University of Southern California Dissertations and Theses

Geothermal resource assessment and reservoir modeling with an application to Kavaklidere geothermal field, Turkey

(USC Thesis Other)

Geothermal resource assessment and reservoir modeling with an application to Kavaklidere geothermal field, Turkey

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Geothermal Resource Assessment and Reservoir Modeling
with an application to Kavaklidere Geothermal Field, Turkey

By

Aysegul BALIKCIOGLU

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
in Partial Fulfillment of the
Requirements for the Degree of
MASTER OF SCIENCE
(PETROLEUM ENGINEERING)

May 2018

Copyright 2018 Aysegul Balikcioglu
ii

The Thesis Committee for Aysegul Balikcioglu
Certifies that this is the approved version of the following thesis:

Geothermal Resource Assessment and Reservoir Modeling
with an application to Kavaklidere Geothermal Field, Turkey

Approved by Committee:

__________________________________________
Professor Fred Aminzadeh (Chair)

__________________________________________
Professor Iraj Ershaghi

__________________________________________
Professor Adam Rose

iii

Dedication

To my beloved late grandfather, A. Yasar Balikcioglu
To my beautiful family

iv

ABSTRACT
This thesis presents the background of conventional geothermal energy and enhanced geothermal
systems (EGS), comparison of volumetric resource assessments and a reservoir model with an
application to the Kavaklidere geothermal field, Turkey. The significant advantages of geothermal energy
resources are reducing the local contributions to global climate change and making use of indigenous
resources to deal with the foreign source dependency. Turkey is rich in geothermal energy and has ever-
growing geothermal developments over the last decade. The Kavaklidere field, covering a 126 km2 area
in Manisa province, Turkey, is the current target for both conventional geothermal and EGS
(enhanced/engineered geothermal system) field development. The field holds Turkey’s highest
geothermal resource temperature, which was recorded as 287 C at 2750 m depth by the General
Directorate of Mineral Research and Exploration (MTA) in 2011.
Geothermal reserve estimation, in general, is the critical part of reservoir engineering to have some
predictions about the producibility of the reserve. The USGS (United States Geologic Survey) (1978) and
MIT (Massachusetts Institute of Technology) (2006) volumetric ‘heat in place’ methods together with
Monte Carlo simulation are widely used for assessing the electrical capacity of a geothermal reservoir.
Garg and Combs (2015) show that these methods overestimate the results because they use arbitrarily
chosen reference temperature and thermal power conversion efficiency values for their estimations
without taking the second law of thermodynamics (exergy) and the installed power cycle system into
consideration. Garg and Combs (2015) proposed a new method, by reformulating USGS volumetric
method, that recoverable heat must be derived by considering specific power cycles, i.e., single-flash or
binary.
The EGS technical and economical electricity production potential of Turkey (3–5 km) is expected to be
25 GWe in the next 25 years (Mertoglu et.al. 2015). The latest development in EGS technology is multi-
zonal stimulation, which is the ability to create multiple permeable zones by hydroshearing in a single
well and increase the amount of produced energy from the well by a factor of two or more (Petty et al.,
2013).
In this thesis, first of all, volumetric estimations of power generation potential based on USGS, MIT and
Garg and Combs (2015) have been evaluated by Monte Carlo simulation with different development
scenarios for the Kavaklidere geothermal field. Applying by Garg and Combs (2015) method, the electric
power capacity of the field is estimated as 182.3 MWe with 90% probability for the conventional
v

hydrothermal case, and as 54.8 MWe with 90% probability for EGS case. Sensitivity analysis for both
cases shows that recovery factor, reservoir area, and thickness have the biggest impact on net power
output among the other input parameters.
Secondly, the existing conditions, such as downhole temperature and pressure and rock density, are
incorporated by using the reservoir simulator TOUGH2 via the graphical interface PetraSim. The
hydrothermal reservoir is shown for 5-spot well patterns in order to estimate the thermal energy from
the production well and demonstrate the temperature change of the reservoir for a 30-year period. The
EGS reservoir with the assumptions and different stimulation scenarios is shown for triplet well pattern
to determine the thermal energy output of the production wells. Also, levelized cost of electricity (LCOE)
is calculated by using GETEM (Geothermal Electricity Technology Evaluation Model) for both
hydrothermal and EGS.

vi

ACKNOWLEDGMENTS

Above all, I give thanks to Allah for his countless blessings.

Firstly, I would like to express my sincerest gratitude to my advisor, Professor Fred Aminzadeh, Director
of Global Energy Network (GEN) and Center for Geothermal Studies (CGS) at the University of Southern
California, for his guidance and help during my research. Getting involved in his great research group
was a privilege for me, and his patient, kindness and extensive knowledge helped me to complete this
study.

I need to give my special thanks to Professor Iraj Ershaghi, Director of the Petroleum Engineering
Program, for his encouragement and guidance during my Master’s study at USC, and for serving as a
member of my thesis committee. A special thanks to Professor Adam Rose, who was another member of
my Master’s thesis committee for his constructive comments.

I would like to express my appreciation to Musa Burcak for providing data about the field in Turkey, and
Joe Iovenitti for sharing his knowledge of Enhanced Geothermal Systems.

I would like to acknowledge my sponsor, The Ministry of National Education, the Republic of Turkey for
giving me the opportunity to pursue a master’s degree and supporting me financially throughout the
process.

I have met lots of beautiful people from all over the world, and my life has always blessed with good
friends, classmates, roommates, and colleagues. My sincere thanks go to all of them.

Lastly, my heartfelt thanks go to my family, my loving mother, Atike, my wonderful father, Seyfullah, my
three beautiful sisters, Nur, Hilal, and Berra, and my beloved brother, Murat, for their unconditional love
and continuous support that always carry me.

vii

TABLE OF CONTENT
ABSTRACT …………………………………………………………………………………………………………………………………iv
ACKNOWLEDGMENTS ………………………………………………………………………………………………………………vi
CHAPTER 1: INTRODUCTION
1.1 Thesis Motivation ………………………………………………………………………………………………………………..1
1.2 Problem Statement and Objectives ……………………………………………………………………………………..7
1.3 Outline of the thesis …………………………………………………………………………………………………………….8

CHAPTER 2: LITERATURE REVIEW
2.1 Geothermal Energy Basics ……………………………………………………………………………………………………9
2.1.1 Geothermal Energy in the World ………………………………………………………………………..12
2.1.2 Geothermal Energy in Turkey ……………………………………………………………………………..15
2.1.3 Geothermal Power Plants ……………………………………………………………………………………19
2.1.4 Enhanced Geothermal Systems – EGS …………………………………………………………………21
2.1.4.1 Hydraulic Stimulation and Hydro-shearing ………………………………………….23
2.1.4.2 Challenges of EGS ………………………………………………………………………………..25
2.2 Kavaklidere Geothermal Field …………………………………………………………………………………………….26
2.2.1 Geological Outlook and Tectonic Setting…………………..…………………………………………27
2.2.2 Hydrogeological Outlook …………………………………………………………………………………….30
2.2.3 Drilling History and Geothermal System ……………………………………………………………..31

CHAPTER 3: METHODOLOGY
3.1 Comparison of Volumetric ‘Heat in Place’ Estimation Methods………..…………………………………33
3.1.1 Garg and Combs Method ……………………………………………………………………………………38
3.2 Probabilistic Assessment …………………………………………………………………………………………………….42
3.2.1 Monte Carlo Simulation ………………………………………………………………………………………42
3.2.2 Input Parameters and Scenarios …………………………………………………………………………45
3.2.3 Results and Discussion ………………………………………………………………………………………..49
3.3 TOUGH2-PETRASIM Reservoir Simulation …………………………………………………………………………..52
3.3.1 Input Parameters and Scenarios for Conventional Reservoir ………………………………54
3.3.2 Results and Discussion …………………………..…………………………..……………………………….55
3.3.3 Input Parameters and Scenarios for EGS …………………………..………………………………..58
3.3.4 Results and Discussion …………………………..…………………………..……………………………….60
3.4 GETEM Simulation for Economic Evaluation …………………………..…………………………..……………..64

CHAPTER 4: CONCLUSION AND RECOMMENDATIONS …………………………..………………………………………66

REFERENCES …………………………..…………………………..…………………………..………………………………………………….69
Appendix A-B …………………………..…………………………..…………………………..…………………………..……………………75

viii

LIST OF TABLES
Table 1.1: Installed capacity of power plants based on energy source in Turkey at the end of 2016
and 2017, MW (TEIAS,2017) .…………………………………………………………………………………………………………..….….2
Table 1.2: Turkey's Vision 2023 targets for the energy sector (INVEST, 2017) …………………………………………6
Table 2.1: Licensed geothermal electric power plants in Turkey by the end of December 2017
(EPDK, 2017) …………………………………………………………………………………………………………………………………………18
Table 2.2: Wells in Kavaklidere Geothermal Field (Burcak,2011; MASPO,2013; Ozdemir et al.,2016) ……31
Table 3.1: Input parameters for normalized work calculations from USGS, MIT and
Garg&Combs Flash Methods ………………….…………………………………………………………………………………………..40
Table 3.2: Normalized available work (WA)n for USGS, MIT, Garg and Combs single flash power cycle
as a function of resource and reference temperatures ………………………………………………………………….40
Table 3.3: Classification of input parameters of PW, MWe, concerning uncertainty (Onur, 2015) ……..…43
Table 3.4: Rock density, porosity and specific heat capacity values for Kavaklidere geothermal field
lithologies (*Gurel et al., 2016, Crain, P. Eng. Crain’s Petrophysical Handbook, Miao, S. Q., Li, H. P., &
Chen, G.,2014) ………………………………………………………………………………………………………………………………………45
Table 3.5: Input parameters for resource assessment of Kavaklidere-Hydrothermal and
Kavaklidere-EGS by using Garg and Combs flash method with Monte Carlo simulation …….………………….47
Table 3.6: Reservoir parameters for Kavaklidere from Basel, 2010 ……………………………………………………….48
Table 3.7: The results of volumetric probabilistic estimations of geothermal reserves by Monte Carlo
Simulation Garg and Combs,2015, USGS and MIT methods for Kavaklidere geothermal field ……………….52
Table 3.8: Simulation input parameters for Kavaklidere-hydrothermal …………………………………………………55
Table 3.9: Input parameters for different simulation models of Kavaklidere-EGS …………………………………60
Table 3.10: Input parameters for GETEM simulation …………………………………………………………………………….65

ix

LIST OF FIGURES
Figure 1.1: Turkey's GDP (PPP) in current international billion US$ change (Wordbank,2016), and Net
electricity consumption between 1990 and 2016 (TUIK, 2016) ……………………………………………………………….2
Figure 1.2: Turkey petroleum and other liquids consumption and production (EIA, 2017) …………………….3
Figure 1.3: Turkey crude and condensate supply by source,2015 (EIA, 2017) …………………………………………3
Figure 1.4: Turkey natural gas consumption and production (EIA, 2017) ………………………………………………..4
Figure 1.5: Turkey’s natural gas supply by source,2015 (EIA, 2017) ………………………………………………………..4
Figure 1.6: CO2 emissions (kt) in the World and Turkey, 1960-2014 (Worldbank, 2017) …………………………5
Figure 1.7: CO2 emissions (metric tons per capita) in the World and Turkey, 1960-2014
(Worldbank, 2017) ……………………………………………………………………………………………………………………..5
Figure 2.1: A Schematic representation of an ideal geothermal system (Liu et al.,2013) ………………….10
Figure 2.2: McKelvey diagram representing a geothermal resource and reserve terminology in the
context of geologic assurance and economic viability (Modified from Muffler and Cataldi (1978)) .……..11
Figure 2.3: Example classifications of geothermal resources by temperature (Celsius)
(Williams et al., 2011) ………………………………………………………………………………………………………………12
Figure 2.4: Geothermal global capacity additions in 2015, by country (WEC,2016) ………………………………13
Figure 2.5: Top countries that utilize the most direct geothermal heat in 2015 (WEC,2016) ………………..14
Figure 2.6: Geothermal direct applications worldwide in 2015, distributed by percentage of total
installed capacity (MWt) (Lund and Boyd, 2015) …………………………………………………………………………….…….14
Figure 2.7: Top 10 geothermal countries based on installed power generation capacity, August 2017
(Source: ThinkGeoEnergy) ……………………………………………………………………………………………………………….15
Figure 2.8: Distribution map of geothermal resources in Turkey concerning plate tectonics (MTA)………16
Figure 2.9: Map of temperature pattern of geothermal resources in Turkey and their
utilization ways (MTA) ……………………………………………………………………………………………………………..16
Figure 2.10: Location map of the main grabens in western Turkey (Yilmaz and Gelisli, 2003) ………….……17
Figure 2.11: Illustration of Single Flash Power Plant (left) and Double Flash Power Plant (right)
(Kagel, 2008) ……………………………………………………………………………………………………………………………20
Figure 2.12: Illustration of Binary Power Plant (Kagel, 2008) …………………………………………………………………20
Figure 2.13: Schematic of an EGS system (MIT, 2006) …………………………………………………………………………..22
Figure 2.14: Types of Geothermal Resources and their Risk/Reward Profile (modified after
Nathwani,2011 and White et al.,2014) ……………………………………………………………………………………………….…23
Figure 2.15: (a) Hydraulic fracturing open new or pre-existing tensile fractures from a small packed-
interval, where the fluid is injected at a pressure higher than the minimum principal stress σ3. (b) Hydro-
shearing aims to reactivate pre-existing natural fractures favorably oriented for shearing, with a fluid
pressure remaining lower than σ3 (Gischig and Preisig, 2015)……………………………………………………………….24
Figure 2.16: Location map of the Kavaklıdere geothermal field (Ozdemir et al., 2016) …………………………27
Figure 2.17: Simplified geological map of the Kavaklıdere geothermal field (Ozdemir et al., 2016) ………28
Figure 2.18: Generalized stratigraphic vertical section of the Kavaklıdere geothermal field
(modified after Ozdemir et al. 2016 and MASPO 2013) ……………………………………………………………29
Figure 2.19: Geological cross-section of the detachment fault and high angle normal faults between
Horzumalayaka and Kavaklıdere (Ozdemir et al.,2016) …………………………………………………………………………30
x

Figure 2.20: Schematic view of scissor faults, faults system, fault set and geothermal reservoirs in the
Kavaklıdere geothermal field (Ozdemir et al.,2016) ………………………………………………………………………………30
Figure 3.1 Schematic of the Monte Carlo uncertainty analysis (modified from Williams et al., 2008) …..44
Figure 3.2: Histogram and Cumulative Probability graph for electrical power generation (MWe) of
Kavaklidere-hydrothermal …………………………………………………………………………………………………………………….51
Figure 3.3: Histogram and Cumulative Probability graph for electrical power generation (MWe) of
Kavaklidere-EGS ……………………………………………………………………………………………………………………………………51
Figure 3.4: Five-spot reservoir configuration and the reservoir model used for
Kavaklidere-hydrothermal ………………………………………………………………………………………………….……55
Figure 3.5: Simulation Results of base case scenario for Reservoir Temperature Distribution after
a) 1-year b) 5-year c) 10-year d) 15-year e) 20-year f) 25-year .………………………………………………..56
Figure 3.6: Simulation results of base case scenario for reservoir temperature distribution at the
end of 30 year ………………………………………………………………………………………………………………………….57
Figure 3.7: Comparison of temperature profiles for two scenarios ……………………………………………….………57
Figure 3.8: Comparison of estimated thermal energy from production well for two scenarios …………….58
Figure 3.9: Reservoir configuration and the reservoir models used for Kavaklidere-EGS ……………………..60
Figure 3.10: Simulation results of five scenarios for reservoir temperature distribution at
the end of 30 year ……………………………………………………………………………………………………..…………….61
Figure 3.11: Comparison of estimated thermal energy from one production well for five scenarios …….64

xi

LIST OF ABBREVIATIONS
ANN – Artificial Neural Network
DOE – U.S. Department of Energy
EGS – Enhanced Geothermal Systems
EIA – U.S. Energy Information Administration
EOS – Equation of State
EPDK - Republic of Turkey Energy Market Regulatory
FIT – Feed-in Tariff
FZI – Fracture Zone Identifier
GDP – Gross Domestic Product
GETEM - Geothermal Electricity Technology Evaluation Model
HDR – Hot Dry Rock
IFDM – Integral Finite Difference Method
LCOE – Levelized Cost of Electricity
MIT - Massachusetts Institute of Technology
MTA - General Directorate of Mineral Research and Exploration
PPP - Purchasing Power Parity
TEIAS - Turkish Electricity Transmission Company
THOUGH – Transport of Unsaturated Groundwater and Heat
TUIK – Turkish Statistical Institute
TZIMs – Thermally-Degradable Zonal Isolation Materials
USC – University of Southern California
USGS - United States Geologic Survey
WEC – World Energy Council
1

CHAPTER 1
INTRODUCTION

1.1 THESIS MOTIVATION
Turkey is a developed country and occupies a unique geographic position, lying partly in Asia and partly
in Europe. Turkey's gross domestic product (GDP) in purchasing power parity (PPP) was estimated at
1.941 trillion US$ in 2016. Data about Turkey's GDP (PPP) change between 1990 and 2016 is shown in
Fig. 1.1 (Wordbank,2016). Historically, economic growth and energy consumption have been highly
correlated for developing and newly developed countries. During the last 25 years, Turkey's energy
consumption increased parallel to its economic growth (Melikoglu,2017). From 2005 to 2015, Turkey’s
4.4% per annum growth rate in primary energy consumption, is the highest in Europe (BP,2017). Net
electricity consumption in Turkey between 1990 and 2016 is shown in Fig. 1.1 (TUIK,2016) as well. As a
result, Turkey's energy infrastructure grew, and power plant capacities increased concurrently. The
installed capacity of Turkey's power plants by energy source at the end of 2016 and 2017 are shown in
Table 1.1 (TEIAS,2017).
However, increased energy consumption parallel to Turkey's economic growth ironically had an adverse
effect on the country's economy, specifically to the country's foreign trade. This is because Turkey is a
net energy importer and most of the country's energy demand is supplied from imported energy
sources, mostly coal and natural gas. Turkey also imports most of the oil it consumes; however, it is
primarily used by the petrochemical industry to produce transportation fuels and petrochemicals. Here
it should be emphasized that this problem is not unique to Turkey. Globally, three types of fossil fuels:
coal, natural gas, and oil provide a majority of the global total primary energy supply (TPES), and the
share of renewable energy sources is still small compared to these fossil fuels (Melikoglu,2017).
2

Figure 1.1: Turkey's GDP (PPP) in current international billion US$ change (Wordbank,2016), and Net
electricity consumption between 1990 and 2016 (TUIK, 2016)

Table 1.1: Installed capacity of power plants based on energy source in Turkey at the end of 2016 and
2017, MW (TEIAS,2017)
2016 2017
Fuel Type Installed
Capacity,
MW
Share,
%
Number of
Power
Plants
Installed
Capacity, MW
Share, % Number
of Power
Plants
Fuel oil + naphtha + diesel 368.7 0.5 14 303.6 0.4 12
Bituminous coal+ lignite + asphaltite 9842.4 12.5 29 9872.6 11.6 30
Imported coal 7473.9 9.5 10 8793.9 10.3 11
Natural gas + LNG 22156.1 28.2 240 23063.7 27.1 243
Renew. + waste + semi waste + pyrolysis oil 467.4 0.6 82 575.1 0.7 98
Multiple fuels solid + liquid 667.1 0.8 23 682.9 0.8 22
Multiple fuels liquid + natural gas 3354.0 4.3 46 3433.6 4.0 47
Geothermal 820.9 1.0 31 1063.7 1.2 40
Hydro (dam type) 19558.6 24.9 116 19776.0 23.2 117
Hydro (river type) 7119.6 9.1 478 7489.7 8.8 501
Wind 5738.4 7.3 148 6482.2 7.6 161
Solar 12.9 0.0 2 17.9 0.0 3
Thermal (unlicensed) 82.1 0.1 33 201.1 0.2 67
Wind (unlicensed) 12.9 0.0 23 34.0 0.0 46
Hydro(unlicensed) 2.9 0.0 3 7.4 0.0 10
Solar (unlicensed) 819.6 1.0 1043 3402.8 4.0 3613
Total 78497.4 100 2321 85200.0 100 5021
0
50
100
150
200
250
0
500
1000
1500
2000
2500
Thousands
Time, Years
GDP, PPP (current international billion US$)
Billions
GDP
Electricity
Consumption
Electricty Consumption, GWh
3

Over the past decade, Turkey’s economy has expanded, and its petroleum and other liquids
consumption have increased. With limited domestic reserves, Turkey imports nearly all its oil
supplies. The country’s production is much less than what the country consumes each year (Fig.
1.2). In 2015, Turkey’s total liquid fuels consumption averaged about 860,000 b/d, and more than
90% of total liquid fuels came from imports. In 2015, most of Turkey’s crude oil imports came from
Iraq and Iran (Fig. 1.3), which supplied slightly more than 60% of the country’s crude oil (EIA,2017).

Figure 1.2: Turkey petroleum and other liquids consumption and production (EIA, 2017)

Figure 1.3: Turkey crude and condensate supply by source,2015 (EIA, 2017)
4

Turkey is increasingly dependent on natural gas imports because its domestic consumption,
especially in the electric power sector, has experienced significant growth. Turkey produces only a
small amount of natural gas, and total production amounted to 14 Bcf in 2015 (Fig. 1.4). In 2015,
Turkey imported 1.7 Tcf of natural gas, accounting for 99% of total natural gas supply, from two
countries Russia and Iran (Fig. 1.5). (EIA,2017)

Figure 1.4: Turkey natural gas consumption and production (EIA, 2017)

Figure 1.5: Turkey’s natural gas supply by source,2015 (EIA, 2017)
5

Globally and countrywide, greenhouse gas emissions mainly CO2 emissions increased immensely over
the last four decades (Fig. 1.6 and Fig. 1.7). Seeking the solutions to decrease greenhouse gas emissions
and the instability of fossil fuels prices encouraged many countries to increase their investment in
renewable energy sources.

Figure 1.6: CO2 emissions (kt) in the World and Turkey, 1960-2014 (Worldbank, 2017)

Figure 1.7: CO2 emissions (metric tons per capita) in the World and Turkey, 1960-2014 (Worldbank,
2017)

0
50
100
150
200
250
300
350
400
1960
1964
1968
1972
1976
1980
1984
1988
1992
1996
2000
2004
2008
2012
0
5000
10000
15000
20000
25000
30000
35000
40000
Turkey, Thousands
Year
CO2 emissions (kt)
World, Thousands
CO2 emissions (kt) World and Turkey, 1960-2014
World
Turkey
0
1
2
3
4
5
6
1950 1960 1970 1980 1990 2000 2010 2020
metric tons per capita
Year
CO2 emissions (metric tons per capita) 1960-2014
Turkey
World
6

In that context, the Turkish government set an ambitious target of producing 30% Turkey's energy
demand in the year 2023 (the 100
th
anniversary of the Turkish Republic) from renewable energy sources.
This is part of Turkey's Vision 2023 energy targets, details of which are given in Table 1.2 (INVEST,2017).
Table 1.2: Turkey's Vision 2023 targets for the energy sector (INVEST, 2017)

The main problem with renewable energy sources is baseload supply. Conventionally, the backbone of
power systems is baseload providers, which are power plants able to generate a constant and
predictable supply of electricity (Pfenninger & Keirstead,2015). In most networks, the primary source of
electrical energy is baseload plants, which use fossil fuels.
Geothermal energy is considered as renewable baseload energy sources that are available 24 hours a
day, 365 days a year, independent of weather conditions. It is clean energy that can be extracted
without burning a fossil fuel such as coal, gas, or oil. Geothermal power is homegrown, reducing our
dependence on foreign oil. (DOE, 2017)
An Enhanced Geothermal System (EGS) is a human-made reservoir, created where there is hot rock but
insufficient permeability or fluid. In an EGS, fluid is injected into the subsurface under carefully
controlled conditions, which cause pre-existing fractures to re-open, creating permeability (DOE, 2017a).
The 2006 MIT study called ‘The Future of Geothermal Energy’ predicted that in the United States alone,
100 GWe of cost-competitive capacity could be provided by EGS in the next 50 years (MIT, 2006). The
EGS technical and economical electricity production potential of Turkey (3–5 km) is expected to be 25
Items Goals
Renewable Energy Sources
Share of renewable sources in energy production 30%
Hydroelectric generation capacity Maximum or 36,000 MW
Wind power installed capacity 20,000 MW
Solar power installed capacity 5000 MW
Geothermal power installed capacity 1000 MW
Biomass 2000 MW
Infrastructure
Length of transmission lines 60,717 km
Reaching a power distribution unit capacity 158,460 MVA
Use of smart grid Established
Natural gas storage capacity 11 billion m3
Energy stock exchange Established
Nuclear power plants 2 operational (3
rd
under construction)
Installed power capacity 120,000 MW
7

GWe in the next 25 years (Mertoglu et al. 2015). Unconventional geothermal areas could be facilitated
by EGS and this helps to extend geothermal energy production countrywide.

1.2 PROBLEM STATEMENT AND OBJECTIVES
Commonly used USGS and MIT methods use arbitrarily chosen reference temperatures such as 15-40-
60-100
o
C values and thermal conversion efficiencies such as 0.12, 0.4, etc., and these values are not
depending on something. They are only subjective guesses. We can see previous similar resource
assessment studies for different geothermal fields in Turkey (Atmaca,2010; Basel,2010; Avsar,2011;
Turan,2016; Akin,2017), and they used the MIT method with different reference temperature such as
60,70,80,100
o
C for power generation purpose. Also in literature, there is a variety of usages of these
values subjectively (Garg and Combs, 2011).
The MIT method does not consider enthalpy and entropy changes of the geothermal system in neither
reservoir conditions nor the surface wellhead conditions. On the other hand, in both methods generally,
a recovery factor value higher than "0" is used even if there is great uncertainty in a geothermal field in
the very early exploration phase.
These reasons result in obtaining overly optimistic estimates of power generation potential of
geothermal systems from the USGS and MIT methods as discussed Garg and Combs (2010, 2011, 2015)
and Grant (2015).
Basel (2010), in her Ph.D. study, used the MIT method coupled with the Monte Carlo simulation to
predict the electricity production potential of 25 geothermal fields of Turkey. Although Kavaklidere field
was one of those 25 fields, she did not give any references or explanations about the reservoir
parameters that were used in the calculations. She used 100
o
C as the reference temperature and used
the formula which was developed by MIT engineers (2006) for only binary power conversion plants for
the conversion efficiency.
Altin (2017) studied three geothermal fields in production of Turkey, which are Kizildere, Germencik,
and Salavatli, by using Garg and Combs method. She stated that, based on the results, the Garg and
Combs method gives perfect matching results in comparison to the real operating conditions of the
fields.
8

The Kavaklidere geothermal field holds Turkey’s highest geothermal resource temperature, which was
recorded as 287 C at 2750 m depth by the General Directorate of Mineral Research and Exploration
(MTA) in 2011. MAK-14 well is assumed of representing the field and has a 98
o
C/1000m temperature
gradient (Burcak,2015). At this point, it is essential to calculate the power production potential of the
field for both hydrothermal and EGS cases, which is one of the primary objectives of this study.
The other objective of this study is to present the reservoir simulation models that incorporate different
production scenarios which help to determine the viability of power production from both hydrothermal
and EGS of Kavaklidere field. Another purpose is to see the reservoir behavior when we adapt the
different hydraulic fracturing stimulations to EGS of the field. For these purposes, TOUGH2 geothermal
reservoir simulator via PetraSim graphical interface is used.

1.3 OUTLINE OF THE THESIS
This thesis is composed of four chapters:
Chapter 1, which is the introduction, explains the motivation behind this study and states the research
problem and objectives.
Chapter 2 presents literature review related to geothermal energy basics, geothermal energy in the
world and Turkey, and geothermal power plants. It summarizes enhanced geothermal systems basics,
technologies, and its challenges. This chapter also focuses on the Kavaklidere geothermal field.
Chapter 3 explains the methodology which is employed during resource assessment and reservoir
simulations of Kavaklidere geothermal field for different scenarios of both hydrothermal and EGS cases.
It also includes the results and discussions.
Chapter 4 draws conclusions and presents suggestions for future studies.

9

CHAPTER 2
LITERATURE REVIEW

2.1 Geothermal Energy Basics
Geothermal energy is generated and stored of the heat energy in the Earth that originates from the
melted magma of the Earth and the decay of radioactive substances. Most geothermal energy gathers
around plate borders where most volcanic eruptions and earthquakes happen (Liu et al.,2013).
According to the United States Geological Survey (USGS) some relevant definitions of geothermal energy
and geothermal systems are as provided (Williams et al., 2011).
“Geothermal resource base is all the thermal energy in the earth’s crust beneath a specific area,
measured from the local mean annual temperature.
Geothermal resource is the fraction of the resource base at depths shallow enough to be tapped by
drilling in the foreseeable future that can be recovered as useful heat economically and legally at some
reasonable future time.
Geothermal reserve is defined as the identified portion of the geothermal resource that can be
recovered economically and legally at present time using existing technology.
Geothermal reservoir is a subsurface system consisting of a large volume of hot water and/or steam
trapped in porous and fractured hot rock underneath a layer of impermeable rock.
Geothermal system is any localized geologic setting where portions of the Earth’s thermal energy may
be extracted from a circulating fluid and transported to a point of use. It includes fundamental elements
and processes, such as fluid and heat sources, fluid flow pathways, and a caprock or seal, which are
necessary for the formation of a geothermal resource.
Enhanced/Engineered geothermal system (EGS) comprises the portion of a geothermal resource for
which a measurable increase in production over its natural state is or can be attained through
mechanical, thermal, and/or chemical stimulation of the reservoir rock.”
The formation of a useful geothermal system needs to possess three essential conditions: an
underground heat source (hot rocks), a heat transfer medium (groundwater) and a heat conducting
10

channel that are the fissures or boreholes communicating the underground heat reservoir and ground
surface (Fig. 2.1) (Liu et.al,2013).

Figure 2.1: A Schematic representation of an ideal geothermal system (Liu et al.,2013)

A schematic describing of above terms is shown in Fig. 2.2 Economic viability of geothermal resource is
determined by drilling depth, fluid quantity and quality, and temperature of the resource (Williams et
al., 2011).
11

Figure 2.2: McKelvey diagram representing a geothermal resource and reserve terminology in the
context of geologic assurance and economic viability (Modified from Muffler and Cataldi (1978))

USGS classifies geothermal systems on the basis of temperature into three categories (White and
Williams, 1975; Muffler, 1979; Williams et al., 2008b):
I. High-temperature systems, with temperature values above 150°C; include both liquid- and steam-
dominated resources.
II. Moderate-temperature systems with a temperature range between 90-150°C; most exclusively liquid-
dominated.
III. Low-temperature systems, having temperature values below 90°C; which are all liquid-dominated
resources.
For direct usage, all three temperature classes are suitable but generally moderate- and high-
temperature systems are viable for electric power generation. If sufficiently low temperatures are
12

available for cooling the working fluid in a binary power plant, the low-temperature range also can be
exploited for electric power generation (Williams et al., 2011).
Other thermal classification systems have been proposed, with most focusing on dividing geothermal
resources into a similar set of three or, more simply, two classes that define a progression from low to
high temperature (or enthalpy) geothermal resources (Fig. 2.3) (Williams et al., 2011).

Figure 2.3: Example classifications of geothermal resources by temperature (Celsius) (Williams et al.,
2011)

2.1.1 Geothermal Energy in the World
The earth has immense natural heat reserves. Electric Power Research Institute (EPRI,1978) estimated
the stored thermal energy down to 3 km within continental crust to be roughly 43 x10
6
EJ. Evaluated by
Stefansson (2005) of the available electrical potential range is from 50 to 200 GW.
While geothermal hot springs have been used since ancient times, the first industrial use of geothermal
resources in the early 1900s where electricity was first generated at Larderello, Italy.
13

Even geothermal energy contributes a small proportion of the world’s primary energy consumption for
both direct and indirect utilization, however, it provides substantially to the nation’s energy supply and
welfare for the countries, such as the Philippines, which lacks homegrown fossil fuels. There were 315
MW of new geothermal power capacity installed in 2015, raising the total capacity to 13.2 GW. Turkey
accounted for half of the new global capacity additions, followed by the United States, Mexico, Kenya,
Japan, and Germany. Fig. 2.4 below shows the share by country of global geothermal capacity additions
in 2015. (WEC, 2016)

Figure 2.4: Geothermal global capacity additions in 2015, by country (WEC,2016)

Regarding the direct use of geothermal heat, the countries with the most extensive utilization that
accounted for roughly 70% of direct geothermal in 2015 are China, Turkey, Iceland, Japan, Hungary, USA
and New Zealand, as shown in Fig. 2.5 below. (WEC,2016)

14

Figure 2.5: Top countries that utilize the most direct geothermal heat in 2015 (WEC, 2016)
Moreover, global direct use of geothermal energy can be further divided into categories of energy
utilization. Fig. 2.6 shows the installed capacity of geothermal direct utilization (MWt) by the types such
as geothermal heat pumps, space heating, greenhouse heating, etc. (Lund and Boyd, 2015)

Figure 2.6: Geothermal direct applications worldwide in 2015, distributed by percentage of total
installed capacity (MWt) (Lund and Boyd, 2015)

Today, Turkey is located at the 4th order in installed electricity capacity in the world in comparison
with the values recorded by the end of August 2017 (Fig. 2.7).
15

Figure 2.7: Top 10 geothermal countries based on installed power generation capacity, August 2017 (Source:
ThinkGeoEnergy)

2.1.2 Geothermal Energy in Turkey
Turkey is located in the Mediterranean part of Alpine-Himalayan Tectonic Belt, which presents crucial
geothermal potential. Active tectonics in the Central and Eastern Anatolia, right lateral-strike slip North
Anatolian Fault Zone and the horst-graben system and widespread volcanism in the Western Anatolia
affect the distribution of the geothermal regions in the country as shown in Figure 2.8 (MTA 2017,
Bozkurt 2001).
According to the study of Geothermal Energy Potential and Exploration in Turkey by MTA, 2017;
Turkey has approximately 1.000 geothermal springs that located all over the country that have various
of temperatures. The capacity of geothermal capacity of Turkey is predicted theoretically as 31.500
MWt. 78% of these geothermal fields are situated in Western Anatolia, 9 in Central Anatolia, 7% in the
Marmara Region, 5% in Eastern Anatolia and 1% in the other regions.

16

Figure 2.8: Distribution map of geothermal resources in Turkey concerning plate tectonics (MTA)

Figure 2.9: Map of temperature pattern of geothermal resources in Turkey and their utilization ways
(MTA)
17

In the resource temperature map by MTA is shown in Fig. 2.9, the hottest areas with red color,
where the resource temperatures are higher than 100
o
C, are in the Western and central parts of
Turkey. 90% of the geothermal areas have low and medium enthalpy which are suitable for direct
applications such as heating and thermal tourism, while 10% of them are ideal for indirect applications
such as electricity generation.
First geothermal exploration and investigations in Turkey started by MTA in 1962. First geothermal
electricity generation held in Kızıldere geothermal field by MTA in 1975 and was initiated by Kızıldere
power plant in Sarayköy district of Denizli with 0,5 MWe power. Geothermal activities which come to a
halt because of some political reasons in the 1990s was accelerated in 2005. So, and drilling geothermal
energy explorations reached from 2.000 meters to 28.000 meters and geothermal budget increased
about ten times for geothermal energy explorations.
Up until now, 1559 geothermal wells have been drilled in Turkey, and 596 of these were by MTA. Drilled
wells are very much denser in the Western part where the Büyük Menderes, Küçük Menderes, and
Gediz grabens are located (Fig. 2.10). 319 geothermal fields are discovered, and the total available heat
capacity from all these wells is stated as 16098.8 MWt. Total direct use installed capacity is reported as
3272 MWt which shows that the utilization of the total available heat is around 20% (Akkus and Alan
2016).

Figure 2.10: Location map of the central grabens in western Turkey (Yilmaz and Gelisli, 2003)
18

By the end of December 2017, there are 30 licensed geothermal electric power plants in operation with
a total capacity of 835.558 MWe, and with a total installed capacity of 1261.853 MWe. Names of the
companies, locations, and other details are listed in Table 2.1. (EPDK, 2017)
Comparison of Geothermal applications in Turkey for the years 2002-2017, number of fields suitable for
electricity production reached from 16 in 2002 to 30 in 2017. In 2002, while 500,000 m
2
of greenhouses
are heated from geothermal energy and now in 2017, 3,931,000 m
2
greenhouses heated with
geothermal energy with an increment of 686%. Residential heating from geothermal energy in reached
from 30,000 residents in 2002 to 114,567 residents in 2017 with an increase of 281%. Installed
electricity production capacity from geothermal energy achieved from 15 MWe in 2002 to 1261.8 MWe
in 2017 end of December with the increment of 8312%. Geothermal heat capacity reached from 3.000
MWt in 2002 to 16,099 MWt in 2017 with an increase of 416%.

Table 2.1: Licensed geothermal electric power plants in Turkey by the end of December 2017 (EPDK,
2017)

Company
Name

Geothermal Power
Plant Name

City - District
Installed
Capacity
(MWe)
Capacity Under
Construction
(MWe)
Capacity in
Operation
(MWe)
Afjet Afjes Afyonkarahisar -
Merkez
2.755 2.755 0

Çelikler
Çelikler Pamukören Jes
Aydin - Kuyucak
67.53 0 67.53
Pamukören Jes 2 22.51 0 22.51
Pamukören Jes 3 22.51 0 22.51
Gürmat Efeler Jes Aydin - Incirliova 162.3 47.4 114.9
Çelikler Çelikler Sultanhisar Jes Aydin - Sultanhisar 13.8 13.8 0
Ken Kipaş Ken Kipaş Santrali Aydin - Merkez 24 0 24
Ken Kipaş Ken 3 Jes Aydin - Efeler 24.8 24.8 0
Limgaz Buharkent Jes Aydin - Buharkent 5 5 0
Greeneco Greeneco Jes-3 25.6 0 25.6
Karyek Karkey Umurlu Jes Aydin - Köşk 12 0 12
Karyek Umurlu-2 Jes 12 0 12
Gümüşköy Gümüşköy Jes

Aydin - Germencik
13.2 0 13.2
Gürmat

Galip Hoca Jes 47.4 0 47.4
Efe 6 Jes 22.6 22.6 0
Beştepeler Kubilay Jes 24 0 24

Maren
Melih Jes 33 33 0
Mehmethan Jes 24.8 0 24.8
Kerem Jes 24 0 24
Maren Santrali 44 0 44
Deniz (Maren II) Jes 24 0 24
Dora-1 Aydin - Sultanhisar 7.951 0 7.951
19

Menderes
Geothermal
Dora IV Jes
Aydin - Köşk
17 0 17
Dora-2 Jes 9.5 0 9.5
Dora III Jes 34 0 34
Turcas Kuyucak Jes Aydin - Kuyucak 18 18 0
Mtn Energy Babadere Jes Çanakkale - Ayvacik 8 0 8
Tuzla
Geothermal
Tuzla 7.5 0 7.5
Akça Tosunlar 1 Jes

Denizli - Sarayköy
3.807 0 3.807
Bereket Kızıldere 6.85 0 6.85
Greeneco Greeneco Jes 25.6 0 25.6
In-Alti Gök Jes 3 3 0
Jeoden Jeoden 2.52 2.52 0

Zorlu Energy
Kızıldere-3 Jes 165 165 0
Kızıldere II Jes 80 0 80
Kızıldere Jes 15 0 15
Enerjeo Enerjeo Kemaliye Jes

Manisa - Alaşehir
24.9 0 24.9
Maspo
Energy
Maspo Jes IV 10 10 0
Mis Energy Mis-1 15 15 0
Sis Energy Özmen-1 Jes 23.52 23.52 0
Türkerler

Türkerler Alaşehir Jes 24 0 24
Alaşehir Jes 2 24 0 24
Zorlu Energy Alaşehir Jes 45 0 45
Alaşehir-2 Jes 24.9 24.9 0
Sanko
Energy
Sanko Jes Manisa - Salihli 15 15 0
Total 1261.853 426.295 835.558

2.1.3 Geothermal Power Plants
Electricity generation is the primary purpose and at the same time more challenging use of geothermal
energy. Different types of power plants are being constructed to be able to convert geothermal heat
into electricity with variable efficiency ratios, which depend on the efficiency of the all components
such as heat exchanger, condenser, turbine, and generator. Geothermal plants depend on one or a
combination of three types of conversion technology – binary, steam, and flash.
In Turkey, more than 80% of power plants are constructed as binary, and the rest is single flash or
double flash. There are no dry steam or gas phase dominated geothermal systems in Turkey, so
there is no example of dry steam power plant.
Flash power plants are the most common and typically require resource temperature more than 180
o
C
(350
o
F). Geothermal hot water quickly loses pressure while rising inside wellbores from deeper in the
earth to the surface, and vaporizes or ‘flashes’ to steam. The steam is separated from the hot water by a
20

separator and then sent to the turbine to run the electricity generator. The remaining liquid is either
sent to another tank to flash for more energy if possible or to a condenser to make it cooler and reinject
it into the ground. Figure 2.10 shows schematically single and double flash-type power plants (Kagel,
2008). The thermal efficiency of these plants is 10-20% (Zarrouk and Moon, 2014). These systems
provide the 63% of produced geothermal energy in the world (Bertani, 2015). In Turkey, Kızıldere (single
flash) and Germencik (double flash) plants are the examples of this technology.

Figure 2.11: Illustration of Single Flash Power Plant (left) and Double Flash Power Plant (right) (Kagel,
2008)
Binary cycle power plants make electricity generation possible from low-temperature geothermal
resources such as 74
o
C to 180
o
C. Geothermal fluid is just used to boil a working fluid which has a lower
boiling point and higher molecular weight than geo-fluid. After being used, geo-fluid is reinjected into
the ground through a closed loop to maintain the reservoir pressure and lifetime. There are different
thermodynamic cycles used in binary power plants such as Organic Rankine Cycle (ORC), which is the
most common and employs butane, propane, pentane and their iso- versions as working fluid. Figure
2.11 shows a binary system for electrical power generation schematically (Kagel, 2008). In Turkey, Tuzla,
Dora-1, and Kızıldere-Binary geothermal plants are
the examples of this technology.

Figure 2.12: Illustration of Binary Power Plant
(Kagel, 2008)

21

2.1.4 Enhanced Geothermal Systems – EGS

The adoption of geothermal energy in the industry has been limited despite its advantages. Only about
0.3% of the world electricity production is accredited to geothermal energy today. The dearth of
geographical locations where geothermal energy is geologically available poses a threat to the
development of the resource and its establishment as the main energy source. Heat, water, and
permeability are the three conditions need to be fulfilled for water to flow through the rock in a
geothermal field. To achieve an increase in the number of locations where geothermal energy can be
produced, the adoption of Enhanced Geothermal Systems (EGS) has been advised (MIT 2006). The
earth’s temperature is high enough to produce geothermal electricity across a significant percentage of
the United States. However, a difficulty is encountered at a high depth where the rock is characterized
by crystallinity and natural permeability. The use of EGS employs pumping water down at high pressures
to create enough force for rock fracture. Eventually, the fractures form conduits for fluid flow, thereby
creating a human-made geothermal reservoir.
Fracturing was innovated in the oil and gas industry as a method of enhancing the recovery and boosting
the efficiency and productivity of wells. The fracturing technology continues to develop as the industry
advances technologically. In the last fifteen years, new fracturing techniques have been introduced that
created a platform to produce large, otherwise uneconomic natural gas in low permeability shale
formations such as the Barnett Shale in Texas. Between 1993 and 2011, an increase in natural gas
production from 11 billion standard cubic feet to 1.93 trillion cubic feet was observed in the Barnett
(Railroad Commission of Texas, 2012). This improvement in fracturing technology in the United States
has a resounding and novel effect on the domestic and global natural gas market.
The use of EGS started from Fenton Hill as HDR (Hot Dry Rock) system, New Mexico in 1970, where a
team of researchers backed by the Department of Energy demonstrated the feasibility of the method.
Afterward, EGS reservoirs have been created in England, France, Germany, Japan and Australia among
others (MIT, 2006).
The procedure in an EGS involves drilling a well to sufficient depth to reach a useful temperature,
creating a large heat-transfer surface area by hydraulically fracturing the rock, and then drilling a second
and sometimes third well into the region where the rock was fractured for production. Furthermore,
cold water is injected into one of the wells drilled, heat then flows through the rock, and it is produced
from the other wells. Figure 2.13 illustrates this example. A team of geothermal experts arranged by MIT
concluded that the total US resource base of EGS is approximated at 14 million exajoules (EJ). One
22

exajoule has about 10
15
Joules. According to this report, the estimated extractable energy available in
the US is around 300,000 EJ. In the United States, about 100 EJ of primary energy is consumed every
year. Tester et al. 2007 estimated the need for an investment of $300 to $400 million over 15 years to
make an early generation EGS competitive, by cost. (MIT, 2006)

Figure 2.13: Schematic of an EGS system (MIT, 2006)

Figure 2.14 shows different types of geothermal resources and potential size in the United States with
their associated risk and reward profile. The most common one is the low temperature and coproduced
geothermal resource which is considered as a low risk, low reward type. The characteristics of the other
intermediate resource types that can benefit from innovative exploration technologies and those
associated with the permeable sedimentary resources. The one with the highest potential for growth is
the EGS, with potential resource size of 16 000 GW in the US. EGS has the highest reward, but at the
same time, it has the highest risk because of the many challenges which are discussed in section 2.1.4.2.

23

Figure 2.14: Types of Geothermal Resources and their Risk/Reward Profile (modified after
Nathwani,2011 and White et al.,2014)

2.1.4.1 Hydraulic Stimulation and Hydro-shearing
To develop an enhanced geothermal system (EGS) in depth reservoirs, hydraulic stimulation is a vital
procedure. The primary purpose is increasing formation permeability, most often in crystalline rock such
as granite, to achieve a sufficient flow rate for economical production.
Two main fracture stimulation mechanisms that may coincide with the hydraulic treatment are
hydrofracturing (HF) and hydro-shearing (HS), but usually, one mechanism is dominant (Claudouhos et
al., 2013).
“1) hydraulic fracturing if tensile fractures develop as the normal stress and tensional strength is
exceeded by the local pore pressure (Figure 2.19a).
2) hydro-shearing, if slip along pre-existing fractures is induced as frictional resistance is overcome due
to increasing pore pressure (Figure 2.19b).”
Both mechanisms have different effects on fracture permeability. The permeability increased by HS is
mostly irreversible due to rearrangement of asperity contacts go with shear dilation. The permeability
enhanced throughout HF reduces nearly reversible after pressurization unless proppant is used to keep
fractures open (Gischig and Preisig, 2015).
24

Currently, in gas and oil reservoirs, the dominant mechanism is opening and propagation of new
fractures (hydro-fracturing), while hydro-shearing is the leading mechanism in EGS projects (Petty et al.,
2013; McClure and Horne, 2013). The main advantages of hydro-shearing are seen in the permanent
permeability increase due to self-propping of fractures, and the lower injection pressure required. The
multi-stage stimulation method has led to higher production rates in shale gas fields. This method has
been considered and adapted in recent EGS projects, such as the Newberry project in Oregon, USA. The
Newberry EGS project efficiently had multiple zones by using thermally-degradable zonal isolation
materials (TZIMs) which are biodegradable polymers (Petty et al., 2013).

Figure 2.15: (a) Hydraulic fracturing open new or pre-existing tensile fractures from a small packed-
interval, where the fluid is injected at a pressure higher than the minimum principal stress σ3. (b) Hydro-
shearing aims to reactivate pre-existing natural fractures favorably oriented for shearing, with a fluid
pressure remaining lower than σ3 (Gischig and Preisig, 2015)
Since fractures provide more porosity and permeability for fluid storage and movement, it is very
important to characterize fracture network in unconventional hydrocarbon and geothermal reservoirs.
Induced seismicity (micro-seismic events) can happen because of creating new fractures or reactivating
the natural fractures in shear zones during the injection or production of fluid. Monitoring these events
can a useful characterization tool to visualize fracture growth and network during injection stimulation.
Using micro seismic data with different methods is a common way to understand the fractures and
25

there are many studies about this. Lately, compressional and shear velocity models, fuzzy cluster
centers, and shear wave splitting are three methods that can be extracted from micro seismic data
directly to obtain reliable characteristics about fractured areas (Aminzadeh et al., 2013).
For improved mapping of natural and induced fractures in unconventional reservoirs, Maity and
Aminzadeh, 2015 introduced a novel integrated approach called hybrid-fracture zone-identifier attribute
(FZI). They used not only micro seismic monitoring data but also 3D surface seismic data and well logs to
better predict the reservoir properties. The FZI makes use of elastic properties of rock derived from
micro seismic as well as log-derived properties within an artificial neural network (ANN) framework.
They demonstrate how to use passive seismic data as a fracture zone identification tool (Maity and
Aminzadeh, 2015).
Furthermore, more data and knowledge gained with these methods can help to eliminate the errors in
locating the fractured areas. It is important to target the stimulated area to optimize drilling targets or
stimulation jobs for the future development plans such as reducing costs and maximizing production.

2.1.4.2 Challenges of EGS
Evidence suggests that EGS is technically feasible, the resource base is readily available in accessible
locations. The economic aspect of EGS is what needs to be overcome. The low productivity of EGS does
not give a justification for investing in the expensive drilling of deep wells. Improved drilling technology
can be employed to reduce costs. Increasing revenue to lower cost could be sought as an alternative.

The amount of electricity produced per well is proportional to the temperature of the produced fluid
and the flow rate. EGS wells typically flow in the range of 20 - 40kg/s. To design and implement EGS
economically on a large scale, flow rates in the 50 - 100 kg/s should be attained (McClure, 2009). On the
other hand, large surface area and fractured volumes are needed to ensure long-term heat transfer at
acceptable rates. For instance, in the Fenton Hill project Phase 1 system heat-transfer surface area was
too small by about a factor of 100 for a commercial-scale system (MIT, 2006).

Another challenge is establishing inter-well connectivity that is a proper connection between the
injector and producer. It can be difficult to achieve and demonstrate. For instance, in the Ogachi project
26

in Japan, the wells were successfully stimulated, but the connection achieved between them was poor.
(MIT, 2006)

It is essential to maintain produced fluid temperature, and because the reservoir is fractured short-
circuiting can be another critical issue. Flow channeling or short-circuiting results in inadequate heat
exchange along the flow path. This happened at the EGS project at Hijiori, Japan, where the produced
temperature dropped from 163°C to 100°C over two years (MIT, 2006). Short-circuiting is always a
concern in fractured reservoirs, including conventional geothermal fields.

One more main challenge is induced seismicity. The stimulation process pumping fluid at pressure
causes micro seismic events which are very small earthquakes which are the order of one or two on the
Richter scale. The largest one with magnitude 3.4 happened in Basel EGS project, Switzerland, caused
the cancellation of the project (MIT,2006). After, EGS drew massive attention from media and caused
public aversion against geothermal projects. Seismicity events at the Geysers geothermal field in
California have been strongly correlated with injection data (Majer and Peterson, 2007). For these
reasons, induced seismicity has become an important subject to study in terms of a monitoring tool and
a potential hazard.
In 2011 DOE introduced a determined program by investing about half a billion dollars in EGS research
and development to address the above challenges. For instance, the University of Southern California
(USC) team received a grant to work on ‘Characterizing Fractures in Geysers Geothermal Field by Micro
Seismic Data, Using Soft Computing, Fractals, and Shear Wave Anisotropy’. Yet, more work needs to be
done to completely point many outstanding complex problems. Sometimes proposed solutions for the
outstanding issues cannot be economically viable (White at al.,2014).

2.2 Kavaklidere Geothermal Field

The Kavaklıdere geothermal field covers a 126 km2 area between Alasehir and Salihli districts in Manisa
province, Turkey. The was tendered by the General Directorate of Mineral Research and Exploration
(MTA) was bought by MASPO Energy, a private company in 2011 (Fig. 2.16). Temperatures in the
Kavaklıdere geothermal field range from 188 to 287
o
C. Permeability in the reservoir consists of a
27

network of fractures in hard metamorphic rocks a short distance above a granitic intrusion (Ozdemir et
al., 2016).

Figure 2.16: Location map of the Kavaklıdere geothermal field (Ozdemir et al., 2016)

2.2.1 Geological Outlook and Tectonic Setting
The Kavaklıdere geothermal field occurs along the active southern margin of the Gediz-Alasehir graben.
Six different units have been recognized in the study area which constitutes the Kavaklıdere geothermal
field. These six units from bottom to top are (1) Metamorphic rocks (gneiss, calc-schist, quartz-schist,
phyllite, mica-schist) of the Precambrian-Middle Triassic Menderes Massif (2) Paleozoic marbles (3)
Granitic rocks (4) Upper Miocene-Lower Pliocene Gediz formation (5) Upper Pliocene-Quaternary
sediments and Kaletepe formation (6) Quaternary alluviums, respectively (Figs. 2.17 and 2.18). (Ozdemir
et al., 2016)
Faults with four different mechanisms were observed in the field (Fig. 2.19). These are detachment
faults, low-angle dip-slip normal faults, high-angle dip-slip normal faults, 1969-earthquake faults, and
strike-slip faults (scissor faults). The development of structural elements in the field has been influenced
by both the detachment fault and 1969 earthquake fault geometry. The listric character of the
detachment fault caused the formation of asymmetric sediment in the basin. 38 fault planes in total,
including 7 detachment faults, 15 high-angle normal faults, 2 low-angle normal faults and 13 shear
fractures/strike-slip faults (scissor faults), were measured at the Kavaklıdere geothermal field. This
analysis of the distribution of the orientation of maximum extension in and across the Alasehir supra-
28

detachment basin shows a dominantly NNE or NNW orientation with minor deviations from the north.
The more north-eastward- directed tension orientations to be a result of differential rotation along the
NNE-directed hinge faults (Ozdemir et al., 2016).
Fractured rocks of the Menderes Massif, such as mica schist, gneiss, and especially marbles, are the
reservoir rocks. Cap rocks for the geothermal fluids include clay-rich intervals within the Neogene
sedimentary units. Most hot springs and hot wells with good flow rates lie near the gently N-dipping
Gediz detachment fault, where it intersects and is cut by ~N–S-striking transverse faults. Karstic marble
and breccia along and near the detachment fault provide good channel ways for flow, possibly
somewhat distal to the main upwelling zone. The geothermal activities in the Kavaklıdere geothermal
field appear to lie at the intersections of minor northerly striking sinistral normal transfer faults and the
Gediz detachment fault (Fig. 2.20) (Ozdemir et al., 2016).

Figure 2.17: Simplified geological map of the Kavaklıdere geothermal field. (Ozdemir et al. 2016)
29

Figure 2.18: Generalized stratigraphic vertical section of the Kavaklıdere geothermal field (modified
after Ozdemir et al. 2016 and MASPO 2013)

30

Figure 2.19: Geological cross-section of the detachment fault and high angle normal faults between
Horzumalayaka and Kavaklıdere (Ozdemir et al.,2016)

Figure 2.20: Schematic view of scissor faults, faults system, fault set and geothermal reservoirs in the
Kavaklıdere geothermal field (Ozdemir et al., 2016)

2.2.2 Hydrogeological Outlook
Thermal waters are hosted by Menderes massif metamorphic rocks, which are made of gneisses, schists,
and marbles. Bulbul el at., 2011 stated that presence of geothermal waters is closely related to normal
fault systems and graben tectonic. Meteoric waters recharging the reservoir rocks are heated at depth
with geothermal gradient. Na-HCO3 water type is dominant for thermal water. They reported that
31

reservoir temperatures of the geothermal system are valued to differ between 125 and 225°C by
mineral equilibria geothermometer, vary between 160 and 240°C by Giggenbach triangular diagram and
differ between 150 and 250°C by silica enthalpy-mixture model (Appendix B). The temperatures
obtained from silica enthalpy model, mineral equilibria geothermometers, Na/K, Na/Li
geothermometers are more useful than others in the study area. They suggested to have reinjection
temperature as approximately 50 to 75°C to avoid scaling problems, because they found that calcite,
aragonite, dolomite, quartz and amorphous silica have scaling tendencies (Bulbul et al.,2011).

2.2.3 Drilling History and Geothermal System
In the study area, Horzumsazdere is the only thermal spring and has temperatures 25–34
o
C with 3–4 l/s
flow rate. Table 2.2 shows the wells in the field with the details.

Table 2.2: Wells in Kavaklidere Geothermal Field (Burcak,2011; MASPO,2013; Ozdemir et al.,2016)

Ownership/year

Well
Name

Well Depth
(m)

Well
Temp
o
C
Temp
Measurement
Place

Flow rate
(l/s)
Municipality of Kavaklidere /1996 AK-1 750 116 BH -
63 WH 3
MTA /2002 KG-1 1447 188 BH 12, A
MTA /2004 AK-2 1507 213 BH 8, A
Ahmet Alduran AA-1 710 67.2 WH 5-7, P
MTA and MASPO /2010 MAK-14 2750 287 BH -
145 WH 35
MTA and MASPO /2010 MAK-15 1750 159 BH 5
80 WH 2
MTA and MASPO /2011 MAK-3 2250 188.5 BH -
175 WH 90
MASPO/2013 MASPO-1 2941 260 BH N/A
MASPO/2013 MASPO-2 3150 256 BH N/A
MASPO/2013 MASPO-3 3420 251 BH N/A
MASPO/2013 MASPO-4 3000 236 BH N/A
MASPO/2013 MASPO-5 2750 200 BH N/A
BH: Bottom Hole, WH: Well Head, A: Artesian, P: Pump, N/A: not available
32

The first reservoir is the Miocene aged of Alasehir Formation which has sandy gravel levels of
Quaternary and Neogene units. AA-1 well at a 710m depth provided 67.2
o
C temperature and 5–7 l/h
compressor flow rate of hot water from the first reservoir. AK-1 well with 116
o
C bottom hole
temperature provided 63
o
C, 3 l/h flow rate of water. The first reservoir has been taken behind a closed
pipe because it is not appropriate to product for electricity aspect of yield and temperature in
Kavaklıdere area. Then the deep geothermal wells drilled by MTA for the target of second reservoir
(Ozdemir et al.,2016).
The second reservoir, which is suitable for electricity and high-temperature hot water, is the Menderes
Massif series, and MAK-14, MAK-15, MAK-3, MASPO-1, MASPO-4, MASPO-5 deep wells were drilled.
Marble-calc-quartzite schist units and the possible faulted-fractured, quartzite and gneiss units have
been discovered. The downhole temperatures were measured 188–287
o
C at different flow rates. The
wells in the second reservoir had higher production efficiency than the first reservoir (Ozdemir et
al.,2016).
Two impermeable cap rock assemblages were found that separate the reservoirs from each other. The
thickness is 2000 m in the middle part of the graben. First cap rock is Kaletepe Formation that has a
sandstone–siltstone-claystone lithologies shows impermeable feature. Despite the claystone and
milestone have a high porosity, they have little or no permeability. Also, during graben, other hard and
brittle lithologies earned secondary permeability, these units preserved fluid and heat-resistant
properties. Second cap rock is the Miocene Gediz Formation units are composed of alternating
claystone-siltstone-clay, marl, and silt (Ozdemir et al.,2016).
The main recharge (water supply) of geothermal reservoirs, the sum of precipitation of meteoric waters
to the basin, and underground and surface waters. In Kavaklidere, the recharge of reservoirs is closely
related to the permeabilities of lithologies at Bozdag horst and Gediz graben. Major East-West trending
graben faults cross the North-South trending faults which had connected each other and from very
remote areas, the recharge can be provided to the neighboring areas by through these faults (Ozdemir
et al.,2016).
In the region the effect of expansion tectonics, while the Menderes Massif continuously rises, N–S
orientation openings are formed and graben. In the grabens depending on thinning at the continental
crust, magmatic events are formed the system’s heat source which was approaching the surface and
rising along zones of weakness. (Ozdemir et al.,2016).
33

CHAPTER 3
METHODOLOGY

The objective of the study is to present the potential conventional and enhanced geothermal system
applications in Kavaklidere region. Tree central parts of the study are geothermal resource assessment,
reservoir simulation, and economic evaluation.
For the resource assessment, volumetric probabilistic methods are engaged rather than deterministic
methods because of the subsurface uncertainties and the related parameters. Different volumetric
methods are compared and examined, and Garg and Combs USGS based volumetric method with Monte
Carlo Simulation was applied to calculate electric power generation potential of the Kavaklidere field for
both conventional and EGS systems.
For the reservoir modeling, PetraSim which is the graphical interface for the TOUGH2 simulator is used
for both conventional and EGS system. TOUGH (Transport of Unsaturated Groundwater and Heat) is the
primary simulator for nonisothermal multiphase flow in fractured porous media.
For the economic evaluation, the Geothermal Electricity Technology Evaluation Model (GETEM) is used
to estimate the Levelized Cost of Energy/Electricity (LCOE) for both conventional and EGS resources.

3.1 Comparison of Volumetric “Heat In-Place” Estimation Methods
During early stages of the exploration and development of oil and gas industry, it is important to apply
some mathematical methods to calculate oil or gas initially in place and their recoverable amount. This
helps the decision makers to have the future targets, economic plans and operation schedules of the
assets. In the same way, for the geothermal area, it is essential to have the reasonable approach to the
stored ‘heat in place’ and its producible and generatable amount.
Different volumetric heat in place approaches have been found in the literature to quantify the
uncertainty of estimates of the geothermal resources associated with an identified hydrothermal
convection system. The USGS method was developed by the US Geological Survey scientists in 1970’s
(e.g., Nathenson, 1975a,b; Nathenson and Muffler, 1975; Muffler and Cataldi, 1977; Brook et al., 1979);
the MIT method in 2006 (MIT, 2006) and Garg & Combs method in 2015.
34

Garg and Combs (2015) showed that usage of the arbitrary values of the reference (ambient or
abandonment) temperatures and conversion efficiencies in USGS and MIT methods overestimates the
electric generation potentials of geothermal reservoirs. To predict the more realistic and suitable values
of electric power generation potential of a geothermal reservoir, Garg and Combs, 2015 suggested that
to consider the thermodynamic properties of specific power plant cycles such as flash or binary based
on exergy analysis (using the 2nd law of thermodynamics).
The common background of all these three methods related to stored heat in place and the recoverable
heat is following;
Subsurface heat in place stored is calculated by using the Eq. 3.1 below (Muffler, 1978);
𝑄 𝑅 = 𝐴 ℎ(𝜌 𝑐 𝑃 )
𝑅 (𝑇 𝑅 − 𝑇 𝑟 ) Eq. 3.1
where Q R=the stored heat in place (kJ), A = Area (m
2
), h=Thickness (m), (𝜌 𝑐 𝑃 )
𝑅 =volumetric, isobaric
specific heat capacity of the liquid saturated rock, T R =resource temperature (
o
C) and T
r
= reference or
abandonment temperature (
o
C).
(𝜌 𝑐 𝑃 )
𝑅 (kJ/m
3
-
o
C) is calculated by using Eq. 3.2 below;
(𝜌 𝑐 𝑃 )
𝑅 = ∅𝜌 𝑙 𝑐 ℎ𝑙 + (1 − ∅)𝜌 𝑟 𝑐 ℎ𝑟 Eq. 3.2
where ∅ = porosity (fraction), 𝜌 𝑙 and 𝜌 𝑟 are liquid and solid rock density respectively (kg/m
3
), 𝑐 ℎ𝑙 and 𝑐 ℎ𝑟
are liquid and solid rock matrix specific heat capacity respectively (kJ/kg-
o
C).
Because Kavaklidere is single-phase liquid-dominated geothermal reservoir, here the steam or gas phase
is disregarded in the calculation of the (𝜌 𝑐 𝑃 )
𝑅 . As noted earlier, there is no steam or gas phase
dominated geothermal systems in Turkey.
It is not possible to produce all stored energy in the reservoir; only recoverable part accounted by the
recovery factor (denoted by 𝑅 𝑓 in this thesis) can be produced at the wellhead. Recovery factor depends
on many parameters and conditions such as permeability, rock matrix and fractures, thermal and
hydraulic boundary conditions, production and injection well depths and patterns etc. (see Garg and
Combs, 2015)
Recovery factor is defined by Eq. 3.3 below;
35

𝑅 𝑓 =
𝑄 𝑤 𝑄 𝑅 Eq. 3.3
Where 𝑅 𝑓 =thermal recovery factor, 𝑄 𝑤 = heat produced at the wellhead (kJ), 𝑄 𝑅 =heat stored in the
reservoir (kJ).
Combining Eq. 3.1, 3.2 and 3.3, the heat recovered at the wellhead Qw can be expressed by Eq. 3.4
below;
𝑄 𝑤 =
𝑅 𝑓 𝐴 ℎ[∅𝜌 𝑙 𝑐 ℎ𝑙 + (1 − ∅)𝜌 𝑟 𝑐 ℎ𝑟 ](𝑇 𝑅 − 𝑇 𝑟 ) Eq. 3.4
Many references discussed on the recovery factor. Grant (2014) recently highlighted that the past values
of recovery factor have been in all cases high in comparison with actual performance. Here referred to
some of the papers examined.
GeothermEx (2004) describes: “Based on our assessment of more than 100 geothermal energy sites
around the world, we have found it more realistic to apply a recovery factor in the range of 0.05 (Min) to
0.2 (Max) without application of a most-likely value”.
C.F.Williams et al. (USGS Open-file Report 2008-1296) describe that the recovery factor “ R f for fracture-
dominated reservoirs is estimated to range from 0.08 to 0.2, with a uniform probability over the entire
range. For sediment-hosted reservoirs, this range is increased from 0.1 to 0.25”.
S.K. Garg and J. Combs (2011) describes: “Prior to geothermal energy well drilling and testing, it will not,
in general, be possible to obtain any reliable estimates of reservoir thickness and thermal recovery
factor. Since it may eventually prove impossible to produce fluid from a geothermal energy reservoir,
the possibility of the thermal recovery factor being zero cannot be discounted during the exploration
phase; therefore, the proper range for thermal recovery factor is from 0 to 0.20 (the latter value is
believed to be the maximum credible value based on worldwide experience with production from liquid-
dominated reservoirs)”.
AGEA compiled by J. Lawless (2010) describes: “In fracture dominated reservoirs where there is
insufficient information to accurately characterize the fracture spacing, adopt the mean USGS value of
14%, or 8 to 20 % with a uniform probability over the entire range when used in probabilistic estimates”.
“In sedimentary reservoirs or porous volcanic-hosted reservoirs, of ‘moderate’ porosity (less than 7% on
average), adopt the mean USGS value of 17.5%, or 10 to 25% with a uniform probability over the entire
range when used in probabilistic estimate”. “In the case of sedimentary or porous volcanic-hosted
36

reservoir of exceptionally high average porosity (over 7%), adopt the empirical criterion of recovery
factor 2.5 times the porosity to a maximum of 50%”.
M.A. Grant (2014) pointed out that there is a wide range of recovery factors: “3-17 % covers the entire
range of observed results. This indicates that any result is subject to an error of at least a factor of 2, or ±
70%. One conclusion is immediate: past recovery factors have been too high, and comparison with the
actual performance show that an average value of 10% should be used.”
Assuming isenthalpic flow in the wellbore and neglecting the work required to raise water to the
wellhead, the enthalpy of produced fluid at the wellhead, hw, is equal to that of liquid water at the
reservoir temperature. Thus by Eq. 3.5 below;
ℎ
𝑤 = ℎ
𝑅 (𝑇 𝑅 ) Eq. 3.5
The amount of fluid produced at the wellhead, mw (kg), is given by Eq. 3.6 below;
𝑚 𝑤 = 𝑄 𝑤 /(ℎ
𝑤 − ℎ
𝑟 ) Eq. 3.6
where hr (kJ/kg) is the enthalpy of liquid water at the reference temperature, Tr.
The concept of availability plays a significant part in the USGS volumetric “heat in place” method.
Availability is defined as the maximum work (or power) output that can theoretically be obtained from a
substance (water) at specified thermodynamic conditions (wellhead) relative to its surroundings.
DiPippo (2008) observes:
“To achieve this ideal outcome, there are two thermodynamic conditions that must be met:
(1) All processes taking place within the system must be perfectly reversible.
(2) The state of all fluids being discharged from the system must be in thermodynamic equilibrium with
the surroundings.”
The first of these conditions amounts to neglecting losses due to friction, turbulence, and other sources
of irreversibility. The second condition requires that any fluids discharged from the system - are in
temperature equilibrium with the surroundings (i.e., reference temperature). None of the real power
cycles can meet these conditions, and the “electrical energy” that is generated is always less than the
37

“available work.” The conversion efficiency, 
𝑐 (utilization factor) is the ratio of the “actual electrical
energy” to the “available work”. (Garg and Combs, 2011)
Neglecting kinetic or potential energy effects, the maximum energy output per unit mass of the
substance e is given by (DiPippo, 2008):
𝑒 = ℎ − ℎ
𝑟 − 𝑇 𝑟𝑘
(𝑠 − 𝑠 𝑟 ) Eq. 3.7
where h and s denote the enthalpy and entropy of the substance (e.g., steam or hydrocarbon vapor) at
turbine inlet conditions with temperature T, Trk is the absolute reference temperature (Kelvin), and sr is
the entropy of liquid phase (water or hydrocarbon liquid) at the reference temperature. For mass m of
the substance, the available work is therefore given by:
𝑊 𝐴 = 𝑚𝑒 = 𝑚 [ℎ − ℎ
𝑟 − 𝑇 𝑟𝑘
(𝑠 − 𝑠 𝑟 )] Eq. 3.8
The available work, as used in the USGS method (Brook et al., 1979), is computed by replacing mass m in
Eq. 3.8 by mw from Eq. 3.6 and combined with Eq. 3.4, as seen by Eq. 3.9 below;
𝑊 𝐴 ,𝑈𝑆𝐺𝑆 = 𝑅 𝑓 𝐴 ℎ[∅𝜌 𝑙 𝑐 ℎ𝑙 + (1 − ∅)𝜌 𝑟 𝑐 ℎ𝑟 ](𝑇 𝑅 − 𝑇 𝑟 ) [1 − 𝑇 𝑟𝑘
(𝑠 𝑅 −𝑠 𝑟 )
(ℎ
𝑅 −ℎ
𝑟 )
] Eq. 3.9
In this Eq. 3.9, TrK is the absolute reference temperature (TrK = 15+273.15=288.15 K) (in the case of
the reference temperature is chosen as 15
o
C), ℎ
𝑅 and ℎ
𝑟 represent the specific enthalpy values of
the liquid at reservoir temperature value and the reference temperature value respectively (kJ/kg).
𝑠 𝑅 and 𝑠 𝑟 represent the specific entropy values of the liquid at the reservoir temperature and the
reference temperature values (kJ/kg-oC) respectively.

The MIT method is based on the first law of thermodynamic disregarding the enthalpy and entropy of
the system. According to the MIT method, for mass m of the substance, the maximum available work
(WA, kJ) is calculated as in Eq. 3.10 below.
𝑊 𝐴 = 𝑚 [ℎ − ℎ
𝑟 ] Eq. 3.10
Using Eqs. 3.4 and 3.6 in 3.10 the resultant Eq. becomes as in Eq. 3.11 as follows;
𝑊 𝐴 ,𝑀𝐼𝑇 = 𝑅 𝑓 𝐴 ℎ(𝜌 𝑐 𝑃 )
𝑅 (𝑇 𝑅 − 𝑇 𝑟 ) Eq. 3.11
38

According to the MIT method, the thermal conversion efficiency for binary systems is calculated as
below in Eq. 3.12 (MIT, 2006);

𝑐 = 0.09345𝑇 𝑅 − 2.32657 Eq. 3.12
where T R is the resource temperature in
o
C.
To calculate the electric production potential (PW, MWe) from the reservoir, we need to use Eq.
3.13 below;
𝑃𝑊 =

𝑐 𝑊 𝐴 10
3
𝐿 𝐹 𝑡 𝑃 Eq. 3.13
where 𝐿 𝐹 = load factor (indicating the percentage of time spent in working condition for the power
plant) (fraction), 𝑡 𝑃 is the project life (indicates the whole lifetime of the power plant) (seconds) and
the 10
3
is the conversion factor from kW to MWe.

3.1.1 Garg and Combs Method
Garg and Combs (2015) have stated that in the usage of USGS volumetric method, arbitrarily chosen Tr
reference temperature values (15 or 40 oC) and inappropriate use of the conversion efficiencies cause
divergence from the realistic values of recoverable heat Qw(kJ) at the wellhead and electricity
production (PW, MWe) results.
They suggest that turbine inlet pressure should be set equal to separator pressure (Garg and Combs,
2015). They introduce the concept of an “abandonment temperature,” which is defined as the
temperature below which a geothermal reservoir will not be produced. Values of abandonment
temperature depend on the specific power cycle. Moreover, they suggest that the reference
temperature values should not be chosen arbitrarily, they must be based on the power plant type.
For a flash type power plant, the lower limit for the abandonment temperature is equal to the
saturation temperature Tsep (oC) corresponding to the saturation pressure Psep (bar) at the separator
(T abn = T r = T sep).
39

In flash type power plants, the reservoir fluid with the average reservoir temperature is separated by
separators at a separator temperature. The separated brine is re-injected to the reservoir while the
steam is used for electricity production (Garg and Combs, 2015).
Garg ve Combs (2015) gave the Eq. 3.14 below for the calculation of available work (W A,FLASH) for flash
type power conversion systems;
𝑊 𝐴 ,𝐹𝐿𝐴𝑆𝐻 =
𝑅 𝑓 𝐴 ℎ(𝜌 𝑐 𝑃 )
𝑅 (𝑇 𝑅 − 𝑇 𝑠𝑒𝑝 )
ℎ
𝑔𝑙
𝑇 𝑠𝑒𝑝 {ℎ
𝑠𝑡𝑚 (𝑇 𝑠𝑒𝑝 ) − ℎ
𝑤 (𝑇 𝑐 ) − 𝑇 𝑐𝐾
[𝑠 𝑠𝑡 𝑚 (𝑇 𝑠𝑒𝑝 ) − 𝑠 𝑤 (𝑇 𝑐 )]} Eq. 3.14
where ℎ
𝑔𝑙
𝑇 𝑠𝑒𝑝 denotes the heat of vaporization (kJ), ℎ
𝑠𝑡𝑚 and 𝑠 𝑠𝑡𝑚
denote the enthalpy (kJ/kg) and
entropy values (kJ/kg-
o
C) of the steam at T sep respectively, ℎ
𝑤 and 𝑠 𝑤 denote the enthalpy (kJ/kg) and
entropy (kJ/kg-oC) values of liquid phase at condenser temperature T c respectively, T cK denotes the
absolute condenser temperature (K).
All liquid and steam properties are evaluated along the saturation line. The values are given by the
International Association for the properties of Water and Steam (IAPWS-IF97, 1997) should be used for
this purpose.
Since the Kavaklidere geothermal field has high reservoir temperature and it is suitable for single or
double flash power cycles, we do not consider binary plants calculations here. For more information
Garg and Combs, 2015 should be revised.
Garg and Combs, 2015 suggested conversion efficiency value should be used as 0.7-0.8; 0.75 as the
average for flash and binary power cycles.
As a demonstrative example, a geothermal resource with 240
o
C is considered. The purpose of this
example is to correlate the available work outputs of this geothermal power plant by using the Garg
and Combs flash methods in comparison to the USGS and MIT methods results.
For the geothermal fluid having 240
o
C resource temperature (TR) which is proper for single flash
conversion system, the turbine inlet pressure is assumed at 5 bars so the corresponding saturation
temperature is 151.831
o
C. Thermodynamic properties are determined as assuming that the turbine
inlet pressure is equal to the separator pressure, and the condenser temperature (Tc) is 40
o
C.
40

Table 3.1: Input parameters for normalized work calculations from USGS, MIT, and Garg&Combs Flash
Methods
Input Parameters for normalized
work (WA)n, kJ
USGS MIT Garg & Combs
Flash
Resource Temperature, (TR, oC)
240 240 240 240
Resource Pressure, (pR, bar)
300 300 - -
Reference Temperature, (Tr, oC)
15 40 15-40 -
Absolute reference temperature (TrK, K)
288.15 313.15 - -
hR (kJ/kg)
1042.61813 1042.61813 - -
hr (kJ/kg)
62.98365 167.54105 - -
sR (kJ/kg-oC)
2.64901 2.64901 - -
sr (kJ/kg-oC)
0.22447 0.57243 - -
Condenser temperature (Tc , oC)
- - - 40
Plant seperator temperature (Tsep, oC)
- - - 151.831
Plant seperator pressure (psep, bar)
- - - 5
hstm (Tsep=151.831 oC) (kJ/kg)
- - - 2748.10141
sstm(Tsep= 151.831 oC) (kJ/kg-oC)
- - - 6.82063
hw (Tc=40 oC) (kJ/kg)
- - - 167.54105
sw(Tc=40 oC) (kJ/kg-oC)
- - - 0.57243
hgl (Tsep= 151.831 oC) =(hstm-hw )
- - - 2580.56036
TcK, K @ (Tc=40 oC)
- - - 313.15

For the simplicity, normalized available work (Eq. 3.15) outputs computed by using USGS (by
assuming two different reference temperatures as 15 and 40
o
C) (Eq. 3.9), MIT (Eq. 3.11) and Garg
and Combs flash (Eq. 3.14) are presented at Table 3.2 below;
(𝑊 𝐴 )
𝑛 = 𝑊 𝐴 /𝑅 𝑓 𝐴 ℎ(𝜌 𝑐 𝑃 )
𝑅 Eq. 3.15

Table 3.2: Normalized available work (WA)n for USGS, MIT, Garg and Combs single flash power cycle
as a function of resource and reference temperatures
Resource Temperature
TR (
o
C)
Reference
Temperature Tr (
o
C)
Normalized Work (WA)n,
o
C
(W A,USGS) n (W A,MIT) n (W A,G&C FLASH) n
240 15 60.30 225 -
40 46.81 200 21.32

41

Based on the results presented on Table 3.2, the normalized available work obtained from the Garg and
Combs method for the single flash is 35.4% of that computed by using USGS method with 15
o
C
reference temperature, 45.5% of that computed by USGS method with 40
o
C reference temperature.
In comparison to the MIT method results, the normalized available work obtained from single flash Garg
and Combs method is 9.5% of that computed by using the MIT method with a 15
o
C reference
temperature, 10.7% of that computed by the MIT method with 40
o
C reference temperature.
These results apparently show that both USGS and MIT methods overestimate the available work if
we do not consider the power plant type used in predicting the electric potential of a geothermal
reservoir as well as the abandonment temperature as the reference temperature for which the
chosen power plant will not be operated.
Normalized available work is calculated as 21.32
o
C by using Eq. 3.15 for single flash power
conversion system where resource temperature is 240
o
C with the input parameters given in Table
3.1. Because of mechanical losses in the generator, only 75% (in average) of the available work
could be converted to electricity. 75% conversion efficiency value determines how much of the
produced work on the wellhead could be converted into energy. Garg and Combs (2015)
recommended taking the conversion efficiency value to be assumed as 75% in the usage of their
approximation. So, at the wellhead, only 15.99
o
C normalized energy could be obtained out of the
total available normalized work (21.32) for the single flash system.
To be able to get approximate results to the Garg and Combs outputs from USGS method for the
producible electricity, one needs to use approximately 34.2% conversion efficiency value with 40
o
C
reference temperature and if one uses MIT method for producible electricity potential then 8%
conversion efficiency should be used with 40
o
C reference temperature regardless from the resource
temperature value. But the best way of estimation of electric potential is the Garg and Combs
(2015) method which gives the most approximate results to reality by considering the real
thermodynamic condition of the power plant conversion cycle with changing approximations
accordingly to resource temperature.

42

3.2 Probabilistic Assessment

Probabilistic method is based upon a range of maximum and minimum values for each property.
Typically, a simple deterministic method is used. However, the probabilistic approach is desirable
because the reservoir is not homogeneous, and there is an uncertainty in the data. In recent times, the
probabilistic approach is being used with increasing frequency.
In reservoir related studies, the degree of uncertainty is introduced in calculations by assigning a range
of probabilities attached to a parameter or event. A probability distribution function can be developed
for a relevant parameter, such as reservoir porosity or field size. This function is based on the frequency
of occurrence of various values of the parameter by observation, experience, rational belief or intuition
(Satter et al.,2008).

3.2.1 Monte Carlo Simulation

There is never enough well, log, and core data available to accurately determine the average input
values, such as porosity and fluid saturation. Unlike the deterministic method, which uses average
properties of the wells to calculate the original oil or gas in place, the probabilistic approach assigns a
range of values for each variable. These values range from a minimum to a maximum, with some
statistical distribution to compute the probability of possible answers. Reservoir parameters such as
porosity, net thickens, and hydrocarbon saturation are found to fall in certain quantifiable probability
distribution patterns. These patterns could be triangular, random, normal or lognormal among others. It
is a common practice in the industry to utilize a Monte Carlo simulation based on distributions of
reservoir properties. This simulation is used to generate a large set of values for the sought result, such
as the oil in place, and then assign a range of probability values to it. (Satter et al.,2008).
It is customary to report predict answers for the 10%, 50% and 90% cumulative probabilities. The
technique needs a sizeable amount of data, which is not available early in the life of the reservoir.
The probabilistic approach differs from the deterministic method as follows (Satter et al.,2008):
- The probabilistic result is a statistically significant distribution of the possible answers.
- An S-curve plot of all the results assigns a probability to each of the possible answers.
43

- Predicted results are usually reported for the 10%, 50% and 90% cumulative probabilities.
- The analysis typically takes 500 to 10,000 simulations to generate a smooth S-curve.
- Variables are confined to limited ranges, often with a most-likely value.
- Random numbers are used to assign precise values for each input variable.
- A different set of random number is used for each successive calculation.
Data have been used to make an approximation of the electric production potential of a liquid-
dominated geothermal reservoir (Eq. 3.13, PW) also includes uncertainties in the input parameters.
However, it could be said that some parameters used in this approximation have much larger
uncertainty ranges in comparison to the other used parameters in some relationship with them.
Table 3.3 classifies the input parameters as “data with large uncertainty ranges” and parameters as
“data with very less or no uncertainty” when predicting the stored heat-in place and electric power
potential of a geothermal system.
Table 3.3: Classification of input parameters of PW, MWe, concerning uncertainty (Onur, 2015)
Data with large uncertainty Data with less or no uncertainty
Reservoir Area (A, m
2
) Reference Temperature (Tr,
o
C)
Reservoir Thickness (h, m) Conversion efficiency ( c, fraction)
Reservoir Temperature (T R,
o
C) Volumetric heat capacity of rock (kJ/m
3 o
C)
Thermal Recovery Factor (Rf, fraction) Plant load factor (L F, fraction)
Plant or project life, (t p, seconds)

Especially In the early phases of exploration or development of the field, the uncertainty is also large for
area, thickness, temperature and recovery factor due to the lack of data and practical field experience
related to reservoir characterization via seismic evaluations, well logs and test data. The lower limit of
the recovery factor (Rf) should be considered as “0” (zero) because it is one of the possible outcomes
that there may be no permeability in the prospective formation (Garg and Combs, 2010).
44

For triangular data distribution, most likely values which are more likely to occur should be
estimated and then the maximum and minimum values should be estimated as the top and base
limits would be used by the software in random number generation process. If the decision makers
do not have any idea about the mostly represented value of the uncertain input parameter, they
should use a uniform distribution by just giving the same chance or probability to each value to
happen in between the top and base limits. Normal or lognormal distributions also could be used if
they have strong ideas about the mean and standard deviation of the uncertain input parameter.
Figure 3.1 shows the general workflow to be followed while working with the Monte Carlo
simulation to make a probabilistic approach to the potential of the electric production potential of a
geothermal system.

Figure 3.1 Schematic of the Monte Carlo uncertainty analysis (modified from Williams et al., 2008)

For this thesis, the Monte Carlo simulation is made by using add-in Palisade@RISK software to Microsoft
EXCEL by making 10000 iterations and by using the flash method of Garg and Combs, 2015. Monte Carlo
simulation gives results by using statistical markers P90 (proved), P50 (probable) and P10 (possible) to
propagate the uncertainty (Sanyal and Sarmiento,2005). These markers represent the probability
45

(chance of occurrence) of the estimated results belong to real geothermal reserve. "P" stands for the
percentile of probability.

3.2.2 Input Parameters and Scenarios

The Kavaklidere geothermal field is considered as conventional/hydrothermal geothermal case and EGS
geothermal case separately to calculate the power generation capacity of each case. The following data
and the assumptions are taken into consideration for the calculations;
Rock Porosity, Density, and Specific Heat Capacity:
As mentioned before in section 2.2.1, fractured rocks of the Menderes Massif, primarily such as marbles
and schists are the reservoir rocks for the hydrothermal case. For EGS case, from 3 to 5 km reservoir
depth, reservoir rock is gneissic granite which has low porosity and permeability. Table 3.4 shows the
rock types in the field and their values for density, porosity and specific heat capacity from the
literature. All of them were defined by triangular distribution.
Table 3.4: Rock density, porosity and specific heat capacity values for Kavaklidere geothermal field
lithologies (*Gurel et al., 2016, Crain, P. Eng. Crain’s Petrophysical Handbook, Miao, S. Q., Li, H. P., &
Chen, G.,2014)
Rock Type Density, g/cm
3
Porosity, frac
Specific Heat
Capacity, kJ/kg
o
C
Marble 2.58-2.74-2.93 0.03-0.07 * 0.88
Schist 2.73-3.19 0.02-0.122 * 1.10
Mica schist 2.79-2.83 0.065 1.00
Phyllite 2.79-2.89 0.08 0.98
Granite 2.45-2.65-2.94 0.005-0.015 0.95
Gneiss 2.64-2.73-2.91 0.005-0.015 1.02
Quartzite 2.54-2.65-2.72 0.001-0.005 0.85
Granodiorite 2.67-2.78 0.02 1.02

When we adapt the values from Table 3.4, the minimum, most likely and maximum porosity values are
taken as 3%, 7% and 12.2% respectively for the hydrothermal case; and are taken as 0.5%, 1.5% and 2%
respectively for EGS case.
46

The minimum, most likely and maximum rock density values are taken as 2640, 2740 and 2940 kg/m
3

respectively for the hydrothermal case; and are taken as 2640, 2730 and 2910 kg/m
3
respectively for
EGS case.
The minimum, most likely and maximum specific heat capacity values are taken as 0.88, 0.95 and 1.1
kJ/kg
o
C respectively for the hydrothermal case; and are taken as 0.95, 0.98 and 1.02 kJ/kg
o
C respectively
for EGS case.
Area and thickness:
For both cases, the minimum, most likely and maximum area values are taken as 40, 43 and 45 km
2

respectively (Ozdemir et al.,2016., Akin,2017).
From the interpretations of MAK-3 and MAK-14 well log data (Appendix A.1 and A.2; Burcak, 2011)
between 1250m-1750m, 1250m-2250m, and 1250m-2750m can be reservoir thickness. So, the
minimum, most likely and maximum reservoir thickness values are taken as 500, 1000 and 1500 m
respectively for the hydrothermal case; are taken as 500, 1500 and 2000 m respectively for EGS case
with considering to reach 5km depth.
Temperature:
In the Kavaklidere field, the minimum bottom hole temperature is 188
o
C (KG-1 well), and the maximum
bottom hole temperature is 287
o
C (MAK-14 well) as seen in Table 2.2. These values are taken as
minimum and maximum reservoir temperature respectively. When we consider the temperatures of the
other wells in the field, also the hydrogeochemical and hydrogeological studies (Appendix B.4, Bulbul et
al.,2011) we assign 240
o
C as most likely value.
Burcak, 2015 stated that the temperature gradient of the field is 98
o
C/1000m and they expect to have
450-500
o
C bottom hole temperature at 5000 m. When we consider this gradient for 3-5km depth, the
temperature range will be 294-490
o
C. For more realistic EGS calculations, the minimum, most likely and
maximum temperature values are taken as 240, 287 and 350
o
C respectively.
Fluid Density and Specific Heat Capacity:
Reservoir fluid density and specific heat capacity changes depending on the reservoir temperature. We
assign the minimum, most likely and maximum values of these parameters considering the reservoir
temperature. For instance, when reservoir temperature is 188
o
C, fluid density is 878 kg/m
3
(Table 3.5).
47

Since EGS reservoir is a hot dry rock, the reservoir fluid will be the fresh water injected from the surface
(Table 3.5). IAPWS-IF97 were used for water properties.
Recovery Factor:
As mentioned in detail in section 3.1, the minimum, most likely and maximum recovery factor values are
taken as 0.07, 0.18 and 0.24 respectively for the hydrothermal case.
There is one published analysis for EGS based upon field data (Grant & Garg 2012), which indicates
lower recovery factor, 1.6%. It would be expected that EGS would have lower recovery than naturally
fractured systems, as the fractures will be less pervasive. Recovery factors for EGS should be at best a
few percents (Grant, 2015). So, the minimum, most likely and maximum recovery factor values are
taken as 0.01, 0.016 and 0.02 respectively for EGS case.
Project life is generally assumed to be 30 years (946728000 seconds). For electricity generation, power
plants are active except the maintenance time since geothermal can provide base load. Thus, the load
factor is taken as 0.95. And, the conversion factor is taken as 0.75 (Garg and Combs, 2015).
Table 3.5 shows all the input parameters mentioned above for both hydrothermal and EGS cases.
Table 3.5: Input parameters for resource assessment of Kavaklidere-Hydrothermal and Kavaklidere-EGS
by using Garg and Combs flash method with Monte Carlo simulation

Case1 - Hydrothermal Case2 - EGS
Parameters PDF Min Mode Max PDF Min Mode Max
Porosity Triangular 0.03 0.07 0.122 Triangular
0.005 0.015 0.02
Rock Density, kg/m3 Triangular 2640 2740 2940 Triangular
2640 2730 2910
Area, km2 Triangular 40 43 45 Triangular 40 43 45
Thickness, m Triangular 500 1000 1500 Triangular
1000 1500 2000
Treservoir, C Triangular 188 240 287 Triangular
240 287 350
Fluid Density, kg/m3 Triangular 737 813 878 Triangular
712 737 813
Recovery factor Triangular 0.07 0.18 0.24 Triangular
0.01 0.016 0.02
Spec Hrock, kJ/Kgc Triangular 0.88 0.95 1.1 Triangular
0.95 0.98 1.02
Spec Hfluid, Kj/Kgc Triangular 4.44 4.77 5.43 Triangular
4.77 5.43 5.75
Proj Life, Seconds Constant 9.46E+08 Constant 9.46E+08
Load Factor Constant 0.95 Constant 0.95
Conversion Factor Constant 0.75 Constant 0.75
Condenser Temperature
(Tc,
o
C)
Constant 40
Plant Seperator Constant 151.831
48

Temperature (Tsep,oC)
Plant Seperator Pressure
(Psep, Bar)
Constant 5
Hstm (Tsep=151.831 oC)
(kJ/kg)
Constant 2748.101
Sstm(Tsep= 151.831 oC)
(kJ/kg-oC)
Constant 6.82063
Hw (Tc=40 Oc) (kJ/kg) Constant 167.5411
Sw(Tc=40 Oc) (kJ/kg-oC) Constant 0.57243
Hgl (Tsep= 151.831 oC)
=(Hstm-Hw)
Constant 2580.56
Tck, K @ (Tc=40 oC) Constant 313.15

For hydrothermal case, 13 different scenarios were run with 10000 iterations each by Monte Carlo
simulation to see the difference between USGC, MIT, and Garg and Combs flash methods.
For scenario #1, the input parameters in Table 3.5 were used for Garg and Combs method.
For scenario #2, uniform distributions were used for each parameter with the same minimum and
maximum values in Table 3.5 to see how the distribution effects on the results.
For scenario #3 and #4, the input parameters from Basel,2010 were used (Table 3.6). For scenario #3,
we considered these parameters as early stage parameters, and the minimum, most likely and
maximum recovery factor values are taken as 0, 0.05 and 0.15 respectively. For scenario #4, the
minimum, most likely and maximum recovery factor values are taken as 0.07, 0.18 and 0.24 respectively.
Table 3.6: Reservoir parameters for Kavaklidere from Basel, 2010

For scenario #5,6,7,8, USGS method was used for different reference temperatures and conversion
factors (Table 3.7). Reservoir parameters are same as scenario #1.
Parameter Distribution Min Mode Max
Reservoir volume (V, m
3
) Triangular 2E+10 4E+10 6E+10
Reservoir Temperature (TR,
o
C) Triangular 180 215 240
Porosity (φ, fraction) Triangular 0.03 0.06 0.08
Fluid density (kg/m
3
) Constant 854.1
Rock density (kg/m
3
) Triangular 2650 2750 2950
Spec Hrock (kJ/kg
o
C) Triangular 0.85 0.90 0.95
Spec Hfluid (kJ/kg
o
C) Constant 4.18
49

For scenario #9,10,11,12,13, MIT method was used for different reference temperatures and conversion
factors (Table 3.7). Reservoir parameters are same as scenario #1.
For EGS case, three different scenarios were run with 10000 iterations each by Monte Carlo simulation
with Garg and Combs flash method.
For scenario #1, the input parameters in Table 3.5 were used.
For scenario #2, we considered ‘0’ recovery factor for early stage scenario with same input parameters
in Table 3.5. Then, the minimum, most likely and maximum recovery factor values are taken as 0, 0.01
and 0.02 respectively.
For scenario #3, we considered project life as 20 years with same input parameters in Table 3.5.

3.2.3 Results and Discussion
Table 3.7 summaries all the results of each case scenarios.
Analysis of the results of Monte Carlo simulation shows that Kavaklidere-hydrothermal geothermal field
has an electrical power capacity of 182.32 MWe with 90% probability (proved), 321,.68 MWe with 50%
probability (probable) and 525.19 MWe with 10% probability (possible) (Scenario #1, Figure 3.2, Table
3.7)
The result of scenario #2 shows that uniform distribution underestimated proved and possible capacity
of the field.
When Basel, 2010 conducted her study, there were only tree wells (AK-1, KG-1, AK-2) drilled in the field.
We considered her input parameters as an early scenario case. The result of scenario #3 shows that with
zero recovery factor, the capacity of the field is low, which is the logical and expected result.
Basel, 2010 used MIT method with 100
o
C reference temperature and 0.17 conversion factor to calculate
electrical power capacity of Kavaklidere-hydrothermal. She found 264.3 MWe (P90), 446.4 (P50) and
695.6 (P10). If she applied Garg and Combs flash method (scenario #4), the result would be 112.49 MWe
(P90), 193.12 MWe (P50) and 305.97 (P10). It shows that she overestimated the capacity of the field.
50

When we use 100
o
C reference temperature and 0.17 conversion factor for today’s parameters (scenario
#13), the results are 299.28 MWe (P90), 493.40 MWe (P50), and 739.13 MWe (P10). This shows we
overestimate the capacity of the field more than 50%.
The results of scenario #5,6,7,8 show that how we can get different results with different reference
temperatures and conversion factors. Some values are underestimated, and some are overestimated
(Table 3.7).
Same with scenario #9,10,11,12, the results show different capacity values for different reference
temperatures and conversion factors (Table 3.7). The results of Scenario #11 (MIT, Tref=40°C, =0.08)
and scenario #12 (MIT, Tref=100°C, =0.12) are the closest results to our base case scenario (Garg and
Combs, scenario#1). This shows if we chose 40
o
C as a reference temperature then we need to choose
0.8 as conversion factor; or if we chose 100
o
C as a reference temperature, then we need to choose 0.12
as a conversion factor. So, there is no specific rule behind it.
At that point, we should point out that the MIT and USGS methods do not consider the real power
conversion system as flash and binary, separator temperature, pressure or enthalpy (for MIT) or
entropy values appropriately designated. So logically the best estimation method is Garg and
Combs approximation considering the real working system thermodynamic properties.

Analysis of the results of Monte Carlo simulation shows that Kavaklidere-EGS geothermal field has an
electrical power capacity of 54.59 MWe with 90% probability (proved), 79.48 MWe with 50% probability
(probable) and 113.84 MWe with 10% probability (possible) (Scenario #1, Figure 3.3, Table 3.7)
The results of scenario #2 show the values for zero recovery factor, and the results of scenario #3 show
the values for 20 years project life.
Also, a sensitivity analysis was run for both cases to see which input parameters have more effect on the
results. It shows that thickness, recovery factor and area have the significant effects on the outputs
among the other input parameters. This is also proof of Table 3.3, which shows Rf, thickness, and area
are the data with large uncertainty. When we have more specific data and information for these
parameters, the results will be more confident and realistic. It is important to update these parameters
with petrophysical, well log and production data.
51

Figure 3.2: Histogram and Cumulative Probability graph for electrical power generation (MWe) of
Kavaklidere-hydrothermal

Figure 3.3: Histogram and Cumulative Probability graph for electrical power generation (MWe) of
Kavaklidere-EGS

Probability, %
Frequency Frequency
Probability, %
52

Table 3.7: The results of volumetric probabilistic estimations of geothermal reserves by Monte Carlo
Simulation Garg and Combs,2015, USGS and MIT methods for Kavaklidere geothermal field

3.3 TOUGH2-PETRASIM Reservoir Simulation
Numerical modeling is a fundamental and strongly interacting instrument in geothermal reservoir
engineering that studies the geothermal reservoir behavior by solving the balance equations of mass,
momentum, and energy in the volume in which hydrothermal circulation of fluid occurs. Consequently,
this tool is very important for the definition of a geothermal reservoir potential, and for understanding
the hydrogeological behavior and heat transport in the reservoir under a specific utilization rate,
allowing us to define and progressively modify the management strategy of the geothermal field, an
optimal location of production and reinjection wells and the best reinjection strategy (Franco & Vaccaro,
2013).
The TOUGH2 (Transport of Unsaturated Groundwater and Heat) numerical simulator (Pruess et al. 1999)
was used for numerical modeling of extraction and recovery periods in the fractured system. TOUGH2 is
a general purpose numerical simulation program for multi-dimensional non-isothermal flows of multi-
phase, multicomponent fluid mixtures in fractured and porous media. The code was developed at the
53

Lawrence Berkeley National Laboratory and is written in standard FORTRAN77. It employs an integral
finite difference method (IFDM) in discretizing the medium, which has the advantage of irregular
discretization in multi-dimensions. Time is discretized fully implicitly as a first-order backward finite
difference and fluxes are computed using upstream weighing (Pruess et al. 1999). PetraSim
(Thunderhead Engineering, Manhattan, KS) was used as pre- and post-processing software at the front
end. PetraSim allows to interactively define the mesh and parameters for the model and then creates an
input file for the TOUGH2 code. After running the TOUGH2 code, PetraSim can be used to process the
simulation results for graphical representation.
This thesis presents the hydrothermal and water-based EGS reservoir simulation. For each model, a
different file was created in PetraSim specifying the simulator, equation of state (EOS) module and
extents of the model. The simulator mode in all cases was set as TOUGH2. In the case of the EOS,
PetraSim supports many EOS options for defining the possible phase condition of the reservoir (Thunder
Head Engineering, 2017). For the study, the ‘EOS1: Water, Non-Isothermal’ type was selected for all the
models, assuming that the geothermal fluid is pure water along the reservoir. All water properties in the
TOUGH2 model obtained from steam table equations as given by the International Formulation
Committee (1997) (Pruess, Oldenburg, & Moridis, 2012).

After selecting the module, there are several steps involved in the development of the reservoir
simulation model. These include;
Step 1 - Selecting the dimensions and shape of the reservoir model: The reservoir shape and dimensions
of the reservoir model are referred to as reservoir domain. This step also involves dividing the domain
into a pre-selected number of layers.
Step 2 - Applying the initial and boundary conditions: Initial and boundary conditions are applied to the
top layer and base layer of the model. For EOS1 single phase initial conditions, temperature (
o
C) and
pressure (Pa) values should be specified.
Step 3 – Creating solution mesh: In the case of the Z dimension, divisions were set up using the
“Regular” function where a fraction of each layer was divided into a specified number of cells. As for the
X and Y dimensions, PetraSim provides three types of solution meshes (Thunder Head Engineering,
2017):
54

• Regular – cells are rectangular hexahedrons.
• Polygonal – uses extruded Voronoi cells to conform to any boundary and support refinement
around wells.
• Radial – represents a slice of an axisymmetric cylindrical mesh.
For the present study, a polygonal mesh was selected, entering the maximum area of the mesh
elements, adjacent wells and boundaries of the model. It was decided to use this type of mesh among
the others, for its fewer requirements of model elements and its shorter run times.
Step 4 – Assigning material properties: Materials are used to define the permeability and other
properties such as rock density, porosity, heat conductivity and specific heat capacity.
Step 5 – Defining cell-specific data: such as material, sources, sinks, and initial conditions.
Step 6 - Setting the solution and output options.

3.3.1 Input Parameters for Conventional Reservoir
These simulations were performed assuming a similar geological volume, consisting of an area of 1 km
2
and a thickness of 500 m. A 3-D, the five-spot configuration was selected for the reservoir, since this is
the geometry used in similar geothermal investigations. The symmetry of the computational grid
reduces the modeling to 1/2th of the system domain. A diagram of the five-spot configuration is
included in Figure 3.4. All sites were simulated as porous media.
For the Kavaklidere-hydrothermal reservoir, data is derived from MAK-14 well. The initial conditions of
temperature and pressure were 287°C and 300 bar, respectively. MAK-14 well production rate is 35 l/s.
Two different production scenarios were modeled and run for comparison. The purpose is to see
different production rates effects on the temperature change of the reservoir and the amount of
thermal energy or heat extraction from the production well. Table 3.8 shows the input simulation
parameters for two scenarios. Simulations were run for a project life period of 30 years.

55

Figure 3.4: Five-spot reservoir configuration and the reservoir model used for Kavaklidere-hydrothermal

Table 3.8: Simulation input parameters for Kavaklidere-hydrothermal

3.3.2 Results and Discussion
After simulating the base case scenario, as seen in Figure 3.5 PetraSim demonstrates the temperature
change of the reservoir for 25 years, and Figure 3.6 shows the reservoir at the end of 30 years.

Simulation input parameters
Base
Scenario

Scenario#2
Rock density, kg/m3 2740
Same as
Scenario#1
Porosity, frac 0.07*
Permeability, mD 1540*
Rock Thermal conductivity, W/mK 2**
Rock Specific Heat capacity, J/kgK 950**
Reservoir temperature,
o
C 287 287
Reservoir pressure, bar 300 300
Injection temperature,
o
C 25 25
Production rate, kg/s 35 70
*Gurel et al.,2016. ** https://www.engineeringtoolbox.com/
56

Figure 3.5: Simulation Results of base case scenario for Reservoir Temperature Distribution after a) 1-
year b) 5-year c) 10-year d) 15-year e) 20-year f) 25-year
We can produce heat/fluid from a geothermal resource at different extraction rates. Excessive
production could bring economic benefits, like earlier return of investment, but could also lead to
resource depletion or even deterioration. However, by using moderate production rates, which consider
the local resource characteristics (field size, natural recharge rate, etc.), the longevity of production can
be secured and sustainable production achieved. (Rybach, 2007)
a) b)
c) d)
e)
f)
57

Figure 3.6: Simulation results of base case scenario for reservoir temperature distribution at the end of
30 year

Figure 3.7 shows that the temperature profile of the reservoir from the production well. After 30 years
with 35kg/s production rate, the reservoir bottom hole temperature is 164
o
C, and with 70 kg/s
production rate it is 112
o
C.

Figure 3.7: Comparison of temperature profiles for two scenarios

Figure 3.8 shows the estimated thermal energy, MW from the production well for both scenarios. When
we have 35kg/s production rate, average thermal energy is 34.2 MW. When we have 70kg/s production
rate, average thermal energy is 79.8 MW. This shows higher production rate provides more thermal
58

energy from the conventional geothermal field, but at the same time, it causes more temperature drop
in the reservoir. So, the optimum value should be found.

Figure 3.8: Comparison of estimated thermal energy from production well for two scenarios

3.3.3 Input Parameters and Scenarios for EGS
Since there is no study yet for EGS in Kavaklidere field, the purpose here is to see how we can adapt
hydraulic fracturing to the field with limited data from the conventional part. The objective of this part is
to demonstrate the potential of energy capacity of the field with different possible production and
stimulation scenarios. After 3000 m depth in the field, reservoir rock is gneissic granite for EGS.
These simulations were performed assuming a similar geological volume 1500x1000x1000, consisting of
an area of 1.5 km2 and a thickness of 1000 m (4 to 5km depth). A 3-D, triplet configuration, one
injection well and two production wells, was selected for the reservoir (Figure 3.9).
The assumptions for calculations and simulation models are following;
- All input parameters were taken from the literature since no data is available.
- The reservoir has initial conditions of 350°C and 65 MPa at 5000 m, and 300
o
C and 60 MPa at
4500 m.
- In all three wells, hydraulic fracturing is executed in the rock formation, so that an entire
-120
-100
-80
-60
-40
-20
0
0.00E+00 2.00E+08 4.00E+08 6.00E+08 8.00E+08 1.00E+09
Energy (MW)
time, s
Energy (MW) for Kavaklidere-Hydrothermal
Scenario#2 Scenario#1
59

connected fractured zone is generated. Although unrealistic, here we assume that zones are
perfectly stimulated with zero water loss. The fractured zone has a height of 250 m (−4700 to
−4950 m) and length of 1500 m (x) and 1000 m (y). The perforation interval for the injection and
production is at a depth of between −4700 and −5000 m.
- MIT, 2006 stated that at least 50 kg/s production flow rate is necessary for an economical
geothermal project. For this reason, production flow rate of 50 kg/s and 100 kg/s per well are
used.
- Water at 25°C is pumped into the injection well with a stimulation wellhead pressure of 15 MPa
(Peluchette,2013).

For scenario#1, fracture permeability was assumed 10 mD, the distance is between a production well
and injection well is 750 m.
For scenario#2, production wells are closer to injection well. The distance is 250m.
For scenario#3, fracture permeability is 100 mD.
For scenario#4, the production rate is higher, 100 kg/s per well.
For scenario#5, two zones are stimulated (Figure 3.9).
Table 3.9 shows the input parameters for all different simulation models.

60

Table 3.9: Input parameters for different simulation models of Kavaklidere-EGS

Figure 3.9: Reservoir configuration and the reservoir models used for Kavaklidere-EGS

3.3.4 Results and Discussion
Figure 3.10 demonstrates the temperature distributions of the reservoir for five different scenarios at
the end of 30 years.
61

62

63

Figure 3.10: Simulation results of five scenarios for reservoir temperature distribution at the end of 30
year

Figure 3.11 shows the comparison of estimated thermal energy from one production well for five
scenarios. The comparison of the result of scenario#2 and scenario#3 shows that stimulated zone with
higher fracture permeability provides more energy. The result of scenario#4 shows that higher
production rate gives higher energy. From the result of scenario#5, even though multiple zones
stimulation gives more energy for first five years, after that, it gives the similar energy as a single zone.
64

Figure 3.11: Comparison of estimated thermal energy from one production well for five scenarios

3.4 GETEM Simulation for Economic Evaluation

The Geothermal Electricity Technology Evaluation Model (GETEM), which is provided by U.S Department
of Energy and Idaho National Laboratory in 2006, is used to estimate the Levelized Cost of
Energy/Electricity (LCOE) for definable geothermal scenarios. This Excel-based tool model evaluates
either a Hydrothermal or an EGS resource type with either a flash-steam or air-cooled binary power
plant based on specific resource parameters by considering financial parameters (debt interest, equity
rate of return, etc.), capital costs (site exploration, drilling and re-drilling, reservoir stimulation, and
surface plant facilities), system performance (thermal drawdown rate or reservoir lifetime, well-flow
rate, number of production and injection wells, etc.), and operating and maintenance costs (DOE,2012).
The levelized cost of electricity (LCOE), also known as Levelized Energy Cost (LEC), is the net present
value of the unit-cost of electricity over the lifetime of a generating asset. It is often taken as a proxy for
the average price that the generating asset must receive in a market to break even over its lifetime.
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
0.00E+00 2.00E+08 4.00E+08 6.00E+08 8.00E+08 1.00E+09
Energy (MW)
time, s
Energy (MW) for Kavaklidere-EGS
Scenaro #1
Scenario#2
Scenario#3
Scenario#4
Scenario#5
65

The reservoir simulation result of Kavaklidere-hydrothermal (scenario#1) shows that production well
provides average 34.2 MW thermal energy. When we considered 0.75 as conversion factor and 0.95 as
load factor, we can get 27 MWe electric power energy from one well. We estimated the electric power
production proved capacity of the field as 182.3 MWe. To reach this capacity, we need 7 production
wells.
The reservoir simulation result of Kavaklidere-EGS (scenario#1) shows that production well provides
average 35.1 MW thermal energy. For this calculation, we considered scenario#1 because it gives a
more stable result for 30 years. When we considered 0.75 as conversion factor and 0.95 as load factor,
we can get 27.7 MWe electric power energy from one well. We estimated the electric power production
proved capacity of the EGS field as 54.6 MWe. To reach this capacity, we need 2 production wells like as
we had in the simulation model.
Table 3.10: Input parameters for GETEM simulation
Input Parameters Hydrothermal EGS
Resource temperature,
o
C 287 350
Resource depth, m 2750 5000
Power plant type Flash Flash
Number of production well 7 2
Production rate, kg/s 35 50

GETEM calculated LCOE for Kavaklidere-hydrothermal as 9.38 ₵/ kWh and 11.91 ₵/ kWh for Kavaklidere-
EGS.
A feed-in tariff (FIT) are fixed electricity prices that are paid to renewable energy producers to accelerate
investment in renewable energy technologies. The feed-in tariff, which is 10.5 USD cent/kWh in Turkey
for geothermal electricity guarantees ten years of power purchase following the commencement of the
operations of the generation facility (YEK Law, 2013). It seems with this incentive an EGS project in
Kavaklidere is doable and economical.

66

CHAPTER 4
CONCLUSION AND RECOMMENDATIONS

Electric power supply capability of the Kavaklidere geothermal field located in Alasehir Graben, West
Anatolia, Turkey is studied considering economic and physical reliability using volumetric heat in place
method. The field does have 182.3 MWe conventional proved capacity and 54.6 MWe EGS proved
capacity. This is the first academic study for the resource assessment and reservoir modeling of the
Kavaklidere geothermal field with the aim of electricity production. In these manners, the findings of
this research are promising.
The main conclusions drawn from this study are summarized as follows:
✓ Garg and Combs (2015) volumetric probabilistic calculation method, which considers the real
operating power conversion system with its thermodynamic condition based on the 2nd law of
thermodynamics, is far away from any arbitrary usage of input parameters. In the USGS and MIT
methods, there is no logical reason for not to choose the reference temperature (15, 40, 100
o
C)
and conversion efficiency (0.08, 0.12, 0.4 etc) values arbitrarily.
✓ Distribution types of input parameters such as porosity, thickness, etc. should be defined by
using core or log data if available, or some analogy field data could be used for this purpose.
✓ The level of sustainable production depends on the utilization technology as well as on the local
geothermal resource characteristics. Its determination needs specific studies, especially model
simulations of long-term production strategies, for which exploration, monitoring and
production data are required.
✓ The simulation study shows that well configuration, production rate, fracture permeability and
multiple stimulated/fractured zones have different effects on thermal energy production of EGS.

This presented study is a preparatory work for future academic studies for the Kavaklidere geothermal
field, especially for enhanced geothermal systems (EGS). Considering the potential of the field and the
country with the future required research and government promotions, these systems could be more
prevalent and commercial. To increase the energy security and sustainability in Turkey, increasing
geothermal energy share with EGS would be a smart approach.
Recommendations to improve the present study for future research are as follows:
67

 More data provide more information and accuracy to estimate the minimum, most likely
and maximum values of reservoir properties which has more uncertainty such as area,
thickness, reservoir temperature and recovery factor. For instance, to determine the
reservoir area more precisely, resistivity maps can be obtained through geophysical studies.
Moreover, in this study, reservoir rock properties such as density, porosity, and specific heat
values were gathered from the literature. Collecting samples from the field and examining
them can help to decrease the uncertainty of the input data. Gaining more information,
data and experience from the reservoir and the production develop the numerical model to
simulate the reservoir in the natural state and the response.
 Since the preferred method in reservoir assessment in the early phases of geothermal
development is the volumetric method, Garg, and Combs (2015) method should be used
instead of USGS and MIT methods to avoid of overestimating the electric generation
potential of geothermal fields. Moreover, in case the possibility of almost “0” permeability
of reservoir, the minimum limit of recovery factor as "0” should be considered at the very
early stage of exploration to expect more realistic primary results.
 In this study only triangular and uniform input data distribution were used in Monte Carlo
Simulation; their interaction and effects to the resultant production potential are discussed
in 3rd part. For future work, normal, log-normal or gamma input data distribution types also
can be used and examined.
 In this study, only water is considered as working fluid for EGS. CO2 should be considered
for the future work. Today, there many research for the use of supercritical CO 2, instead of
water, as the geothermal working fluid. Even though CO2 is much more expensive and
somewhat more challenging to work with than water, there are many encouraging
advantages for EGS. For instance, it provides much more power output, minimizes parasitic
losses from pumping and cooling. There should be more rock-fluid interaction studies with
CO2 because it has lower tendency to dissolve minerals and other substances than water,
which helps reducing corrosion and scaling of system components. Moreover, the most
promising advantage of using CO2 is carbon sequestration.
 Because the most of geothermal fields are corporatized in Turkey, it is hard to access the
current data and information since they are confidential. That is the reason this study was
conducted with the limited data. To expand the extent of the further studies, there should
68

be a national database for geothermal as well as for other energy sources which is free for
research purposes.

69

REFERENCES

Akin, S. (2017). Geothermal Resource Assessment of Alasehir Geothermal Field. Proceedings, 42nd
Workshop on Geothermal Reservoir Engineering Stanford University, Stanford, California, February 13-
15, 2017.
Akkus, İ. and Alan, H. (2016). Türkiye’nin Jeotermal Kaynakları, Prospeksiyonlar, Sorunlar ve Öneriler
Raporu (Rapor No. 123). Ankara: TMMOB Jeoloji Mühendisleri Odası Raporu. (in Turkish)
Altin, M. (2017). Probability Based Volumetric "Heat In-Place" Methods for Predicting Power (Electricity)
Generation Potential of Liquid-Dominated Geothermal Systems. (Master’s Thesis, Istanbul Technical
University). Istanbul.
Aminzadeh, F., Tafti, T. A., and Maity, D. (2013). An integrated methodology for sub-surface fracture
characterization using microseismic data: a case study at the NW Geysers. Computers &
Geosciences, 54, 39-49.
Atmaca, İ. (2010). Resource Assessment in Aydin-Pamukören Geothermal Field (Master's Thesis,
Ankara/ Middle East Technical University, 2010). Ankara.
Avsar, O. (2011). Geochemical Evaluation and Conceptual Modeling of Edremit Geothermal Field. (Ph.D.
Thesis, Ankara/ Middle East Technical University, 2011). Ankara.
Basel, E. D. K. (2010). Investigation of Turkey’s geothermal Potential, Ph.D. Dissertation (in Turkish),
Istanbul Technical University, Graduate School of Science, Engineering and Technology, Istanbul, Turkey,
310 p.
Bertani, R. (2015). Geothermal Power Generation in the World. 2010–2015 Update Report World
Geothermal Congress, Melbourne, Australia.
Bozkurt, E. (2001) Neotectonics of Turkey – a synthesis. Geodinamica Acta, 14:1-3, 3-30
BP (2017). New energy supplies for Turkey and Europe: a visual guide to the Southern Gas Corridor. 12
October 2017. https://www.bp.com/en/global/corporate/bp-magazine/locations/visual-guide-to-
europe-southern-gas-corridor-tanap-turkey.html. [Accessed January 27, 2017].
Brook, C. A. Mariner, R.H., Mabey, D.R., Swanson, J.R., Guffanti, M., and Muffler, L.J.P., (1979).
Hydrothermal convection systems with reservoir temperatures ≥90◦C, In:Muffler, L.J.P. (Ed.),
Assessment of Geothermal Resources of the United States—1978, 790. U.S Geological Survey Circular,
170p
Bulbul, A., Ozen, T., and Tarcan,G. (2011). Hydrogeochemical and hydrogeological investigations of
thermal waters in the Alasehir-Kavaklidere area (Manisa-Turkey). African Journal of
Biotechnology, 10(75), 17223-17240.
Burcak, M. (2011). MAK-2011-03 Well Log and MAK-2010-14 Well Log. Appendix A.1 and A.2
Burcak, M. (2015). Kizgin Kuru Kaya (Kkk) Ve Geliştirilebilir Jeotermal Sistemler (Gjs) Ve Türkiye’de
Araştirmaya Uygun Bölgeler. MTA 80. Yıl Sempozyumu, 01-03 Aralık,2015, Ankara. (in Turkish)
70

Crain, P. Eng. Crain’s Petrophysical Handbook [Internet]. 2018. IGNEOUS and METAMORPHIC
RESERVOIRS. [cited 01/01/2018].
DOE (2012). GETEM Manuals and Revision Notes.
https://www.energy.gov/sites/prod/files/2014/02/f7/geothermal_electricity_technology_evaluation_m
odel_2012.pdf [Accessed February 28,2018]
DOE (2017). Office of Energy Efficiency & Renewable energy. Geothermal FAQs.
https://energy.gov/eere/geothermal/geothermal-faqs#why_geothermal_energy_renewable [Accessed
January 27, 2017].
DOE (2017a). Office of Energy Efficiency & Renewable energy. Enhanced Geothermal System Fact Sheet.
https://www1.eere.energy.gov/geothermal/pdfs/egs_basics.pdf [Accessed January 27, 2017].
EIA (2017). Turkey Energy Profile: Important Transit Hub for Oil and Natural Gas – Analysis. February 4,
2017. http://www.eurasiareview.com/04022017-turkey-energy-profile-important-transit-hub-for-oil-
and-natural-gas-analysis/#comments [Accessed January 27, 2017].
EPDK (2017) Republic of Turkey Energy Market Regulatory. Electricity Generation Licenses.
http://lisans.epdk.org.tr/epvys-web/faces/pages/lisans/elektrikUretim/elektrikUretimOzetSorgula.xhtml
[Accessed December 31, 2017]
EPRI (1978) Geothermal Energy Prospects for the Next 50 Years. special report ER-611-SR, February
1978.
Franco, A., & Vaccaro, M. (2013). Numerical simulation of geothermal reservoirs for the sustainable
design of energy plants: A review. Pisa, Italy: Elsevier.
Garg, S. K., & Combs, J. (2010). Appropriate use of USGS volumetric “heat in place” method and Monte
Carlo calculations. In Proceedings 34th Workshop on Geothermal Reservoir Engineering, Stanford
University, Stanford, California, USA.
Garg, S. K., and Combs, J. (2015). A reformulation of USGS volumetric “heat-in-place” resource
estimation method, Geothermics, Vol. 55, pp. 150-158.
Garg, Sabodh, K. and Combs, J. (2011). A Reexamination of USGS Volumetric "Heat in Place" Method.
Stanford, CA, USA: Proceedings, 36th Workshop on Geothermal Reservoir Engineering, Stanford
University.
Geraud, Y., Rosener, M., Surma, F., Place J., Le Garzic E., and Diraison M. (2010). “Physical properties of
fault zones within a granite body: Example of the Soultz-sous-Forêts geothermal site.” Geoscience, vol.
342, p. 566-574.
Gischig,V. S., & Preisig, G. (2015). Hydro-Fracturing Versus Hydro-Shearing: A Critical Assessment of Two
Distinct Reservoir Stimulation Mechanisms. International Society for Rock Mechanics and Rock
Engineering.
Grant, M. A. (2015). Resource assessment, a review, with reference to the Australian
code. Resource, 19, 25.
71

Grant, M. A., & Garg, S. K. (2012). Recovery factor for EGS. In Proceedings of the 37th Workshop on
Geothermal Reservoir Engineering, Stanford University, Stanford, California, USA (pp. 738-740).
Gurel, E., Coskuner, Y., and Akin, S. (2016). Fractal Modeling of Outcrop Fracture Patterns in Alasehir
Geothermal Reservoir Turkey. 41st Workshop on Geothermal Reservoir Engineering Stanford University.
Volume: SGP-TR-209
IAPWS-IF97 (1997). The International Association for the Properties of Water and Steam. IAPWS
Industrial Formulation 1997 for the Thermodynamic Properties of Water and Steam.
INVEST (2017). Invest in Turkey. Investment Support and Promotion Agency of Turkey. Energy and
Renewables. http://www.invest.gov.tr/en-US/sectors/Pages/Energy.aspx [Accessed January 27, 2017].
Kagel, A. (2008). The State of Geothermal Technology, Part II: Surface Technology. January 2008. U.S.
Geothermal Energy Association.
Liu, J., Chen, X., and Guo, W. (2013). Volcanic Natural Resources and Volcanic Landscape Protection: An
Overview. Intech 2013.
Lund, W. J., and Boyd, L. T. (2015). Direct utilization of geothermal energy 2015 worldwide review,
Proceeding World Geothermal Congress, Melbourne, Australia, April 19-25.
Majer, E. L., & Peterson, J. E. (2007). The impact of injection on seismicity at The Geysers, California
Geothermal Field. International Journal of Rock Mechanics and Mining Sciences, 44(8), 1079-1090.

MASPO (2013). Maspo Jeotermal Enerji Santrali 35 MWe, CED Raporu. April 2013. (in Turkish,
Unpublished)

Maity, D., and Aminzadeh, F. (2015). Novel fracture zone identifier attribute using geophysical and well
log data for unconventional reservoirs. Interpretation, 3(3), T155-T167.
McClure, M. W. (2009). Fracture stimulation in enhanced geothermal systems (Doctoral dissertation,
Stanford University).
McClure, M. W., and Horne, R. N. (2013). Conditions required for shear stimulation in EGS.
In Proceedings of the 2013 European Geothermal Congress, Pisa, Italy (Vol. 37).

Melikoglu, M. (2017). Geothermal energy in Turkey and around the World: A review of the literature
and an analysis based on Turkey's Vision 2023 energy targets. Renewable and Sustainable Energy
Reviews, 76, 485-492.

Mertoglu, O., Basarir, N., Saracoglu, B. (2015). Turkey’s Geothermal Potential on EGS. Proceeding
World Geothermal Congress, Melbourne, Australia, April 19-25.

Miao, S. Q., Li, H. P., and Chen, G. (2014). Temperature dependence of thermal diffusivity, specific heat
capacity, and thermal conductivity for several types of rocks. Journal of Thermal Analysis and
Calorimetry, 115(2), 1057-1063.
MIT (2006). The future of geothermal energy – Impact of enhanced geothermal systems (EGS) on the
United States in the 21stCentury, Massachusetts Institute of Technology.
72

MTA (2017) Geothermal Energy Potential and Exploration in Turkey. November 2017. (in Turkish)
http://www.mta.gov.tr/v3.0/arastirmalar/jeotermal-enerji-arastirmalari#haberler [Accessed February
10, 2018]
Muffler, L. P. J., and Cataldi, R. (1978). Methods for regional assessment of geothermal resources. U.S.
Geological Survey Open-File Report 77-870, 78p.
Nathwani, J. (2011). EGS Subprogram Overview. Geothermal Technologies Program Peer Review,
Bethesda, MD.
Onur, M. (2015). Geothermal resource assessment for predicting power generation by volumetric
methods, In: Proceedings of Inerma – International Energy Raw Materials and Energy Summit 2015, 1-3
October, Istanbul, Turkey, 14 p.
Ozdemir, A., Yasar, E., & Cevik, G. (2016). An importance of the geological investigations in Kavaklıdere
geothermal field (Turkey). Geomechanics and Geophysics for Geo-Energy and Geo-Resources, 3(1), 29-
49.
Peluchette, J. (2013). Optimization of integrated reservoir, wellbore, and power plant models for
enhanced geothermal systems. West Virginia University.
Petty, S., Nordin, Y., Glassley, W., Cladouhos, T. T., & Swyer, M. (2013). Improving geothermal project
economics with multi-zone stimulation: results from the Newberry Volcano EGS demonstration.
In Proceedings of the 38th Workshop on Geothermal Reservoir Engineering, Stanford, CA (pp. 11-13).
Pfenninger, S., & Keirstead, J. (2015). Comparing concentrating solar and nuclear power as baseload
providers using the example of South Africa. Energy, 87, 303-314.
Pruess, K., C. Oldenburg, et al. (1999). TOUGH2 User's Guide, Version 2.0. Berkeley, CA, Lawrence
Berkeley National Laboratory.
Pruess, K., Oldenburg, C., & Moridis, G. (2012). TOUGH2 User's Guide Version 2.1. Berkeley, California:
Earth Sciences Division, Lawrence Berkeley National Laboratory, University of California.
Railroad Commission of Texas (2012). Barnett Shale Gas Well Production 1993-2011.
http://instituteforenergyresearch.org/wp-content/uploads/2012/08/Barnett-Shale-Fact-Sheet.pdf
[Accessed February 20, 2018]
Rybach, L. (2007) Geothermal Sustainability. Proceedings European Geothermal Congress 2007
Unterhaching, Germany, 30 May-1 June 2007
Sanyal, S. K., & Sarmiento, Z. (2005). Booking Geothermal Energy Reserves. GRC Transactions, 29.
Satter, Abdus Iqbal, Ghulam M. Buchwalter, James L. (2008). Practical Enhanced Reservoir Engineering
- Assisted with Simulation Software. PennWell.
Seyitoglu G, Tekeli O, Cemen İ, Sen S, İsik V (2002). The role of the flexural rotation/rolling hinge model
in the tectonic evolution of the Alasehir graben, western Turkey. Geological Magazine,139: 15-26.
Stefansson, V. (2005) World Geothermal Assessment. Proceedings World Geothermal Congress 2005,
Antalya, Turkey.
73

TEIAS (2017). Turkey’s Installed Capacity by sources in 2016 and 2017.
https://www.teias.gov.tr/sites/default/files/2018-01/Kguc2017.pdf (in Turkish) [Accessed January 27,
2017].
Tester, J. W., Anderson, B. J., Batchelor, A. S., Blackwell, D. D., DiPippo, R., Drake, E. M., ... & Petty, S.
(2007). Impact of enhanced geothermal systems on US energy supply in the twenty-first
century. Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and
Engineering Sciences, 365(1853), 1057-1094.
ThinkGeoEnergy (2017). Breaking News: Turkey breaks into the 1 GW Geothermal Country Club.
http://www.thinkgeoenergy.com/breaking-news-turkey-breaks-into-the-1-gw-geothermal-country-club/
[Accessed January 25, 2017]
Thunder Head Engineering (2017) PetraSim User Manual. Manhattan, United States: Thunder Head
Engineering. https://www.thunderheadeng.com/wp-
content/uploads/dlm_uploads/2015/04/PetraSimManual-3.pdf
TUIK (2016). Distribution of net electricity consumption by sectors; 2016. http://www.turkstat.gov.tr/
[Accessed January 27, 2017].
Turan, A. (2016). The Feasibility Study For An Enhanced Geothermal System Application In Dikili-Izmir
Region. (Master’s Thesis, Northern Cyprus/Middle East Technical University, 2016)
WEC (2016). World Energy Resources Geothermal 2016. https://www.worldenergy.org/wp-
content/uploads/2017/03/WEResources_Geothermal_2016.pdf [Accessed January 27, 2017].
White, L., Aminzadeh, F., & Rose, A. (2014). A Post-Audit of Geothermal Development in California’s
Imperial Valley. Journal of Sustainable Energy Engineering, 2(2), 166-191.
Williams, C.F., Reed, M.J., and Anderson, A.F. (2011). Updating the Classification of Geothermal
Resources. Proceedings of the 36th Workshop on Geothermal Reservoir Engineering, Stanford
University, California, January 31 – February 2.
Williams, C.F., Reed M.J., and Mariner., R.H. (2008). A Review of Methods Applied by the U.S.
Geological Survey in the Assessment of Identified Geothermal Resources. USA: Open-FileReport 2008-
1296, U.S. Department of the Interior, U.S. Geological Survey.
Worldbank (2016). GDP, PPP (current international $). World Bank Data 2016.
https://data.worldbank.org/indicator/NY.GDP.MKTP.PP.CD?end=2016&locations=TR&start=1990&view
=chart [Accessed January 27, 2017].
Worldbank (2017). Data. CO2 emissions. Last Updated:01/25/18
https://data.worldbank.org/indicator/EN.ATM.CO2E.KT [Accessed January 27, 2017].
YEK Law (2013). The Law on Use of Renewable Energy Resources for Electricity Energy Generation.
Regulation on Unlicensed Electricity Generation in the Electricity Market. Official Gazette-28783 dated
02.10.2013.
Yilmaz, M., & Gelisli, K. (2003). Stratigraphic–structural interpretation and hydrocarbon potential of the
Alaşehir Graben, western Turkey. Petroleum Geoscience, 9(3), 277-282.
74

Zarrouk, J. S., and Moon, H. (2014). Efficiency of geothermal power plants: A worldwide review.
Elsevier, 51, 142-153.

75

APPENDIX A
A.1 MAK-2011-03 Well Log

76

A.2 MAK-2010-14 Well Log

77

APPENDIX B
B.1 Physical measurements and chemical analyses of the water samples from Alasehir and Kavaklidere geotermal areas. T (°C): Values are
measured as outlet temperatures (deep well temperatures; *63°C, **183°C and ***213°C). pH: standard unit at 25°C, EC: electrical conductivity
( μS/c m ) (Bulbul et al. 2011)

78

B.2 Distribution of thermal waters from the study area in a B.3 Geological map and water samples (Seyitoglu et al., 2002)
Piper diagram (Bulbul et al. 2011)

79

B.4 Silica mixing models for some thermal waters (number 7, 17, 21 and 109) in the area (Bulbul et al., 2011)

80

B.5 Stick diagram of mean saturation index for various minerals in the study area (Bulbul et al., 2011)

81

Abstract (if available)

Abstract This thesis presents the background of conventional geothermal energy and enhanced geothermal systems (EGS), comparison of volumetric resource assessments and a reservoir model with an application to the Kavaklidere geothermal field, Turkey. The significant advantages of geothermal energy resources are reducing the local contributions to global climate change and making use of indigenous resources to deal with the foreign source dependency. Turkey is rich in geothermal energy and has ever-growing geothermal developments over the last decade. The Kavaklidere field, covering a 126 km² area in Manisa province, Turkey, is the current target for both conventional geothermal and EGS (enhanced/engineered geothermal system) field development. The field holds Turkey’s highest geothermal resource temperature, which was recorded as 287℃ at 2750 m depth by the General Directorate of Mineral Research and Exploration (MTA) in 2011. ❧ Geothermal reserve estimation, in general, is the critical part of reservoir engineering to have some predictions about the producibility of the reserve. The USGS (United States Geologic Survey) (1978) and MIT (Massachusetts Institute of Technology) (2006) volumetric ‘heat in place’ methods together with Monte Carlo simulation are widely used for assessing the electrical capacity of a geothermal reservoir. Garg and Combs (2015) show that these methods overestimate the results because they use arbitrarily chosen reference temperature and thermal power conversion efficiency values for their estimations without taking the second law of thermodynamics (exergy) and the installed power cycle system into consideration. Garg and Combs (2015) proposed a new method, by reformulating USGS volumetric method, that recoverable heat must be derived by considering specific power cycles, i.e., single-flash or binary. ❧ The EGS technical and economical electricity production potential of Turkey (3–5 km) is expected to be 25 GWe in the next 25 years (Mertoglu et.al. 2015). The latest development in EGS technology is multi-zonal stimulation, which is the ability to create multiple permeable zones by hydroshearing in a single well and increase the amount of produced energy from the well by a factor of two or more (Petty et al., 2013). ❧ In this thesis, first of all, volumetric estimations of power generation potential based on USGS, MIT and Garg and Combs (2015) have been evaluated by Monte Carlo simulation with different development scenarios for the Kavaklidere geothermal field. Applying by Garg and Combs (2015) method, the electric power capacity of the field is estimated as 182.3 MWe with 90% probability for the conventional hydrothermal case, and as 54.8 MWe with 90% probability for EGS case. Sensitivity analysis for both cases shows that recovery factor, reservoir area, and thickness have the biggest impact on net power output among the other input parameters. ❧ Secondly, the existing conditions, such as downhole temperature and pressure and rock density, are incorporated by using the reservoir simulator TOUGH2 via the graphical interface PetraSim. The hydrothermal reservoir is shown for 5-spot well patterns in order to estimate the thermal energy from the production well and demonstrate the temperature change of the reservoir for a 30-year period. The EGS reservoir with the assumptions and different stimulation scenarios is shown for triplet well pattern to determine the thermal energy output of the production wells. Also, Levelized Cost of Electricity (LCOE) is calculated by using GETEM (Geothermal Electricity Technology Evaluation Model) for both hydrothermal and EGS.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Integrated reservoir characterization for unconventional reservoirs using seismic, microseismic and well log data

PDF

Assessing induced seismicity rate increase based on deterministic and probabilistic modeling

$History matching for fractured reservoirs with Hough-transform-based parameterization$

PDF

History matching for fractured reservoirs with Hough-transform-based parameterization

PDF

Continuum modeling of reservoir permeability enhancement and rock degradation during pressurized injection

PDF

Optimization of coupled CO₂ sequestration and enhanced oil recovery

PDF

Integration of multi-physics data into dynamic reservoir model with ensemble Kalman filter

PDF

Deep learning architectures for characterization and forecasting of fluid flow in subsurface systems

PDF

Developing frameworks to quantify the operational and environmental performance of energy systems within the context of climate change

PDF

Modeling and simulation of complex recovery processes

PDF

Deep learning for subsurface characterization and forecasting

PDF

Economic model predictive control for building energy systems

PDF

Asphaltene deposition during co₂ injection in enhanced oil recovery applications

PDF

Modeling and analysis of propulsion systems and components for electrified commercial aircraft

PDF

Modeling and simulation of multicomponent mass transfer in tight dual-porosity systems (unconventional)

PDF

Computational validation of stochastic programming models and applications

PDF

High throughput computational framework for synthesis and accelerated discovery of dielectric polymer materials using polarizable reactive molecular dynamics and graph neural networks

PDF

GaN power devices with innovative structures and great performance

PDF

Phase change heterostructures for electronic and photonic applications

Asset Metadata

Creator Balikcioglu, Aysegul (author)

Core Title Geothermal resource assessment and reservoir modeling with an application to Kavaklidere geothermal field, Turkey

School Viterbi School of Engineering

Degree Master of Science

Degree Program Petroleum Engineering

Publication Date 04/26/2018

Defense Date 03/21/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag enhanced geothermal systems (EGS),geothermal resource assessment,Kavaklidere geothermal field,levelized cost of electricity (LCOE),net power output,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Aminzadeh, Fred (committee chair), Ershaghi, Iraj (committee member), Rose, Adam (committee member)

Creator Email balikcio@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-495281

Unique identifier UC11265818

Identifier etd-Balikciogl-6273.pdf (filename),usctheses-c40-495281 (legacy record id)

Legacy Identifier etd-Balikciogl-6273.pdf

Dmrecord 495281

Document Type Thesis

Format application/pdf (imt)

Rights Balikcioglu, Aysegul

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA