratory.
Other necessary measurements to monitor the system are background measurements,
called “blank”, where no sample is inserted into the spectrometer and AEW samples.
The later are water samples, taken in the IUP from an aquarium containing water with
known temperature and air pressure. This water stays a few weeks in the aquarium and
42
4.2 Noble gases: Mass spectrometry
Figure 4.1: Schematic description of MM5400 mass spectrometer, used for noble gas measurements in the IUP, adopted from Friedrich [2007]. Gas molecules are flowing
through the inlet and are ionized. The ions change their curve under magnetic
field and are detected by one of the two detectors. The distinction between
different ions is based on the mass and radius of the curve, following Equation
4.1.
Figure 4.2: Measurement process of the mass spectrometer, modified from Wieser [2011].
exchanges gases with the surrounding air until an equilibrium is reached. The measured
noble gas concentrations are compared with the theoretical concentrations.
As of today, eight noble gas isotopes are measured by the mass spectrometer:
3He, 4He, 20Ne, 22Ne, 36Ar, 40Ar, 84Kr and 132Xe. Further detailed information about
the mass spectrometer in the IUP is found in Friedrich, 2007 ; Wieser, 2011 and Kaudse,
2014.
4.2.3 Data evaluation
The calculation of gas amount from the measured signal is done by the WuCEM software,
developed by Michael Jung from the IUP. The analysis is done by a few steps. On the first
step, the measured signal is extrapolated to the time of the gas inlet, t = 0. The signal
amplitudes, measured at different times, are plotted against the time and are fitted, usually
with a linear curve. The intercept of the fitted curve is taken as the signal strength at the
time of inlet. This is done for the samples and the all the “Cals” and “blanks” that are
measured close to the time of the samples measurement and to all the measured isotopes
43
4 Methods
and their corresponding “fastcals”. On the second step, the measured isotopes signals
are corrected with the calibrations and blank measurements. The resulting gas amounts
are given in units of [ccSTP], which is the volume (in cubic centimeters) that a given
amount of gas occupies as STP: Standard Temperature and Pressure, of T0 = 273.15 K
and p0 = 1 atm.
From gas amount to concentration
The concentration of each isotope is obtained by dividing the amount by the total mass
of the water in the sample, measured by weighting the sample before and after the measurement. The concentration of each element is calculated from the concentrations of the
isotopes composing it, weighted with their atomic abundance factors. That way, the total element concentration can be calculated only from part of the isotopes, thus it is not
required to measure them all. The atomic abundance factors of the isotopes measured in
the IUP are listed in Table 4.1 and the calculation is done as follows:
P
P
Isotopen
n
n [Isotope]
P
[Element] =
= n P
,
(4.2)
m · n fn
n fn
where [...] marks the concentration of an element or an isotope, n refers to a single isotope
species, i.e. 4He, fn is the atomic abundance factor and m is the sample mass.
The total helium concentration is calculated only from 4He, since the amount of 3He is
negligible:
[He] ' [ 4He]
The concentration of argon is usually calculated as follows:
[Ar] =
[ 36Ar] + [ 40Ar]
0.003364 + 0.996
As explained in Chapter 2, in case the ratio
atmosphere, [Ar] is calculated only from 36Ar:
[Ar] =
[ 36Ar]
0.003364
36Ar/ 40Ar
is smaller than its ratio in the
(4.3)
NGTs Fitting
The calculated noble gas concentrations can be fitted with one of the excess air models to
find the NGT, as explained in the theory chapter. This is done here using Panga, a fitting
software developed by Michael Jung. The concentrations are given to the software, as well
as water salinity and estimated air pressure at the time of recharge. The software fits the
given parameters to one of the excess air models and determines the NGT.
Samples with air contamination or degassing should be treated differently than other
samples. Since it can be assumed that neon originates from the atmosphere and since its
solubility does not change so much with the temperature, relatively to the other noble
gases, it can be used as an indicator of air contamination or degassing. The average
neon concentration of air equilibrated water is ∼2 · 10−7 ccSTP/g. Much higher or lower
44
4.3 Isotopic composition
Table 4.1: Atomic abundance factors of the measured noble gas isotopes from Kipfer et al.
[2002].
Isotope
Factor
He3
He4
0.0000014
1.0000000
Ne20
Ne22
0.9050000
0.0923000
Ar36
Ar40
0.0033640
0.9960000
Kr84
0.5700000
Xe132
0.2689000
concentrations indicate excess air or degassing, respectively and may lead to a wrong NGT.
A more qualitative way to identify degassing is the parameter ∆Ne, defined as the relative
part of excess air derived neon [Aeschbach-Hertig and Solomon, 2013]:
Fitting results that are not optimal, for example extremely low NGTs or high uncertainties, can be improved with the help of Monte Carlo simulations [Jung et al., 2012]. These
randomly produce fitting results that distribute with certain expectation value and a statistical deviation. In some cases the Monte Carlo simulations succeed to produce physical
results even though the fitting results are not physically reasonable.
4.3 Isotopic composition
Samples of tritium, 13C and stable isotopes are taken in glass bottles. The bottles should be
filled with water almost completely, leaving a minimal portion of air for a safe transport.
4.3.1 Tritium
Tritium was measured in the tritium laboratory at the IUP, as described by Grothe [1992].
9 ml water is heated to 600 ◦ C and reduced by magnetism to produce H2 gas. The gas is
then measured in a low counting chamber for 48 hours. This system can detect tritium
from 2 TU, therefore results that show smaller amounts can only be qualitatively treated,
which is sufficient for this study.
4.3.2 δ 13 C
During the sampling, before sealing the bottles, one drop of of Micropur (Katadyn Products Inc.), which is used to purify water, was added to the water to stop any biological
activity and prevent further fractionation.
The measurements were carried out in the carbon laboratory at the IUP. The process
consists of extraction of the DIC from the water, followed by a measurement in an accelerated mass spectrometer (AMS). For more information about the extraction line see Unkel,
45
4 Methods
2006 and Kreuzer, 2007. The 13C measurement is further described by Wieser [2011]. The
precision of the results is 0.03 h.
4.3.3 Stable isotopes
The stable isotopes were measured with a Picarro L2120-i system by the BGR, the laboratory of the Federal Institute for Geosciences and Natural Resources, in Hannover, Germany.
According to the manufacturer, the measurement precision of δ 18 O is better than 0.1 h
and of δ 2 H is better than 0.5 h.
46
5 Results
This chapter shows and describes the results of the thermal waters sampling in Israel.
The results are divided into two sections. The first section deals with noble gas temperatures and the the following sections present the results of helium isotopes in the thermal
waters, in connection with other environmental isotopes and noble gases. Measurement
uncertainties and errors appear only in tables and are omitted from graphs for clarity.
Thermal waters from 13 wells and springs were sampled along the Dead Sea Transform in
Israel in December 2012. The sampling sites spread from the Hula Valley in the north,
to the Arava Valley in the south. All locations are marked on the map in Figure 3.5
and described in Table 3.1. Water temperature, electrical conductivity and alkalinity are
presented in Table 5.1.
Table 5.1: Water temperature, electrical conductivity and alkalinity. H.Tveria corresponds
to Hamat Tveria. The corresponding precisions are: Temperature: ±2 ◦ C,
electrical conductivity: ±10 µS/cm and alkalinity: 0.07 meq/l for values under
3.9 meq/l and 0.1 meq/l for higher values.
Sample
T [◦ C]
E.C [µS/cm]
Alkalinity [meq/l]
Gofra
Shamir
H. Tveria
Meitzar 2
Meitzar 3
Makla
Balzam
Mineral
Qedem
Yahav 6
Yahav 16
Yahav 116
Tzofar 20
30.8
45.4
58.9
65.3
40.2
48.5
41.7
42.7
43.8
43.5
35.0
31.0
43.3
6840
1237
37800
1835
660
1895
1393
148100
148500
2690
2780
2170
9630
2.62
3.94
2.73
4.68
0.79
0.56
0.79
1.54
2.73
4.62
3.30
0.90
3.08
Noble gas isotopes of helium, neon, argon, krypton and xenon were measured by the
mass spectrometer of the Groundwater and Paleoclimate Research Group at the Institute
of Environmental Physics in Heidelberg University. Several samples were collected from
each site and the measurements were carried out in three runs. The samples from the
first run, which went without significant complications, are used for further analysis. The
second and third runs are not included in the analysis due to an electrical blackout during
the measurements. Their results are found in Appendix A.7. The measured isotopes
concentrations from the first run are listed in Table 5.2. The total amount of each element
was calculated from the isotopes composing it, using Equation 4.2 and is shown in Table
5.3. Ein Makla, Hamat Tveria and Ein Qedem have high 36Ar/ 40Ar ratios (see Appendix
47
5 Results
3.378E-12
7.963E-12
1.105E-12
1.525E-11
4.328E-11
2.501E-11
1.241E-12
6.204E-12
3.670E-12
9.174E-12
4.120E-11
3.322E-14
7.729E-12
He
∆ 36Ar
1.426E-13
3.458E-13
4.920E-14
7.722E-13
1.890E-12
1.036E-12
4.729E-14
2.430E-13
1.457E-13
3.564E-13
1.839E-12
1.339E-15
2.919E-13
∆ 3He
3
Ar
2.385E-08
2.682E-08
2.431E-08
2.465E-08
2.978E-08
2.315E-08
2.444E-08
2.542E-08
2.396E-08
1.988E-08
1.759E-08
1.162E-08
2.048E-08
36
9.681E-07
1.254E-06
1.116E-06
1.014E-06
1.303E-06
1.030E-06
1.111E-06
1.172E-06
1.082E-06
8.088E-07
6.494E-07
3.009E-07
2.054E-07
1.424E-06
2.046E-06
4.614E-07
5.068E-06
1.515E-05
9.880E-06
2.376E-06
9.235E-06
6.772E-06
1.567E-05
2.826E-05
2.786E-08
9.132E-06
He
∆ 40Ar
8.651E-09
1.054E-08
2.464E-09
3.596E-08
8.904E-08
5.963E-08
1.234E-08
5.569E-08
4.224E-08
8.123E-08
1.494E-07
2.840E-10
5.518E-08
∆ 4He
4
Ar
6.929E-07
8.924E-07
8.199E-07
7.859E-07
9.090E-07
7.652E-07
7.887E-07
8.350E-07
7.544E-07
6.354E-07
5.846E-07
3.527E-07
3.789E-07
40
2.742E-04
3.704E-04
3.339E-04
2.982E-04
3.881E-04
3.074E-04
3.299E-04
3.494E-04
3.112E-04
2.449E-04
2.087E-04
8.424E-05
1.060E-04
1.256E-07
2.218E-07
1.976E-07
1.612E-07
1.680E-07
1.483E-07
2.360E-07
2.212E-07
2.067E-07
1.521E-07
9.213E-08
6.767E-08
3.961E-08
Ne
∆ 84Kr
5.118E-10
9.040E-10
8.297E-10
7.227E-10
6.997E-10
6.313E-10
1.008E-09
9.411E-10
8.605E-10
6.175E-10
3.843E-10
2.863E-10
1.774E-10
∆ 20Ne
20
Kr
7.041E-10
8.288E-10
7.542E-10
7.149E-10
8.377E-10
7.331E-10
7.212E-10
7.653E-10
7.236E-10
6.296E-10
6.123E-10
4.527E-10
4.590E-10
84
3.675E-08
4.950E-08
4.243E-08
3.897E-08
5.016E-08
4.026E-08
4.024E-08
4.079E-08
3.744E-08
2.849E-08
2.871E-08
9.614E-09
1.366E-08
Ne
∆ 22Ne
22
4.718E-11
7.234E-11
4.773E-11
4.945E-11
6.524E-11
4.999E-11
6.151E-11
4.901E-11
5.358E-11
3.420E-11
3.785E-11
1.465E-11
2.157E-11
∆ 132Xe
6.088E-11
9.623E-11
9.809E-11
7.747E-11
8.021E-11
6.903E-11
1.122E-10
9.874E-11
9.163E-11
7.167E-11
5.177E-11
3.985E-11
3.556E-11
Xe
2.350E-09
3.227E-09
2.587E-09
2.463E-09
3.132E-09
2.506E-09
2.511E-09
2.444E-09
2.354E-09
1.753E-09
1.868E-09
5.142E-10
8.737E-10
132
1.282E-08
2.268E-08
2.018E-08
1.654E-08
1.720E-08
1.520E-08
2.412E-08
2.266E-08
2.111E-08
1.560E-08
9.439E-09
6.918E-09
4.027E-09
Table 5.2: Noble gas isotopes concentrations in thermal waters along the Israeli side of the Dead Sea Transform, in [ccSTP/g].
Sample
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
Hamat Tveria
Mineral
Qedem
Sample
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
Hamat Tveria
Mineral
Qedem
48
49
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
H.Tveria
Mineral
Qedem
Sample
1.424E-06
2.046E-06
4.614E-07
5.068E-06
1.515E-05
9.880E-06
2.376E-06
9.235E-06
6.772E-06
1.567E-05
2.826E-05
2.786E-08
9.132E-06
He
8.651E-09
1.054E-08
2.464E-09
3.596E-08
8.904E-08
5.963E-08
1.234E-08
5.569E-08
4.224E-08
8.123E-08
1.494E-07
2.840E-10
5.518E-08
∆He
1.388E-07
2.451E-07
2.184E-07
1.782E-07
1.857E-07
1.639E-07
2.608E-07
2.445E-07
2.284E-07
1.681E-07
1.018E-07
7.479E-08
4.376E-08
Ne
5.183E-10
9.144E-10
8.400E-10
7.306E-10
7.083E-10
6.386E-10
1.019E-09
9.515E-10
8.702E-10
6.253E-10
3.899E-10
2.904E-10
1.818E-10
∆Ne
2.754E-04
3.719E-04
3.352E-04
2.994E-04
3.897E-04
3.063E-04
3.312E-04
3.508E-04
3.125E-04
2.458E-04
1.930E-04
8.459E-05
6.107E-05
Ar
6.939E-07
8.936E-07
8.210E-07
7.870E-07
9.104E-07
6.883E-06
7.898E-07
8.362E-07
7.555E-07
6.362E-07
5.229E-06
3.532E-07
6.089E-06
∆Ar
6.448E-08
8.684E-08
7.444E-08
6.837E-08
8.800E-08
7.063E-08
7.060E-08
7.156E-08
6.568E-08
4.997E-08
5.037E-08
1.687E-08
2.396E-08
Kr
1.235E-09
1.454E-09
1.323E-09
1.254E-09
1.470E-09
1.286E-09
1.265E-09
1.343E-09
1.270E-09
1.105E-09
1.074E-09
7.941E-10
8.053E-10
∆Kr
8.739E-09
1.200E-08
9.620E-09
9.158E-09
1.165E-08
9.321E-09
9.337E-09
9.091E-09
8.754E-09
6.521E-09
6.947E-09
1.912E-09
3.249E-09
Xe
1.754E-10
2.690E-10
1.775E-10
1.839E-10
2.426E-10
1.859E-10
2.288E-10
1.823E-10
1.992E-10
1.272E-10
1.408E-10
5.448E-11
8.021E-11
∆Xe
Table 5.3: Elemental noble gas concentrations in thermal waters along the Israeli side of the Dead Sea Transform, in [ccSTP/g]. H.Tveria
refers to the spring of Hamat Tveria. Argon, which was used in this study only for NGT determination, is corrected to atmospheric
component, as explained in sections 2.2 and 4.2.
5 Results
A.6, therefore their argon concentrations were calculated only from
sections 2.2 and 4.2.
36Ar,
as explained in
5.1 Noble gas temperatures
The concentrations from Table 5.3 were fitted to different excess air models to obtain
NGTs and the atmospheric component of each noble gas, as explained in chapters 2 and
4.
Examination of the neon concentrations in Table 5.3 reveals that Ein Gofra, Ein Balzam,
Meitzar 2, Ein Makla, Tzofar 20, Hamat Tveria, Mineral and Ein Qedem contain less than
2 · 10−7 ccSTP/g, therefore it is expected to see degassing effects in the fit results. Mineral
and Ein Qedem, both from the Dead Sea area, have very low neon concentrations and
generally lower noble gas concentrations than the rest of the samples. This is probably
due to the very high salinities there (∼200 g/kg [Stern, 2010]), leading to a low solubility
in these cases. The high salinities are taken into account in the preparation of the samples
for the fit, since at least part of the water discharging from these locations originates from
Dead Sea water [Starinsky, 1974] and not fresh water. The initial salinity in Panga was
hence set to 200 g/kg for the Mineral and Ein Qedem samples.
Noble gas concentrations from all the samples were fitted with the UA model, which is the
most simple model (Equation 2.16). Ein Gofra, Ein Balzam, Ein Makla, Hamat Tveria
and Ein Qedem show negative excess air fractions, which indicates degassing processes.
However, this model is very simplified, thus the CE model is used.
The CE model was chosen to fit the samples of this study, since it can be adopted for
degassed samples as well. In a first step, all the samples where fitted with initial F = 0.5
(see Equation 2.20). Only Meitzar 3 and Ein Yahav 16 give reasonable results, with
0 < A < 0.05 and 0 < F < 0.1, therefore another fit was performed, this time with F > 1,
or to be more concrete F = 2, for the samples that show possible degassing from their
neon concentrations and the fit results of the UA model. The CE model fit results are
shown in Table 5.4.
The results were optimized using Monte Carlo simulations, as explained in chapter 4. The
samples can be divided into four groups of different results characteristics. Examples of
Monte Carlo results from each group are shown in Figure 5.1.
1. Normal, reasonable results: There are two samples in this group. Ein Yahav 16
is the only sample which did not need to be optimized. Ein Yahav 6, as well, gave
reasonable fit parameters (F < 1 and A << 1).
2. Degassing: This group consists of Ein Gofra, Ein Makla, Ein Balzam, Hamat
Tveria, Tzofar 20 and Mineral. They are characterized by F >> 1 and A <<
1, indicating degassing. The CE model is not aimed to estimate the amount of
degassing, therefore in such case the model parameters have big uncertainties. As
explained in chapter 2, the excess air composition is determined by the fractionation
F . In case of a degassed sample, the model calculated a “reversed” fractionation, so
that air would be degassing from the sample with the same composition of the excess
air and there is a big uncertainty in qualifying the degree of degassing. The samples
in this group are the same samples that have a relatively low neon concentration.
50
5.1 Noble gas temperatures
6577_EIN_YAHAV_6_2
6577_EIN_YAHAV_6_2
1,200
30
70
1,000
28
60
800
50
Counts
40
Counts
T [°C]
26
600
24
30
400
20
22
200
10
20
0
0.02
0.04
0.06
0.08
0
0.1
20
22
24
26
A [ccSTP/g]
28
30
32
T [°C]
6585_TZOFAR_20_1
6585_TZOFAR_20_1
500
33.5
300
33
250
32.5
400
200
Counts
150
Counts
T [°C]
300
32
200
31.5
100
31
100
50
30.5
0
-0.02
0
0.02
0.04
0.06
0.08
0.1
0
0.12
30.5
31
31.5
32
A [ccSTP/g]
32.5
33
33.5
34
T [°C]
6546_SHAMIR_121210_1
6546_SHAMIR_121210_1
3,000
2,500
13
2,500
2,000
12.5
Counts
12
Counts
T [°C]
2,000
1,500
1,500
1,000
11.5
1,000
500
11
10.5
500
0
0.002
0.004
0.006
0.008
0
0.01
10.5
11
11.5
12
A [ccSTP/g]
12.5
13
T [°C]
6606_EIN_QEDEM_1
6606_EIN_QEDEM_1
1,200
21.5
200
1,000
21
20.5
150
800
Counts
100
Counts
T [°C]
20
600
19.5
400
19
50
18.5
200
18
0
0.019
0.02
0.021
0.022
A [ccSTP/g]
0.023
0.024
0.025
0
18
18.5
19
19.5
20
20.5
21
21.5
T [°C]
Figure 5.1: Representative results of Monte Carlo simulations, that were performed in
order to optimize the NGT fits with the CE model (Equation 2.20). The
results are listed in Table 5.4. The simulated temperature against the excess
air fraction in each sample is plotted on the left column and a histogram of the
simulated temperatures is plotted on the right column. Each of the samples
represents a different type of results obtained in this study. From the top to
bottom the samples are: Ein Yahav: Reasonable excess air fraction, Tzofar
20: Degassing, Shamir: Quasi equilibrated and Ein Qedem: Results with low
probability.
51
5 Results
3. Results with a low probability: Meitzar 2 and Ein Qedem have F >> 1 and
A << 1 like the second group, but their χ2 are extremely high. The NGT obtained
for Meitzar 2 is only 2.6 ◦ C, indicating that something went wrong with this sample.
However, there were no special events during the sampling, accept of small bubbles
seen in the sampling tube (see Appendix A.5), but these were also formed in other
samples. Ein Qedem is the saltiest sample and has generally lower noble gas concentrations that the rest of the samples (Table 5.3), suggesting that massive degassing
might took place before the sample was taken.
4. quasi-equilibrated samples: This group consists Shamir, Meitzar 3 and Ein Yahav
116, with F ; A << 1. However, there is a possibility that these parameters are
positive, due to the big uncertainties. Meitzar 3 indeed shows reasonable results
with the UA. It is inferred that these samples are close to equilibrium state.
Overall, the CE model gives reasonable NGTs for most samples, ranging from 11 ◦ C in
Shamir to 31 ◦ C in Tzofar 20 in the north and south, respectively. Meitzar 2 and Ein
Qedem are excluded from the discussion due to big uncertainties of their results. The
average NGT in the northern part of the sampling area is 17.8 ◦ C and in the southern,
the average is 23.9 ◦ C. These temperatures are comparable with the present annual mean
air temperatures. According to data from the Israeli Meteorological Service, an annual
average air temperature of 18 ◦ C was measured between 1983 – 2000 in Kfar Blum, north
of Lake Kinneret. In Sodom, south of the Dead Sea, an annual average air temperature
of 23 ◦ C was measured in 1983 – 2000. Hamat Tveria and Tzofar 20 have slightly higher
NGTs than the rest of the samples in their regions, north and south respectively. This is
further discussed in chapter 6.
52
5.1 Noble gas temperatures
Table 5.4: Results of fit to the CE model. The upper table shows the fitted NGTs and the
atmospheric component of 3He and He (equilibrium + excess air). The lower
table shows the fit parameters. H.Tveria refers to the spring of Hamat Tveria.
The uncertainties of the parameters are the calculated standard deviations of
the Monte Carlo simulations. Ein Yahav 16 was not optimized by Monte Carlo
and has a chi square in [%], refers to the probability to obtain the listed fit
results.
Sample
NGT
[◦ C]
∆NGT
[◦ C]
Shamir
Gofra
Meitzar 2
Meitzar 3
Makla
Balzam
H. Tveria
Mineral
Qedem
Yahav 6
Yahav 16
Yahav 116
Tzofar 20
11.09
19.24
2.64
18.80
16.19
17.60
23.85
21.88
19.71
24.00
22.31
20.13
31.23
0.18
0.42
0.11
0.90
0.44
0.15
1.70
1.90
0.45
1.46
0.93
0.64
0.35
3
He atm (CE)
[ccSTP/g]
∆ 3He atm (CE)
[ccSTP/g]
Heatm (CE)
[ccSTP/g]
∆Heatm (CE)
[ccSTP/g]
5.672E-16
2.478E-16
3.155E-16
4.116E-16
3.040E-16
3.271E-16
1.864E-16
2.415E-15
8.647E-17
4.560E-16
7.696E-16
1.157E-15
2.727E-16
6.171E-08
3.233E-08
4.168E-08
5.349E-08
3.854E-08
4.305E-08
2.447E-08
1.948E-08
1.139E-08
6.098E-08
5.817E-08
6.877E-08
4.289E-08
3.919E-10
1.740E-10
2.272E-10
2.877E-10
2.186E-10
2.354E-10
1.348E-10
1.539E-09
6.302E-11
3.364E-10
5.111E-10
7.533E-10
1.967E-10
8.461E-14
4.386E-14
5.671E-14
7.294E-14
5.240E-14
5.866E-14
3.325E-14
2.662E-14
1.548E-14
8.314E-14
7.956E-14
9.449E-14
5.847E-14
Sample
A
[ccSTP/g]
∆A
[ccSTP/g]
F
∆F
χ2
∆χ2
Shamir
Gofra
Meitzar 2
Meitzar 3
Makla
Balzam
H.Tveria
Mineral
Qedem
Yahav 6
Yahav 16
Yahav 116
Tzofar 20
1.234E-04
1.888E-03
1.055E-04
1.432E-02
8.110E-07
-6.735E-07
1.904E-02
-3.070E-03
2.156E-02
4.346E-02
3.449E-03
1.200E-03
6.458E-03
5.287E-04
1.614E-03
4.997E-03
7.195E-03
7.210E-06
1.137E-06
1.112E-02
1.664E-03
7.615E-04
1.319E-02
3.021E-03
1.450E-03
7.772E-03
-3043.66
4.88
958.12
-3.22
947.41
1001.41
3.79
102.16
1.52
0.63
0.09
-1416.44
148.40
989.82
4.83
90.10
102.95
174.82
336.34
93.55
334.55
0.01
0.01
0.57
1878.52
574.43
5.3
1.8
159.1
3.4
7.9
3.9
6.4
2.3
645.5
1.1
0.2
1.3
3.9
4.2
2.3
25.3
3.4
5.3
3.4
4.9
2.7
50.7
1.5
69.1
1.7
4.5
53
5 Results
5.2 Helium isotopes
Helium isotopes, 3He and 4He, are used as tracers of intrusion of volatiles from the mantle
into the crust on the Israeli side of the Dead Sea Transform. The concentrations of both
isotopes are plotted against each other in Figure 5.2. The results show two main branches
Figure 5.2: 3 He concentration versus 4 He concentration. Samples from the same geographic area are marked with the same color. Note the different trends of
samples from the northern part of the sampling area and samples from southern part of the sampling area.
and the sample from Hamat Tveria, which lies between them. The upper branch consists
of samples from the northern regions of the sampling area (Hula Valley, Lake Kinneret and
Yarmouk Valley) and the lower branch consists of samples from the southern regions (Dead
Sea and Arava Valley), meaning that northern samples tend to contain more 3He relative
to the southern ones. This might be an evidence of an enhanced mantle derived helium
component there, since 3He predominantly originates from the mantle. Interesting is that
samples that are located in the same field have different isotope compositions. Meitzar 2
and Meitzar 3 wells are a few meters away from each other, but contain different amounts
of 3He and 4He. Ein Makla and Ein Balzam, the sampled springs of Hamat Gader, lie
5 km away from Meitzar and also have significantly different amounts of 3He and 4He. The
same is true for the three wells from Ein Yahav. Mineral and Ein Qedem are located by
the Dead Sea, only 4 km away from each other and show different isotope concentrations.
Hamat Tveria lies on the western shore of Lake Kinneret and has much higher isotope
concentrations than the rest of the samples. Looking more closely at Figure 5.2, it can
54
5.3 Helium components
be noticed that the samples from Tzofar 20 is shifted from the other samples as well, so
that the two might form a new branch. Interesting is that Tzofar 20 and the samples from
Lake Kinneret shores (Hamat Tveria and Gofra) show a correlation between their helium
concentration (≈ 4He) and their electrical conductivity, which is used here as an indication
for water salinity (Figure 5.3).
Figure 5.3: Helium concentration versus electrical conductivity, which is related to the
water salinity. The conductivity is presented on a logarithmic scale. Samples
from the same geographic area form clusters and show different trends, except
of the sample from Tzofar 20, which lies closer to the samples from the eastern
and western Lake Kinneret.
5.3 Helium components
The fractionation ratio 3He/ 4He is calculated and separated into tritiogenic, atmospheric
and terrigenic, consists of radiogenic and mantle components, as shown in Figure 2.6. The
decomposition is done in steps. First of all, the tritiogenic component is discussed, then
terrigenic helium is separated from atmospheric helium and finally mantle derived helium
and crustal helium are calculated from the terrigenic helium.
55
5 Results
5.3.1 Tritiogenic helium:
Tritium was measured in order to determine the tritiogenic helium amount in each sample.
The results are given in Table 5.5. Values under 2 TU are considered as background noise
since it is the detection limit of the measurement system.
Table 5.5: Tritium concentrations in the thermal waters of the Dead Sea Transform. The
concentrations are calculated from equation 2.13. Values smaller than 2 TU
are under the detection limit of the measurement system, therefore they are
interpreted as 0 TU.
Sample
Gofra
Meitzar 2
Balzam
Shamir
Meitzar 3
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
H.Tveria
Mineral
Qedem
3H
[TU]
1.01
0.91
0.12
0.62
0.66
0.66
1.90
6.52
0.25
0.83
0.74
3.34
0.08
∆ 3H [TU]
0.92
0.96
0.97
1.01
1.00
0.95
1.02
1.10
1.04
1.06
1.03
1.09
0.87
Thermal waters are generally considered to be old enough, so that no tritium should be
found in them. Hence, finding tritium in such samples indicates mixing with younger
water. Ein Yahav 6 and Mineral are the only samples that contain a significant amount
of tritium. However, the amount of 6.5 TU, measured in Ein Yahav 6, corresponds to
∼100 TU≈2 · 10−13 ccSTP/g of tririogenic 3He produced in the water since the tritium
bomb peak in 1963, calculated from Equation 2.13. This is one order of magnitude lower
than the total concentration of 3He in the sample, therefore is neglected in the further
analysis. The case of Mineral is different, since the 3He concentration of this sample is
∼3 · 10−14 , which is lower than the tritiogenic helium concentration that should have been
dissolved there of ∼51 TU≈1 · 10−13 ccSTP/g. It can be explained by mixing of the water
in this well with younger water from the Dead Sea or the local fresh groundwater. However,
the initial amount of tritium infiltrated during the mixing is not known, therefore it is not
clear how much tritiogenic helium presents in this sample. Hence, tritiogenic helium in
the sample from Mineral is neglected too.
5.3.2 Terrigenic versus atmospheric helium:
The ratios 3He/ 4He and Ne/He were calculated for each sample in or order to distinguish
between atmospheric and terrigenic helium in the water. The results are given in Table
5.6.
Figure 5.4 shows the relationship between the atmospheric and terrigenic components. It
is noticeable that samples from the same geographic area lie close to each other. Waters
56
Rs
2.371E-06
3.893E-06
2.394E-06
3.008E-06
2.856E-06
2.531E-06
5.224E-07
6.718E-07
5.419E-07
5.853E-07
1.458E-06
1.192E-06
8.463E-07
Sample
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
H. Tveria
Mineral
Qedem
1.011E-07
1.702E-07
1.074E-07
1.538E-07
1.258E-07
1.059E-07
2.008E-08
2.661E-08
2.177E-08
2.293E-08
6.553E-08
4.959E-08
3.236E-08
∆Rs
1.713
2.813
1.730
2.173
2.063
1.829
0.377
0.485
0.392
0.423
1.053
0.862
0.611
Rs /Ra
0.073
0.123
0.078
0.111
0.091
0.077
0.015
0.019
0.016
0.017
0.047
0.036
0.023
∆(Rs /Ra )
1.714
2.815
1.734
2.174
2.064
1.829
0.377
0.485
0.391
0.423
1.053
0.858
0.611
Rcorr /Ra
0.073
0.123
0.078
0.111
0.091
0.077
0.015
0.019
0.016
0.017
0.047
0.037
0.023
∆(Rcorr /Ra )
0.097
0.120
0.473
0.035
0.012
0.017
0.110
0.026
0.034
0.011
0.004
2.685
0.005
Ne/He
0.001
0.001
0.003
0.000
0.000
0.000
0.001
0.000
0.000
0.000
0.000
0.029
0.000
∆ (Ne/He)
Table 5.6: 3He/ 4He and Ne/He ratios. Rs corresponds to the measured 3He/ 4He. Rcorr is the air corrected 3He/ 4He ratio, from Equation
2.30, using the measured Ne/He ratios. The 3He/ 4He ratios are normalized to 3He/ 4He in air, Ra , which is: (1.384 ± 0.006)·10−6
[Clarke et al., 1976].
5.3 Helium components
57
5 Results
from the north have the highest 3He/ 4He ratios, between 1 Ra in Hamat Tveria and 2.8 Ra
in Shamir. Ein Qedem and Mineral from the Dead Sea have 3He/ 4He ratios of 0.6 Ra
and 0.9 Ra , respectively and southernmost samples from the Arava Valley show a range of
3He/ 4He from 0.4 R to 0.5 R .
a
a
All the samples have 3He/ 4He ratios that are higher than the typical crustal ratio (see
Section 2.4), indicating that they all contain some fraction of mantle derived helium. As
for the atmospheric component, all the samples, except of Mineral and Meitzar 3, lie
almost exactly on the mixing line between crustal and mantle derived helium endmembers
and contain a very minimal amount of Ne/He. This indicates that helium in the samples is
mostly of terrigenic origin and only a very small amount comes from atmospheric origin.
This even holds for the samples from springs, that usually experience more ad-mixing
compared to wells and therefore would be expected to contain higher atmospheric helium.
In contrast, Mineral and Meitzar 3 have a considerable atmospheric component. These
were also the samples that were most affected by air contamination during sampling, as
described in Appendix A.5.
Another interesting sample is Hamat Tveria, that shows a 3He/ 4He ratio which is very
close to the atmospheric ratio (Rs ≈ 1). One could think that it is a result of a prominent
atmospheric component, but this is not the case. Hamat Tveria has a very low neon to
helium ratio (Table 5.6), due to the high 4He concentration in this sample shown in Figure
5.2. During the time of recharge, this water contained the helium and neon concentrations
of the atmospheric endmember, which are marked as “Atmosphere” in Figure 5.4. The
water got older and incorporate helium from the crust and from the mantle, causing
it to move along a mixing line between atmospheric and local terrigenic endmembers. It
happened to be that the final helium isotope composition is very similar to the atmospheric
ratio. This case demonstrates how useful it is to examine the neon concentration together
with the helium isotopes concentrations, because they can be a big help to distinguish
between atmospheric and terrigenic influences.
58
5.3 Helium components
Air correction:
The separation to components was performed in two ways: By calculation of the mixing
line intercept (Equation 2.35) and by using the fitting results of Panga from Table 5.4, for
the atmospheric component (Equation 2.38).
The first method is demonstrated in Figure 5.4 for two of the samples, that have the highest
Ne/He ratio. Since each sample is located on a mixing line between the atmospheric end
member and a terrigenic end member, the uncontaminated 3He/ 4He can be determined.
Such a mixing line is drawn for Mineral and Meitzar 3, as shown in Figure 5.4. The
intercept of this line with the y axis represents the terrigenic component of the sample. In
case of Mineral, this value is almost identical to the terrigenic component of Ein Qedem,
which is located only about 5 km away from Mineral. If Mineral would really contain a
significant amount of tritiogenic helium, it would not lie so close to the point of Ein Qedem,
after the air correction. Therefore it is deduced that the sample of Mineral consists of a
big atmospheric helium component, a smaller terrigenic one and no further components.
Applying the same correction for Meitzar 3 brings it close to Meitzar 2, which is located
in the same well field.
The differences between the results of the two methods do not exceed the 10 % except for
Mineral, having about a 39 % difference. The higher differences occur in samples that have
Figure 5.4: 3He/ 4He versus the concentration ratio of Ne/He. Rs is normalized to the isotopic ratio in the atmosphere, Ra = 1.384 · 10−6 from Clarke et al. [1976]. The
solid lines represent mixing lines between the atmosphere, crust and mantle
endmembers, following Figure 2.7. The dashed lines represent the air corrections of Mineral and Meitzar 3 (see the text for more information). The data
for this graph is taken from Table 5.6.
59
5 Results
a more significant atmospheric component, or better to say, a high Ne/He ratio, as seen in
Figure 5.4. Since most of the samples in this study have small atmospheric components,
both methods are proper for the air correction, except of the sample from Mineral. Since
not all the samples could be well fitted, as shown in Section 5.1, the first method of
intercept determination was chosen for the further analysis. The corrected 3He/ 4He ratios
are given in Table 5.6.
5.3.3 Separation of atmospheric, crustal and mantle derived helium
The relative parts of atmospheric, crustal and mantle derived helium components were
calculated for each sample using equations 2.44 and 2.42. The results are given in Table
5.7. Figure 5.5 combines all the three components. It is actually an extension of Figure
5.4, where the terrigenic component is now split into crustal and mantle components.
Figure 5.5: The relative parts of mantle, atmospheric and crustal helium components in
each sample. The data for the graph is taken from Table 5.7. The mantle,
atmospheric and crustal endmembers are marked on the graph and each axis
represents a mixing line between two of them. The grid lines of the mantle
component are marked with a solid line, the grid lines of the atmospheric
component are marked with a dashed-dotted line and the grid lines of the
crustal component are marked with a dashed line. The percentage of each
component in a sample is read parallel to these lines. For example, the sample
from Mineral contains 69.2 % atmospheric 4He, 2.29 % 4He from a mantle origin
and 28.5 % crustal helium.
60
5.3 Helium components
It is seen in Table 5.7 that the fraction of 3He from the mantle is above 90 % in almost all
samples. One might be surprised to see that Tzofar, for example, has about 96 % mantle
3He, a bit higher than Meitzar 3, while Figure 5.4 shows that the mantle component of
Meitzar 3 is bigger than that of Tzofar 20.
The mantle 4He percentage in Table 5.7 indicates that Meitzar 3 indeed contains more
mantle helium than Tzofar 20 (22 % compared to 6 %). However, it is a direct consequence
of the mixing model used here. As explained in Section 2.4.1, only 4He includes useful
information about the partition of atmospheric, crustal and mantle derived helium in the
water.
All the samples, except of Mineral, contain more than 90 % terrigenic helium. This is
visualized in Figure 5.5, where all data points are concentrated at the bottom of the triangle
and show low atmospheric fractions. Mineral lies far away from the other samples, with
69.2 % atmospheric helium, which is an evidence to the air contamination that occurred
during sampling (see Appendix A.5 for more information). Therefore it is excluded from
the general discussion later.
Mantle fraction of the total He in the sampled waters ranges from 4 % in Ein Yahav 116
to 39 % in Shamir. It can be noticed here as well, that northern samples contain more
mantle helium.
61
5 Results
4 He
∆4 HeA
1.32
1.07
6.70
0.39
0.13
0.21
7.67
1.36
2.20
0.65
0.08
81.17
0.20
3 He
A
0.02
0.02
0.09
0.01
0.00
0.00
0.04
0.01
0.01
0.00
0.00
5.57
0.00
∆3 HeA
0.06
0.05
0.30
0.02
0.01
0.01
0.31
0.06
0.09
0.03
0.00
7.29
0.01
4 He
M
23.25
38.39
22.20
29.82
28.38
25.11
4.64
6.44
5.11
5.63
14.39
2.19
8.26
∆4 HeM
1.01
1.71
1.08
1.54
1.26
1.06
0.20
0.27
0.22
0.23
0.66
0.91
0.32
3 He
M
98.05
98.63
92.74
99.15
99.37
99.20
88.79
95.87
94.33
96.14
98.74
18.36
97.63
∆3 HeM
74.48
58.59
66.20
69.33
71.35
74.50
92.47
92.90
94.03
94.10
85.52
27.87
91.61
4 He
C
5.98
6.15
6.12
7.21
6.22
5.90
5.15
5.49
5.53
5.44
6.32
7.66
5.35
∆4 HeC
0.63
0.30
0.55
0.46
0.50
0.59
3.54
2.77
3.47
3.22
1.17
0.47
2.17
3 He
C
1.01
1.71
1.08
1.54
1.26
1.06
0.20
0.27
0.22
0.23
0.66
5.65
0.32
0.03
0.02
0.03
0.03
0.02
0.03
0.14
0.11
0.14
0.13
0.05
0.10
0.08
∆3 HeC
Table 5.7: Atmospheric, crust and mantle derived helium components of the total helium concentration in the water. Their
calculation is explained in Section 2.4.1.
Sample
2.27
3.02
11.59
0.85
0.28
0.39
2.89
0.66
0.86
0.27
0.09
69.94
0.12
A
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav1 16
Tzofar 20
H. Tveria
Mineral
Qedem
62
5.4 Carbon isotopes
5.4 Carbon isotopes
δ 13 C is analysed in order to gain further information about volatile fluxes from the mantle.
The results are given in Table 5.8. The isotopic composition of the samples varies from
the most depleted with −11.8 h (Meitzar 3) to Mineral with +5.5 h, which is the only
positive value and about 6 h higher than the second enriched sample, Hamat Tveria,
with −1.8 h. Most of the samples, except of Mineral, have δ 13 C values in the range of
groundwater, between −20 h and 0 h. The extreme high δ 13 C in the sample from Mineral
can be due to the fact that the hyper saline water of this spring consists of some fraction of
Dead Sea water, which also shows positive δ 13 C values, up to 4 h [Barkan et al., 2001].
Table 5.8: Carbon isotopes and concentrations in thermal waters in the Israeli side of the
Dead Sea Transform. CO2 / 3He data are taken from Torfstein et al. [2013]. M, L
and S correspond to mantle, limestone and sediment derived CO2 components,
which were calculated from the mixing model explained in Section 2.4.2. The
uncertainty of these components is determined from the range of the isotope
signature of CO2 from mantle, limestone and sediment origin and are explained
in the text.
Sample
δ 13 C [h]
∆ δ 13 C [h]
CO2 / 3He
M [%]
L [%]
S [%]
Meitzar 3
Balzam
Makla
Meitzar 2
Qedem
Gofra
Tzofar
H. Tveria
Yahav 6
Yahav 16
Yahav 116
Mineral
Shamir 1
-11.793
-8.446
-7.633
-3.399
-1.096
-8.382
-7.150
-1.847
-5.714
-3.396
-6.203
5.543
-1.535
0.004
0.028
0.013
0.003
0.019
0.019
0.026
0.001
0.002
0.025
0.002
0.010
0.007
1.190E+09
7.540E+08
2.315E+08
2.300E+09
1.254E+09
2.330E+09
1.580E+07
1.873E+08
-
1.26
1.99
6.48
0.65
1.20
0.64
94.94
8.01
-
59.70
70.29
69.48
88.16
95.41
71.56
1.80
87.57
-
39.04
27.72
24.04
11.19
3.39
27.80
3.26
4.42
-
Ein Makla, Tzofar, Ein Yahav 6 and Ein Yahav 116 are the only samples that lie in
the typical δ 13 C range in the upper mantle: −8 h < δ 13 C < −5 h [Sano and Marty,
1995]. Since both helium and carbon are still escaping from the mantle, it would be
expected to find more mantle helium in samples that have δ 13 C typical for the mantle,
but Figure 5.6 shows no such correlation. This is probably due to the fact that δ 13 C
could only be measured from the total carbon in the water, including all species of carbon.
Furthermore, in contrast to 3He/ 4He, no correlation between δ 13 C and geographic locations
is seen(Figure 5.6).
In order to have a better understanding of the carbon origin, the δ 13 C data are combined
with CO2 / 3He data from [Torfstein et al., 2013]. These authors suggest that the origin of
the carbon is mostly from reactions in the crustal sediments and mention the option to
use the isotopic signatures in order to gain more information. After finding out that only
a part of the helium in the waters originates from the mantle, it can be expected to see a
mixing between mantle carbon and carbon from crustal sources.
63
5 Results
(a) δ 13 C versus terrigenic helium from Table 5.6.
(b) δ 13 C versus mantle derived helium fraction from Table 5.7.
Figure 5.6: Relationship between δ 13 C and mantle derived helium in the samples.
The shading marks the typical signature of mantle derived carbon of
−5 h<δ 13 C<−8 h.
64
5.4 Carbon isotopes
In order to estimate the fraction of mantle derived carbon in the water samples, a mixing
model between three end members of the upper mantle (MORB), marine carbonate and
sediments is used, as explained in Section 2.4.2. The results are listed in Table 5.8. The
isotopic signatures and CO2 / 3He taken for the end members are given in Table 2.2. These
values are not discrete, but have a range. Hence, they are varied here in order to determine
their influence on the results.
CO2 / 3He ratios in limestone and in sediments are varied between 1 · 1012 and 1 · 1014 .
This changed the calculated endmember fractions of all samples by only a few percent.
Changing the δ value of the mantle affects the results by about ±1 %. The change of
δ value of the limestone leads to a change of ±10 % of the limestone and the sediments
endmembers fractions. varying the δ value of the sediments leads to a change of ±50 %
in sediment fractions and ±30 % for the limestone fraction in all the samples, except
for limestone in Tzofar 20, which its limestone varies in ±90 %. The CO2 / 3He ratio of
the upper mantle was the parameter that mostly affected the results. Taking a value of
7.5 · 108 instead of 1.5 · 109 reduces the results to about a half and taking 3 · 109 doubles
the results.
It is important to notice that the relative share of CO2 from the mantle is determined
only from the CO2 / 3He ratios (Equation 2.50). Looking at the CO2 / 3He ratios of the
samples and the endmembers shows that they are too low relative to the CO2 / 3He ratio
of the mantle endmember, thus it can be expected to see mantle derived CO2 fractions
that exceed the 100 %. Therefore, the analysis of the carbon origin in the thermal waters
from mantle, limestone and sediment sources can be done only qualitatively. Some of
the endmembers percentages obtained from the mixing model are negative or bigger than
100 %, which makes no sense. In order to achieve reasonable values, the ratio of CO2 / 3He
in the mantle (Table 2.2) was reduced to 1.5 · 107 , which is 100 times smaller than the
original value of 1.5 · 109 and the maximal value that gives reasonable results. Further
reducing it lowers the mantle fractions, but leaves similar proportions between the end
members.
The calculated mantle derived CO2 is found in a range of 0.64 % in Ein Gofra to more
than 94 % in Tzofar 20, which is the southernmost sample. Tzofar 20 and Ein Makla,
which have δ 13 C values that are typical to the mantle, show bigger mantle components in
Figure 5.7. However, Hamat Tveria has a δ 13 C signature higher than −5 h and also show
a relatively big mantle component. No correlation is seen between the mantle components
and the geographic location of the samples, but it might be also due to the lack of CO2 / 3He
data for the southern samples. As for the limestone and sediment components, most of
the samples have qualitatively more limestone carbon than sedimental organic carbon, but
this is not necessarily the true, due to big uncertainties of ±20 %.
Since a much lower CO2 / 3He of was considered for the mantle endmember, it cannot be
considered as “mantle” any more. This means that the CO2 components distribution
shown in Figure 5.7 corresponds to three non-mantle end members. Moreover, CO2 / 3He
of the samples is on average lower than the global mean value of (1.5 ± 1.1) · 1010 in
groundwater [Sano and Marty, 1995 ; Sano and Williams, 1996 and de Leeuw et al., 2007.
This might be due to different fractionation ratios of carbon compared to helium isotopes
during gas-water exchange [Van Soest et al., 1998]. The solubility of carbon dioxide is
about 40 times bigger than the solubility of helium in thermal waters [Stephen and Stephen,
1963], means that the ratio of CO2 in gas and water phase is much smaller than the ratio
of helium in gas and water phase. It is therefore expected to measure a smaller fraction
65
5 Results
Figure 5.7: Mixing model between mantle derived carbon (M), marine carbonate in limestone (L) and carbon from sediments (S) components in the samples. The
samples are ordered by geographic locations, from north (top) to south (bottom). δ 13 C and CO2 / 3He of the end members from Table 2.2 are inserted into
the model equations described in Section 2.4.2. 1.5 · 107 is taken for the mantle
end member, which is 100 times smaller than the actual value, measured by
[Marty and Jambon, 1987], since the later yields endmembers percentages that
are negative or bigger than 100 %.
of CO2 in a gas phase, relative to the fraction of helium in a gas phase. This element
fractionation explains why Tzofar 20 contains so much more “mantle” derived carbon
than the other samples. This sample has the smallest CO2 / 3He, of 1.58 · 107 whereas
the rest of the samples have ratios of ∼ 108 to ∼ 109 . The crust underlies the southern
Dead Sea Transform is about 10 km thicker than the crust underlies the northern Dead
Sea Transform (Figure 3.3), suggesting that the water in the south can flow longer until
it reaches the surface, thus enhanced element fractionation occurs there.
Despite the uncertainties and especially the big variations of CO2 / 3He ratios in limestone
and sediment, the calculation is still useful since it shows that the carbon in the water
samples from the Israeli side of the Dead Sea Transform does not originate from mantle
sources. This confirms the findings of Torfstein et al. [2013], that the CO2 in the water
does not originate from the mantle.
66
5.5 Stable isotopes
5.5 Stable isotopes
δ 18 O and δ 2 H are measured in order to examine possible exchange of δ 18 O between the
groundwater and the local rocks in high temperatures, as a results of heat flux from the
mantle. The results are compared to the Global Meteoric Water Line (GMWL) and the
local Eastern Mediterranean Meteoric Water Line (EMWL). They are given in Table 5.9
and shown in Figure 5.8. The measured values are spread over a large range and with the
same enrichment level for δ 18 O and δ 2 H, in other words samples that are more enriched
in δ 18 O are also more enriched in δ 2 H. Tzofar 20, from Arava Valley, is the most depleted
sample, with δ 18 O = −8.38 h and δ 2 H = −57.6 h. Mineral, from the Dead Sea basin, is
the most enriched sample, with δ 18 O = −4.2 h and δ 2 H = +1.08 h.
Table 5.9: Stable isotopes.
Sample
δ 18 O
∆ δ 18 O
δ2H
∆ δ2H
Meitzar 3
Shamir
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Gofra
Tzofar 20
H. Tveria
Mineral
Qedem
-5.69
-7.03
-5.93
-7.02
-6.04
-6.60
-7.51
-6.91
-4.60
-8.38
-3.57
1.08
0.92
0.10
0.07
0.09
0.15
0.12
0.06
0.11
0.10
0.07
0.05
0.10
0.06
0.10
-28.68
-36.02
-30.74
-36.63
-31.41
-39.56
-47.83
-42.38
-21.84
-57.58
-16.00
-4.24
-4.49
0.41
0.11
0.56
0.90
0.53
0.45
0.66
0.54
0.12
0.13
0.50
0.40
0.57
The data in Figure 5.8 can be divided into three clusters: Northern cluster, Dead Sea cluster and Arava Valley cluster, each of them shows a different trend of the stable isotopes.
The stable isotopes of the northern cluster lie close to the EMWL, indicating that the the
water is of meteoric origin. The samples in this cluster lie on a line that is shifted from
the EMWL towards the GMWL, indicating evaporation. Mazor et al. [1973] attribute this
shift to a mixing between endmembers with different temperatures and salinities. Arad
and Bein [1986] indicate that these end members that form the thermal water in Meitzar
and Hamat Gader groups originate from different recharge zones in the north and south of
Golan Heights, from different recharge heights, therefore the depletion is due to elevation
effect, as explained in Section 2.4.3. The most depleted water in this cluster, from Meitzar
2 and Shamir, originates from the northern, higher parts of the Golan Heights and they
are also among the deepest sampled wells, thus contain a higher amount of mantle derived
helium, as explained before.
The stable isotopes of Ein Qedem and Mineral, in the Dead Sea cluster, lie on a mixing
line between the EMWL and the mean value of δ 18 O in the Dead Sea (4 h) [Gat and
Dansgaard, 1972], showing evaporation effect. It is deduced from chemical analyse that
the thermal springs along the Dead Sea consist of a mixing between local fresh water and
ancient Dead Sea water, namely evaporated sea water [Möller et al., 2007b], which can
explain the enrichment in δ 18 O and δ 2 H of the samples from the Dead Sea area.
67
5 Results
Figure 5.8: Stable isotopes signatures. The Global Meteoric Water Line (GMWL) [Craig,
1961a] and Eastern Mediterranean Meteoric Water Line (EMWL) [Gat and
Carmi [1970], cited by e.g. [Gat and Dansgaard, 1972]] are plotted for a better
orientation, where the EMWL is considered as the local meteoric water line in
Israel (LMWL), e.g. Gat and Dansgaard, 1972 and Möller et al., 2007a. The
data for this graph is given in Table 5.9.
The samples from Arava Valley are shifted from the EMWL and lie pretty close the the
GMWL. An explanation is found by Gat et al., 1969 ; Gat and Dansgaard, 1972 and Möller
et al., 2006, who refer to the water as ancient fossil water, suggesting that it might had
recharged as the climate in the area was cooler and more humid than known in the present
time. Additional trend which can be seen within this cluster is that Tzofar 20 lies away
from the other samples from Ein Yahav. Gat et al., 1969 and Möller et al., 2007a found
that the chemical composition of Tzofar 20 is different from the composition of the wells of
Ein Yahav, but interestingly similar to the composition of the water from Hamat Tveria in
the north. They indicate that these waters were formed from the same primordial brine,
that had splayed over over wide parts of the Dead Sea Transform. Interesting to notice
is that these two samples distinct from the others by their relationship between helium
isotopes and electrical conductivity, which support the findings of these authors.
Non of the samples is enriched in δ 18 O relative to other samples of the same geographic
area, indicating no evidence of water-rock interaction, thus no evidence of high water
temperatures. This supports the findings of Gat et al. [1969], who only found evidence of
δ 18 O exchange in a depth of 3150 m, which is much deeper than the sampling sites in this
work (Table 3.2).
68
6 Discussion
6.1 Spatial distribution of 3He/ 4He
The isotope ratio 3He/ 4He plays an important role in the determination of the origin of
helium in the thermal waters along the Dead Sea Transform. The samples were taken
from different geographic areas along the Israeli side of the Dead Sea Transform, from the
Hula Valley in the north to the Arava Valley in the south, showing a trend in the 3He/ 4He
ratios. 3He/ 4He is the highest in the northern samples and decreases towards the south. In
order to visualize effects of geographic location on the helium isotopes, the helium isotopes
concentrations and the ratio 3He/ 4He are plotted and referred to a specific location. The
city of Eilat is chosen, which is the southernmost city in Israel and is located on the Gulf
of Eilat, also known as the Gulf of Aqaba (Figure 6.1). The data are arranged in three
main clusters of three main regions: Samples from the north part of the sampling site,
samples from the Dead Sea area and samples from Arava Valley.
Figure 6.1: The distance from Eilat, at the Gulf of Aqaba, versus the helium isotopes
concentrations and their ratio. The locations of the sampling sites can be
found in Figure 3.5.
While the isotopes concentrations do not enable to distinguish between samples from
different clusters, the 3He/ 4He ratios show a distinct range in each cluster. The samples
from the Arava Valley have the lowest 3He/ 4He ratios, followed by the samples from the
Dead Sea area, with medium 3He/ 4He values and the samples from the north, which have
the highest 3He/ 4He ratios from all samples. Hence, 3He/ 4He ratio, compared to the
individual 3He and 4He concentrations, has a higher sensitivity to trace mantle derived
helium.
69
6 Discussion
The correlation between 3He/ 4He and the geographic location of a sample increases from
the south to the north of the Dead Sea Transform, leading to a larger diversity of 3He/ 4He
ratios in the northern samples compared to the southern ones (Figure 6.1). It can be
explained by the fact that the samples from the south are located closer by each other
compared to the samples from the north. Hamat Tveria and Shamir, for example, are
located 50 km away from each other and have a difference of ∼2.4 · 10−6 in their 3He/ 4He
values while Tzofar 20 and Ein Yahav, that are located less than 20 km apart, have similar
3He/ 4He values, with a difference of only ∼6.5 · 10−8 . Moreover, it is interesting to compare
the two wells from Meitzar, in the north, with the three wells from Ein Yahav in the south.
Even though the wells in each field contain water from different depths and aquifers, the
3He/ 4He levels of the samples within each of these fields (Table 5.6) vary from each other
by only 10 %.
6.2 Mantle, crustal and atmospheric components distribution
Figure 6.2: Percentages of mantle, crustal and atmospheric helium in the different regions
along the Dead Sea Rift: Hual Valley, Yarmouk Valley, eastern and western
shores of Lake Kinneret, the Dead Sea and Arava Valley. The data are taken
from Table 5.7. Notice the north to south gradient of the mantle fraction. All
the samples, aside from Mineral, contain a very low amount of atmospheric
helium.
The total concentration of helium in each sample can be separated to contributions from
70
6.3 The geothermal structure of the Dead Sea Transform
atmospheric, crustal and mantle sources (Figure 5.5), using the measured 3He/ 4He ratios
and the typical ratios of each of the endmembers (Table 2.1), as explained in Chapter 2.
The distribution of these components in different regions along the Dead Sea Transform
is shown in Figure 6.2. All samples, except of Mineral, contain mostly helium of crustal
origin, which makes sense since these thermal waters circulate in the crust. Mineral and
Meitzar 3 contain a significant contribution of atmospheric helium due to air infiltration
during the sampling process (see Appendix A.5). The rest of the samples in all regions
show less than 5 % atmosphere derived helium, meaning that more than 90 % of the helium
is terrigenic. In addition, no difference is seen in the share of the atmospheric helium
component in springs compared to wells. The increase mantle derived helium from south
to north is noticed in Figure 6.2, as well as an increase from the western to eastern sides
of Lake Kinneret. The maximal 3He/ 4He of 2.8 Ra is obtained in Shamir. For comparison,
Torfstein et al. [2013] measured a ratio 3He/ 4He of (6.6 ± 0.7)Ra in basalts from the Golan
Heights.
6.3 The geothermal structure of the Dead Sea Transform
Thermal springs are formed when water that flows in warm, usually deep layers discharges
at the surface. Ascent can happen when high pressure pushes up the water or in areas
where the crust is cracked, called break up points. Therefore thermal waters are usually
found in tectonically active regions. The heat may come from different sources and as
a result of different processes. The volatiles dissolved in the water can be useful for the
understanding of the origin of this heat.
The thermal water of the Dead Sea Transform gains heat mostly from the normal geothermal gradient of ∼25 K/km [Starinsky et al., 1979 and Arad and Bein, 1986], which leads
to higher temperatures in greater depths. The temperatures of the sampled wells are analysed in respect to their average screens, which correspond to the depth of ascent. Springs
are not analysed since the original depth from which water is ascending is not known.
Figure 6.3 shows that the sampled wells from the southern part of the sampling area have
more or less a normal behaviour, which means that water from higher depths have higher
temperatures. On the other hand, the samples from the northern part of the sampling
area are scattered and show no correlation between the depth and water temperature.
The wells presented in Figure 6.3 can be divided into two groups with different trends:
The wells of Meitzar and the rest of the samples, which lie on a line with a different
slope. The data in each group were linearly fitted and the corresponding temperature
gradients were calculated. The southern samples, together with the sample of Shamir
from the north, yield a temperature gradient of 14 K/km and the samples from Meitzar
well field yield a gradient of 59 K/km. For comparison, Arad and Bein [1986] reported
a temperature gradient of 40 K/km in the area of Meitzar and the Golan Heights. This
enhanced temperature gradient is in good agreement with the heat flux anomaly found
by Shalev et al. [2013] and shown in Figure A.12 in Appendix A.8. This additional heat
source is attributed to magma cooling, during young magmatic activity according to Arad
and Bein, 1986 ; Bein and Feinstein, 1988 ; Bajjali et al., 1997 ; Shalev et al., 2013 and
Roded et al., 2013.
The relationship between helium isotopes and heat is examined in Figure 6.4. No correlation is seen between 3He/ 4He ratio and water temperature, which supports the fact
71
6 Discussion
Figure 6.3: Depth of the sampled wells against water temperature. The depth corresponds
to the averaged screen of the well (see Table 3.2). Note the different groups
of samples from Meitzar and the rest of the samples. The samples in each
group were linearly fitted (solid grey lines) in order to calculate temperature
gradients. The calculated temperature gradients are 59 K/km for the samples
from Meitzar and 14 K/km for the rest of the samples.
that not only thermal waters have relatively high 3He/ 4He ratios, but also colder water.
Notice the isotopic ratio of the spring Ein Gofra, on the east shore of Lake Kinneret. This
sample has a ratio of 3He/ 4He that is similar to the 3He/ 4He ratios of the thermal springs
in Yarmouk Valley (Figure 5.4), in spite of its relatively low water temperature (30.8 ◦ C).
Two explanations are suggested: 1) The magma in the mantle underlying this spring has
already cooled down, but helium is still escaping from it into the groundwater in this area.
2) Mantle derived helium does not affect the water temperature, but is increased in deeper
layers. This indicates that the old water of Ein Gofra might come from a depth of at
least a few hundred of meters below the surface and retains the mantle derived helium
incorporated at that depth. The lower temperature in Ein Gofra might be a result of a
slow ascending of the water in the ground.
A change in the helium flux from the mantle with the depth of the well is noticed. It
can be seen in Figure 6.4(b) that the 3He/ 4He ratio of the northern samples is increasing
with depth and is pretty constant for the southern samples. It is shown in Figures 6.2
and 5.5 that the major amount of helium in most samples is of crustal origin, therefore
it is suggested that 3He/ 4He decreases in the north due to dilution of the mantle derived
72
6.3 The geothermal structure of the Dead Sea Transform
(a)
(b)
Figure 6.4: 3He/ 4He ratio versus water temperature (a) and depth of the sampled wells
(b), determined by the average screen of the well.
73
6 Discussion
helium with radiogenic helium from the crust. Since all samples contain some fraction of
mantle derived helium, it is not clear if the mantle intrusion into the crust is constant
along the Dead Sea Transform or increases from south to north.
Torfstein et al. [2013] suggest two explanations to the differences between the contributions
of mantle derived helium in the north and south of Israel. The first possibility is that there
is a constant helium flux from the mantle into the crust along the Dead Sea Transform,
which is more diluted by radiogenic helium from the crust in the southern regions of
the Dead Sea Transform. The second possibility suggests that more helium is escaping
from the upper mantle in the northern regions of the Dead Sea Transform. According
to Torfstein et al. [2013], young magmatic activity took place in all the sampling areas,
in the north and the south, therefore it cannot explain the higher 3He/ 4He ratios in the
northern part of the sampling area. It is interesting to look at the samples from Meitzar
and Ein Yahav fields in Figure 6.4(b), located in the Yarmouk Valley and Arava Valley,
respectively. Two trends can be seen: 1) The samples from Meitzar 2 and Meitzar 3 show
an increase in the mantle derived helium with depth, while Ein Yahav 6, Ein Yahav 16
and Ein Yahav 116 have a similar mantle derived helium component. 2) Meitzar 3 is only
about 150 m deeper than Ein Yahav 116 and has a value of 3He/ 4He which is about 4.5
times higher. On one hand, the first trend supports the idea of dilution of the mantle
derived helium in the northern samples by helium from local groundwater. On the other
hand, the second trend supports the possibility of an enhanced mantle derived helium
component in the north. The lack of evidence for dilution in the south can be explained
by the fact that the crust which underlies the Arava Valley is about 1.4 times thicker than
the crust underlies the north part of the sampling area, as indicated by Torfstein et al.
[2013] and shown in Figure 3.3. Therefore the samples there might be more diluted and
are less affected by mixing.
It is not so clear whether higher 3He/ 4He ratios in the northern sampling sites are due to
enhanced intrusion of mantle derived helium there or stronger dilution by 3He/ 4He ratios
in the south, which reduces the measured 3He/ 4He ratios in the groundwater there. It is
possible that a different faulting pattern in the north enables more mantle derived helium
to reach the groundwater.
The examination of the change in the 3He/ 4He ratio with the well depth confirms the
idea suggested by Torfstein et al. [2013], of two possible patterns of mantle derived helium
intrusion into the crust. However, 3He/ 4He ratios in thermal waters from the Jordanian
side of the Dead Sea Transform, measured by Kaudse [2014], show a different trend. No
decrease in the measured 3He/ 4He ratios from north to south is observed, but rather a
local anomaly in the 3He/ 4He ratios in the area between the Dead Sea and Lake Kinneret.
Moreover, Kaudse [2014] also found a correlation between high measured 3He/ 4He ratios
and local heat flux.
6.4 Other isotopes as tracers for mantle flux
6.4.1 Carbon isotopes
The δ 13 C values are in good agreement with measurements from other studies [Mazor
et al., 1973 and Rosenthal, 1994]. The results show no correlation between δ 13 C and
geographic locations of the samples. It is not so simple to interpret the origin of CO2 in
74
6.4 Other isotopes as tracers for mantle flux
the water only from δ 13 C since the isotope signatures of the different sources overlap, as
shown in Figure 2.8.
The data of CO2 / 3He ratios from Torfstein et al. [2013], combined with δ 13 C in the three
endmember mixing model (Figure 5.7) show no mantle derived CO2 component in the
water (Section 5.4). δ 13 C was indeed not necessary for the determination of the mantle
endmember share in the water and could be calculated by Torfstein et al. [2013], by
inserting CO2 / 3He into Equation 2.50.
6.4.2 Stable isotopes
The measured δ 18 O and δ 2 H are similar to values obtained by other studies [Mazor et al.,
1973 ; Mazor et al., 1980 ; Rosenthal, 1994 and Möller et al., 2006]. No evidence was found
of interaction between water and crustal rocks.
As mentioned in Section 5.5, the sample from Tzofar 20 was attributed to recharge in
a cooler and more humid climate in ancient time. In contrast, the NGT calculated for
Tzofar 20 is more than 30 ◦ C, which is even more than the annual mean temperature
at the present time. The reason for this contradiction is not very clear, but a possible
explanation is the existence of a deep unsaturated zone at the recharge area. In such a
case, the water keeps exchanging noble gases with air in greater depth, where the ground
temperature increases due to the normal geothermal gradient. Therefore, the NGT can
be higher than the local air temperature.
75
7 Summary and outlook
Thermal waters in Israel were analysed in order to identify local intrusions of volatile
gases from the mantle into the crust. Thirteen samples were taken from thermal wells and
springs along the Israeli side of the Dead Sea Transform. The sampling area included,
from north to south: Hula Valley, Lake Kinneret, Yarmouk Yalley, the Dead Sea area
and Arava Valley. The sampling sites are shown in Figure 3.5. Helium, which is among
the gases that are escaping from the mantle, was used as the main tracer of the mantle
intrusions. The isotope ratio 3He/ 4He of the water was compared to the typical ratio in
the upper mantle in order to determine the amount of mantle derived helium within the
sampling area.
Water temperature, electrical conductivity and alkalinity were measured directly in the
field. Samples of groundwater for analysis of 3H, δ 13 C and stable isotopes (δ 18 O and
δ 2 H) were collected in bottles. Groundwater for noble gas samples was collected in copper
tubes. The noble gas concentrations were measured by mass spectrometry at the IUP.
The measured noble gas isotope concentrations were used for analysing 3He/ 4He ratios
and for Noble Gas Temperature (NGT) determination.
To determine the different helium sources, the total helium concentration was decomposed.
Only small amounts of tritium were found in the samples. Thus, the tritiogenic component
could be neglected. By comparing the ratios of 3He/ 4He and Ne/He, terrigenic helium
was separated from atmospheric helium. Finally, atmospheric, crustal and mantle derived
helium could be distinguished from each other, considering the typical 3He/ 4He ratios of
these three reservoirs.
Different geographic areas show different 3He/ 4He ratios. The highest ratios were found
in samples from the northern part of the sampling area (Hula Valley, Yarmouk Valley
and Lake Kinneret), followed by lower 3He/ 4He ratios measured in the Dead area and the
southernmost samples, from Arava Valley, which showed the lowest 3He/ 4He ratios. The
analysis revealed that the helium origin is predominantly terrigenic, containing mostly
radiogenic helium from the crust. Mantle derived helium was found in all the samples,
ranging from 2.2 % to 38.4 % of the total helium concentration. Only the sample from
Mineral well consisted mostly of atmospheric derived helium, but this is attributed to
atmospheric contamination during sampling. The results confirm the findings of Torfstein
et al. [2013] who saw a similar trend of enhanced 3He/ 4He values in the northern samples
relative to the southern samples. However, thermal waters from the Jordanian side of the
Dead Sea Transform show local variations of 3He/ 4He ratios, but no clear north-south
trend [Kaudse, 2014].
Furthermore, the relationship between water temperature, depth of ascent and the 3He/ 4He
ratio in wells was examined. All the analysed samples, except of both wells from Meitzar,
have a temperature-depth dependency corresponding to a temperature gradient which is
lower than the normal geothermal temperature gradient. The analysis of water from the
wells of Meitzar implies a temperature gradient which is much higher than the normal
77
7 Summary and outlook
geothermal gradient, indicating a local heat anomaly. This is supported by data of heat
flux in Israel, reported by Shalev et al. [2013], who indicated an elevated heat flux around
the Yarmouk Valley. No correlation was found between the 3He/ 4He ratio and the water temperature. The relationship between 3He/ 4He ratios in wells and the depth of the
wells is different for samples from the northern and southern parts of the sampling area.
Whereas the northern samples show a clear increase of the 3He/ 4He ratio with the well
depth, 3He/ 4He ratios of the southern samples do not vary much. This difference between
the north and south, together with north-south trend seen for the 3He/ 4He ratios, suggest
enhanced intrusions of helium from the mantle in the northern areas along with a stronger
dilution of the 3He/ 4He ratio by the crust in southern areas.
In addition to the analysis of mantle derived helium, δ 13 C was measured in order to trace
mantle derived CO2 . Only four samples out of thirteen show a typical δ 13 C isotope signature of mantle CO2 . Thus, data of CO2 / 3He from Torfstein et al. [2013] were combined
with the δ 13 C data to analyse mixing between mantle, limestone and sediment derived
CO2 components. A model from Sano and Marty [1995] was used to estimate the relative
contributions of the different components to the total concentration of CO2 . However, the
model yielded unreasonable values. Therefore, a lower CO2 / 3He ratio was inserted into
the model equations, which led to reasonable results, showing clearly that the CO2 in the
water is not of a mantle origin. These results are in agreement with findings of Torfstein
et al. [2013], who found no evidence of mantle CO2 in their samples.
Stable isotope data shows that all waters are of meteoric origin. Their distribution reflects
different recharge conditions, i.e. recharge altitudes and/or recharge times. Furthermore,
no evidence for clear groundwater-rock interaction is found, which would take place at
temperatures above 100 ◦ C. This means that all sampled water did not circulate deep
enough, where such effects could impact the stable isotope composition.
More information of the recharge conditions was obtained by NGT determination. The
measured noble gas concentrations were fitted with the CE excess air model. The resulting
temperatures are close to the mean annual air temperatures in the Dead Sea Transform at
the present time. The fact that the NGTs of the samples from Arava Valley correspond to
the annual mean temperature or are even higher contradicts the explanation of the stable
isotope trend there, which suggests recharge in a cooler climate. A possible explanation
would be that the unsaturated zone in the recharge area was deep enough, that the water
temperature was influenced by the geothermal gradient before it lost contact with the
atmosphere, which had led to a higher NGT and therefore do not reflect the mean annual
surface temperature but the temperature at a deep groundwater table.
Since in this work almost all hot springs and wells in the Israeli side of the Dead Sea
Transform were analysed, further investigations should be performed for cold springs in
the same region. Based on the measurement from Ein Gofra, where the water temperature
was quite low (30 ◦ C), it can be expected to find 3He/ 4He ratios in cold waters, which are
similar to the 3He/ 4He ratios measured in the thermal waters from the same geographic
area. Moreover, the analysis of water from cold springs could provide important and
interesting new information, especially in the area between Lake Kinneret and the Dead
Sea, where no thermal wells or springs are accessible.
78
A Appendix
A.1 Ostwald solubility calculation
Table A.1: Coefficients for solubility calculation (Equation 2.8) from Benson and Krause
Jr. [1976].
Noble gas
a0 [-]
a1 [K]
a2 [K2 ]
He
Ne
Ar
Kr
Xe
-5.0746
-4.2988
-4.2123
-3.6326
-2.0917
-4127.8
-4871.1
-5239.6
-5664.0
-6693.5
627250
793580
995240
1122400
1341700
A.2 Mixing ratios of noble gases
Table A.2: Atmospheric volume mixing ratios of noble gases in dry air.
Noble gas
mixing ratio
He
Ne
Ar
Kr
Xe
(5.24±0.05)·10−6
(1.818±0.0004)·10−5
(9.34±0.01)·10−3
(1.14±0.01)·10−6
(8.7±0.1)·10−8
A.3 Stable isotopes notation
The isotope ratio R is defined as:
R≡
Abundance of rare isotopes
Abundance of common isotopes
The delta notation is defined as:
Rsample − Rstandard
δ≡
=
Rstandard
Rsample
− 1 · 1000 h
Rstandard
79
A Appendix
A.4 Locations in Israel
Table A.3: Locations in Israel, marked on the map in figure A.1 from Rosenthal et al.
[2009].
Aqaba (57)
Ashdod (37)
Banias River (3)
Beer Tuvia (45)
Beer Sheva (47)
Bet She’an Valley (19)
Betlehem (39)
Dan River (2)
Eilat (56)
Ein Yahav (52)
Fuliya springs (8)
Haon (10)
Harod Valley (18)
Hasbani River (4)
Hatira (50)
80
Hazeva (51)
Hebron (44)
Hula Valley (5)
Jenin (21)
Jericho (35)
Jerusalem (36)
Jordan Valley, upper (6)
Judea syncline (40)
Lisan Peninsula (48)
Makhtesh Ramon (54)
Menashe reservoir (16)
Mt. Carmel (14)
Mt. Gilboa (20)
Mt. Hermon (1)
Naaman River (13)
Nablus (25)
Paran-Zin syncline (55)
Pezael (30)
Qalqiliya (28)
Qane spring (42)
Qilt (34)
Ramallah (33)
Salfit (26)
Samar spring (43)
Samaria (24)
Samaria syncline (23)
Samia (31)
Sdom (49)
Shefela (38)
Shiqma reservoir (46)
Tabigha springs (7)
Taninim river (15)
Tiberias hot springs (9)
Tubas (22)
Tul Karem (27)
Uja (32)
Yahav wells (53)
Yarmouk River (11)
Yizrael Valley (17)
Zemach (12)
Zofar (53)
Zukim springs (41)
A.4 Locations in Israel
Figure A.1: General location map of Israel and the Jordan Rift Valley from Rosenthal
et al. [2009]. For the list of locations see Table A.3
81
A Appendix
A.5 A detailed description of the sampling sites and the
sampling process
The purpose of this part is to give the reader more information about the main characteristics of the sampling sites sampled in the frame of this work and the problems that
have arisen during the sampling process. The sampling sites are presented in an alphabetic order. Note that a few of the names start with “Ein”, which is the Hebrew word for
“spring”. However, one should pay more attention, because not all the places that start
with “Ein” are springs. A good example for it are the three wells of Ein Yahav in the
Arava in the south of Israel. Generally, the natural springs become fewer in the south of
Israel, in the Negev desert.
Ein Qedem
This is one of the thermal springs that were exposed during the decrease of the sea level. It
is located about 30 meters away from the Dead Sea, 6 km northern from Kibutz Ein Gedi.
The spring is surrounded with sink holes and a moist soil. There is one main spring and
another few smaller springs that appear and disappear every few months. The water had
a strong smell of sulfur and the flow rate was low. Therefore, the noble gas sample was
taken by sucking up the water through the water hose and the copper tube. This is the
saltiest water that was sampled in this work, with salinity of about 190 gr/kg [Starinsky,
1974; Stern, 2010]. It was sampled right after Mineral, which has a similar salinity. As
a result, the sensor used to measure electrical conductivity couldn’t reach zero in the air,
even after it was washed in fresh water. However, it was back to normal on the next day.
(a) Ein Qedem spring on the view of the Dead (b) The spring pool (photograph of Paul Bauer)
Sea and the mountains of Jordan
Figure A.2: Ein Qedem springs
82
A.5 A detailed description of the sampling sites and the sampling process
Ein Yahav
This well field is located in the northern Arava, about 30 km away from the Dead Sea. It
consists of three artesian wells, numbered 6, 16 and 116, with different depths, screens and
different water temperatures between 31 ◦ C and 43.5 ◦ C. Samples were taken from all the
three. Ein Yahav 6 and Ein Yahav 116 had a sampling tap and Ein Yahav 16 was broken
at that time, so the water was delivered to an alternate point, about 10 meters away from
the drill (figute A.3).
(a) Ein Yahav 6
(b) Ein Yahav 16 (left) and Ein Yahav 116 (c) The alternative out(right)
flow of Ein Yahav 16, due
to a problem in the drilling
Figure A.3: Ein Yahav wells
The water of Ein Yahav 6 was the warmest and had a light smell of sulfur. It became grey
after about 15 minutes in the air. A water hose with a copper tube was connected to the
sampling tap and noble gas samples, as well as the other samples, were taken without any
problems.
The pipe of Ein Yahav 16 was under construction, therefore the water was delivered to
another pipe a few meters away, so that it could still be used. As can be seen in Figure
A.3 there was no sampling tap and the flow was pretty strong. The water had a smell of
unrecognised chemicals. The samples were taken directly from the pipe. For the noble gas
samples, the water hose was placed in the outflow pipe. Small bubbles were formed in the
hose.
83
A Appendix
Ein Yahav 116 had the coldest water out of the three. The sampling process was similar
to that of Ein Yahav 6, with no special events.
Gofra
Gofra is the name of one of the beaches on the east side of Lake Kinneret. It is called after
the word Gofrit, which means sulfur in Hebrew and describes very well the strong smell
that comes from the spring Ein Gofra on the beach (picture A.4). This was the site with
the lowest water temperature (30.8 ◦ C). The water springs under a rock and is collected
in a small shallow pool. The water in the pool contains white soft sulfur sediments and
bubbles arise from the bottom. The discharge pace is low, therefore the noble gas samples
were taken by inserting the water hose under the rock, as close as possible to the spring
itself and sucking water until it filled the copper tube. A few smaller unpermanent springs
were observed around the main spring.
Hamat Gader
Hamat Gader is a spring field of five hot springs, on the southern slopes of Golan Heights,
on the border with Jordan. The word Hamat comes from the Hebrew “Ham”, which
means “hot” and Gader means “fence”, named after the fence of the border with Jordan,
that passes right at the side of this field. The place is also known from the big resort
there. Samples were taken from two of the springs, Ein Makla and Ein Balzam, that have
the highest temperatures. According to information from Hamat Gader Resort, it can be
assumed that the water springs from a depth of about 2 km.
Ein Makla is a very hot spring, with a temperature of 48.5 ◦ C, that is located in an
Figure A.4: Ein Gofra spring at Gofra Beach, Lake Kinneret
84
A.5 A detailed description of the sampling sites and the sampling process
(a) The archaeological site from the Roman time. (b) Ein Makla spring. Notice the sulfur sediments
Ein Makla is situated between the walls on the in the water
right side. Behind are the highlands of Jordan
(c) Ein Balzam spring. This is also one of the hot
pools in Hamat Gader resort. The floor was left as
it is and covered with gravel in order to enable the
water to flow out freely
Figure A.5: Ein Makla and Ein Balzam in Hamat Gader
archaeological site from the Roman time (see picture A.5). The spring forms a pool,
which has a depth of about 3 m and a diameter of about 5 m. The pool is covered by a
black sheet. There are sulfur sediments on the water with a very strong smell. Bubbles
were rising from the bottom of the pool, each time from a different point and the water
came out from the ground probably in more than one points in the pool. Since the water
was too hot, it was only possible to stand in it with special boots that blocked the heat
for a while. Gloves were also needed in order to fill up the bottles. Since it was not clear,
where exactly the spring point is, the water hose of the noble gas sample was thrown into
the middle of the pool and we made sure that no sulfur pieces would block it. My father
held the other side and sucked water into the copper tube. I knocked on the sides of the
copper tube to make sure there were no air bubbles trapped inside and closed it with the
ratchet. Four noble gas samples were taken in Ein Makla for surly having good samples
from this site.
The other spring that was sampled in Hamat Gader is called Ein Balzam. It is also called
”The spring of scents”, probably because of the prominent sulfur smell there. According
85
A Appendix
to information from Hamat Gader Resort, the flow rate there is 500 - 700 m3 /h and it is
opened for the visitors of the resort. The place of the spring is built as a roofed bathing
pool, with built walls, and the natural bottom, with gravel on it, so that the water can
come out freely. The pool is about 10 x 5 m big. The water is about 42 ◦ C so that it was
possible to go inside for about ten minutes to take the samples. The lifeguard showed
us where it is believed that the water comes out. The noble gas samples were taken in
a similar way to Ein Makla. I was holding one side of the water hose with the foot, so
that it stayed close to the bottom and my father sucked water form the other side. It was
helpful for the sucking to first blow air out of the hose as it was already in the water.
Hamat Tveria
This ancient spring field is located in the city of Tiberias, on the west side of Lake Kinneret.
A medical bath from ancient time is found at this site. There are 17 hot springs, some of
them are under the ground and nowadays there is a park for visitors and thermal pools
that use the springs water. The water is very hot (56 ◦ C) and contains sulfur, which gives
nice colours to the canals in the park. The sample was taken in the main spring, which
appears in Figure A.6. The water springs from the hole in the built wall and is delivered
to the hot pools of the resort. In order to take the samples, the water hose of the noble
gas sample was inserted a few centimeters into the spring and was held by a stone that
was inserted there too. Three noble gas samples were taken, due to a stuck screw in one
of the aluminium racks. I noticed small bubbles that were formed in the bottles and hoses
as the water came out of the spring, which might be an indication of degassing.
(a) The main spring
(b) The Roman spring
Figure A.6: Hamat Tveria springs
Meitzar
Meitzar is a well field on the southern slopes of the Golan Heights on the border with
Jordan, close to Hamat Gader. Two out of the three boreholes there were sampled:
Meitzar 2 and Meitzar 3. Both are artesian and have a sampling tap close to the drilling
point, where the water hose was connected (see picture A.7).
Meitzar 2 was the sampling site which had the highest temperature of 65.3 ◦ C. Little
bubbles were forming during the sampling, making a bigger bubble at the end of the
copper tube. The water hose was therefore bent a little bit and the big bubble was gone.
The smaller bubbles remained and might be a result of degassing.
The sampling process of Meitzar 3 was similar to that of Meitzar 2. The pressure in the
86
A.5 A detailed description of the sampling sites and the sampling process
sampling tap was pretty low, so that less water came out. As a result, bubbles were formed
in the water hoses and the copper tubes. I waited until the bubbles were minimal and
closed the copper tubes.
(a) Meitzar 2
(b) Meitzar 3
Figure A.7: Meitzar well field
87
A Appendix
Mineral
Mineral is a beach on the northern-east side of the Dead Sea. The well is artesian and
provides water to the thermal pool located on the beach. As shown in picture A.8 water
from the well is pumped to the upper part of the containers and then flows inside. Air
contact obviously occurs in the containers. In addition, the pumping is done alternately,
which means that water is not always flowing in the containers. The water had a smell of
sulfur and it contained small unrecognized soft white pieces. It was only possible to take
the samples from an outflow pipe that was located at the bottom of one of them. The
sampling is shown in Figure A.8c. The water hose was inserted into the outflow pipe and
a stone was pushed there, to prevent the hose form slipping out of the pipe, due to the
strong flow. Some bubbles were formed on the outflow side of the copper tube, so we bent
the water hose close to the connection with the tube and the bubbles were gone.
(a) The drilling site of Mineral beach. The square
marks the outflow pipe, which can be seen more detailed in (b)
(b) The inside of one of the containers
(c) The outflow pipe, from which the samples were
taken
Figure A.8: Mineral borehole site
88
A.5 A detailed description of the sampling sites and the sampling process
Shamir
Shamir is the northernmost site that was sampled. It is located on the northern slopes of
the Golan Heights, close to Kibutz Shamir. There are three wells in this field, one of them
is thermal. This thermal well was sampled. It is an artesian well with a maximal flow rate
of 500 m3 /h. There was a sampling tap close to the actual drilling point (see Figure A.9).
Little bubbles were formed in the hoses.
Figure A.9: Shamir well on the northern slopes of the Golan Plateau. The noble gas
sample is connected to the sampling tap, close to the drilling point, which is
the vertical black column.
89
A Appendix
Tzofar
Tzofar 20 is a thermal well that belongs to Tzofar well field in the Arava. It was the
southernmost site that was sampled for this work. The well is artesian and is about
1000 m deep. As shown in Figure A.10, water comes out of the ground and flows through
a 1.5 m pipe into a pool. The tap on the pipe is most of the time closed, so that water
doesn’t come out all the time. It was opened a few minutes before samples were taken
in order to wash out water that might have been in contact with air. The flow was not
regulated and bubbles were flowing through the water hose of the noble gas samples.
Smaller bubbles started to form in the hose, a few seconds after it came out of the well.
Figure A.10: Tzofar 20
90
A.6 Neon and argon isotopic ratios
A.6 Neon and argon isotopic ratios
Table A.4: Neon and argon isotopic ratios calculated directly from WuCEM.
Sample
Gofra
Shamir
Meitzar 3
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 6
Yahav 16
Tzofar 20
Hamat Tveria
Meinral
Qedem
20Ne/ 22Ne
∆ 20Ne/ 22Ne
36Ar/ 40Ar
∆ 36Ar/ 40Ar
9.790
9.793
9.791
9.790
9.803
9.764
9.793
9.792
9.787
9.768
9.773
9.784
9.845
0.005
0.004
0.005
0.005
0.005
0.004
0.004
0.004
0.005
0.005
0.005
0.004
0.006
287
296
299
298
298
302
297
298
296
289
324
288
316
6
6
6
6
6
6
6
6
6
7
6
6
7
A.7 Noble gas concentrations of the second and third
measurement runs
The tables containing the noble gas concentrations obtained from the second and third
measurement runs can be found on the following two pages.
91
A Appendix
Date
14.05.2013
14.05.2013
15.05.2013
21.05.2013
21.05.2013
22.05.2013
23.05.2013
24.05.2013
24.05.2013
12.06.2013
13.06.2013
14.06.2013
Date
14.05.2013
14.05.2013
15.05.2013
21.05.2013
21.05.2013
22.05.2013
23.05.2013
24.05.2013
24.05.2013
12.06.2013
13.06.2013
14.06.2013
He
∆ 3He
3
3.138E-13
6.185E-13
1.836E-12
1.059E-12
5.725E-14
1.351E-13
1.473E-13
5.591E-12
1.809E-12
∆ 36Ar
1.803E-12
2.935E-15
3.845E-13
8.146E-12
1.657E-11
4.791E-11
3.360E-11
1.203E-12
3.650E-12
3.920E-12
1.244E-10
4.412E-11
4.713E-11
3.397E-14
8.512E-12
Ar
36
5.562E-08
–
3.740E-08
1.537E-07
3.243E-08
3.083E-08
2.914E-08
7.972E-08
2.450E-08
4.179E-08
2.690E-08
2.779E-08
2.169E-06
–
1.337E-06
6.264E-06
1.103E-06
1.022E-06
9.294E-07
3.179E-06
6.911E-07
7.364E-07
3.174E-07
3.597E-07
∆ 4He
2nd measurement run
He
1.354E-08
3.194E-08
9.280E-08
5.740E-08
1.426E-08
4.414E-08
9.729E-09
1.324E-06
1.745E-07
4
2.274E-06
4.979E-06
1.572E-05
8.463E-06
2.390E-06
6.622E-06
1.468E-06
2.172E-04
2.820E-05
3.257E-07
1.906E-10
7.533E-08
3rd measurement run
2.856E-05
2.734E-08
9.878E-06
∆ 40Ar
2nd measurement run
Ar
1.952E-06
–
1.228E-06
5.927E-06
1.052E-06
1.011E-06
8.842E-07
3.110E-06
6.775E-07
40
6.384E-04
–
3.967E-04
1.903E-03
3.297E-04
3.084E-04
2.773E-04
1.006E-03
2.101E-04
1.344E-06
5.448E-07
6.813E-07
3rd measurement run
2.103E-04
8.394E-05
1.066E-04
Ne
2.634E-09
–
5.804E-10
1.068E-08
8.244E-10
7.079E-10
4.436E-10
4.008E-09
3.357E-10
∆ 20Ne
9.844E-09
6.905E-09
3.914E-09
7.436E-08
–
1.677E-08
3.052E-07
2.407E-08
2.067E-08
1.280E-08
1.147E-07
9.537E-09
Ne
5.562E-11
3.943E-11
2.499E-11
2.586E-10
–
5.667E-11
1.040E-09
8.021E-11
6.905E-11
4.335E-11
3.934E-10
3.551E-11
∆ 22Ne
22
7.335E-07
–
1.646E-07
2.997E-06
2.356E-07
2.023E-07
1.252E-07
1.123E-06
9.312E-08
4.943E-10
3.522E-10
2.044E-10
20
9.613E-08
6.770E-08
3.853E-08
7.801E-11
–
7.340E-11
1.397E-10
6.975E-11
5.341E-11
5.109E-11
9.662E-11
4.762E-11
∆ 132Xe
∆ 84Kr
3.508E-09
–
3.054E-09
6.257E-09
2.553E-09
2.366E-09
2.275E-09
3.788E-09
1.902E-09
4.301E-11
2.949E-11
4.821E-11
Xe
Kr
9.103E-10
–
7.439E-10
2.028E-09
6.006E-10
5.411E-10
5.268E-10
1.173E-09
3.995E-10
1.949E-09
4.690E-10
7.455E-10
132
6.537E-08
–
5.064E-08
1.489E-07
3.986E-08
3.781E-08
3.762E-08
8.651E-08
2.892E-08
4.066E-10
1.646E-10
2.325E-10
84
2.892E-08
9.343E-09
1.397E-08
Table A.5: Concentrations of noble gas isotopes from the second and third measurements periods in [ccSTP/g]. Only helium isotopes could
be measured in Ein Balzam, since the heating element was accidentally turned on in the middle of the measurement.
Sample
Shamir
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 16
Gofra
Tzofar 20
Hamat Tveria
Hamat Tveria
Mineral
Qedem
Sample
Shamir
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 16
Gofra
Tzofar 20
Hamat Tveria
Hamat Tveria
Mineral
Qedem
92
2.274E-06
4.979E-06
1.572E-05
8.463E-06
2.390E-06
6.622E-06
1.468E-06
2.172E-04
2.820E-05
He
1.354E-08
3.194E-08
9.280E-08
5.740E-08
1.426E-08
4.414E-08
9.729E-09
1.324E-06
1.745E-07
∆He
8.101E-07
–
1.819E-07
3.311E-06
2.603E-07
2.236E-07
1.384E-07
1.241E-06
1.029E-07
Ne
6.410E-04
–
3.982E-04
1.910E-03
3.310E-04
3.096E-04
2.784E-04
1.010E-03
2.109E-04
Ar
∆Ar
1.954E-06
–
1.230E-06
5.934E-06
1.054E-06
1.012E-06
8.854E-07
3.114E-06
6.784E-07
3rd measurement run
2.664E-09
–
5.872E-10
1.083E-08
8.339E-10
7.162E-10
4.487E-10
4.057E-09
3.398E-10
∆Ne
1.147E-07
–
8.884E-08
2.613E-07
6.993E-08
6.633E-08
6.600E-08
1.518E-07
5.073E-08
Kr
1.597E-09
–
1.305E-09
3.558E-09
1.054E-09
9.493E-10
9.242E-10
2.058E-09
7.009E-10
∆Kr
1.304E-08
–
1.136E-08
2.327E-08
9.495E-09
8.800E-09
8.459E-09
1.409E-08
7.072E-09
Xe
2.901E-10
–
2.730E-10
5.196E-10
2.594E-10
1.986E-10
1.900E-10
3.593E-10
1.771E-10
∆Xe
Hamat Tveria 2.856E-05 3.257E-07 1.063E-07 4.998E-10 2.111E-04 1.346E-06 5.073E-08 7.134E-10 7.248E-09 1.600E-10
Mineral
2.734E-08 1.906E-10 7.480E-08 3.559E-10 8.431E-05 5.459E-07 1.639E-08 2.888E-10 1.744E-09 1.097E-10
Qedem
9.878E-06 7.533E-08 4.256E-08 2.068E-10 1.071E-04 6.823E-07 2.451E-08 4.080E-10 2.772E-09 1.793E-10
Shamir
Balzam
Meitzar 2
Makla
Yahav 116
Yahav 16
Gofra
Tzofar 20
Hamat Tveria
Sample
2nd measurement run
Table A.6: Noble gas concentrations from the second and third measurement periods in [ccSTP/g], ordered by measurement date. The
concentrations from the first run are listed in chapter 5. Only helium could be calculated in Ein Balzam, since only helium
isotopes were measured there (see table A.5).
A.7 Noble gas concentrations of the second and third measurement runs
93
A Appendix
A.8 Additional figures
(a) WTW sensor used for the temperature and (b) Alkalinity measurement with Salifert kit.
electrical conductivity measurements.
(c) Noble gas sample.
Figure A.11: Additional photos from the field and the laboratory.
94
A.8 Additional figures
Figure A.12: Calculated geothermal heat flux in Israel, adopted from [Shalev et al., 2013].
The black dots refer to temperature measurements in boreholes. Notice the
flux anomaly east of Lake Kinneret, in the Golan Heights.
95
List of Figures
1.1
Interior layers of Earth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
Solubility versus temperature and salinity . . . . . . . . . . . .
Noble gas components in groundwater . . . . . . . . . . . . . .
Excess air models . . . . . . . . . . . . . . . . . . . . . . . . . .
CE model derivation . . . . . . . . . . . . . . . . . . . . . . . .
MORB and OIB . . . . . . . . . . . . . . . . . . . . . . . . . .
Helium components . . . . . . . . . . . . . . . . . . . . . . . .
Air correction explanation: 3He/ 4He versus Ne/He in a sample
δ 13 C signatures of different sources of carbon in groundwater .
Stable isotopes: GMWL . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16
17
19
21
23
24
25
30
32
3.1
3.2
3.3
3.4
3.5
The Levant Rift . . . . . . . .
Pull apart basin scheme . . .
The crust under the Dead Sea
Stratigraphic column of Israel
Sampling sites locations . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
34
35
36
38
40
4.1
4.2
Schematic description of MM5400 spectrometer . . . . . . . . . . . . . . . . 43
Measurement process of the mass spectrometer . . . . . . . . . . . . . . . . 43
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
NGTs results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4He concetration versus 3He concentration . . . . . . . . . . . . . . . . . . .
He concentration versus electrical conductivity . . . . . . . . . . . . . . . .
Ra versus Ne/He . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Atmospheric, crustal and mantle derived helium . . . . . . . . . . . . . . .
δ 13 C versus 3He/ 4He and mantle derived helium percentage . . . . . . . . .
Distribution of mantle, carbonate and sediments derived CO2 in the samples
Stable isotopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
54
55
59
60
64
66
68
6.1
6.2
3He/ 4He
69
6.3
6.4
. . . . . . .
. . . . . . .
Transform
. . . . . . .
. . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
and helium isotopes concentrations versus the distance from Eilat
Atmospheric, crustal and mantle derived helium in different regions along
the Dead Sea Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Depth versus water temperature of the sampled wells . . . . . . . . . . . . .
3He/ 4He ratios versus temperature and depth of the sampled wells . . . . .
A.1 Israel locations . . . . . . . . . . . . . .
A.10 Tzofar 20 . . . . . . . . . . . . . . . . .
A.11 Additional photos from the field and the
A.12 Heat flux map of Israel . . . . . . . . . .
. . . . . . .
. . . . . . .
laboratory
. . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
70
72
73
81
90
94
95
97
List of Tables
2.1
2.2
Average 3He/ 4He in the atmosphere, crust and mantle . . . . . . . . . . . . 29
Isotopic signature of the main sources of carbon in groundwater . . . . . . . 31
3.1
3.2
General information about the sampling sites . . . . . . . . . . . . . . . . . 39
Wells data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.1
Atomic abundance factors of the measured noble gas isotopes . . . . . . . . 45
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
Water parameters . . . . . . . . . . . .
Noble gas isotopes concentrations . . .
Elemental noble gas concentrations . .
NGTs results . . . . . . . . . . . . . .
Tritium results . . . . . . . . . . . . .
3He/ 4He and Ne/He ratios . . . . . . .
Atmospheric, crust and mantle derived
δ 13 C and CO2 / 3He . . . . . . . . . . .
Stable isotopes . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
47
48
49
53
56
57
62
63
67
A.1
A.2
A.3
A.4
A.5
A.6
Vant Hoff equation coefficients . . . . . . . . . . . . . . . . . . .
Atmospheric volume mixing ratios of noble gases in dry air . . .
Israel locations . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Neon and argon isotopic ratios . . . . . . . . . . . . . . . . . . .
Noble gas isotopes concentrations: Additional measurement runs
Noble gas concentrations: Additional measurement runs . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
79
79
80
91
92
93
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
helium components
. . . . . . . . . . . .
. . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
99
Acknowledgments
First of all I would like to thank my supervisor, Prof. Dr. Werner Aeschbach Hertig, for
the helpful explanations and ideas you provided me during the work. Thank you for giving
me the opportunity, to carry out a field trip in Israel.
Tillmann Kaudse, thank you very much for your significant contribution to this work as
my tutor, always making sure that things went on the right side, from the beginning to
the end.
Special thank to Dr. Joseph Guttmann from Mekorot, for enabling me to take samples at
the drilling wells of Meitzar, Ein Yahav and Tzofar and for providing necessary literature.
I would like to thank to all other sites owner, for allowing me to take samples there:
Hilik from Mei Golan (Shamir drillings), Tami and Sharon from Israel Nature and Park
Authority (Hamat Tveria springs), Oren and Ibrahim from Mineral Beach, Andrea and
Moti from Hamat Gader Resort and Gofra Beach camping site.
Paul Königer from BGR in Hannover, special thanks for kindly measuring my stable
isotopes samples. Adrian Walser and Giles Möhl, thank you for measuring δ 13 C and
3H.
I would like to thank to the people from my research group, especially Tillmann, Simon
and Florian, for your help in the organization of the field trip, during the work in the
laboratory, reviewing my thesis before the submission and making a great team work and
atmosphere.
Special gratitude to Prof. Dr. Emanuel Mazor from Weitzmann Institute, for the explanations about thermal water formation and groundwater analyzing.
My dear friends, Paul Bauer and Svenja Reith, thank you so much for reviewing my thesis
with a lot of attention and care. Special thanks to my boyfriend Clemens Schwingshackl. I
really appreciate your help with my thesis and the support, especially during busy times.
At last, I would like to thank my parents, Raya and Shraga Tsur and my brother, Yuval
Tsur, for driving with me to all the sites and helping with the sampling.
All of you, who are not mentioned here, I would like to thank you too for your contribution
to my work.
101
Bibliography
Aeschbach-Hertig, W. (2005). A comment on “Helium sources in passive margin aquifers
- new evidence for a significant mantle 3 He source in aquifers with unexpectedly low in
situ 3 He/ 4 He production” by MC Castro [Earth Planet. Sci. Lett. 222 (2004) 897–913].
Earth and Planetary Science Letters, 240(3):827–829. 27, 28
Aeschbach-Hertig, W., El-Gamal, H., Wieser, M., and Palcsu, L. (2008). Modeling excess
air and degassing in groundwater by equilibrium partitioning with a gas phase. Water
Resources Research, 44(8). 18, 20, 21
Aeschbach-Hertig, W., Peeters, F., Beyerle, U., and Kipfer, R. (1999). Interpretation
of dissolved atmospheric noble gases in natural waters. Water Resources Research,
35(9):2779–2792. 22
Aeschbach-Hertig, W. and Solomon, D. K. (2013). Noble gas thermometry in groundwater
hydrology. In The Noble Gases as Geochemical Tracers, pages 81–122. Springer. 15, 18,
20, 22, 45
Allègre, C., Staudacher, T., and Sarda, P. (1987). Rare gas systematics: formation of the
atmosphere, evolution and structure of the Earth’s mantle. Earth and Planetary Science
Letters, 81(2):127–150. 11
Arad, A. and Bein, A. (1986). Saline-versus freshwater contribution to the thermal waters
of the northern Jordan Rift Valley, Israel. Journal of Hydrology, 83(1):49–66. 67, 71
Bajjali, W., Clark, I. D., and Fritz, P. (1997). The artesian thermal groundwaters of
northern Jordan: insights into their recharge history and age. Journal of hydrology,
192(1):355–382. 71
Ballentine, C. and Hall, C. (1999). Determining paleotemperature and other variables
by using an error-weighted, nonlinear inversion of noble gas concentrations in water.
Geochimica et Cosmochimica Acta, 63(16):2315–2336. 22
Barkan, E., Luz, B., and Lazar, B. (2001). Dynamics of the carbon dioxide system in the
Dead Sea. Geochimica et Cosmochimica Acta, 65(3):355 – 368. 63
Barnes, I., Irwin, W. P., and White, D. E. (1978). Global distribution of carbon dioxide
discharges and major zones of seismicity. Survey Water Resources Inv., Open-file rept,
pages 78–39. 29
Bayer, H.-J., Hötzl, H., Jado, A., Röscher, B., and Voggenreiter, W. (1988). Sedimentary
and structural evolution of the northwest Arabian Red Sea margin. Tectonophysics,
153(1):137–151. 34
Bein, A. and Feinstein, S. (1988). Late Cenozoic thermal gradients in Dead Sea transform
system basins. Journal of Petroleum Geology, 11(2):185–192. 71
103
Bibliography
Benson, B. B. and Krause Jr., D. (1976). Empirical laws for dilute aqueous solutions of
nonpolar gases. Journal of Chemical Physics, 64(2):689–709. 16, 79
Benson, B. B. and Krause Jr., D. (1980). Isotopic fractionation of helium during solution:
A probe for the liquid state. Journal of Solution Chemistry, 9(12):895–909. 23
Bergelson, G., Nativ, R., and Bein, A. (1999). Salinization and dilution history of ground
water discharging into the Sea of Galilee, the Dead Sea Transform, Israel. Applied
Geochemistry, 14(1):91–118. 12
Bröder, L. (2011). A paleoclimate record from groundwater of the Great Artesian Basin
in Australia. Master’s thesis, Heidelberg University. 32
Clark, I. and Fritz, P. (1997). Environmental isotopes in hydrology. 30
Clarke, W. B., Jenkins, W., and Top, Z. (1976). Determination of tritium by mass spectrometric measurement of 3 He. The international journal of applied radiation and isotopes,
27(9):515–522. 23, 57, 59
Cook, P. G. and Herczeg, A. L. (2000). Environmental tracers in subsurface hydrology,
volume 529. Kluwer Academic Publishers Boston. 16, 22
Craig, H. (1961a). Isotopic variations in meteoric waters. Science, 133(3465):1702–1703.
31, 68
Craig, H. (1961b). Standard for reporting concentrations of deuterium and oxygen-18 in
natural waters. Science, 133(3467):1833–1834. 31
de Leeuw, G., Hilton, D., Fischer, T., and Walker, J. (2007). The He–CO2 isotope and
relative abundance characteristics of geothermal fluids in el salvador and honduras:
New constraints on volatile mass balance of the central american volcanic arc. Earth
and Planetary Science Letters, 258(1):132–146. 65
El-Isa, Z., Mechie, J., Prodehl, C., Makris, J., and Rihm, R. (1987). A crustal structure
study of Jordan derived from seismic refraction data. Tectonophysics, 138:235–253. 35
Elliott, T. (2009). Earth science: Restoration of the noble gases. Nature, 459:520–521. 23
Freund, R. (1965). A model of the structural development of Israel and adjacent areas
since upper cretaceous times. Geological Magazine, 102:189–205. 34
Freund, R., Garfunkel, Z., Zak, I., Goldberg, M., Weissbrod, T., Derin, B., Bender, F.,
Wellings, F. E., and Girdler, R. W. (1970). The Shear along the Dead Sea Rift [and
discussion]. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 267(1181):107–130. 34
Friedrich, R. (2007). Grundwassercharakterisierung mit Umwelttracern: Erkundung
des Grundwassers der Odenwald-Region sowie Implementierung eines neuen EdelgasMassenspektrometersystems. Inaugural dissertation, Universität Heidelberg. 43
Garfunkel, Z., Zak, I., and Freund, R. (1981). Active faulting in the Dead Sea Rift.
Tectonophysics, 80:1 – 26. 34, 35
Gat, J. and Carmi, I. (1970). Evolution of the isotopic composition of atmospheric waters
in the Mediterranean Sea area. Journal of Geophysical Research, 75(15):3039–3048. 31,
68
104
Bibliography
Gat, J. R. and Dansgaard, W. (1972). Stable isotope survey of the fresh water occurrences
in Israel and the northern Jordan Rift Valley. Journal of Hydrology, 16(3):177–211. 31,
67, 68
Gat, J. R., Mazor, E., and Tzur, Y. (1969). The stable isotope composition of mineral
waters in the Jordan Rift Valley, Israel. Journal of Hydrology, 7(3):334–352. 31, 68
Giggenbach, W. (1992). Isotopic shifts in waters from geothermal and volcanic systems
along convergent plate boundaries and their origin. Earth and Planetary Science Letters,
113(4):495–510. 32
Ginzburg, A., Makris, J., Fuchs, K., Prodehl, C., Kaminski, W., and Amitai, U. (1979).
A seismic study of the crust and upper mantle of the Jordan-Dead Sea Rift and their
transition toward the Mediterranean Sea. Journal of Geophysical Research: Solid Earth,
84(B4):1569–1582. 35
Gomez, F., Karam, G., Khawlie, M., McClusky, S., Vernant, P., Reilinger, R., Jaafar, R.,
Tabet, C., Khair, K., and Barazangi, M. (2007). Global positioning system measurements of strain accumulation and slip transfer through the restraining bend along the
dead sea fault system in Lebanon. Geophysical Journal International, 168(3):1021–1028.
35
Graham, D. W. (2002). Noble gas isotope geochemistry of mid-ocean ridge and ocean island
basalts: Characterization of mantle source reservoirs. In Porcelli, D., Ballentine, C., and
Wieler, R., editors, Noble Gases in Geochemistry and Cosmochemistry, volume 47 of
Rev. Mineral. Geochem., pages 247–317. Mineralogical Society of America, Geochemical
Society, Washington, DC. 23, 24
Griesshaber, O. and Oxburgh (1992). Helium and carbon isotope systematics in crustal
fluids from the Eifel, the Rhine Grabben and Black Forest, F.R.G. Chemical Geology,
99:213 – 235. 29
Grothe, J. (1992). Datenerfassung und Datenauswertung am Heidelberger Low-LevelTritium-Messsystem. Diploma Thesis, Universität Heidelberg. 45
Heaton, T. H. E. and Vogel, J. C. (1981). “excess air” in groundwater. Journal of
Hydrology, 50:201–216. 17
Herzberg, O. and Mazor, E. (1979). Hydrological applications of noble gases and temperature measurements in underground water systems: Examples from Israel. Journal of
Hydrology, 41:217–231. 12
Hoefs, J. (2009). Stable isotope geochemistry. Springer. 29, 32
Hoetzl (2009). The water of the Jordan Rift Valley. Springer-Verlag Berlin Heidelberg.
33, 36
Hoffmann, E. and Stroobant, V. (2007). Mass spectrometry principles and applications.
England: West Sussex. 42
Horowitz, A. (2001). The Jordan Rift Valley. A.A. Balkema Publishers. 33, 34, 35, 36
Jung, M., Wieser, M., von Oehsen, A., and Aeschbach-Hertig, W. (2012). Properties of the
closed-system equilibration model for dissolved noble gases in groundwater. Chemical
Geology. 45
105
Bibliography
Kaudse, T. (2014). Noble gases in groundwater of the Azraq Oasis, Jordan and along the
central Dead Sea Transform. In preparation, Heidelberg University. 12, 16, 43, 74, 77
Kipfer, R., Aeschbach-Hertig, W., Peeters, F., and Stute, M. (2002). Noble gases in lakes
and ground waters. In Porcelli, D., Ballentine, C., and Wieler, R., editors, Noble gases
in geochemistry and cosmochemistry, volume 47 of Rev. Mineral. Geochem., pages 615–
700. Mineralogical Society of America, Geochemical Society, Washington, DC. 17, 18,
20, 45
Kreuzer, A. (2007). Paläotemperaturstudie mit Edelgasen in Grundwasser der Nordchinesischen Tiefebene. Inaugural dissertation, Universität Heidelberg. 46
Lange, T., Hammerschmidt, K., Friedrichsen, H., Marei, A., and Weise, S. M. (2008).
Groundwater in the shallow aquifer of the jericho area, jordan valley – noble gas evidence
for different sources of salinization. In Zereini, F. and Hötzl, H., editors, Climatic changes
and water resources in the Middle East and North Africa, pages 469–495. Springer,
Berlin, Heidelberg. 12
Lupton, J. E. (1983). Terrestrial inert gases-isotope tracer studies and clues to primordial
components in the mantle. Annual Review of Earth and Planetary Sciences, 11:371–414.
23, 24
Mamyrin, B. and Tolstikhin, I. (1984). Helium isotopes in nature, volume 3. Elsevier. 12,
18, 23
Mart, Y. (1991). The Dead Sea Rift: from continental rift to incipient ocean. Tectonophysics, 197:155 – 179. 33
Mart, Y., Ryan, W. B., and Lunina, O. V. (2005). Review of the tectonics of the Levant
Rift system: the structural significance of oblique continental breakup. Tectonophysics,
395(3):209–232. 34
Marty, B. and Jambon, A. (1987). C/ 3He in volatile fluxes from the solid earth: implications for carbon geodynamics. Earth and Planetary Science Letters, 83(1):16–26. 29,
66
Mayer, S. (2012). Analyzing short-term fluctuations of noble gas concentrations in soil air
and groundwater. Master’s thesis, Heidelberg University. 19
Mazor, E. (1972). Paleotemperatures and other hydrological parameters deduced from
gases dissolved in groundwaters, Jordan Rift Valley, Israel. Geochimica et Cosmochimica
Acta, 36:1321–1336. 12
Mazor, E., Kaufman, A., and Carmi, I. (1973). Hammat Gader (Israel): Geochemistry of
a mixed thermal spring complex. Journal of Hydrology, 18(3):289–303. 12, 67, 74, 75
Mazor, E., Levitte, D., Truesdell, A. H., Healy, J., and Nissenbaum, A. (1980). Mixing
models and ionic geothermometers applied to warm (up to 60 ◦ C) springs: Jordan Rift
Valley, Israel. Journal of Hydrology, 45(1-2):1–19. 75
Möller, P., Dulski, P., Salameh, E., and Geyer, S. (2006). Characterization of the sources of
thermal spring-and well water in jordan by rare earth element and yttrium distribution
and stable isotopes of H2 O. Acta hydrochimica et hydrobiologica, 34(1-2):101–116. 68,
75
106
Bibliography
Möller, P., Rosenthal, E., Geyer, S., and Flexer, A. (2007a). Chemical evolution of saline
waters in the Jordan-Dead Sea transform and in adjoining areas. International Journal
of Earth Sciences, 96(3):541–566. 68
Möller, P., Rosenthal, E., Geyer, S., Guttman, J., Dulski, P., Rybakov, M., Zilberbrand,
M., Jahnke, C., and Flexer, A. (2007b). Hydrochemical processes in the lower Jordan
valley and in the Dead Sea area. Chemical Geology, 239(1):27–49. 38, 67
O’Nions, R. and Oxburgh, E. (1988). Helium, volatile fluxes and the development of
continental crust. Earth and Planetary Science Letters, 90(3):331–347. 29
Picard, L. (1987). The Elat (Aqaba)-Dead Sea-Jordan subgraben system. Tectonophysics,
141:23 – 32. 33
Quennell, A. . (1959). The structural and geomorphic evolution of the Dead Sea Rift.
Quart. J. Geol. Soc. London, 114:1–24;432–433. 34
Quennell, A. (1956). Geologic map of the Kingdom of Trans Jordan. 34
R., F. (1970). Plate tectonics of the Red Sea and East Africa. Nature, 228:453. 34
Roded, R., Shalev, E., and Katoshevski, D. (2013). Basal heat-flow and hydrothermal
regime at the Golan-Ajloun hydrological basins. Journal of Hydrology, 476:200–211. 71
Roedel, W. and Wagner, T. (2010). Physik unserer Umwelt: Die Atmosphäre. Springer
Verlag, 4 edition. 16
Rosenthal, E. (1994). The hydrogeothermal provinces of Israel. In Risler, J.-J. and Simmers, I., editors, Hydrogeothermics (International contributions to hydrogeology), pages
113–135. A A Balkema Publishers, Leiden, London, New York, Philadelphia, Singapore.
74, 75
Rosenthal, E., Guttman, J., Sabel, R., and Möller, P. (2009). Limiting hydrochemical
factors for sustainability of water resources: The Cisjordanian experience. Chemie der
Erde-Geochemistry, 69(3):191–222. 80, 81
Sano, Y. and Marty, B. (1995). Origin of carbon in fumarolic gas from island arcs. Chemical
Geology, 119(1):265–274. 29, 31, 63, 65, 78
Sano, Y. and Wakita, H. (1988). Precise measurement of helium isotopes in terrestrial
gases. The Chemical Society of Japan, 61:1153 – 1157. 12, 26
Sano, Y. and Williams, S. N. (1996). Fluxes of mantle and subducted carbon along
convergent plate boundaries. Geophysical Research Letters, 23(20):2749–2752. 65
Schlosser, P., Stute, M., Sonntag, C., and Otto Münnich, K. (1989). Tritiogenic 3 He in
shallow groundwater. Earth and Planetary Science Letters, 94(3):245–256. 18
Shalev, E., Lyakhovsky, V., Weinstein, Y., and Ben-Avraham, Z. (2013). The thermal
structure of Israel and the Dead Sea Fault. Tectonophysics. 71, 78, 95
Starinsky, A. (1974). Relationship between Ca-chloride brines and sedimentary rocks in
Israel. PhD thesis. 37, 50, 82
Starinsky, A., Katz, A., and Levitte, D. (1979). Temperature-composition-depth relationship in Rift Valley hot springs: Hammat Gader, northern Israel. Chemical Geology,
27(3):233–244. 71
107
Bibliography
Stephen, H. and Stephen, T. (1963). Solubilities of inorganic and organic compounds,
volume 1. Pergamon Press Oxford. 65
Stern, O. (2010). Geochemistry, hydrology and paleo-hydrology of Ein Qedem spring
system. Master’s thesis, The Hebrew University. 37, 50, 82
Sun, T., Hall, C. M., and Castro, M. C. (2010). Statistical properties of groundwater
noble gas paleoclimate models: Are they robust and unbiased estimators? Geochemistry
Geophysics Geosystems, 11(2):Q02002. 22
Torfstein, A., Hammerschmidt, K., Friedrichsen, H., Starinsky, A., Garfunkel, Z., and
Kolodny, Y. (2013). Helium isotopes in Dead Sea Transform waters. Chemical Geology.
12, 36, 37, 63, 66, 71, 74, 75, 77, 78
Unkel, I. (2006). AMS-14C-Analysen zur Rekonstruktion der Landschafts- und Kulturgeschichte in der Region Palpa (S-Peru). Inaugural dissertation, Universität Heidelberg. 45
Van Soest, M., Hilton, D., and Kreulen, R. (1998). Tracing crustal and slab contributions
to arc magmatism in the lesser antilles island arc using helium and carbon relationships
in geothermal fluids. Geochimica et Cosmochimica Acta, 62(19):3323–3335. 65
Wieser, M. (2011). Imprints of climatic and environmental change in a rgional aquifer
system in an arid part of India using noble gases and other environmental tracers. PhD
thesis, Heidelberg University. 19, 43, 46
Wilson, J. (1965). A new class of faults and their bearing on continental drift. Nature,
207(4995):343–347. 34
Yechieli, Y., Starinsky, A., and Rosenthal, E. (1992). Evolution of brackish groundater in a
typical arid region: Northern Arava Rift Valley, southern Israel. Applied Geochemistry,
7(4):361 – 374. 37
108
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising