GEOTHERMAL TRAINING PROGRAMME Reports 2013

Orkustofnun, Grensasvegur 9, Number 12

IS-108 Reykjavik, Iceland

COMPARISON, CHARACTERIZATION, AND INTERPRETATION

OF GEOTHERMAL FLUID GEOCHEMISTRY IN THE SEDIMENTARY
ENVIRONMENTS OF KIBIRO, PANYIMUR, AND ÖXARFJÖRDUR

Grace Lajwe
Ministry of Energy and Mineral Development
Department of Geological Survey and Mines
P.O. Box 9, Entebbe
UGANDA
[email protected]

ABSTRACT

Geochemical data interpretation of representative chemical, analytical, and isotopic

results for geothermal waters and gas in sedimentary environments obtained by
standard analytical methods were studied in this research work. Techniques like
chemical geothermometers (silica and cation geothermometers) and gas
geothermometers were used in predicting the variation in temperatures in three
geothermal systems, i.e. Kibiro and Panyimur in the East Africa Rift System, and
Öxarfjördur in Iceland. The speciation programme WATCH was used to calculate
the activity of the different species of anions and cations, Na/K, Quartz and
Chalcedony temperatures, a Ternary diagram for Na/1000-K/100-√Mg based on the
equilibration conditions for these cations in these waters at particular temperatures,
and a Cl-SO4-HCO3 ternary plot were constructed. Different mixing models and
plots of different constituent relationships were used in the comparison,
characterization and interpretation of the geochemistry of these different geothermal
areas. The geochemistry of magmatically driven geothermal systems may be used
to develop effective exploration techniques for the geothermal systems that are
driven by temperature gradients like the East African Rift System (EARS), the
western branch of which Kibiro and Panyimur are part.

The chemical and isotopic composition of the water samples from Kibiro, Panyimur
and Öxarfjördur, suggest mixing of cold groundwater with geothermal waters.
Major ions and isotope parameters were used to characterize the waters from Kibiro
and Panyimur as Na-Cl-HCO3 waters, and water from Öxarfjördur as Na-Cl water.
The main processes affecting the geothermal system in the sedimentary geological
environment are: dilution and conductive cooling.

1. INTRODUCTION

For future exploration, building a conceptual model, and deciding on the utilization of geothermal
resources in the area where they exist, a sound understanding of the chemistry of the geothermal fluids
needs to be established in the beginning of the exploration phase for better facilitation and approaches

201
Lajwe 202 Report 12

(Bruhn et al., 2010). Water and gas chemistry have been used for the last decades to: find the origin of
the geothermal fluids, understand the mixing effects in the geothermal system, evaluate subsurface
temperature, and predict corrosion and scaling problems during the use of these fluids (Fournier, 1977;
Rybach, and Muffler, 1981; Arnórsson, et al., 1983; Arnórsson, 2000; Giggenbach, 1988).

The chemical composition of geothermal steam and water from natural manifestations provides initial
constraints on reservoir temperature, and helps identify production properties and potential
environmental concerns related to the utilization of the geothermal fluid. These can only be attained by
carrying out geochemical exploration, one of the exploration techniques applied in the search for any
geothermal resource for any utilization.

This project aims to investigate the geochemical relationship of the geothermal waters of Kibiro and
Panyimur geothermal areas in East Africa and compare them with the low temperature geothermal
system of the Öxarfjördur geothermal field which is located in North Iceland. The formation conditions
of the geothermal fluids differ from one area to another (Hjartarson et al., 2005); this can only be proven
by analysing the chemical characteristics and comparing the findings with the different formation
conditions. The geology of these geothermal areas is briefly described, together with a classification of
the geothermal waters. The speciation program WATCH (Bjarnason, 2010), and geothermometers were
mainly used for the determination of the reservoir temperatures and equilibrium conditions.

Geothermal systems all over the world are mostly associated with a rift system where a powerful heat
source is close to the surface of the earth (Björnsson et al., 2005). The geothermal activity in Kibiro
and Panyimur is largely restricted to the rift system (Figure 1) of the western branch of the EARS, with
geothermal manifestations like fumaroles, hot springs and travertine occurring along some of the border
faults of the rift valley and within the volcanic fields (Data and Bahati, 2003). The geothermal system
of the Öxarfjördur area in north Iceland is also within a major graben which is associated with current
volcanic activity.

1.1 Objectives of the study

There are two objectives: 1) to understand the geochemical evolution of geothermal waters in two
different sedimentary geological environments by carrying out a classification and an interpretation of
their geochemical analytical results; 2) to develop better geothermal exploration and exploitation
methods and techniques for geothermal waters occurring in sedimentary environments.

2. LITERATURE REVIEW

2.1 Geology of the Western branch of the East African Rift System

The western arm of the EARS, where the Kibiro and Panyimur areas of this study are located, has
Cenozoic rift faults (Atekwana et al., 2004) starting from the north along the Sudan border, and then
curving to the west and southwest along the border with the Democratic Republic of Congo, and south
to Rwanda and Burundi for about 100 km. The western branch of the EARS is considered to be at an
early stage of development, and is younger, that is Late Miocene to Recent, than the more mature eastern
branch (Morley, 1999). The rifting region has a higher heat flow than the surrounding Pre-Cambrian
terrain. Two different en echelon strands are found in the Western Rift, separated by the Rwenzori
Mountains which rise from a base of less than 1000 m in the Rift to over 5000 m elevation. Within the
rift valley there are thick layers of late Tertiary and Quaternary sediments, fresh water and saline crater
lakes, volcanics, and plutonic bodies identified beneath L. Albert and L. Edward (EDICON, 1984). The
majority of the geothermal areas of Uganda are found in the Western Rift Valley (Figure 1).
Report 12 203 Lajwe

The western branch of

the rift runs along the
western side of Lake
Victoria and along the
edge of the East African
plateau (Figure 1). The
geography of the western
branch is typically half
grabens characterized by
high angle normal rift
faults. The rift is
characterized by a
paucity of volcanism
relative to the Kenyan
and Ethiopian rifts
(Ebinger et al., 1989).
The Lake Albert Rift was
initiated in early
Miocene and is
dominated by a thick
sequence of sediments. It
is characterized by an
abundance of potassic
alkaline rocks that
consist of basalts,
carbonatites, ultrapotas-
sic mafic rocks and
potassic mafic-felsic
lava.

2.1.1 Geology of Kibiro FIGURE 1: Modified EARS (Wood and Guth, 2012)
The area (Appendix I) has some volcanic rocks injected through Pleistocene lacustrine rocks in the rift
valley and granitic gneisses on and above the escarpment at Kibiro; big dykes of basic intrusive rocks
can be traced in some cases for miles in the argillites and gneisses trending NNE to SSW. The
mineralogical composition of the rocks in this area is: porphyritic augites, pyroxexes, ilmenite, and
titaniferous minerals. In addition to the more basic intrusives, there are also other granitic types like
charnockites found 4.8 km southeast of Kibiro in a very large intrusion running east and west, presumed
to be a late date intrusion related to faults (Kato, 2003).

The geology of the Kibiro geothermal area (Figure 2) is quite unique in its tectonic setting (Gíslason et
al., 2005). Geological and geophysical studies carried out in this area revealed that the geothermal
activities may be associated with the block faulted granites some distance from the main rift system.

2.1.2 Geology of Panyimur

The area (Appendix I) is characterized by fractured crystalline basement rocks such as coarse hornblende
gneisses, coarse hornblende garnet rocks, talcose rocks and pegmatitic veins in a gorge that dips into the
escarpment. In the area there is foliation/basement schistose trending- NNE-SSW. In addition, there is
another schistose trend almost NE, parallel to the major local faults (Natukunda, 2013). The major rocks
in the area include crystalline basement rocks, which are coarsely jointed granitic-gneiss with outcrops
to the west and Pleistocene sediments to the east of the Rift fault boundary. There are some mylonite
gneisses which are intruded by pegmatitic dykes, quartz veins and garnetiferous amphibolite dykes on
Lajwe 204 Report 12

the escarpment, with the minor faults cross cutting the rift faults (upper Panyimur fault and the lower
Panyimur fault, both striking northeast).

2.2 Geology of Öxarfjördur

The Öxarfjördur area in NE-Iceland is characterized by historical lavas of younger to older age, with
thick sediment of about 1000 m (Georgsson et al., 2000) from a glacial river and by fissure swarms from
active central volcanoes further inland (Figure 2). It is located at the junction between the NE-SW
spreading zone which crosses Iceland, and a right – lateral transform zone, the Tjörnes Fracture Zone
(Ólafsson, et al., 1993). Today the waters around the island are oxygen saturated. This could be due to
the fact that major climate changes during the Pleistocene could have changed the paths of ocean
currents and maybe that of the oxygen cycle, too.

The Öxarfjördur sedimentary graben falls in a transform zone of lateral transform movement, spreading,
and volcanism in the zone between the North Iceland Volcanic Zone in Öxarfjördur and the spreading
axis along the Kolbeinsey Ridge, (Jónsson et al., 1991).

The geological model of the Öxarfjördur sedimentary trough, based on the tectonic settings and
geophysical data, indicates that it is younger than 1 Ma, with a probable thickness of up to 1 km at the
shore inside the Krafla and Theistareykir fissure swarms, while thinning southwards where the
sediments are expected to be more frequently interbedded with lavas and hyaloclastites (Georgsson et
al., 2000).

FIGURE 2: Geological maps of Kibiro and Öxarfjördur (DGSM, 2013 and ISOR)

2.3 Surface geothermal manifestations in Kibiro, Panyimur and Öxarfjördur

2.3.1 Surface geothermal manifestations in Kibiro and Panyimur

The hot springs in the Panyimur geothermal area are located within a stretch about 1.5 km along the
main fault (Data and Bahati, 2003) and all are aligned in a northeast direction in the western escarpment
of the Rift Valley, which probably means that they are controlled by the intersection of the traverse fault
Report 12 205 Lajwe

together with the main rift. In addition to that, deposits of travertine, sulphurous algae, the smell of
hydrogen sulphide and all the discharges are characterized by gas bubbling (Bahati, 2007).

The manifestations of Kibiro are mainly concentrated at Kibiro and Kachuguru, west of the escarpment,
on the shores of Lake Albert; some fumarolic activity can be seen at Kachuru and along the escarpment
for approximately 1.5 km from the intersection of the Kachuguru fault and the main escarpment to the
Kibiro hot springs (Bahati, 2012). There are some calcite deposits observed in cracks and fissures in
many locations along the escarpment, the Kachuguru fault, the Kitawe fault and in the crystalline rocks
located south-southeast of Kibiro, indicating extinct thermal fluid discharges. The hot springs at Kibiro
are apparently related to a secondary fault, oblique to the main Rift fault, and most likely controlled by
their intersection (Figure 2). The total flow measured from the hot springs is approximately 7 l/s and the
maximum surface temperature is 86.4°C (Gíslason et. al., 1994).

2.3.2 Surface geothermal manifestations in Öxarfjördur

The surface geothermal manifestations in the Öxarfjördur region are meagre, mainly in the form of warm
springs and warm ground within the active fissure swarms. Geothermal manifestations are known at 13
locations, which can be divided into 3 groups (Georgsson et al., 1993). These groups are: 1) The Krafla
fissure swarm which has the most important sites like Skógalón where water at a shallow level has
temperatures up to 100°C, covering an area of several hundred square metres, and Bakkahlaup with
warm ground covering a wide area with temperatures of up to 80°C in the uppermost part, considerably
lower before the Krafla fires. 2) The Theistareykir fissure swarm in the western part of the Öxarfjördur
area, with the most important one at Ytra Lón close to the coast, with recently measured temperatures
of 50°C; older records show temperatures above 80°C. 3) Geothermal sites found outside the active
fissure swarms.

2.4 The chemistry of the geothermal fluids

2.4.1 Brief overview of geothermal fluid chemistry

The chemistry of geothermal fluids is diverse and largely reflects the geological setting of a particular
area (Henley and Ellis, 1983). Most of these chemical differences depend upon the source of recharge
waters with some contribution of gases from magmatic o r metamorphic sources. Geothermal fluids
contain a number of different constituents with varying concentrations (Ármannsson, 2013), with
important chemical parameters which characterize geothermal fluids, such as pH which describes the
acidity or alkalinity of the fluid, and TDS which gives a measure of the amount of chemical constituents
dissolved in the fluids.

The constituents of geothermal water can be grouped into two categories with respect to their properties,
i.e.:

1) Reactive constituents, which are also used as geo-indicators due to the fact that they tend to
equilibrate with some of the minerals in the geothermal system (Arnórsson, 2000a).
Information on temperature and the steam fraction, which together constitute the physical state
of the geothermal system, may be provided by these constituents.
2) Conservative constituents such as Br, Cl, B, N2,2H, 13C, 40Ar, and 18O (Arnórsson, 2000a) may
be used to trace the origin and flow of the geothermal fluids, as they do not take part in water-
rock interaction.

The geothermal fluids or natural waters may be classified using the Cl-SO4-HCO3 ternary plot diagram
(Giggenbach, 1991). This diagram is used to distinguish a number of different types of thermal waters
like: immature unstable waters, giving an indication of the mixing relationships or the geological
groupings; steam heated waters; volcanic waters; and peripheral waters. And this leads to the provision
Lajwe 206 Report 12

of a trend and preliminary statistical evaluation of the grouping of geothermal waters (Giggenbach,
1988). Water circulating in high-enthalpy geothermal reservoirs is mostly of meteoric origin but, in
some geothermal systems, it may be oceanic or evolved connate waters (Bruhn et al., 2010). For
geothermal systems which are located along the convergent plate boundaries and are close to the
volcanic-magmatic association, the deep magmatic heat source tends to add some acidic gases like CO2,
SO2, HF, H2S and HCl, with some andesitic waters in the system.

In high-enthalpy geothermal systems, four different types of geothermal waters have been classified
(Truesdell, 1991; Giggenbach, 1988; Henley et al., 1984; Ellis and Mahon, 1977):

1) Acid sulphate water: this is the type of thermal water which is mainly found in the upflow part
of the geothermal system, (Bruhnet al., 2010). This type of geothermal water is characterized by
a low pH of 0 to 3 and a low concentration of chloride. It reacts very fast with the host rocks,
mainly giving advanced argillic alteration paragenesis. Composed of alunite and kaolinite. CO2
and H2S are the two main types of gases in this geothermal environment. The H2S may undergo
oxidization in the upflow zone to form SO4 near the surface and contains constituents mostly
dissolved from the surrounding rock with which it is in contact (Ármannsson, 2009).

2) Sodium – chloride water: water with a chloride concentration of up to 10,000 mg/kg, and pH of
approximately ±1 or ±2 units to a neutral pH at depth, for example 5.5-5.6 at a temperature of
200oC to 300oC. This type of water contains the following constituents in higher concentrations
than in cold water: Na, B, F, Li, Si, K, Ca, but with low Mg and with CO2 and H2S as the main
dissolved gases (Bruhn et al., 2010).

3) Acid chloride – sulphate water: this type of water is found in the deepest part of the connecting
neutral pH that is a NaCl system, there and in the inflow of magmatic gases rich in sulphur and
hydrochloride species, namely H2S and SO2; this produces a very reactive aqueous solution that
dissolves the rock or leaches the cations from them, leading to the deposition of pyrite, alunite,
kaolinite and anhydrite (look at the chapter in Ármannsson, 2010). The acid Chloride – Sulphate
water is mainly found in crater lakes (Bruhn et al., 2010). The interaction of the magmatic gases
with rock masses and water is greater at depth, with higher temperatures and lasts for a longer
period than in the crater lakes. This leads to the formation of neutral NaCl waters (Reed, 1991;
Giggenbach, 1997), due to more neutralization in the system.

4) Sodium bicarbonate water: found in: the zone of condensation for a vapour-dominated
geothermal reservoir; the marginal regions of the liquid-dominated geothermal reservoirs; in the
deep geothermal reservoir located in sedimentary rocks and/or metamorphic rocks. The
condensation of geothermal steam in a free oxygen groundwater in a relatively deep zone may
be the dissolution of CO2, present in some of the gases in the system, which can lead to the origin
of bicarbonate rich waters (Bruhn al., 2010). The lack of oxygen in this system prevents the
oxidation of H2S, meaning that the acidity in the aqueous solution is from the dissociation of the
H2CO3 as shown in Equation 1.

⇌ (1)

The conversion of feldspars to clay is aided by this weak acid, especially in medium to high
temperature geothermal reservoirs, gives rise to neutral aqueous solutions.

2.4.2 Water-rock interaction

Geological processes are mostly a result of reactions which take place between the minerals in the rock
and the fluids (Driesner, 2013). When geothermal water or fluids move through the rock, they may
chemically react with the rock with which they come into contact, (and some of the rocks themselves
are chemically complex). This may lead to the following: 1) dissolution of some selective minerals
Report 12 207 Lajwe

from the rock; 2) Precipitation of some minerals from the solution; and 3) substitution of certain
chemical elements in the fluids by other elements in the minerals. All three processes may lead to
secondary mineral formation by depletion or enrichment of some of the chemical components from the
fluids (Rodríguez, 2011), or removal of chemical components from the rock due to the dissolution of
the primary minerals. A series of chemical reactions, associated with the interactions between the water,
the chamber rock and the path by which the water reaches the earth’s surface (Arnórsson et al., 2007)
may occur, for example reactions which typically involve sulphur and /or metal cations. Water - rock
interactions generally take place slowly, that is primary rock dissolves very slowly in water but, like
most reactions, the process is faster at higher temperatures. The changes in the chemical and/or
mineralogical composition may or may not lead to volume changes in the rock (Arnórsson et al., 2002).
If the volume of the rock is changed, then porosity and permeability may be affected, too. When the
reaction between water and rock is prolonged, some of the ionic species and saturated silica are dissolved
(Equation 2). Na+(aq) and Ca 2+(aq) also dissolve in the geothermal water.

Physico-chemical conditions like temperature may, if decreased, lead to the precipitation of minerals
due to low solubility (Equation 4). Boiling, upon pressure decreases, may cause re-partitioning of the
forces between the vapour and the liquid phase leading to solubility decreases for some metals, for
example, gold.

2 2 9 → 2 4 (2)
Feldspar rock kaolinite clay dissolved silica

Water may be in equilibrium with the rock for sufficiently long to be used to calculate the temperature
of the water. In effect, the ratios of the concentrations of the liberated cations like: Ca2+, Na+, Mg2+and
K+ are temperature dependent. The most important process governing their relative contents is rock
dissolution and equilibration (Giggenbach, 1984).

The solubility of silica is also temperature dependent, but this equilibrium is attained rather more rapidly
than the dissolution of most other silicate minerals.

2 → (3)
Dissolved silica siliceous sinter

This reaction may be used as the basis of a geothermometer (Afeworki, 2010). Any difference in
temperatures obtained from the two geothermometers is related to the sampling of the water relative to
the deep aquifer: waters that have moved away from the aquifer are likely to have lower temperatures
derived from silica concentrations than temperatures derived from concentrations of Na, K and Ca. This
may either be due to faster equilibration in the silica dissolution than the ion-exchange reactions or to
mixing with dilute water. Some ion exchange reactions are, however, very fast such as the K-Mg
reaction, and a geothermometer based on it may even show a lower temperature than a silica
geothermometer.

Iron and manganese ions may be liberated from silicate minerals and ultimately precipitate as their
highly coloured oxides. The example below shows the oxidative liberation of iron from a silicate mineral
under acid conditions. Iron (II) is oxidized to iron (III).

7 2 → 2 3 2 (4)
Geothermal water Fe-bearing silicate clay (kaolinite) hematite geothermal water
(ferric oxide)

As the water rises to the surface it cools, and silica minerals are precipitated along with other minerals.
This may help "cap" the aquifer, and prolong the geothermal system's existence. Water that does reach
 Lajwe 208 Report 12

the surface through faults and fractures will cool further and produce the sinter mounds and terraces that
are typical around geysers and hot springs.

When water in a geothermal system has attained a high temperature that is above 100°C, it will react
with the rock to a greater extent, leading to the production of significant oxygen shift, which can be
proven by Equation 4 (Arnórsson, 2000b); this may tell us if the geothermal water shows precipitation
which fell during the present day climate or under different climatic conditions.

2.4.3 Mixing in the upflow zones

Geothermal waters may be mixed with cold water after a certain period of conductive cooling of the hot
water, before, after or during boiling (Arnórsson, 2000b). Depending on pressure and temperature
conditions, the main component of geothermal fluids, H2O, can be present in different physical states
(Bruhn et al., 2010). The presence of a single liquid phase in the geothermal reservoir is the most
frequent situation, but it is not the only one, since either a two-phase liquid vapour mixture or a single
vapour phase can be present in the reservoir. These possibilities can be ascertained by accurate enthalpy
data or by gas geochemistry (Giggenbach, 1980; Bertrami et al., 1985).

3000 Large variations in the

temperature and flow rates
Steam
of thermal springs in a
particular field that can be
linked with the parallel
variations in the
concentrations of the non-
2000 reactive components in the
water, like Cl, usually
constitute the best
Enthalpy (J/g)

evidence that mixing has

taken place (Marini and
Cioni, 1985). Mixing
Aquifer fluid (265°C) models have been
developed to allow the
1000 Steam loss estimation of the hot water
g
component in mixed
n
ixi waters emerging in springs
t erm
a or discharged from shallow
l dw Aquifer fluid
Co flashed to drill holes (Truesdell and
100°C Fournier, 1977). Three
mixing models are most
0 frequently applied: 1) the
chloride-enthalpy model;
0 1000 2000
Chloride (mg/kg) 2) the silica-enthalpy
mixing model; and 3) the
FIGURE 3: Enthalpy versus chloride for boiling and dilution silica – carbonate mixing
(Foumier, 1979) model.

The enthalpy versus chloride plot is a suitable tool for distinguishing the effects of boiling (steam loss)
and mixing, since both steam and cold waters, which generally have low chloride contents, are
characterized by very different enthalpy values (Fournier, 1979). The enthalpy-chloride plot (Figure 3)
shows that boiling moves the liquid from the point representative of the 265°C geothermal liquid
towards higher chloride contents and lower enthalpies, whereas the addition of cold, dilute waters causes
a decrease in both enthalpy and chloride. If the discharged water is cooled, mainly through conductive
heat loss, the chloride concentration of the deep hot water remains unchanged (Bruhn et al., 2010).
Report 12 209 Lajwe

The linear relationship between the concentrations of conservative components like CI and B or CI and
δ2H are generally considered to constitute the best evidence for mixing (Bruhn et al., 2010). The
magnitude of the oxygen shift depends on the extent of the water - rock interaction. Generally, increasing
temperature enhances chemical reaction rates (Stefánsson, et al., 2007) including reactions involving
water and rock, hence increasing the 18O isotope shift (see Chapter 2.4.5). Mixing of geothermal waters
with local groundwater may manifest itself in a linear relationship between the δ values for 2H and 18O
or between these values and the aqueous concentrations of conservative elements such as Cl.

Hot water ascending to the surface from the geothermal reservoir tends to have a relatively high
concentration of total dissolved solids (Árnason and Gíslason, 2009), but cold groundwater contains
relatively low total dissolved solids, so if such cold groundwater mixes with the hot water, it leads to
dilution which reduces the concentration of the total dissolved solids (Rodríguez, 2011). Chloride,
which is a non-reactive component in water, tends to give the best evidence of mixing in water systems,
depending on the variation of temperature and the flow rates of the geothermal hot or cold water.

In many geothermal hot springs, mixing of the hot and cold water leads to dilution and cooling (Noda
and Shimada, 1993); to prove this by using the different mixing model techniques, the chemical
composition of each component in the geothermal water needs to be evaluated thoroughly. And for
better use of mixing models, to establish whether there is mixing in the up flow zone of a geothermal
system, a large number of samples from the geothermal area is required. The chloride enthalpy plot is
used in mixing prediction because chloride concentration increases gradually with continuous rock
dissolution with an increase in temperature (Yohannes, 2004).

2.4.4 Mineral saturation

This is one of the techniques used in the evaluation of mineral equilibrium, using the assumption of
mineral dissolution in solution, taking into account the different physical conditions in which this
mineral dissolves (Arnórsson and Bjarnason, 2007). Determining the chemical equilibrium between the
aqueous solution and the mineral in any natural system needs to be carried out with an idea of the
activities of the different species present in that solution (Arnórsson et al., 2002).

2.4.5 Isotope chemistry of geothermal fluids

Isotopic chemistry is one of the most important tools used in the exploration for geothermal resources
today. To gain representative results for a particular geothermal system, the isotope values need to be
interpreted in conjunction with other chemical techniques (Giggenbach et al., 1984; Giggenbach., 1991;
Craig., 1963). It is used mainly to establish the isotopic characteristics (that is δ18O and δ2H) of the
precipitation in relation to the altitude and latitude, together with continental effects. There is a general
decrease in the values of the δ18O and δ2H for samples collected from geothermal areas that are further
inland, at a higher elevation and at higher latitudes. And these values are more closely related to the
meteoric water line through the equation:

δ H 8δ O d (5)

where d = 10 and for each isotope δ = [(Rsample – RSMOW)/RSMOW] ×1000, reported in ‰, and R = the ratio
of the number of atoms of a given isotope to the number of atoms of the element’s most abundant
isotope, i.e. R = 18O/16O for oxygen-18 and R =2H/1H for deuterium and SMOW is the Standard Mean
Ocean Water as defined by the IAEA.

In a high-temperature geothermal system, there may be isotopic exchange between rock minerals which
are rich in δ18O and water, causing the δ18O values to increase and fall to the right of the world meteoric
line, or the so called oxygen isotope shift (Giggenbach, 1991). Along the convergent plate boundaries
of a volcanic system, Giggenbach (1992) showed that there is a linear relationship between the values
of δ18O and δ2H, which is also the case in numerous geothermal systems showing a mixing trend in
Lajwe 210 Report 12

common with a magmatic source with δ2H of about -20 ± 10‰ and δ18O of about +10 ‰. This
phenomenon is common in subduction systems and explains the isotope signature of andesitic waters,
but is unlikely to be applicable in the systems presently studied.

2.4.6 Geothermometers in geothermal exploration

This has proved to be one of the major exploration tools used in the search for geothermal energy. This
method is based on the isotopic and the chemical composition of the fluids that discharge from natural
hot springs, cold springs and wells (D´Amore and Arnórsson, 2000b). They are used for the following:
1) to predict or estimate the subsurface temperature of the geothermal system which is expected to be
encountered in case of any drilling in the area for exploration or production purposes; 2) during the late
stages of geothermal development and utilization, especially in establishing the horizons in productive
wells; and 3) the chemical state of the recharging water in the vicinity of a well as a result of cooling or
boiling, due to depressurization, can be determined by this technique (Bruhn et al., 2010).

In a geothermal system, most of the chemical reactions are temperature dependent (Ármannsson, 2009)
and their kinetics are not fast at a relatively low temperature, hence the equilibrium characteristics are
preserved even when the water flows to the surface and cools down. Using the assumption of the
conservation of mass and heat and determining the chemical composition of the surface geothermal
fluid, the subsurface temperature may be determined.

Geothermometers are divided into basically three categories:

1. Isotope geothermometers
These are based on the fractionation of the isotopes of light elements (given by Equation 5) between
gaseous and aqueous portions of the geothermal fluids. This is mainly temperature dependent making
it very important:

1000 (6)

where ln αAB =Isotopic fractionation in two phases;

δA = Concentration of constituent A;
δB = Concentration of constituent B.

2. Water or solute geothermometers

These are based on the fact that the concentration and any ion ratio for an aqueous species may be used
as a geothermometer, taking into consideration that equilibration is attained for the reaction (Arnórsson
and Svavarsson, 1985). The most common types of such geothermometers are the silica (quartz and
chalcedony), the Na–K and the Na–K–Ca (Fournier, 1973; Fournier and Truesdell, 1973; Ellis and
Mahon, 1977; D’Amore and Arnórsson, 2000). For this particular research work, the water or solute
geothermometers used are shown by the equations below.

Arnórsson et al., 1983 (Cation):

933
273.13 (7)
0.993 log

Truesdell and Fournier, 1976 (Cation):

856
273.13 (8)
0.857 log
Report 12 211 Lajwe

Giggenbach et al., 1994 (Quartz):

4410
273.13 (9)
14.0

Fournier, 1977 (Quartz):

1522
273.13 (10)
5.75

Fournier, 1977 (Quartz):

1309
273.13 (11)
5.19

Arnórsson et al., 1983 (Quartz):

1112
273.13 (12)
4.91

Fournier, 1977 (Quartz):

1032
273.13 (13)
4.69

Fournier and Truesdell, 1973 (Cation):

1647
273.13 (14)
√
log 2.06 2.47

Arnórsson, 2000 (Quartz):

55.3 0.36590 5.3954 10 5.5132 10 74.360 (15)

Fournier and Plotter, 1982 (Quartz):

53.5 0.11236 0.5559 10 0.1772 10 88.390 log (16)

Arnórsson, 2000 (Quartz):

66.9 0.13780 4.9727 10 5.5132 10 87.841 log (17)

Verma and Santayo, 1997 (Quartz):

44.119 0.24469 1.7414 10 79.305 log (18)
(19)
2.8 1.6
(20)
0.8 0.2 5.4 2
4410 (21)
Lajwe 212 Report 12

3. Gas geothermometers
These are applicable for gas and steam samples since in many geothermal fields there are fumaroles, hot
ground and acidic surface waters. These types of geothermometers are grouped into three as follows:

1) The mineral gas equilibria geothermometers;

2) Gas – gas equilibria geothermometers; and
3) Mineral – gas equilibria geothermometers involving CO2, CH4 and H2S, which need to be
externally fixed, using empirical methods.

3. METHODS

3.1 Data sources

The geochemical data of selected hot springs used in this research work were collected from the
department of geological survey and mines in Uganda, (GSMD). These data were acquired during the
UNDP/UNDESD project funded by OPEC and the Iceland Government, and data for the Öxarfjördur
area were collected from Iceland GeoSurvey, (ISOR), by the geochemistry department; these results
were acquired as part of the early studies for the search for oil. The data set includes analysed samples
from selected hot springs, a cold spring and wells listed in Table 1.

3.2 Field sampling methodology

In all these areas of research work, geochemical sampling of the hot springs, and the wells was carried
out by the GSMD geochemistry team, together with Dr. Halldór Ármannsson from the Iceland
GeoSurvey, in Uganda (GSMD, 2008), while the sampling from the Öxarfjördur area, north Iceland,
was carried out by the Iceland GeoSurvey geochemistry team. Measurements of the physical parameters
and an analysis of the volatile components of the water samples collected were carried out in the field.
Rock samples were collected for mineralogy analysis to determine the source of salinity in the
geothermal waters from Kibiro and Öxarfjördur, although these were not used in this research work.

Untreated samples were collected for pH, carbonate, H2S and conductivity measurements and also for
the analysis of Mg and SiO2. During the sampling, the concentrations of SiO2 were expected to be more
than 100 ppm, so the samples were diluted with distilled or deionized water to bring the concentrations
into the range of 30 to 100 ppm in triplicate and the dilution factor was recorded. Samples were also
collected and anions were filtered; for the cations, the filtered samples were acidified with nitric acid.
All the samples used in the determination of sulphate were precipitated with Zn (CH3COO)2 and then
the sulphate was analysed after filtration of the samples. For the analysis of stable isotopes (δ2H, δ18O),
the isotope laboratory needed 60 ml of filtered sample. A summary of the sample treatment is given in
Table 2 and the preservation method for the sampling of hot spring and hot water wells are presented in
Table 3.

These samples had to be preserved both by chemical and physical processes, as shown in Table 3, in
order to ensure no changes in the chemical and physical characteristics of these waters and the gases.
Preservation techniques were carried out both in the field and in the laboratory, depending on the
handling of the different constituents in these samples.
TABLE 1: Chemical composition (mg/kg) of selected samples from the
Öxarfjördur, Kibiro and Panyimur and (isotope values in ‰ SMOW)
Report 12
213
Lajwe
Lajwe 214 Report 12

TABLE 2: Treatment and sub-samples from hot springs and hot water wells
(Ármannsson and Ólafsson, 2012)

Capacity
Specification Container Treatment Used for
(ml)
pH, CO2, H2S
None; amber glass bottle with ground
Ru Glass 250-300 (if not in field)
glass stopper
and conductivity
Mg, SiO2
Ru Plastic 200 None
if < 100ml
Dilution; 0ml of sample added to 50 ml of
Rd (1:1) Plastic 3×100 SiO2 if < 100ml
distilled, deionized water
Fu Plastic 200 Filtration Anions
Filtration; 0.8ml of conc. HNO3 acid
Fa Plastic 200 Cations
added to 200 ml of sample
Filtration; 2ml of 2M ZnAc2 added to
200, SO4, δ34S
Fp, Fpi Plastic sample in 100 ml volumetric glass flask
>500 andδ18O in SO4
and ≥10 ml to ≥500 ml bottle containing
≥25mg SO4 to precipitate sulphide
Filtration; 1, 60ml, and 2, 1000ml amber δ2H, δ18O, 13C
Fui, Fuc,Fut Glass 60, 1000
glass bottles and 3H
Note: Fp =Filtered precipitated; Fpi =Filtered precipitated for isotope; Fui =Filtered untreated;
Fu =Filtered untreated; Fa =Filtered acidified; Ru =Raw untreated; Rd =Raw diluted.
TABLE 3: Preservation methods for geothermal samples (Ármannsson and Ólafsson, 2006)

Type Method Purpose Used for

Physical Filtration with 0.45 or Prevent interaction with suspended Anions, Cations, TDS
0.22 μ membrane filter matter
Freezing Prevent biological activity Nutrients
Airtight container Prevent interaction with atmospheric air
On-site analysis Prevent reactions of reactive Reactive constituents
constituents
Chemical Base addition Absorption of acid gases CO2, H2S in steam,
δ34S in H2S in vapour
Acidification Prevent adsorption on the walls of Cations
containers
Precipitation Prevent a constituent from reaction to Sulphate to fix
change the concentration of another Sulphide
constituent
Sterilization Prevent biological activity, using HgCl
or formaldehyde
Dilution Prevent polymerization and Silica
precipitation
Redox To change oxidation state of volatile Hg
constituent to make it less volatile
Ion exchange Concentrate and further prevent Trace cations
adsorption on the walls of container of
trace constituents
Extraction Concentrate and further prevent Trace cations
adsorption on the walls of container of
trace constituents
Report 12 215 Lajwe

3.3 Analytical methods

The Department of Geological Survey and Mines carried out the determination of the TDS and
conductivity of the geothermal waters in their laboratory at Entebbe. The samples were then divided
into different portions for analysis. Then the samples were sent for complete chemical analysis to the
Institute of Geological and Nuclear Science, New Zealand (Ármannsson, 2010), but the samples
collected from Öxarfjördur area were analysed in the Orkustofnun Laboratory. The methods and
instruments summarized in Table 4 were used in the analysis of these samples.

TABLE 4: Analytical methods (Ármannsson and Ólafsson, 2006)

Constituents Laboratory Method LOD PR RSD%

Conductivity GSMD 4 electrode
pH GSMD GE
CO2 and H2S GSMD Titration 0.02 – 1.0 3.8 – 13.7
OS and
SiO2, B and NH3 SP 0.01- 0.5 0.7 – 8.3
GSMD
Na, K, Ca, Mg, Sr, Li OS AAS 0.0005-0.1 1.5 – 2.7
SO4, Cl and Br OS IC 0.02 – 0.025 0.8 – 5.4
TDS OS GR 2.5 3.4
F OS ISE 0.05 0.9
Fe, Al and Mn OS FAAS 0.001
δ18O and δ2H USI MS
CO2, n.c., H2S, n.c., CH4, H2,
OS GC 0.01 – 0.02
N2, Ar, and C2H6

4. RESULTS AND DISCUSSION

4.1 The aqueous chemistry of the geothermal fluids in these areas

The results for selected hot springs and hot water wells presented in Table 1 from Kibiro, Panyimur and
Öxarfjördur are generally in the range of neutral – alkaline with pH of 6.7- 9.5. A general summary of
the different chemical component ranges in the samples from Kibiro, Panyimur and Öxarfjördur are
presented in Table 5. Samples from Öxarfjördur contain very low H2S (<0.03 – 0.07 mg/kg), CO2 (5.1
– 30.8 mg/kg); the δ18O is -12 - -9.3 ‰, and δ2H -114.1 - -87.8 ‰, and TDS is relatively high (1283 –
6401 mg/kg). Samples from Kibiro also contain relatively high CO2 (115 – 367 mg/kg) and H2S (0.0 –
17.3 mg/kg). Panyimur samples, in general, contain relatively low concentrations of all the chemical
components. And, in general, the samples from Panyimur and Öxarfjördur contain lower concentrations
of Mg than the samples from Kibiro. There is a relatively higher concentration of fluoride in some of
the samples from Panyimur, compared to the samples from Kibiro and Öxarfjördur. The chemistry of
these water samples differs mainly in the samples from Öxarfjördur which, in general, contain almost
none or very little H2S, but contain the highest amount of Na, K, TDS, and concentrations of SiO2, Na,
Cl, CO2, Mg and TDS.
Lajwe 216 Report 12

TABLE 5: Summary of the different constituents in mg/kg and stable isotopes

in ‰ SMOW of water samples of Öxarfjördur, Kibiro and Panyimur

T(°C) PH CO2 H2S Cl SO4

Gen 24 -132 6.26-9.5 5.1-367 <0.03-17.3 5.2-3539 1.9-352
Öxa 78.5-132 7.3-9.5 5.1-30.8 <0.03-0.07 658-3539 46.1-232
Kib 23.6-86.5 6.26-8.93 115-367 0-17.3 5.2-2580 5.3-227
Pan 35-60 7.56-9.06 71-216 <0.1-5.61 51-470 4.4-352
K Ca Mg SiO2 B TDS
Gen 2.6-169 1.4-400 0.07-39.5 54-214 <0.1-16.7 124-6401
Öxa 41-98 6.1-400 0.18-1162 1.6-214 0.53-1.77 104-36110
Kib 2.6-470 9.75-75.9 0.99-39.5 0.5-147 2.0-2.47 124-4810
Pan 2.6-19.2 1.4-24 0.04-3.1 54-129 <0.1-0.76 337-890
Na Br δ2H δ18O F
Gen 12.4-10113 <0.02-17.3 -100.5-39.8 -0.8-5.47 0.12-13.8
Öxa 427-10113 2.2-12.1 -100.5-7.1 -12--0.8 0.12-0.7
Kib 12.4.2163 0.13-17.3 -3.9-39.8 -1.01-5.47 5.12-7.2
Pan 110-352 <0.02-1.5 -16.6-2.5 -18.8--2.0 2.4-13.8

4.2 Use of ternary diagram in classifying geothermal waters of Kibiro, Panyimur and Öxarfjördur

4.2.1 Cl-SO4-HCO3 ternary diagram

The Cl-SO4-HCO3 ternary

plot diagram has been used
in the classification of
geothermal and natural
waters (Giggenbach, 1991).
Composition of many water
types is shown as one of the
following in the triangular
plot (Figure 4): mature
waters, steam-heated waters,
volcanic waters and
peripheral waters. From the
Cl-SO4-HCO3 ternary plot,
the degree of separation of
the plotted points for
high-chloride and
bicarbonate waters shows
some indication of a r elative
degree of interaction of t h e
fluids at lower temperature
with CO2, and of the HCO3
contents increasing with
time and distance travelled
FIGURE 4: Cl-SO4-HCO3 ternary plot (Giggenbach, 1991) for water
by the waters underground
samples from Kibiro, Panyimur and Öxarfjördur
(Giggenbach, 1988).

In Figure 4, for the waters from Kibiro, Panyimur and Öxarfjördur, there is a great degree of variation
among the HCO3 and Cl waters. The waters from the Öxarfjördur area plotted in the volcanic region of
the diagram, with two samples in the Cl portions, one in the SO4 and another one in the HCO3 part. Four
of the Kibiro samples plotted in the chloride portion of the diagram, and three in the HCO3 region, but
Report 12 217 Lajwe

more towards the SO4 and the HCO3 section. Two of the Panyimur samples plotted in the Cl area, and
two in the peripheral water zone.

4.2.2 The Na-K-Mg triangular diagram

This technique for the derivation of Na-K-Mg-Ca geoindicators was initiated by Giggenbach (1988) as
one of the classification techniques for different kinds of water depending on whether they are: 1)
immature water; 2) partially equilibrated water; or 3) fully equilibrated water. Using the plots, one may
apply geothermometers to t h e equilibrated and partially equilibrated water only. The triangular
diagram, w h i c h is temperature dependent, is based on two reactions, Equations 17 and 18.
(22)
2.8 1.6 0.8 0.2 5.4 2 (23)

Equations 22 and 23 are used in the derivation of the geothermometer used.

In the triangular plot, the area of partial equilibrium indicates that there could be some mineral that
has dissolved, though it has not yet attained equilibrium or it may be that it is geothermal water,
but has been diluted with water that has not yet reached equilibrium, for example cold
groundwater. Points which plotted close to the √Mg corner indicate that there is a very high
proportion of relatively cold groundwater mixed in the water system.

In Figure 5, 4 samples from Panyimur and 6 samples from Öxarfjördur plotted close to the full
equilibrium line but 4 samples from Panyimur and 1 sample from Öxarfjörður plotted in the partially
equilibrated region of the diagram, probably meaning that there may be some recharge from cold
groundwater into the geothermal systems in those areas. In Kibiro, two of the samples plotted in the
immature region, hence contain cold groundwater, while six of the samples plotted in the partially
equilibrated portion. The
tkm temperature for
samples from Kibiro
varied, ranging between
160°C to 140°C, and then
10

100°C to <100°C; the tkn

0

12
0
was from 100°C to 240°C; 14
for Panyimur, the tkm 0
16
ranged from 120°C to 18
0

70°C and the tkn from 0

100°C to 160°C; for 22

0
24
Öxarfjördur, tkm varied in 26
0
0
the range of <100°C to 28
0
200°C and the tkn from
100°C to 200°C. The tkm
temperature for all these
samples varied
considerably. This could
be due to the fact that the
K-Mg geothermometer
tends to react faster to
change than the K-Na
geothermometer in the
same system.
FIGURE 5: Na-K-Mg triangular diagram for samples from Kibiro,
Panyimur and Öxarfjördur
Lajwe 218 Report 12

4.3 Geothermometer data output

Sixteen different geothermometer Equations (eight for quartz geothermometers, two for chalcedony,
three each for Na/K, K-Mg and Na-K-Mg) were applied in the prediction of the subsurface temperatures
of the geothermal water from Kibiro, Panyimur and Öxarfjördur. The calculated geothermometer
temperatures for eleven geothermometers out of the sixteen are presented in Table 6. The average
calculated quartz geothermometer temperature for hot springs and wells in Öxarfjördur was 150°C, in
Kibiro samples it was 123°C and in Panyimur it was 129°C. The Na-K geothermometer temperatures
for Kibiro and Panyimur were high while the Na-K-Ca geothermometer gave very high temperatures
for Öxarfjördur. This may due to loss of CO2 leading to calcite precipitating out of the solution. This
affects the concentration of calcium in the solution, leading to abnormally high geothermometer
temperatures. The calculated chalcedony geothermometer temperatures for these samples were: for
Öxarfjördur 116°C, for Kibiro 114°C, and for Panyimur 101°C. The calculated Na-K geothermometer
temperatures were: for Öxarfjördur 148°C, for Kibiro 197°C and for Panyimur 116°C. Appendix II
shows comparison plots of calculated temperature results obtained from sixteen different equations of
geothermometers for Öxarfjördur, Kibiro and Panyimur.

TABLE 6: Calculated values for 11 equations for solute geothermometers temperatures (in °C)
for samples from Öxarfjördur, Kibiro and Panyimur

TNa-K TNa-K 1 TK-Mg Tqz, 1 Tqz, 2 Tqz, 3 Tqz, 4 Tch 1 Tch 2 Tqz, 5 Tqz, 6
LOC
ISK 139.6 129.77 132.6 152 153.80 150.2 143.08 136.42 132.7 160.33 160.39
ISG 144.4 134.79 197.4 142.6 144.1 137.82 132.77 123.60 121 148.95 148.94
IBK 191.1 183.73 130 132.9 134.1 125.35 122.23 110.68 109.2 137.36 137.39
IA1 150.3 140.89 191.9 148.8 150.5 145.97 139.57 132.03 128.7 156.46 156.48
IA3 136.9 127.01 153.1 145.1 146.7 141.10 135.5 126.99 124.1 151.98 151.98
IA4 140.8 131.01 161.5 171.9 174.2 176.99 165.12 164.22 157.9 185.61 185.08
IJÖ 162.1 153.24 35.97 63.23 56.68 41.81 43.03 24.71 28.64 57.01 57.12
ISW 115.6 104.91 98.38 1.29 -35.28 -39.53 -48.75 -43.10 -36.85 -10.61 -27.54
KIB
KM1 205.36 198.79 148.7 145.1 146.7 141.05 135.47 126.94 124.1 151.93 151.93
KM2 204.98 198.39 148.00 143.5 145.1 139.01 133.76 124.83 122.1 150.05 150.04
KM3 206.4 199.84 145.7 142.3 143.8 137.45 132.46 123.22 120.7 148.61 148.60
KM4 210.6 204.31 150.1 147.3 149.00 144.03 137.95 130.03 126.90 154.67 154.69
KM4 182.30 174.43 45.87 131.7 131.9 131.85 131.85 128.08 103.4 104.41 131.77
KM6 285.1 284.34 40.18 118.8 119.00 119.00 119.00 117.11 90.25 90.24 119.05
KM7 238.90 234.54 45.34 122.5 122.7 122.69 122.69 120.27 94.02 94.29 122.70
KM8 532.3 564.09 92.86 -34.8 65.25 65.25 65.25 -21.62 -59.76 -66.38 -67.87
PAN
PA1 99.73 88.62 110.9 118.4 119.1 107.31 106.58 91.94 91.84 120.34 120.59
PA2 96.83 85.64 98.15 116 116.5 104.30 103.90 88.81 88.93 117.46 117.77
PA3 137.9 128.06 72.33 105.7 105.5 91.79 92.57 75.78 76.80 105.44 105.97
PA4 83.04 71.54 101.5 116.6 117.2 105.06 104.58 89.6 89.67 118.19 118.48
PA5 108.1 97.21 74.13 115.4 115.9 103.53 103.21 88.01 88.18 116.72 117.04
PA6 141 131.22 96.34 140.7 142.2 135.34 130.68 121.03 118.7 146.66 146.64
PA7 147.3 137.81 106.7 145.1 146.7 141.05 135.47 126.94 124.1 151.93 151.93
PA8 153.9 144.70 110.6 145.1 146.7 141.05 135.47 126.94 124.1 151.93 151.93
Note: TNa-K=Na-K cation geothermometer; TK-Mg=K-Mg cation geothermometer;
Tqz (1-6) =Quartz geothermometers; Tch (1 and 2)=Chalcedony geothermometer.

In many instances there was not a good agreement between the Na-K temperatures and the silica
temperatures for samples from Öxarfjördur, but the cation geothermometers seemed to agree quite
closely. The quartz geothermometer temperatures for Kibiro seemed to agree quite well, and for the
Panyimur samples, there were great variations in all the calculated geothermometer temperatures. The
magnesium concentrations were high in these samples, suggesting a substantial influence of cold
Report 12 219 Lajwe

groundwater, rendering the Na/K geothermometer unreliable; at low temperatures, the silica
geothermometers are not dependable either.

The anomalously high estimates from the Na-K-Ca geothermometer for all the samples could be due to
calcium carbonate deposited during the ascent which prevents aqueous K and Na from interacting with
the country rock (Fournier and Truesdell, 1973).

Geothermometers take advantage of specific mineral –solute reactions (Karingithi, 2009) which are slow
to equilibrate at cooler temperatures, especially under conditions where the fluid is effectively separated
from the minerals which control the equilibria. The result is that a hot equilibrium temperature is “frozen
in” to the fluid, reflected by solute concentrations or solute ratios.

4.4 Mixing model interpretations

Assuming that the silica

has not precipitated before
or after mixing and that
there has been no
conductive cooling of the
water, then the dissolved
silica concentration of a
mixed water and a silica –
enthalpy diagram may be
used to determine the
temperature of the hot
water component.
However, it is not easy to
determine whether a given
spring is a mixture of hot
and cold spring water and
also to prove that silica has
not precipitated or that
conductive cooling has not
occurred (Truesdell and
Fournier, 1977).

From the enthalpy – silica

mixing model (Figure 6),
the samples from Kibiro
indicate a geothermal
system with a temperature FIGURE 6: Silica-enthalpy mixing model applied to water samples
of about 191-220°C; from Kibiro, Panyimur and Öxarfjördur
samples from Panyimur
indicate temperatures of about 150-160°C, and samples from Öxarfjördur show temperatures of 190-
220°C.

The silica – carbonate plot (Figure 7) for all the samples from Panyimur indicates that boiling has not
occurred; for Kibiro, two of the samples showed boiling in the hot spring while some showed
equilibration and others suggested no boiling or mixing. Samples from Öxarfjördur suggested that
boiling had occurred, meaning that some of the CO2 had been lost in the steam phase, hence lowering
the CO2 concentrations in some of those samples.
Lajwe 220 Report 12

4.5 Gas geothermometer temperature results based on this analysis

Gas geothermometer tem-

peratures were calculated
for one sample from Kibiro
and the results were: TCO2
=199°C, TCH4/C2H6 = 208°C
and TH2/Ar = 221°C. For the
samples from Öxarfjördur,
the following estimates of
TCH4/C2H6 were made on
three well samples; these
SiO2 (mg/kg)

showed the following

temperature ranges: Well
AER-01 = 150-170°C;
Well AER-03 = 175-
180°C; and Well AER-04 =
200-220°C. This was done
basically to confirm the
variations of gas
geothermometers and the
chemical or solute
geothermometers.

4.6 Binary plots

FIGURE 7: Silica versus carbon dioxide
A mixing trend can be
clearly seen in all the areas under study as shown in Figure 8a, K+ vs. Cl- and in Figure 8b, Na+ vs. Cl-,
with positive correlations for Öxarfjördur and Kibiro, which may mean that there is a high Cl-
concentration possibly being added into these systems. In Öxarfjördur, it could be that seawater is
mixing with the geothermal waters. In Kibiro, the source may be from the salt mining in that area and/or
sodium chloride geothermal waters. For Panyimur, sodium chloride geothermal waters may be mixing
at depth.

The concentration of HCO3, shown in Figure 8c, in four samples from Kibiro and in two samples from
Panyimur is relatively higher than in other samples, suggesting that there could be CO2 absorption from
the gases that the fluid carries or from the condensation of geothermal steam containing CO2 in the
particular geothermal water samples containing high concentrations of HCO3. There is an increase in
the concentration of chloride in the samples from Öxarfjördur which may mean that there is either
mixing of the geothermal waters with sea water or with deep geothermal fluids.

Looking at the relationships of different elements in the geothermal waters from Kibiro, Panyimur and
Öxarfjördur (Figure 9), the following conclusions may be drawn from these results. There are some
linear relationships between B and Cl (Figure 9a) which may mean that there is mixing of cold water
with hot geothermal fluid in the upflow zones in these systems. For the Kibiro samples, there seems to
be a linear relationship between Li and Cl (Figure 9b) that may be due to the uptake of lithium from the
waters by the surrounding rocks in the area, suggesting that there could be some water rock interaction
in the system

The relationships between Na – B and Ca – B (Figures 9c and d) in which B is a conservative element,

and a linear relationship in the plots of these samples may mean that Ca and Na are present in the
environment in which these waters occur; the increase in the B concentration suggests that there is water-
Report 12 221 Lajwe

200
Öxarfjördur (a)
2000
Kibiro
(b) rock interaction since Ca
Kibiro
Panyimur
Panyimur
Öxarfjördur
and Na are less mobile
160 1600
compared to B, and may
r = 0.999 be removed from the water
r = 0.967
120 1200
r = 0.992
during secondary mineral
precipitation out of the
80 800 solution.
r = 0.314

40
r = 0.939
400
Clear linear relationships
r = 0.04 are observed for δ18O and
0 0
Cl in samples from
0 1000 2000 3000 4000 0 1000 2000 3000 4000 Öxarfjörður (Figure 8d)
Cl (mg/kg) Cl (mg/kg)
which may be due to
mixing of saline and non-
saline component and/or
water-rock interaction and
B and Cl in samples from
all three areas (Figure 9a)
HCO3 (mg/kg)

suggesting water-rock
interaction.

3000
(e) 8000
(f)

6000
2000
SO4 (mg/kg)

4000

1000

2000
rav = 0.996

0
0
0 100 200 300 400
Ca (mg/kg) 0 1000 2000 3000 4000
Cl (mg/kg)
40
(g)
Öxafjördur
Kibiro
Panyimur

30
Ca (mg/kg)

20

10

0 1000 2000 3000 4000

Cl (mg/kg)

FIGURE 8: Binary plots of K vs. Cl, Na vs. Cl, HCO3vs. Cl, Cl vs.
δ18O, SO4vs. Ca, TDS vs. Cl, Mg vs. Cl, and Ca vs. Cl
Lajwe 222 Report 12

Li (mg/kg)
2000 400
(c) Öxarfjördur (d)
Kibiro
Panyimur

1600
300

1200

200

800

100
400

0 0

0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5

B (mg/kg) B (mg/kg)
20 4000
(f)
Öxarfjördur (e) Öxarfjördur
Kibiro Kibiro
Panyimur Panyimur
16

3000

12

8 2000

1000

0
-4
0 50 100 150 200 250
0 1000 2000 3000 4000
SiO2 (mg/kg)
Cl (mg/kg)

FIGURE 9: Binary plots of B vs. Cl, Li vs. Cl, Na vs. B,

Ca vs. B, B vs. Cl and Cl vs. SiO2
Report 12 223 Lajwe

4.7 Mineral saturation

In the log (Q/K) plots showing the equilibrium state of some of the minerals calculated by the WATCH
programme for Kibiro samples (Figure 10), most of the curves intersect the line of log (Q/K) equal to
zero for Kibiro area in the temperature range 140 – 160°C, and some of the minerals converge below
the log (Q/K) equal to zero line at 180 ->200°C. For Panyimur, most of the curves cross the log (Q/K)
equal to zero line at 120 – 160°C and converge below it at 180 – > 200°C. For Kibiro and Panyimur, it
may be inferred that there is some mixing of cold water with a geothermal component. The anhydrite
in the samples from Kibiro and Panyimur are undersaturated. Samples from Öxarfjördur (Figure 11)
show that most of the minerals have not yet reached equilibrium. The curves for the different minerals
intersect the log (Q/K) line equal to zero at temperatures of about 120 – 150°C. And some of the curves
converge below the log (Q/K) line equal to zero at temperatures of about 90 – 180°C, with some curves
converging above 200°C. There is some mixing of cold water with a geothermal component in the
Öxarfjördur area. And most of the minerals are undersaturated, that is their solubility, for example, of
hydrate, talc, and calcite decreases (Stefánsson et al., 2000) with increasing temperature leading to their
being precipitated out of the solution. This may be due to mixing of hot water with cold water in the
area.

4.7.1 More output from the WATCH programme

The WATCH speciation program (Bjarnason, 2010; Arnórsson et al., 1983a) was used in the calculation
of the mineral equilibria, based on the fact that different minerals have different reaction quotients and
equilibrium constants. The ionic balance values for two samples from Öxarfjördur, four samples from
Kibiro and two samples from Panyimur are above 5%, but the rest are below 5%.

4.8 Isotope data interpretation

In the Öxarfjördur area, the δ18O ranges from -12.0 to -0.8 ‰ and δ2H from -100.4 to -7.1 ‰; in Kibiro:
δ18O from 5.5 to -1.0 ‰ and δ2H from +39.8 to - -3.9 ‰, and for Panyimur: δ18O values range from -
18.8 to -2.0 ‰, then δ2H from 2.5 to -16.6 ‰. There is a significant variation in the δ18O values of these
three areas.

All Kibiro samples plotted close to: the WML, KRL and CARL lines, with some slight shift from the
WML line (Figure 12) which may mean that there is some oxygen shift due to water rock interaction
within the system as a result of a low water rock ratio or a high temperature. All samples from Panyimur
plotted above the lines, suggesting that there could be some good permeability in the area; this could
also mean that there is excess deuterium.

The samples from Öxarfjördur all plotted below the KRL and CARL lines, but close to the WML line.
This could mean that the water samples from Öxarfjördur have undergone water rock interaction for a
relatively long time and there may be a high temperature at depth, hence leading to oxygen shift due to
the exchange of oxygen between water and the rock.
Lajwe 224 Report 12

a) b)
log (Q/K)

8
(d) anhydrite
c) d) microline
prehite
quartz
calcite
muscovite
albite
4

-4

40 80 120 160 200

Temperature (°C)
12
(e)
e) anhydrite
muscovite f)
prehite
quartz
calcite
8
microline
albite
log (Q/K)

-4

40 80 120 160 200

Temperature ( °C)

FIGURE 10: Mineral equilibrium plots for Kibiro samples

(a – d) and Panyimur samples (e and f)
Report 12 225 Lajwe
log (Q/K)

8
anhydrite (4)
sil.amorph
calcite
fluorite
6 talc
quartz
wollastonite

4
log (Q/K)

-2

40 80 120 160 200

Temperature (°C)

FIGURE 11: Mineral equilibrium plots for Öxarfjördur

5. CONCLUSIONS

 The geothermal waters of Kibiro, Panyimur and Öxarfjördur are classified as sodium-chloride-
bicarbonate waters. The geothermal samples from Öxarfjördur are more mature than the water
samples from Kibiro and Panyimur.
 From geothermometer temperature prediction and interpretation, quartz and Na/K
geothermometers are applicable in a very wide range of geological environments, for example
from rhyolitic, classic sediments to a basaltic geological setting.
 The mineral saturation plots for Kibiro and Panyimur showed that all waters had almost reached
equilibration with most minerals, but in Öxarfjördur they have not yet attained equilibrium.
 The chemical composition of the water samples from Kibiro, Panyimur and Öxarfjördur, with
different sedimentary geological settings, is much affected by the strata through which these fluids
flow.
 Despite the fact that the Öxarfjördur geothermal system is located in a sedimentary graben, there
is considerable water- rock interaction at depth beneath the sedimentary system compared to the
situation at Kibiro and Panyimur.
Lajwe 226 Report 12

40
Lw

20 Continental African
rainline

Kenya rainline
0 (Clarke et al. 1990)

-20

Öxarfjördur
-40
Kibiro
Panyimur

-60
World meteoric line
(Craig 1961)

-80

-15 -10 -5 0 5 10

O (‰)

FIGURE 12: Plot of δ2H vs δ18O for samples from Kibiro, Panyimur and Öxarfjördur

5.1 Recommendations and future research work

 More geochemical exploration work needs to be carried out in geothermal systems occurring in
sedimentary geological settings for better interpretation of the geochemistry of these systems.

 Combined geochemical, geological and geophysical studies are required for better understanding
of geothermal systems in sedimentary settings.

 A distribution map for deuterium ratios and full geological maps of the geothermal systems in
Kibiro and Panyimur are required.

 The geochemistry of the hot spring discharges in Kibiro and Panyimur may be controlled by the
interaction of the fluids with the crystalline basement, and is characterized by low B/Cl
conservative element ratios and a high pH. This needs more research.
Report 12 227 Lajwe

 From this study, the application of geothermometry techniques clearly differ from one place to
another due to the different environmental conditions, such as physical processes like conductive
cooling and mixing with cold water. Boiling during upflow may affect possible water-rock
equilibria in the deep reservoir, and this can result in unreliable temperature predictions.
Furthermore, it is very useful to employ different geothermometers and methods for studying
equilibrium states in order to check each and to exclude unreliable samples and non-applicable
geothermometers.

ACKNOWLEDGEMENTS

To the United Nation University Geothermal Training Program together with the Government of
Iceland, I hereby express my gratitude for the opportunity granted to me for the participation in this six
months Geothermal Training Programme. I am indebted to former Director, Dr. Ingvar B. Fridleifsson
and the new Director, Mr. Lúdvík S. Georgsson, for UNU, for their undivided dedication to the success
of this training.

Special thanks to my supervisors, Dr. Halldór Ármannsson and Mr. Dadi Thorbjörnsson for their
assistance and critical advice including provision of data, both analytical and literary
information during the preparation of this report. I would like to thank all the ISOR staff, especially
those who gave lectures in the different fields that build up the geothermal training programme.

I am grateful to the following: Ms. Thórhildur Ísberg, Mr Ingimar G. Haraldsson, Ms. Málfrídur
Ómarsdóttir and Mr. Markús A.G. Wilde for the commitment that they offered us during this training
period. And not forgetting all the other UNU staff for their coordination of the training activities and for
guidance that led to the successful completion of this training.

To my late mum and my son (Obiro Anthony Claret), I dedicate this small piece of my work to you.
And I thank the Department of Geological Survey and Mines for allowing me to come for this training.

Finally, I would like to give thanks to the Almighty God for his generous support and encouragement
in all that I do.

NOMENCLATURE

EARS = East Africa Rift System

ISOR = Iceland GeoSurvey
MEMD = Ministry of Energy and Mineral Development
CARL = Continental African rain line
WML = World meteoric line
KRL = Kenyan rain line
GSMD = Geological Survey and Mines Department
UNU = United Nations University
TDS = Total dissolved solids
OS = Orkustofnun
ICEIDA = Icelandic International Development Agency
IAEA = International Atomic Energy Agency
LOC = Location
KIB = Kibiro
ÖXA = Öxarfjördur
PAN = Panyimur
Lajwe 228 Report 12

AAS = Atomic absorption spectrophotometry

IC = Ion chromatography
ISE = Ion specific electrode
FAAS = Flameless atomic absorption spectrophotometry
GC = Gas chromatography
USI = University of Iceland science institute, Reykjavik
n.c. = non-condensable
MS = Mass spectrometry
T = Temperature in (oC)
Tqz (1-8) = Temperature obtained using quartz geothermometer
Tch (1-2) = Temperature obtained using chalcedony geothermometer
TK-Mg, TNa-K, TNa-K and Na-K-Ca are different cation geothermometers used to obtain
the temperature

REFERENCES

Afeworki O., M., 2010: Analysis of temperature and pressure characteristics of the Hverahlíd
geothermal field in the Hengill geothermal system, SW-Iceland.Report 20 in: Geothermal training in
Iceland 2010. UNU-GTP, Iceland, 1-28.

Ármannsson, H., 2009: Application of geochemical methods in geothermal exploration. Proceedings

of the “Short Course IV on Exploration for Geothermal Resources” organized by UNU-GPT, KenGen
and GDC, at Lake Naivasha, Kenya, 8 pp.

Ármannsson, H., 2010: Investigations of geothermal areas in Uganda other than Katwe-Kikorongo,
Buranga and Kibiro. ÍSOR – Iceland GeoSurvey, prepared for ICEIDA, report ÍSOR-2010/003, 1-17
pp.

Ármannsson, H., 2013: Fluid chemistry and utilization. UNU-GTP, Iceland, unpublished lecture notes,
22 pp.

Ármannsson, H. and Ólafsson, M., 2006: Collection of geothermal fluids for chemical analysis. ÍSOR
– Iceland GeoSurvey, Reykjavík, report ISOR-2006/016, 17 pp.

Árnason, K., and Gíslason, G., 2009: Geothermal surface exploration. Paper presented at “Short
Course on Surface Exploration for Geothermal Resources”, organized by UNU-GTP and LaGeo, Santa
Tecla, El Salvador, 7 pp.

Arnórsson, S. (ed.), 2000a: Isotopic and chemical techniques in geothermal exploration, development
and use. Sampling methods, data handling, and interpretation. International Atomic Energy Agency,
Vienna, 351 pp.

Arnórsson, S., 2000b: The quartz and Na/K geothermometers. I. New thermodynamic calibration.
Proceedings of the World Geothermal Congress 2000, Kyushu-Tohoku, Japan, 929-934.

Arnórsson, S., and Bjarnason, J.Ö., 2007: Fluid-fluid interaction in geothermal systems. Reviews in
Mineralogy & Geochemistry, 65, 229-312.

Arnórsson, S., and Svavarsson, H., 1985: Application of chemical geothermometry to geothermal
exploration and development. Geoth. Res. Council, Transactions, 9-1, 293-298.
Report 12 229 Lajwe

Arnórsson, S. (ed.), D’Amore, F., and Gerardo-Abaya, J., 2000: Isotopic and geochemical techniques
in geothermal exploration, development and use: Sampling methods, data handling, and interpretation.
International Atomic Energy Agency, Vienna, 351 pp.

Arnórsson, S., Gunnarsson, I., Stefánsson, A., Andrésdóttir, A., and Sveinbjörnsdóttir, Á.E., 2002:
Major element chemistry of surface- and ground waters in basaltic terrain, N-Iceland. I. Primary mineral
saturation. Geochim. Cosmochim. Acta, 66, 4015-4046.

Arnórsson, S., Gunnlaugsson, E., and Svavarsson, H., 1983: The chemistry of geothermal waters in
Iceland. III. Chemical geothermometry in geothermal investigations. Geochim. Cosmochim. Acta, 47,
567-577.

Atekwana, E.A., Hogan, J.P., Kampunzu, A.B., and Modisi, M.P., 2004: Early structural evolution of
the nascent Okavango rift zone, NW Botswana. Proceedings of the East African Rift System:
Geodynamics, Resources, and Environment Conference, Addis Ababa, 12-16.

Bahati, G., 2007: Status of geothermal energy exploration in Uganda. Paper presented at “Short course
II on surface exploration for geothermal resources, Naivasha Kenya”, organized by UNU-GTP,
Iceland, and KenGen, Naivasha, Kenya, CD SC-05, 10 pp.

Bahati, G., 2012: Status of geothermal exploration and development in Uganda. Paper presented at
“Short Course VII on Exploration for Geothermal Resources, organized by UNU-GTP, GDC and
KenGen, at Lake Bogoria and Lake Naivasha, Kenya, 14 pp.

Bertrami, R., Cioni, R., Corazza, E., D´Amore, F., and Marini, L., 1985: Carbon dioxide in geothermal
gases. Reservoir temperature calculations at Larderello (Italy). Geothermal Research Council,
Transactions, 9, 299-303.

Bjarnason, J.Ö., 2010: The speciation program WATCH, Version 2.4. The Iceland Water Chemistry
Group, Reykjavík.

Björnsson, A., Eysteinsson, H., and Beblo, M., 2005: Crustal formation and magma genesis beneath
Iceland: magnetotelluric constraints. In: Foulger, G.R., Natland, J.H., Presnall, D.C., Anderson, D.L.
(eds.), Plates, plumes and paradigms. Geological Society of America, Spec. Pap. 388,665–686.

Bruhn, D., Manzella, A., Vuataz, F., Faulds, J., Moeck, I., and Erbas, K., 2011: Exploration methods.
In: Huenges, E. (ed.), Geothermal energy systems. Wiley-VCH Verlag GmbH, Weinheim, 465 pp.

Craig, H., 1963: The isotopic geochemistry of water and carbon in geothermal areas. In: Tongiorgi, E.
(ed.), Nuclear geology on geothermal areas. Consiglio Nazionale delle Ricerche, Laboratorio di
Geologia Nucleare, Pisa, 17-53.

D’Amore, F., and Arnórsson, S., 2000: Geothermometry. In: Arnórsson, S. (ed.), Isotopic and chemical
techniques in geothermal exploration, development and use. Sampling methods, data handling,
interpretation. International Atomic Energy Agency, Vienna, 152-199.

D’Amore, F., Krajca, J., Michard, G., Nuti, S., Ólafsson, M., Paces, T., Zhaoli, S., Wei, T., and Zhifei,
Z., 1991: Fluid sampling for geothermal prospecting. UNITAR/UNDP Centre on Small Energy
Resources, Rome, 93-117 pp.

Data, G., and Bahati, G., 2003: The chemistry of geothermal waters from areas outside the active
volcanic belt, SW-Uganda. Proceedings of the 2nd KenGen Conference, Nairobi, Kenya, April 2003, 24-
29.
Lajwe 230 Report 12

Driesner, T, 2013: Geothermal activities in the Main Ethiopian Rift: Hydrogeochemical characterization
of geothermal waters and geothermometry applications (Dofan-Fantale, Gergede-Sodere, Aluto-
Langano). Geothermics, 47, 1-12.

Ebinger, C.J., Deino, A.L., Drake, R.E., and Tesha, A.L., 1989: Chronology of volcanism and rift basin
propagation: Rungwe volcanic province, East Africa. J. Geophys. Res., 94-B11:15, 785-803.

EDICON, 1984: Aeromagnetic interpretation of Lake Albert/Edward portion of the Western Rift Valley.
EDICON, Inc., Denver, Colorado, unpublished report.

Ellis, A.J., and Mahon, W.A.J., 1977: Chemistry and geothermal systems. Academic Press, New York,
392 pp.

Fournier, R.O., 1977: Chemical geothermometers and mixing model for geothermal systems.
Geothermics, 5, 41-50.

Fournier, R.O., 1979: A revised equation for Na-K geothermometer. Geoth. Res. Council, Trans., 3,
221-224.

Fournier, R.O., and Potter, R.W., 1982: An equation correlating the solubility of quartz in water from
25° to 900°C at pressures up to 10,000 bars. Geochim. Cosmochim Acta, 46, 1969-1973.

Fournier, R.O., and Truesdell, A.H., 1973: An empirical Na-K-Ca geothermometer for natural waters.
Geochim. Cosmochim. Acta, 37, 1255-1275.

Georgsson, L.S., Árnason, K., and Karlsdóttir, R., 1993: Resistivity sounding in high-temperature areas
in Iceland, with examples from Öxarfjördur, N-Iceland and Brennisteinsfjöll, SW-Iceland. Proceedings
of the 14thPNOC-EDC Geothermal Conference, Manila, 9 pp.

Georgsson, L.S., Fridleifsson, G.Ó., Ólafsson, M., and Flóvenz, Ó.G., 2000: The geothermal
exploration of the Öxarfjördur high-temperature area, NE-Iceland. Proceedings of the World
Geothermal Congress 2000, Kyushu-Tohoku, Japan, 1157-1162.

Giggenbach, W.F., 1980: Geothermal gas equilibria. Geochim. Cosmochim. Acta, 48, 2693-2711.

Giggenbach, W.F., 1984: Mass transfer in hydrothermal alteration systems – A conceptual approach.
Geochim. Cosmochim. Acta, 48, 2693–2711.

Giggenbach, W.F., 1988: Geothermal solute equilibria. Derivation of Na-K-Mg-Ca geoindicators.

Geochim. Cosmochim. Acta, 52, 2749-2765.

Giggenbach, W.F., 1991: Chemical techniques in geothermal exploration. In: D’Amore, F.

(coordinator), Application of geochemistry in geothermal reservoir development. UNITAR/UNDP
publication, Rome, 119-144.

Giggenbach, W.F., 1992: Isotope shift in waters from geothermal and volcanic systems along
convergent plate boundaries and their origin. Earth and Planetary Sci. Lett., 113, 495-510.

Giggenbach, W.F., Sheppard, D.S., Robinson, B.W., Stewart, M.K., and Lyon, G.M., 1994:
Geochemical structure and position of the Waiotapu geothermal field, New Zealand. Geothermics, 23,
599-644.

Gíslason, G., Ngobi, G., Isabirye, E., and Tumwebaze, S. 1994: An Inventory of three Geothermal Areas
in West and Southwest Uganda. Prepared by the United Nations for a Project of the UNDP.
Report 12 231 Lajwe

Gíslason, G., Ívarsson, G., Gunnlaugsson, E., Hjartarson, A., Björnsson, G., and Steingrímsson, B.,
2005: Production monitoring as a tool for field development; A case history from the Nesjavellir field,
Iceland. Proceedings of the World Geothermal Congress 2005, Antalya, Turkey, 9 pp.

GSMD, 2008: Exploration report by the geological team of DGSM. DGSM, Entebbe, unpublished
report.

Henley, R.W. and Ellis, A.J. 1983: Geothermal systems ancient and modern: a geochemical review.
Earth Science and Reviews, 19, 1-50.

Henley, R.W., Truesdell, A.H., and Barton, P.B. Jr., 1984: Fluid-mineral equilibrium in hydrothermal
systems. Society of Economic Geologists, Reviews in Economic Geology, 1, 267 pp.

Hjartarson, A., Axelsson, G., and Xu Y., 2005: Production potential assessment of the low-temperature
sedimentary geothermal reservoir in Lishuiqiao, Beijing, P.R. of China, based on a numerical simulation
study. Proceedings of the World Geothermal Congress 2005, Antalya, Turkey, 12 pp.

Jónsson, G., Kristjánsson, L., and Sverrisson M., 1991: Magnetic surveys of Iceland. Tectonophysics
189, 229-247.

Kato, V., 2003: Chemistry of thermal springs in Uganda. Geological Survey and Mines Department,
Entebbe, report KVK/14, 2 pp.

Karingithi, C.W., 2009: Chemical geothermometers for geothermal exploration, Paper presented at
Short Course IV on Exploration for Geothermal Resources, organized by UNU-GTP, KenGen and GDC,
at Lake Naivasha, Kenya, 12 pp.

Marini, L., and Cioni, R., 1985: A chloride method for the determination of the enthalpy of steam/water
mixtures discharged from geothermal wells. Geothermics, 14, 29-34.

Mosley, P.N., 1993. Geological evolution of the late Proterozoic “Mozambique belt” of Kenya.
Tectonophysics, 221, 223-250.

Natukunda.J.F., 2013: Exploration report on the follow up of the geology of Panyimur by the geological
team of DGSM.DGSM, Entebbe, unpublished report.

Noda, T., and Shimada K., 1993: Water mixing model calculation for evaluation of deep geothermal
water. Geothermics, 22, 165-180.

Ólafsson, M., Fridleifsson, G.Ó., Eiríksson, J., Sigvaldasson, H., and Ármannsson, H., 1993: On the
origin of organic gas in Öxarfjördur, NE-Iceland. Orkustofnun, Iceland, report OS-93015/JHD-05, 76
pp.

Reed, M.H., 1991: Computer modelling of chemical processes in geothermal systems: Examples of
boiling, mixing and water-rock reaction. In: D´Amore, F. (coordinator), Applications of geochemistry
in geothermal reservoir development. UNITAR/UNDP publication, Rome, 275-297.

Rodríguez, A., 2011: Water-rock interaction of silicic rocks: an experimental and geochemical
modelling study. University of Iceland, MSc thesis, UNU-GTP, Iceland, report 1, 44 pp.

Rybach, L., and Muffler, L.J.P. (eds.), 1981: Geothermal systems, principles and case histories. John
Wiley & Sons, Ltd., Chichester, 359 pp.
Lajwe 232 Report 12

Stefánsson, A., Gíslason, S.R., and Arnórsson, S., 2000: Dissolution of primary minerals in natural
waters II. Mineral saturation state. Chemical Geology, 172, 251-276.

Stefánsson, A., Gunnarsson, I., and Giroud, N., 2007: New methods for the direct determination of
dissolved inorganic, organic and total carbon in natural waters by Reagent-Free™ IonChromatography
and inductively coupled plasma atomic emission spectrometry. Analytical ChimicaActa, 582, 69-72.

Truesdell, A.H., 1991: Origin of acid fluids in geothermal reservoirs. Geotherm. Res. Counc.,Trans.,
15, 289-296.

Truesdell, A.H., and Fournier, R.O., 1976: Calculations of deep temperatures in geothermal systems
from the chemistry of boiling spring waters of mixed origin. Proceedings of 2nd U.N. Symposium on the
Development and Use of Geothermal Resources, San Francisco, 1,837-844.

Truesdell, A.H., and Fournier, R.O., 1977: Procedure for estimating the temperature of a hot water
component in a mixed water using a plot of dissolved silica vs. enthalpy. U.S. Geol. Survey J. Res., 5,
49-52.

Verma, S.P., and Santayo, E., 1997: New improved equations for Na/K, Na/Li and SiO2
geothermometers by outlier detection and rejection. J. Volcanol. Geotherm. Res., 79, 9-23.

Wood, J., and Guth, A., 2012: East Africa's Great Rift Valley, a complex rift system. Geology.com,
webpage: geology.com/articles/east-africa-rift.shtml.

Yohannes B., E. 2004: Geochemical interpretation of thermal water and gas samples from Krýsuvík,
Iceland and Alid, Eritrea. Report 18 in: Geothermal training in Iceland 2004. UNU-GTP, Iceland, 403-
438.
Report 12 233 Lajwe

APPENDIX I: Maps of Kibiro and Panyimur (source DGSM, 2013)

Lajwe 234 Report 12

APPENDIX II: Comparison plots of calculated temperature results obtained from sixteen
different equations of geothermometers for Öxarfjördur, Kibiro and Panyimur