Embed Size (px)
Citation preview
PROCEEDINGS, Thirty-Ninth Workshop on Geothermal Reservoir Engineering
Stanford University, Stanford, California, February 24-26, 2014
SGP-TR-202
1
Updated Supply Characterization of Geothermal District Heating and Cooling Application in
United States
Xiaoning He, Brian J. Anderson
Department of Chemical Engineering, West Virginia University, Morgantown, WV, 26506, USA
Keywords: Supply curve, geothermal district heating and cooling, sensitivity analysis
ABSTRACT
This paper is an update of an earlier supply analysis, He, et al., (2013), of low-temperature utilization of geothermal resources using
geothermal district heating and cooling systems. Along with the use of identified hydrothermal resources, it updates the supply curve for
direct-use utilization using undiscovered hydrothermal resources and near-hydrothermal EGS resources. It also incorporates an
improved design of the heating system, so that the energy production of the system matches the continuously changing energy demand
due to changes in the ambient temperature. Due to uncertainties in characterizing the geothermal reservoir and the related costs, Monte
Carlo simulations are used for both the estimates of thermal potential and levelized cost. The thermal potential of identified
hydrothermal resources is identified with a mean of 72,577 MWth, with a lowest cost at $6.74/MMBtu, while that of the undiscovered
hydrothermal resources has a mean of 240,860 MWth, with a lowest cost at $7.25/MMBtu. The thermal potential of near-hydrothermal
EGS has a mean of 41,035 MWth, with a lowest cost at $7.87/MMBtu. At last, sensitivity analyses are conducted to target the metrics
with the most effective influence in decreasing the levelized cost of such applications.
1. INTRODUCTION
Geothermal energy has the potential to significantly decrease humankind’s dependence on hydrocarbon fuels and greenhouse gas
emissions. Geothermal district heating and cooling (GDHC) systems are a highly-efficient way to utilize geothermal energy. It uses a
warm water source, directly from a geothermal reservoir or cascading from other high-temperature geothermal process, to provide
heating in the winter and cooling in the summer via a sophisticated design of a series of heat exchangers and a pipeline network. As of
2010, the installed capacity of geothermal district heating and cooling is about 215 MWth, Lund, et al., (2010), in the U.S., from which
the energy saving amounts to 4 million barrels of equivalent oil, and prevents 0.5 million tonnes of carbon and 2 million tonnes of CO2
being released to the atmosphere every year, calculated using data by Lawrence Livermore Laboratories, Kasameyer, (1997).
Supply analysis has been accepted as a powerful tool in both traditional and renewable energy research for the estimation of energy
reserves, as well as for estimating the cost of energy, e.g. Blair, (1978), Kilian, ( 2009), Gordon, (1975), Karki, et al., (2004), and Cook,
et al., (2010). Traditionally, supply analyses of geothermal energy have been concentrated only on geothermal power generation, e.g.
Petty, et al., (2007), Augustine, (2011). This paper is an update of an earlier supply analysis of GDHC systems, He, et al., (2013). It
expands its focus on not only the hydro-geothermal resources, but also to include enhanced geothermal systems (EGS). Besides, this
paper optimizes the methods identifying the thermal potential of geothermal energy by introducing the Monte Carlo simulations, and
improves the heat exchangers’ design to let the production of geothermal water dynamically change with the changing energy demands
due to ambient temperature change.
2. METHOD
This study includes two parts, 1) the estimation of the thermal potential and 2) the determination of the levelized cost of heat. The
geothermal resources can be categorized into four types by the reservoir quality or water availability: identified hydro-geothermal,
undiscovered hydro-geothermal, near hydro-thermal EGS and deep EGS resources. This paper only focuses on the first three categories,
because of the uncertainty currently associated with cost and achieved flow rates from green-field geothermal. The methods to estimate
the potential of different resources are different, but the same levelized cost model is used for all the three categories.
2.1 Thermal potential estimation and reservoir characteristics
2.1.1 Identified hydro-geothermal resources
The thermal potential estimate of this category has been discussed in the former supply analysis, He, et al., (2013). It assumed the
thermal potential is equivalent to the potential water can deliver to the surface, and calculated by Equation 1:
dTCm=E PWH (1)
The former report identified 251 moderate and high temperature geothermal resources in the western U.S., and retrieved the reservoirs’
temperature and depth data from the USGS reports, e.g. White, et al., (1975), Muffler, (1979) and Reed, (1983). However, very little
information about the reservoirs’ flow rate was provided in the USGS reports.
He and Anderson
2
This paper updates the method to determine the flow rate of each reservoir. It is derived from the volume method, which was used in the
past USGS assessments for evaluating the power potential of geothermal energy, e.g. Nathenson, (1975), Muffler, et al., (1978),
Muffler, (1979), Lovekin, (2004), Williams, (2004). The volume method develops a mathematic process to transfer the thermal energy
in the reservoir to the electricity production by Equation 2:
000 ssTmTTCVRt
W WHWHRgu
e
(2)
In this Equation, t is the lifetime of the project; ρC is the volumetric specific heat of the reservoir rock; V is the volume of the reservoir;
TR is the reservoir temperature; T0 and s0 is the temperature and the entropy per unit mass of water at the reference state; W Hm ̇ and
sWH are the mass flow rate of water and the entropy per unit mass of water at the well head; Rg represents the recovery factor from the
thermal energy stored in the rock; and ηu represents the utilization efficiency from exergy to electricity. Here it is assumed that the total
mass flow rate ̇ obtained at the well head depends only on the volume of the reservoir V, because with larger reservoir volume, the
fractures in the reservoir are larger, allowing more flow rate of water through the reservoir. With this relation between ̇ and V, for
any reservoir with reservoir temperature TR given, the power potential will depend only on the mass flow rate ̇ Equation 2
provides the theoretical basis of the potential estimate in the Geothermal Electricity Technology Evaluation Model (GETEM) developed
by U.S. Department of Energy. GETEM will be used to find the mass flow rate with power potential data retrieved from the USGS
Energy Data Finder by Equation 3:
ReWH TWGETEMm , (3)
At this point, the thermal energy potential can be calculated by using Equation 1, by assuming the returning temperature at 40°C. With
the presence of possible values of reservoir temperature and mass flow rate, the Monte Carlo simulation will be used during the
calculation.
2.1.2 Undiscovered hydrothermal resources
Due to the uncertainty of the potential and location of undiscovered hydrothermal resources, there is not a way to target such kind of
resource one by one and give estimation of them. However, the geologic condition of undiscovered hydrothermal resources is assumed
to be the same with that of the identified hydrothermal resources. For example, from analysis of the known geothermal systems, young
felsic magmatism has a strong spatial correlation with geothermal energy, Smith and Shaw, (1975); higher underground heat flow is
usually relevant to larger possibility of geothermal reservoir occurrence; all the Quaternary faults have a strong statistical significance
for the correlation within 4 km distance of geothermal occurrences, Williams and DeAngelo, (2008), etc. Such indicators are all with
known data. For each indicator, by assigning strength values for different locations based on the relation between that indicator and the
occurrence of geothermal energy, it becomes an “evidence layer”. By integrating and overlapping several evidence layers, with
statistical analysis, a favorability factor of geothermal resources’ occurrence map can be generated. Such indicator favorability theory
has been used to estimate undiscovered resources in many areas including geothermal energy, Coolbaugh and Shevenell, (2004), etc.
In the latest USGS geothermal power potential report, Williams chose 28 evidence layers of different geologic metrics which are
directly correlated to the occurrence of geothermal energy, weighed the favorability of each layer, integrated the results by the use of
ArcGIS, and derived a geothermal favorability map of mid-western U.S., Williams, et al., (2009). During the study, he applied the
Bayesian statistical theory to analyze the relation between the evidence layers and the existing geothermal resources, and to estimate the
occurrence probability of geothermal resources in unexplored regions. The favorability of undiscovered hydrothermal resources is in the
form of the ratio of occurrence density (number of geothermal systems per unit area) of undiscovered resources to that of the identified
resources. As stated in the last part, the power potential ( eW ) is linear to the thermal potential ( E )by the factor of utilization factor, ηu,
as shown in Equations 4 and 5:
For discovered resources
EW ue (4)
For undiscovered resources
'' EW ue (5)
Also, as assumed in the power potential estimation, Augustine, (2011), the undiscovered resources can be directly calculated from the
occurrence probability ratio (α), shown in Equation 6:
ee WW ' (6)
Using Equations 4, 5, and 6, the thermal potential of undiscovered hydrothermal resources can be calculated by Equation 7:
He and Anderson
3
EE ' (7)
As for the reservoir characteristics, a single reservoir depth, temperature, and mass flow rate will be assigned to all undiscovered
resources in each state. For every state, the undiscovered resources are more similar to the large identified resources than the small ones
in reservoir characteristics. So the reservoir depth, temperature, and mass flow rate is measured by calculating the mean-thermal
capacity-weighted average of these parameters of the identified resources, as shown in Equations 8, 9, 10, and 11:
n
i i
ii
E
E
(8)
n
i iidd )(' (9)
n
i iiTT )(' (10)
n
i iimm )(' (11)
Where d’, T’, and 'm and d, T, m represent the depth, temperature and mass flow rate of undiscovered and identified resources, β is the
thermal potential weight factor for each resource, and i is the index of each identified hydrothermal resource in that state.
2.1.3 Near-hydrothermal EGS resources
Reservoir temperature and mass flow rate are the most important factors when determining the economics of a geothermal resource. The
near-hydrothermal EGS is defined as the geothermal resource around the hydrothermal site but lack of sufficient permeability to let
water through. Because of increased certainty surrounding the geologic environment and the potential for leveraging existing
infrastructure, it is the least expensive EGS resource, and should be exploited first. Recently the U.S. Department of Energy and Ormat
Technologies, Inc. announced the first EGS power plant connected to the electricity grid, which belongs to the near-hydrothermal EGS
resources, and produces 1.7 MWe of electricity.
To the best of our knowledge, there is not a formal assessment of near-hydrothermal EGS resources. Therefore, an estimation of its
thermal potential is preliminary and based on many assumptions. In this paper, the reservoir temperature of near-hydrothermal resource
is assumed as the same temperature with that hydrothermal resource which is in the nearby vicinity. Since the thermal potential is
determined by temperature and mass flow rate, as shown in Equation 1, and the mass flow rate is the only one to distinguish the near-
hydrothermal EGS from the hydrothermal resources, the thermal potential of each near-hydrothermal EGS is assumed to be the
difference between the mean and high-end estimates of thermal potential of its corresponding hydrothermal resources, shown in
Equation 12:
percentilepercentilenear EEE 5095 (12)
As for estimation of mass flow rate, Darcy’s Law describes that the flow rate through a porous medium is determined by the permeabil-
ity, the viscosity, the pressure gradient, and the drainage cross section. The viscosity is easy to determine, however the other parameters
are very much related to the subsurface characteristics and the hydraulic fracturing process used to create the reservoir. Thus, the
estimation of the flow rate of a near-hydrothermal EGS reservoir is very site-specific. Exploration data is very crucial for mass flow rate
estimation of each EGS reservoir. With absence of exploration data, McVeigh assumed the mass flow rate in his research as 54 kg/s
based on the current hydraulic fracture technology, McVeigh, et al., (2007). Augustine assumed a flow rate of 30 kg/s and 60 kg/s for
the current and improved technology scenario, Augustine, (2011). Both of their researches aim to provide the risk and supply analysis of
geothermal energy for the DOE’s Geothermal Technologies Program. So in this paper, regardless of the subsurface characteristics, the
mass flow rate is assumed as the available flow rate at the well head, and is directly set as possible values, 40 kg/s, 60 kg/s, 80 kg/s,
representing the current achieved rate and the increasing rate because of the technology improvement in the future. The advantage is to
provide a wide range of the flow rate, and future work can refer to different flow rate scenarios for different EGS reservoir and verify
the levelized cost, while the disadvantage is that the flow rate of a specific EGS reservoir is unknown until the exploration stage is
finished, after that, the result in this study can be used.
2.2 System design and cost estimate
A GDHC system aims to provide both residential and commercial heating and cooling demands. Once a geothermal reservoir is
confirmed to be applicable, the design of the surface heating and cooling network is also very important. The GDHC system is very site
specific, because the geothermal hot water must be produced and consumed at the same location to avoid energy loss during
transmission. Thus, the design of the GDHC system must be consistent with both the geothermal reservoir’s quality and the local energy
demands.
He and Anderson
4
The former supply analysis, He, et al., (2013), estimated the residential heating and cooling demand of the areas where the identified
hydro-geothermal reservoirs are located, by using the historical energy consumption data from U.S. EIA. This paper updates the
commercial heating and cooling demand of the same areas from the U.S. EIA commercial building energy consumption survey
(CBECS). In order to represent the total energy demand changes due to the daily ambient temperature change, the year-round ambient
temperature (Tab) change is assumed to follow the sine function as shown in Equation 13:
minminmax365
sin abababab TnTTT
(13)
With the fact that heating and cooling demand is a function of ambient temperature, the heating ( ̇) and cooling ( ̇) demand can be
represented as Equation 14 and Equation 15:
C3.18,0
C3.18,365
cosmax
ab
ab
TifH
TifnHH
(14)
C3.18,0
C3.18,365
cosmax
ab
ab
TifC
TifnCC
(15)
Figure 1 gives an example of using Equations 13, 14, and 15 to represent a day-to-day heating and cooling demand change, as well as
the ambient temperature change. In Figure 1, the maximum and the minimum ambient temperature year round is set at 33°C and -7°C.
The maximum heating and cooling demand is set at 38 MWth and 9 MWth.
Figure 1: Year-round heating and cooling demand change with the ambient temperature change.
The fundamental relation of the heating and cooling process is the heat transfer Equation in Equation 1, where the left side represents
the energy demand, and the right side represents the energy provided by geothermal water, in this case. The former supply analysis, He,
et al., (2013), used an empirical function to address the issues regarding to the heat transfer efficiency when dealing with a real heating
or cooling process. The disadvantage is that it averages any energy demand variation and uses the maximum energy demand to design
the system. It may cause the waste of geothermal water and accelerate the depletion of a geothermal reservoir.
This paper updates a new method which is able to process a system with continuous changing energy demand. The basic geothermal
heating unit consists of an indoor radiator and an outdoor heat exchanger, as shown in Figure 2.
He and Anderson
5
Figure 2: Schematic of a basic geothermal heating unit.
The streams TP and TR is the production and return stream of the geothermal water, and the loop of stream TS and TO is the secondary
flow inside the building. This method consists of two states of design, which are the reference state and the normal operation state. A
summary of the assumptions and statements used is listed in Table 1.
Table 1: Assumptions and statements used in the heat exchanger design.
Parameter Reference Normal
TP Treservoir Treservoir
TS 80 Calculate
ΔTradiator 40 40
TO Calculate Calculate
TR =TO+3 =TO+3
The reference state is chosen at the system peak operation, and Equation 16 developed by Valdimarsson, (1993), provides the
connection between the reference and normal operation state.
For the radiator
0
3.1
0 LMTD
LMTD
H
H
(16)
In this Equation, the LMTD0 can be calculated with the reference state assumptions and Equation 17. 0H is the peak heating demand.
H is the heating demand at normal operation state, which can be calculated by Equation 14. Thus the LMTD of the radiator at any
operation state can be calculated. Then Excel-Solver will be used to calculate the radiator streams’ temperature (TS and TO) by
Equation 17 and 18:
)/()(ln
)()(
roomroom
roomroom
TTOTTS
TTOTTSLMTD
(17)
40TOTS (18)
With other heat exchanger principles, the other streams in the system can also be solved.
The other aspects of the GDHC system, such as the pipeline network design and sizing, and the economic analysis will follow the same
procedure as the former supply analysis, He, et al., (2013).
3. RESULTS AND DISCUSSION
3.1 Identified Hydro-geothermal Resources
Since some of the parameters contain probability distribution, the Monte Carlo simulation is used in this study. The iteration is set at
1000 times, and the simulations are processed by Excel @Risk add-in.
The available water mass flow of each resource is first calculated by the use of GETEM. By the @Risk’s distribution fitting, all of them
are fitted with the triangular distribution, with significantly small root mean square error (RMSE). Ten sets of the flow rate results are
randomly selected from the total of 251 resources, shown in Table 2. The resources with top five flow rate are all in the state of
He and Anderson
6
California, and they are Salton Sea area, Geysers area, Medicine Lake, Coso area, and Brawley. Their flow rates are so large that their
thermal potentials are also the top five among all the resources.
The thermal potential is then calculated by Equation 1. The total thermal potential of the 251 identified hydrothermal resources has a
mean of 72,577 MWth, and with a 95% probability of only 33,250 MWth and a 5% probability of up to 113,535 MWth. Identified
resources are concentrated in the state of California, Nevada, and Alaska. Since quite a few numbers of resources have already been
under development or in operation, mainly for geothermal power generation and space heating, the existing energy applications should
be subtracted from the total thermal potential. The existing power generation capacity is 2,479 MWe, Augustine, (2011), and that of
space heating is around 215 MWth, Lund, (2010).
Table 2: Triangular distribution fitted by @Risk using the ten randomly selected flow rate results, flow rate in kg/s.
Name 5%th
flow rate Mean flow rate 95%th
flow rate RMSE×10-9
Abraham, UT 11.60 52.78 122.29 7.89
Akutan
Fumaroles, AK 231.66 687.19 1405.68 9.96
Bailey Bay, AK 53.15 225.17 511.27 7.98
Big Bend HS, CA 16.58 62.21 139.75 9.88
Big Creek HS, ID 77.42 307.66 681.30 9.84
Blue Mt., NV 267.85 911.79 1969.27 9.70
Chena, AK 9.04 36.38 81.84 7.59
Coso Area, CA 2062.87 3613.73 4818.69 6.61
Cove Fort, UT 10.76 22.87 37.68 7.23
Dixie HS, NV 48.14 138.12 274.76 9.10
If assuming an overall 10% efficiency of thermal energy to electricity, the available thermal potential from identified hydro-geothermal
resources is with a mean total of 47,566 MWth.
As stated before, the GDHC system must be built at a populated location, so that the geothermal hot water can be consumed without
much loss of energy. There are some of the geothermal resources with very few people living near them. Based on the levelized cost
model, the levelized cost at such locations are extremely high, more than $150/MMBtu. In the following analysis, those locations with
unreasonably high levelized cost will be neglected. The levelized cost of the identified hydro-geothermal resources is estimated on a
site-by-site basis. Since the presence of some inputs’ probability distribution, the levelized cost is also with a probability distribution.
The supply curve can be plotted based on each resource’s thermal potential and levelized cost (LCOH). Here the supply curve will be
truncated to show the first 60 GWth of thermal potential to emphasize the resources with lowest levelized cost, which are likely to be
developed first, as shown in Figure 3. It also emphasizes the part of the supply curve with levelized cost lower than $40/MMBtu, as
shown in Figure 4. Among all the resources, the Weiser area, ID is with the lowest levelized cost, a mean of $6.74/MMBtu. This lowest
LCOH is lower than the average cost of residential heating using natural gas according to the EIA (2013), $9.2/MMBtu; however, most
of the identified resources would be found with an LCOH above the average heating cost. All the resources with competitive levelized
cost can be characterized as with median or high reservoir temperature, median or low drilling depth, and the most importantly, with
high energy demand. Figure 4 also tells that there are more than 8 GWth of thermal energy can be utilized under $40/MMBtu with 10%
of confidence, and more than 4 GWth of thermal energy can be utilized under $40/MMBtu with 90% of confidence.
Figure 3: Supply curve of the identified hydrothermal resources, truncated at 60 GWth, in comparison with the current cost of
heating by natural gas, which is $9.2/MMBtu.
He and Anderson
7
Figure 4: Partial enlargement of the identified hydrothermal resources’ supply curve, with levelized cost lower than
$40/MMBtu.
To determine the sensitivity of the model, seven technical or economic inputs from the model are selected. They are the system’s energy
demand, reservoir temperature, drilling cost, project lifetime, discount rate, pipeline capital, and heating and cooling facility capital.
Several simulations will be run to ensure each input varies -50%, -25%, +25%, and +50% in every simulation. The results of the
sensitive analysis are shown in Figure 5. The energy demand has the most negative influence and the drilling cost has the most positive
influence on levelized cost. Increasing the energy demand is the most effective way to decrease the levelized cost.
Figure 5: Sensitivity analysis of the levelized cost model with the identified hydrothermal resources data.
To further identify the relation between the levelized cost and the system’s energy demand, the population data (P) of each location
versus the levelized cost (LC) for every resource is plotted, as shown in Figure 6 in blue. With the least absolute error regression
analysis of the population and the levelized cost data, the relation can be expressed in Equation 18, as shown in Figure 6 in red:
6469.110202 7495.0 PLC (19)
He and Anderson
8
Figure 6: The population data at each geothermal reservoir versus the calculated levelized cost is plotted in blue. The levelized
cost function derived by regression analysis is plotted in red.
3.2 Undiscovered Hydro-geothermal Resources
The favorability of hydrothermal resources presents the ratio of the undiscovered hydrothermal resources’ occurrence to the discovered
hydrothermal resources’ occurrence. It is a statistical result derived from quantifying the preferable geologic conditions of hydrothermal
resources, such as quaternary magmatic activity, heat flow, seismicity, etc. In this paper, several reports e.g. Augustine, (2011),
William, et al., (2008), (2009) are consulted for the electricity potential to estimate the favorability for each state. The results of the
favorability data are shown in Table 3:
Table 3: Favorability data of different geothermal regions.
Region 5%
favorability 50%
favorability 95%
favorability
Alaska 2.28 2.64 3.13
Hawaii 9.78 13.45 17.00
Others 1.60 3.60 6.70
Based on Equation 7, the thermal potential of the undiscovered hydrothermal resources is calculated, with a mean total of 240,860
MWth, and with a 95% probability of only 48,930 MWth and a 5% probability of up to 579,660 MWth. The distribution of the
undiscovered hydrothermal resources’ thermal potential of each state is shown in Figure 7.
Figure 7: The total thermal potential of the undiscovered hydrothermal resources has a mean 241 GWth. The distribution in
each state is shown above.
As stated before, the reservoir characteristics of each state are calculated by a mean thermal potential weighted average. Thus the
reservoir characteristics are more like the large identified geothermal reservoir in each state. But populated place not necessarily locates
near the large identified geothermal resource in every state. In this study, for each state’s undiscovered hydrothermal resource, the
GDHC system’s energy demand is assumed to be the largest energy demand in that state. The supply curve is shown in Figure 8. The
He and Anderson
9
undiscovered resource with lowest levelized cost is the resource under 150°C in California. The cost is $7.25/MMBtu with 50% of
confidence.
Figure 8: Supply curve of the undiscovered hydrothermal resources, in comparison with the current cost of heating by natural
gas, which is $9.2/MMBtu.
Since the same model is used to calculate the levelized cost, the sensitive analysis for both the identified and the undiscovered
hydrothermal resources are the same. For undiscovered resources, the energy demand still has the most negative influence and the
drilling cost still has the most positive influence on levelized cost. And increasing the energy demand is still the most effective way to
decrease the levelized cost.
3.3 Near-hydrothermal EGS
The near-hydrothermal EGS has the same reservoir depth and temperature with the hydrothermal reservoir which it is close to. The
availability for water pathway is the difference between these two resources. Based on Equation (12), and the thermal potential
estimation of the identified hydrothermal resources, the total thermal potential of near-hydrothermal EGS is 41,035 MWth. The top five
resources with the most thermal potential are all in California, which are Salton See area (8,555 MWth), Geysers area (3,416 MWth),
Brawley (1,646 MWth), Coso area (1,448 MWth), and the Medicine Lake (1,393 MWth).
With the uncertainty of the mass flow rate of each EGS reservoir, 40 kg/s, 60 kg/s, 80 kg/s are used as flow rate inputs for each case.
The supply curve is shown in Figure 9. Among all the resources, the Weiser area, ID is still with the lowest levelized cost, a mean of
$7.87/MMBtu. The cost increase is mainly due to the hydro-fracture stimulation cost to create the EGS reservoir. Theoretically with
increase mass flow rate, the levelized cost will decrease because of the decreased number of production well. The several resources at
the left end of the supply curve in Figure 9 do perform such character. The levelized cost of the 80 kg/s scenario is the lowest, while that
of the 40 kg/s scenario is the highest. But for the other resources in this study, most locations are with small energy demand, usually
below 10 MWth. As a result, one production well with 40 kg/s water flow rate is sufficient to provide the energy demand. Increase in
mass flow rate will not efficiently decrease the levelized cost. That’s why in Figure 9 the supply curves for three different mass flow
rate scenarios overlap for most parts.
Figure 9: Supply curve of the near-hydrothermal EGS resources, with levelized cost lower than $100/MMBtu, in comparison
with the current cost of heating by natural gas, which is $9.2/MMBtu.
He and Anderson
10
Finally, Figure 10 shows the supply curve which integrates all the three categories of geothermal resources. Figure 11 is a
partial enlargement to show the thermal potential with acceptable levelized cost. The combined supply curve shows the order in which
resource should be developed first based on the calculated levelized cost. Figure 10 tells over 80% of the thermal potential is with a
levelized cost lower than $80/MMBtu. In Figure 11, except the starting point is the identified hydrothermal resource (Weiser area, ID)
and its corresponding near-hydro EGS, over 50,000 MWth of potential is undiscovered, with an LCOH that is lower than the natural gas-
based heating. Besides, there are another 100,000 MWth of the undiscovered hydrothermal resources with levelized cost between $20
and $30/MMBtu. The near-hydro EGS is the least expensive type of EGS resource. The levelized cost of near-hydro EGS is a little
higher than its corresponding identified hydrothermal resources. Thus in the supply curve, the near-hydro EGS and the identified
hydrothermal resource are usually together. In fact, there is not much thermal potential available from identified hydrothermal
resources. From Figure 3, there is only 50,000 MWth of potential with levelized cost lower than $80/MMBtu, while the existing
geothermal application has consumed the half of such potential, which is also the potential with lowest levelized cost. As a result, the
near-hydrothermal EGS corresponding to those most competitive identified hydrothermal resources may be a good choice for expanding
the existing system.
Figure 10: Supply curve integrated with three categories of geothermal resources.
Figure 11: Partial enlargement of the integrated supply curve, in comparison with the current cost of heating by natural gas,
which is $9.2/MMBtu.
He and Anderson
11
4. CONCLUSION
This paper focuses on the supply analysis of geothermal district heating and cooling systems (GDHC). This study categorizes the
geothermal resources into four types: identified hydro-geothermal resources, undiscovered hydro-geothermal resources, near-hydro EGS
and deep EGS. The thermal potential of the first three categories is estimated in this study. The thermal potential of the 251 identified
hydrothermal resources is with a mean of 72,577 MWth, and with a 95% probability of only 33,250 MWth and a 5% probability of up to
113,535 MWth. The thermal potential of the undiscovered hydrothermal resources is with a mean of 240,860 MWth, and with a 95%
probability of only 48,930 MWth and a 5% probability of up to 579,660 MWth. The thermal potential of the near-hydrothermal EGS
resources is about 41,035 MWth.
This paper also describes the levelized cost of a GDHC system using different types of geothermal resources. The cost of near-
hydrothermal EGS is usually higher than the hydrothermal resources, due to the hydro-shear or stimulation processes. The location with
the lowest levelized cost is the Weiser area, in Idaho, for both identified hydrothermal or near-hydrothermal EGS resources. The
levelized cost is $6.74/ MMBtu using identified hydrothermal resources, and $7.87/MMBtu using a near-hydrothermal EGS system.
The undiscovered hydrothermal resources in California are the most preferable among all the undiscovered hydrothermal resources. The
levelized cost is $7.25/MMBtu. According to the sensitivity analysis of the cost model, the energy demand of the GDHC system has the
most decreasing influence and the drilling cost has the most increasing influence on levelized cost. Therefore, increasing the energy
demand is the most effective way to decrease the levelized cost of a GDHC system. Further investigation reveals that the levelized cost
follows a power function with the population, which directly influences the energy demand.
The supply curves show the order in which resource should be first developed. Since currently over half of the competitive identified
hydrothermal resources has been developed, the undiscovered hydrothermal and the near-hydrothermal EGS resources are noteworthy.
There are over 150,000 MWth of potential undiscovered with levelized cost lower than $30/MMBtu, about 30% of such resources are
estimated to have an LCOH lower than the EIA average for natural gas heating. And with a little higher cost, the near-hydrothermal
EGS may be a good choice to expand the existing system which is based on hydrothermal resources.
5. ACKNOWLEDGEMENT
The authors would like to acknowledge the help by project partners Mr. M. G. Bedre, graduate committee members Dr. T. Carr, Dr. R.
Jackson, and Dr. R. Turton, and Dr. J. Zondlo, lab workers Mr. M. Gaddipati, Mr. S. Velaga, Ms. N. Garapati, express thanks to Mr. C.
Augustine and the Department of Energy’s Geothermal Technologies Program, Project EE0002745 for funding.
REFERENCES
Augustine, C., “Updated U.S. Geothermal Supply Characterization and Representation for Market Penetration Model Input,” (2011),
NRTL/TP-6A20-47459.
Blair, J. M., “Control of oil,” Vintage Books, (1978).
Cook, T. R., Dogutan, D. K., Reece, S. Y., “Solar energy supply and storage for the legacy and nonlegacy worlds,” Chemical
reviews 110.11 (2010): 6474-6502.
Coolbaugh, M. F., Shevenell, L. A., “A method for estimating undiscovered geothermal resources in Nevada and the Great Basin,”
Geothermal Resources Council Transaction, 28(2004) 13-18.
Gordon, R. L., “Economic analysis of coal supply: an assessment of existing studies,” Research Project 335, Key Phase Report.
Pennsylvania State University, (1975).
He, X., Anderson, B. J., “Supply characterization of hydro-geothermal resources in western U.S.,” Proceeding to the Thirty-Eighth
Workshop on Geothermal Reservoir Engineering, Stanford University, CA, February 11-13, 2013.
Karki, R., Billinton, R., “Cost-effective wind energy utilization for reliable power supply,” Energy Conversion, IEEE Transactions
on 19.2 (2004): 435-440.
Kasameyer, R., “Brief guidelines for the development of inputs to CCTS from the technology working group, Working draft,”
Lawrence Livermore Laboratory, Livermore, CA, (1997).
Kilian, L., “Not all oil price shocks are alike: Disentangling demand and supply shocks in the crude oil market,” The American
Economic Review, 99.3 (2009): 1053-1069.
Lovekin, J., “Geothermal inventory,” Geothermal Resources Council Transaction, 33-6 (2004) 242-244.
Lund, J. W., “Direct utilization of geothermal energy,” Energies, 3 (2010) 1443-1471.
McVeigh, J., Cohen, J., Vorum, M., Porro, G., Nix, G, "Preliminary technical risk analysis for the geothermal technologies program,”
(2007), NREL/TP-640-41156.
Muffler, L. P. J., “Assessment of Geothermal Resources of the United States,” U.S. Geological Survey Circular 790 (1979), 163p.
Muffler, L. P. J., Cataldi, R., “Methods for regional assessment of geothermal resources,” Geothermics, 7(1978) 53-89.
Nathenson, M., “Physical factors determining the fraction of stored energy recoverable from hydrothermal convection systems and
conduction-dominated areas,” U.S. Geological Survey Open-File report 75-525, 50 p, (1975).
He and Anderson
12
Petty, S., Porro, G., “Updated U.S. Geothermal Supply Characterization,” Proceeding to the Thirty-Second Workshop on Geothermal
Reservoir Engineering, Stanford University, CA, January 22-24, 2007. NREL/CP-640-41073.
Reed, M. J., “Assessment of Low-temperature Geothermal Resources of the United States,” U.S. Geological Survey Circular 892
(1983), 73p.
Smith, R. L., Shaw, H. R., “Igneous-related geothermal systems,” in White, D. F. and Williams, D. L. eds., “Assessment of geothermal
resources of the United States,” U. S. Geological Survey Circular 726, 155p, (1975).
U.S. Energy Information Administration Heating Fuels Comparison Calculator, (2013).
Valdimarsson, P., “Modelling of geothermal district heating systems,” University of Iceland, (1993).
White, D. E., Williams, D. L., “Assessment of Geothermal Resources of the United States,” U.S. Geological Survey Circular 726
(1975), 155p.
Williams, C. F., “Development of revised techniques for assessing geothermal resources,” Proceeding to the Twenty-Ninth Workshop
on Geothermal Reservoir Engineering., Stanford University, CA, January 26-28, 2004.
Williams, C. F., DeAngelo, J., “Mapping geothermal potential in the western United States,” Geothermal Resources Council
Transaction, 32(2008) 181-187.
Williams, C. F., Reed, M. J., Mariner, R. H., DeAngelo, J., Peter, G. S., “Quantifying the undiscovered geothermal resources in the
United States,” Geothermal Resources Council Transaction, 33(2009) 995-1002.
Williams, C. F., Reed, M. J., Mariner, R. H., DeAngelo, J., Galanis, S. P. J., “Assessment of moderate- and high-temperature
geothermal resources of the United States,” U.S. Geological Survey Fact Sheet 2008-3082, U.S. Department of Interior
