ArticlePDF Available

Analysis of the impact of storage conditions on the thermal recovery efficiency of low-temperature ATES systems

  • Delft University of Technology & KWR Water research institute

Abstract and Figures

Aquifer thermal energy storage (ATES) is a technology with worldwide potential to provide sustainable space heating and cooling using groundwater stored at different temperatures. The thermal recovery efficiency is one of the main parameters that determines the overall energy savings of ATES systems and is affected by storage specifics and site-specific hydrogeological conditions. Although beneficial for the optimization of ATES design, thus far a systematic analysis of how different principal factors affect thermal recovery efficiency is lacking. Therefore, analytical approaches were developed, extended and tested numerically to evaluate how the loss of stored thermal energy by conduction, dispersion and displacement by ambient groundwater flow affect thermal recovery efficiency under different storage conditions. The practical framework provided in this study is valid for the wide range of practical conditions as derived from 331 low-temperature (< 25 °C) ATES systems in practice. Results show that thermal energy losses from the stored volume by conduction across the boundaries of the stored volume dominate those by dispersion for all practical storage conditions evaluated. In addition to conduction, the displacement of stored thermal volumes by ambient groundwater flow is also an important process controlling the thermal recovery efficiencies of ATES systems. An analytical expression was derived to describe the thermal recovery efficiency as a function of the ratio of the thermal radius of the stored volume over ambient groundwater flow velocity (Rth/u). For the heat losses by conduction, simulation results showed that the thermal recovery efficiency decreases linearly with increasing surface area over volume ratios for the stored volume (A/ V), as was confirmed by the derivation of A/V-ratios for previous ATES studies. In the presence of ambient groundwater flow, the simulations showed that for Rth/u <1 year, displacement losses dominated conduction losses. Finally, for the optimization of overall thermal recovery efficiency as affected by these two main processes, the optimal design value for the ratio of well screen length over thermal radius (L/Rth) was shown to decrease with increasing ambient flow velocities while the sensitivity for this value increased. While in the absence of ambient flow a relatively broad optimum exists around an L/Rth-ratio of 0.5–3, at 40 m/year of ambient groundwater flow the optimal L/Rth-value ranges from 0.25 to 0.75. With the insights from this study, the consideration of storage volumes, the selection of suitable aquifer sections and well screen lengths can be supported in the optimization of ATES systems world-wide.
Content may be subject to copyright.
Contents lists available at ScienceDirect
journal homepage:
Analysis of the impact of storage conditions on the thermal recovery
eciency of low-temperature ATES systems
Martin Bloemendal
, Niels Hartog
Department of Water Management, Delft University of Technology, Delft, The Netherlands
KWR, Watercycle Research Institute, Nieuwegein, The Netherlands
Faculty of Geosciences, Utrecht University, Utrecht, The Netherlands
Aquifer Thermal Energy Storage (ATES)
Recovery eciency
Well design
Storage conditions
Aquifer thermal energy storage (ATES) is a technology with worldwide potential to provide sustainable space
heating and cooling using groundwater stored at dierent temperatures. The thermal recovery eciency is one
of the main parameters that determines the overall energy savings of ATES systems and is aected by storage
specics and site-specic hydrogeological conditions. Although benecial for the optimization of ATES design,
thus far a systematic analysis of how dierent principal factors aect thermal recovery eciency is lacking.
Therefore, analytical approaches were developed, extended and tested numerically to evaluate how the loss of
stored thermal energy by conduction, dispersion and displacement by ambient groundwater ow aect thermal
recovery eciency under dierent storage conditions. The practical framework provided in this study is valid for
the wide range of practical conditions as derived from 331 low-temperature (< 25 °C) ATES systems in practice.
Results show that thermal energy losses from the stored volume by conduction across the boundaries of the
stored volume dominate those by dispersion for all practical storage conditions evaluated. In addition to con-
duction, the displacement of stored thermal volumes by ambient groundwater ow is also an important process
controlling the thermal recovery eciencies of ATES systems. An analytical expression was derived to describe
the thermal recovery eciency as a function of the ratio of the thermal radius of the stored volume over ambient
groundwater ow velocity (R
/u). For the heat losses by conduction, simulation results showed that the thermal
recovery eciency decreases linearly with increasing surface area over volume ratios for the stored volume (A/
V), as was conrmed by the derivation of A/V-ratios for previous ATES studies. In the presence of ambient
groundwater ow, the simulations showed that for R
/u < 1 year, displacement losses dominated conduction
losses. Finally, for the optimization of overall thermal recovery eciency as aected by these two main pro-
cesses, the optimal design value for the ratio of well screen length over thermal radius (L/R
) was shown to
decrease with increasing ambient ow velocities while the sensitivity for this value increased. While in the
absence of ambient ow a relatively broad optimum exists around an L/R
-ratio of 0.53, at 40 m/year of
ambient groundwater ow the optimal L/R
-value ranges from 0.25 to 0.75. With the insights from this study,
the consideration of storage volumes, the selection of suitable aquifer sections and well screen lengths can be
supported in the optimization of ATES systems world-wide.
1. Introduction
World-wide eorts aim to reduce greenhouse gas emissions and to
meet energy demands sustainably (EU, 2010; SER, 2013; UN, 2015).
Global demand for heating and cooling in the built environment ac-
counts for about 40% of the total energy consumption (EIA, 2009; Kim
et al., 2010; RHC, 2013). In reducing this demand, the use of Aquifer
Thermal Energy Storage
(ATES) systems for space heating and cooling
has a high potential in the many regions worldwide that have sub-
stantial seasonal, or sometimes diurnal, variations in ambient air tem-
perature combined with favorable geohydrological conditions
(Bloemendal et al., 2015).
Although much of the early ATES research has focused on storage at
high temperatures (Molz et al., 1983, 1978; Nagano et al., 2002;
Réveillère et al., 2013; Tsang, 1978 ), most practical experience with
seasonal ATES systems has in recent years been gained in particularly
Received 14 June 2017; Received in revised form 9 September 2017; Accepted 17 October 2017
Corresponding author at: Delft University of Technology, Department of Water Management, PO Box 5048, 2600 GA, Delft, The Netherlands.
E-mail address: (M. Bloemendal).
Also often referred to as open loop geothermal storage systems. Closed loop or borehole heat exchangers also have a high potential for energy savings. In this paper the focus is on
ATES systems because they provides a more (cost) eective option for large scale cooling and heating in urban areas mainly for utility buildings and large scale housing complexes.
Geothermics 71 (2018) 306–319
0375-6505/ © 2017 Elsevier Ltd. All rights reserved.
several European countries (Eugster and Sanner, 2007; Fry, 2009;
Haehnlein et al., 2010; Willemsen, 2016). These ATES systems sea-
sonally store thermal energy at relatively low temperatures (< 25 °C)
alternating between cooling and, assisted by a heat pump, heating
mode (Fig. 1). The number of ATES systems has grown rapidly in the
past decade particularly in The Netherlands (Fig. 2), a country with a
moderate climate and widespread presence of thick sedimentary aqui-
fers. The introduction of progressively stricter energy eciency re-
quirements for buildings (Energy Performance Coecient (EPC)), sti-
mulated the adoption of ATES in the built environment. As a result,
there are currently almost 2000 systems in operation in relatively
shallow sandy aquifers (typically 20150 m.b.g.l.).
For both an optimal energy performance of an ATES system as well
as minimal eect on the subsurface, the thermal energy recovery e-
ciency needs to be as high as possible. Under these conditions, the
electricity required for groundwater pumping and heat pump (Fig. 1)is
Previous studies have shown that the thermal recovery eciency of
ATES systems are negatively aected by thermal energy losses from the
stored volume by conduction, diusion and dispersion (Doughty et al.,
1982; Sommer et al., 2014). While for high temperature (> 45 °C)
ATES systems, the negative impact of the buoyancy of the stored hot
water on thermal recovery eciency typically needs to be considered
(Lopik et al., 2016; Zeghici et al., 2015), this can be neglected for low
temperature ATES systems (Doughty et al., 1982; Zuurbier et al., 2013).
However, as these low temperature ATES systems are typically
targeting relatively shallow aquifers, the impact of stored volume dis-
placement by ambient groundwater ow requires consideration. Al-
though the impact of ambient groundwater ow on injected and re-
covered water volumes has been studied (Bear and Jacobs, 1965; Ceric
and Haitjema, 2005), the impact of ambient groundwater ow on
thermal recovery eciency in ATES systems, has thus far not been
explored. Moreover, it is unclear how the combined impact of these
processes (dispersion, conduction and advection) aects the thermal
recovery eciency of ATES systems under practical conditions and how
the eciency can be optimized.
Therefore, the aim of this study is to use analytical methods to
elucidate the impact of ambient groundwater ow and conduction and
dispersion on the thermal recovery eciency of ATES systems and to
use numerical methods to assess how the combined heat loss by mul-
tiple processes can be minimized. As a practical framework for the
conditions investigated, the wide range of ATES system characteristics
and hydrogeological conditions in the Netherlands was used. The re-
sulting insights are meant to provide a useful basis to enable the opti-
mization of the thermal recovery eciency of ATES systems and to
further optimize development for sustainable heating and cooling of
buildings world-wide.
ASurface area of the heat storage in the aquifer [m
αDispersivity [m]
Volumetric heat capacity of water; 4.2 × 10
Volumetric heat capacity of saturated porous medium;
2.8 × 10
Eective dispersion [m
Thermal dispersion [m
Average temperature dierence between warm and cold
well [°C]
EEnergy [J]
Thermal eciency []
iGroundwater head gradient []
kHydraulic conductivity [m/d]
Thermal conductivity of water and particles; 2.55 [W/m/
LWell screen length [m]
nPorosity; 0.3 []
Q Pumping rate of ATES wells [m
ρWater density; 1000 [kg/m
RThermal Retardation factor []
Thermal radius [m]
Hydraulic radius [m]
τDimensionless time of travel parameter []
Length of storage period [d]
TTemperature [°K]
t Time step [d]
uAmbient groundwater ow velocity [m/d]
vFlow velocity of the groundwater [m/d]
Velocity of the thermal front [m/d]
VYearly (permitted or actual) storage volume groundwater
Fig. 1. Illustration of the basic working principle of a low-temperature
seasonal ATES system. Left: in direct cooling mode while storing heat
for winter. Right: vice-versa in heating mode supported by a heat
pump while storing cooling capacity for summer.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
2. Materials and methods
2.1. Theory of heat transport and recovery during ATES
2.1.1. Denition of thermal recovery eciency for ATES systems
The thermal energy stored in an ATES system can have a positive
and negative temperature dierence between the inltrated water and
the surrounding ambient groundwater, for either heating or cooling
purposes (Fig. 1). In this study the thermal energy stored is referred to
as heat or thermal energy; however, all the results discussed equally
apply to storage of cold water used for cooling. As in other ATES studies
(Doughty et al., 1982; Sommer, 2015), the recovery eciency (η
an ATES well is dened as the amount of injected thermal energy that is
recovered after the injected volume has been extracted. For this ratio
between extracted and inltrated thermal energy (E
), the total
inltrated and extracted thermal energy is calculated as the cumulated
product of the inltrated and extracted volume with the dierence of
inltration and extraction temperatures (ΔT=T
) for a given
time horizon (which is usually one or multiple storage cycles), as de-
scribed by:
== =
ΔT Q dt
ΔT Q dt
out out
in in
with, Qbeing the well discharge during time step tand ΔTthe weighted
average temperature dierence between extraction and injection. In-
jected thermal energy that is lost beyond the volume to be extracted is
considered lost as it will not be recovered. To allow unambiguous
comparison of the results the simulations in this study are carried out
with constant yearly storage and extraction volumes (V
2.1.2. Loss of heat due to displacement by ambient groundwater ow
Signicant ambient groundwater ow is known to occur at ATES
sites (Bonte et al., 2013b; Groot, 2013; Hartog et al., 2013), which leads
to displacement of the injected volumes (Bear and Jacobs, 1965; Bonte
et al., 2013a). This may lead to signicant reduction in the thermal
energy recovery eciency of ATES systems as ambient groundwater
ow (u) contributes to thermal losses by displacing the injected water
during storage. The heat transport velocity (u
) is retarded with respect
to ambient groundwater ow (Doughty et al., 1982; Hecht-Mendez
et al., 2010) due to heat storage in the aquifer solids. The thermal re-
tardation (R) depends on porosity (n) and the ratio between volumetric
heat capacities of water (c
) and aquifer (c
, with c
and c
the solids volumetric heat capacity), following:
== ≈
10. 5
aq (2)
Resulting in a heat transport velocity at approximately 50% of the
groundwater ow velocity (u). Under conditions of ambient ground-
water ow, thermal energy stored in an aquifer will thus be displaced
and can only be partly (Bear and Jacobs, 1965) recovered.
2.1.3. Loss of heat by dispersion and conduction
Mechanical dispersion and heat conduction spread the heat over the
boundary of the cold and warm water bodies around the ATES wells. As
a consequence of the seasonal operation schedule, diusion losses are
Fig. 2. Top: increase of number of ATES systems
during recent years in the Netherlands along with the
decreasing EPC-standard for dwellings, The EPC
value is a normalized value of the expected energy
use of a building (CBS, 2016a; LGR, 2012; Ministry-
of-Internal-Aairs, 2012). Bottom: The increasing
percentage of new buildings build with ATES system
(CBS, 2016a,b).
Fig. 3. Simplied presentation of the resulting subsurface thermal and
hydrological storage cylinder for an ATES system for homogeneous
aquifer conditions.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
negligible (Anderson, 2005; Bear, 1979). Both other processes are de-
scribed by the eective thermal dispersion (D
) which illustrates the
relative contribution of both processes to the losses, following:
nc αv
where, the rst term represents the conduction, which depends on the
volumetric heat capacity (c
) of water and the thermal conductivity
) and porosity (n) of the aquifer material which are considered to
remain constant at about 0.15 [m
/d] in a sandy aquifer with porosity
of 0.3. The rate at which conduction occurs can be determined by the
increasing standard deviation: =
Dt2T, with D
, the eective
thermal dispersion (the left hand term of Eq. (3) and tthe storage time).
For half a year storage period the rate at which heat moves through
conduction is about 7 m.
The second term of Eq. (3) represents the mechanical dispersion,
which depends on the dispersivity (α) of the subsurface, porosity and
the ow velocity of the water (v), which is the sum of the force con-
vection due to the inltration and extraction of the well, as well as the
ambient groundwater ow (u). For ATES wells that fully penetrate an
aquifer conned by aquitards, the dispersion to cap and bottom of the
thermal cylinder (Fig. 3) is negligible due to the lack of ow (Caljé,
2010; Doughty et al., 1982). With regularly applied values of 0.55 for
the dispersivity (Gelhar et al., 1992), the dispersion is in the same order
of magnitude as the conduction at ow velocities of 0.010.1 m/d.
Since losses due to mechanical dispersion and conduction occur at
the boundary of the stored body of thermal energy, the thermal re-
covery eciency therefore depends on the geometric shape of the
thermal volume in the aquifer (Doughty et al., 1982). Following
Doughty et al. (1982), the inltrated volume is simplied as a cylinder
with a hydraulic radius (R
)dened as:
and for which the thermal radius (R
)isdened as:
R10.66 .
th win
hh h (5)
The size of the thermal cylinder thus depends on the storage volume
(V), screen length (L, for a fully screened aquifer), porosity (n) and
water and aquifer heat capacity (Fig. 3). This equation is approximate
because heterogeneities and partially penetration of the screens are
ignored. Doughty et al. (1982) introduced a dimensionless ratio of
screen length and the thermal radius (L/R
) as a parameter to describe
thermal recovery eciency of ATES systems for a particular stored
thermal volume. They found that the ATES recovery eciency has a at
optimum between a value of 1and 4 for this ratio.
Losses due to interaction between ATES systems are not taken into
account in this research. Also interaction between the warm and cold
well of the same system is not taken into account as this is prevented by
the permitting requirement to ensure sucient separation distance
(three times the thermal radius).
2.2. Numerical modeling of ATES
As losses due to conduction, dispersion and displacement occur si-
multaneously, MODFLOW (Harbaugh et al., 2000) simulations is used
to evaluate their combined eect on recovery eciency. For the si-
mulation of ambient groundwater ow and heat transport under var-
ious ATES conditions, a geohydrological MODFLOW model (Harbaugh
et al., 2000) coupled to the transport code MT3DMS (Hecht-Mendez
et al., 2010; Zheng and Wang, 1999). These model codes use nite
dierences methods to solve the groundwater and (heat) transport
equations. This allows for simulation of inltration and extraction of
groundwater in and from groundwater wells and groundwater tem-
perature distribution, as was done in previous ATES studies e.g. (Bonte,
2013; Caljé, 2010; Sommer, 2015; Visser et al., 2015). In the dierent
modeling scenarios the storage volume is varied between 12,000 and
300,000 m
with ow rates proportionally ranging from 8 to 200 m
screen lengths between 10 and 105 m and ambient groundwater ow
velocities between 0 and 50 m/y following the characteristics from
Dutch practice as will be introduced in the next section. Density dif-
ferences are neglected as this is considered a valid assumption (Caljé,
2010) for the considered ATES systems that operate within a limited
temperature range (< 25 °C). The parameter values of the model are
given in Table 1, the following discretization was used:
- Model layers; the storage aquifer is conned by two 10 m thick clay
layers. The storage aquifer is divided in 3 layers, a 5 m thick upper
and lower layer, the middle layersthickness is changed according to
the required screen length of the modeled scenario.
- The spatial discretization used in horizontal direction is 5 × 5 m at
well location, gradually increasing to 100 × 100 m at the borders of
the model. A suciently large model domain size of 6 × 6 km was
used to prevent boundary conditions aecting (< 1%) simulation
results. The gradually increasing cell size with distance from the
wells results the cell size of 15 m at 200 m of the well. This dis-
cretization is well within the minimum level of detail to model the
temperature eld around ATES wells as was identied by Sommer
et al. (2014).
- A temporal discretization of one week is used, which is suciently
small to take account for the seasonal operation pattern and re-
sulting in a courant number smaller than 0.5 within the area around
the wells where the process we care about occur. The simulation has
a horizon of 10 years, suciently long to achieve stabilized yearly
recovery eciencies.
The PCG2 package is used for solving the groundwater ow, and the
MOC for the advection package simulating the heat with a courant
number of 1. To set the desired ambient groundwater ow velocity for
the dierent scenarios simulated, the constant hydraulic head bound-
aries were used to set the required hydraulic gradient. In the aquifer an
ATES doublet is placed with a well distance of ve times the maximum
thermal radius of the wells to avoid mutual interaction between the
warm and cold storage volumes. In scenarios with groundwater ow,
the ATES wells are oriented perpendicular to the ow direction.
The energy demand prole of ATES systems varies due to variations
in weather conditions and building use which is of importance for the
actual value of the thermal eciency. For 12 varying scenarios the
eciencies are determined for both a weather dependent and the reg-
ular energy demand prole, showing that the eciencies of the corre-
sponding conditions dier. However, they show the same relation ac-
cording to the changes in conditions; the Pearson correlation coecient
of the two simulation result collections is 0,97. Based on this evaluation
all simulations are done with one basic energy demand prole, to allow
for comparison with the analytical solutions also the constant storage
volume energy demand pattern will be used; heat injection, storage,
Table 1
MODFLOW simulation parameter values (Caljé, 2010; Hecht-Mendez et al., 2010).
Parameter value
Horizontal conductivity aquifers 25 m/d
Horizontal conductivity aquitards 0.05 m/d
Longitudinal dispersion 1 m
Transversal dispersion 0,1 m
Bulk density 1890 kg/m
Bulk thermal diusivity 0.16 m
Solid heat capacity 880 J/kg °C
Thermal conductivity of aquifer 2.55 W/m °C
Eective molecular diusion 1·10
Thermal distribution coecient 2·10
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
extraction and again storage during 13 weeks each as is commonly done
in other ATES research (e.g. (Sommer et al., 2014; Zuurbier et al.,
2.3. Characteristics and conditions of ATES systems in The Netherlands
2.3.1. Characteristics of the ATES systems
Data on the location, permitted yearly storage volume, pump ca-
pacity and screen length of 331 ATES systems in The Netherlands (15%
of total number of systems) were obtained from provincial databases
that keep combined records for ATES characteristics of interest for this
research (Provinces of Gelderland, Noord-Brabant, Noord-Holland,
Utrecht and Drenthe, Fig. 4).
2.3.2. Geohydrological conditions at ATES systems
For a geographically representative subset of 204 ATES systems
(Fig. 4) it was possible to extract available aquifer thickness and derive
estimates on the ambient groundwater ow, as this additional data are
not available in the provincial databases. These estimates are based on
hydraulic conductivity and head gradients derived from the Dutch
geologic databases (TNO, 2002a) for the coordinates of these ATES
systems. The groundwater head gradient is read from equipotential
maps (TNO, 2002a) while the hydraulic conductivity and aquifer
thickness is obtained from local soil proles in the REGIS II (TNO,
2002a,b) subsurface model of the Netherlands and literature values for
hydraulic conductivity (Bear, 1979; Kasenow, 2002) corresponding to
the soil proles from the bore logs. The data are abstracted and pro-
cessed for the aquifer regionally targeted for ATES systems, therefore,
ATES systems with wells installed in other aquifers are excluded from
the local analysis. Legal boundaries are also taken into account, in
Noord-Brabant for instance it is not allowed to install ATES systems
deeper than 80 m below surface level, so any aquifer available below
80 m is disregarded for the systems in this province. For all locations a
porosity value of 30% is assumed, a value common for Dutch sandy
aquifers (Bloemendal et al., 2015; NVOE, 2006; SIKB, 2015a).
3. Results
3.1. ATES system properties in The Netherlands
3.1.1. Permitted capacity and screen length
The permitted capacity of the ATES systems ranges up to
5,000,000 m
/year but most (70%) are smaller than 500,000 m
year (Fig. 5,Table 2). The observed dierences in ATES system char-
acteristics for the dierent provinces were limited and therefore not
presented separately.
To be able to evaluate the resulting geometry of the storage volume
in evaluating dispersion and conduction losses it is assumed that the
Fig. 4. Locations of selected ATES systems from 5
provincial databases. Other provinces have ATES
systems as well but in their databases some char-
acteristics required for this evaluation were missing,
Open circles indicate locations for which ATES
characteristics were available. Filled circles indicate
locations for which also the local geohydrological
conditions were available.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
thermal energy is stored in a single cylindrical volume. Most ATES
systems in the Netherlands are single doublet systems or multiple
doublet systems with clustered warm and cold wells. However, parti-
cularly for some larger systems, warm and cold wells are not clustered,
due to for example spatial planning or geohydrological and/or geo-
technical reasons (Bloemendal et al., 2015). Unfortunately the pro-
vincial data did not include the number or type of well pairs. Therefore
the data was ltered for the systems for which a multiple number of
well pairs or other deviating aspects could be conrmed. Those systems
mostly belong to the largest 10% of the systems, or belong to outliers in
the data distribution of screen length over stored volume, and were
therefore excluded. For the largest systems, multiple doublets were
conrmed for several systems (e.g. C, D, F, G, H, I). In addition, some
errors were found in the data of the provincial databases, inconsistent,
incomplete entries (e.g. E) with errors (e.g. impossible short or long
screen lengths), such as monowell systems with only one very long
screen which should be divided in two screens (A and B in Fig. 6). As a
result of this validation of the dataset, 331 systems were selected for
further evaluation (Fig. 6). The data used for analysis represents about
15% of the approximately 2000 systems operational in the Netherlands
(Willemsen, 2016).
3.1.2. Geohydrological conditions
Table 3 shows the overall geohydrological characteristics at the
location of 204 ATES systems. Both hydraulic conductivity and ambient
groundwater ow velocity show a wide range.
3.2. Analytical evaluation of ATES thermal recovery
3.2.1. Loss of thermal energy due to dispersion and conduction
Both conduction and dispersion losses occur at the boundary of the
stored thermal cylinder. Following Eq. (3); near the well, where ow
velocity of the inltrated water (v) is high, dispersion dominates the
conduction term, while further from the well, the eects of dispersion
decreases. Eq. (3) and the values for the dispersion and aquifer prop-
erties in Table 1 are now used to identify the distance from the well at
Fig. 5. Frequency distribution of dataset according to permitted
yearly storage volume of groundwater. Distribution of well design
metrics of selected data is shown separately.
Table 2
ATES system characteristics in provincial datasets selected for this study.
Number of ATES systems Permitted capacity (V)[m
/y] Installed screen length (L)[m]
0.25 perc. Average 0.75 perc. 0.25 perc. Average 0.75 perc.
Initial data 434 90,000 539,000 674,000 20 37 45
selected data 331 80,000 244,000 320,000 20 32 40
Fig. 6. Dataset characteristics; outliers are excluded from the dataset.
A, B = monowells with only top of upper and bottom of lower lter in
the data, C = University Campus 6 doublets, D = Oce with 3
doublets, E = Oce building with only extracted volume of one year
available in data, unrealistically small for size of building, F = oce
with 4 doublets, G = Hospital with 4 doublets, H = conference center
with 2 monowells, I = Oce with 3 doublets.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
which the dominating process contributing to loss, changes from dis-
persion to conduction, Fig. 7. The pump capacity data of the ATES
systems together with the storage volume and screen length are used to
plot the thermal radii of the systems with respect to their maximum
specic discharge, showing that even assuming a relatively high dis-
persivity of 5 m, beyond 10% of permitted storage volume inltration,
conduction is dominating in the dispersivity equation, indicating that at
full storage capacity conduction losses will be dominating.
When the inltration continues, the movement of the thermal front
is dominated by the advective heat transport of the injection., The
(high) dispersion losses that occur at the high ow velocities close to
the well are overtakenwhen inltration of heat continues, resulting
in sharp heat interface as the inltration volume increases. This sharp
interface remains sharp during inltration because the heat injected by
the well travels faster than the standard deviation for the conduction
T. During storage and extraction the interface will become
less sharp due to respectively conduction and the opposite eect of
these mechanisms. The heat that thus stays behind causes that e-
ciency improves and stabilizes over multiple storage cycles. From which
it is concluded that losses can be minimized by minimizing the total
surface area of the circumference and the cap and bottom of the
thermal cylinder (A) of the stored heat volume (V) in the aquifer. This
can be done by identifying an appropriate screen length according to
the required storage volume and local conditions, in order to minimize
the surface area volume ratio;
πR πR L
22 22
th th
th th
For any given storage volume an optimal screen length exists at
which conduction and dispersion losses are minimal at the screen
length thermal radius ratio (L/R
) is 2, when the diameter of the
cylindrical storage volume is equal to its screen length. From Fig. 8 can
be seen that for larger storage volumes the A/V-ratio is smaller, and less
sensitive at larger screen lengths, exhibiting a relatively at minimum
compared to small storage volumes. Although the absolute losses in-
crease with increasing storage volume, the relative losses are smaller.
To identify the optimal screen length the derivative for surface area
of the thermal cylinder is equated to zero, which results in an expres-
sion for optimal screen length as a function of required storage volume;
=+ →
cL πcV
πc L LA πc V
cL πcV
πc LL
22 21
Consequently, relatively small storage volumes experience higher
losses due to dispersion losses. Because there is no or little ow to and
from the conning layers of an ATES well, conduction losses along the
interface with the conning soil layers may dier from the ones around
the circumference. Therefore Doughty et al. (1982) distinguished be-
tween the two in their research to optimize well design, to account for
the reduced conduction losses to conning layers after several storage
cycles. Their Simulation showed that eciency increases with the rst
number of storage cycles and found that the optimal ratio between
screen length and thermal radius (L/R
) has a at optimum around 1.5
when taking into account dierent thermodynamic properties of aqui-
fers and aquitards. Substituting the expression for the thermal radius
, Eq. (5)) in the optimal relation of L/R
=1.5 gives the optimal
screen length (L) as a function of storage volume (V);
2.25 1.02
This shows that the solution for the screen length results in the same
third root of the storage volume, only with a smaller constant 1.02 []
instead of 1.23 [] as was derived from the optimal A/V-ratio solution,
Eq. (7) &(8). This is the case because over multiple cycles, the con-
duction losses to cap & bottomdecrease; losses from earlier cycles
dampen the losses during following cycles.
From the lines for L/R
is 1.5 it can be seen that on average, screen
lengths are designed far from optimal with respect to minimizing
conduction losses. Doughty et al. (1982) however, found a at optimum
for L/R
-value, thus it may also be acceptable when the L/R
-value is
between 1 and 4, based on the moment of deection of the L/R
constructed by Doughty et al. (1982). However most systems have L/
-values lower than 1, indicating that screen lengths used in Dutch
practice are relatively short (Fig. 9). Analysis shows that 56% of the
ATES systems with an L/R
<1 have insucient aquifer thickness
available for longer screens.
3.2.2. The eect of ambient groundwater ow on recovery eciency
For the analysis of the impact of ambient groundwater ow on the
recovery eciency, it is assumed that a cylindrical shape of the injected
volume is maintained during displacement. Ceric and Haitjema (2005)
determined that this assumption is valid for conditions where their
dimensionless time of travel parameter τ,(Ceric and Haitjema, 2005)is
smaller than one;
==τπki Lt
πnu Lt
2() 2
sp sp
Table 3
Ranges in geohydrological characteristics of the 204 ATES systems under consideration,
for which geohydrological conditions could be retrieved.
Available aquifer
thickness range
Hydraulic conductivity
Groundwater ow range
30180 545 3100
Fig. 7. Lines: the relation between specic well discharge and radial
distance at which the radial ow velocities where conduction and
dispersion are equal (Eq. (3)) for the outer-bounds of the range of
thermal dispersivity regularly applied in literature. Open circles the
thermal front of the ATES systems in the data at dierent storage
capacities related to their specic well discharge.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
The groundwater head gradient (i), hydraulic conductivity (k),
screen length (L) and pumping rate (Q) of the ATES systems in the data
are used to determine the time of travel parameter for each system. The
only unknown is the length of storage period (t
). With an average
storage period of 183 days (half a year) only one of calculated τvalues
for the 204 ATES systems was larger than one; a very small system in
high ambient groundwater ow velocity. On top of meeting the re-
quirement of Ceric and Haitjema, the thermal retardation also causes
the heat to ow at half the speed of water, which then makes the as-
sumption of preservation of a cylindrical shape during displacement an
acceptable simplication. These conditions allow the denition of the
recovery eciency as a function of the overlapping part of the cylin-
ders, with and without the displacement induced by ambient ground-
water ow. Assuming that the ambient groundwater ow is horizontal,
the surface area of the thermal footprints before and after displacement
with the groundwater ow represents this eciency, Fig. 10 (top).
Goniometric rules are used to express the overlapping surface area
) of the thermal footprint as a function of groundwater ow
velocity and thermal radius, as follows:
Rtu R tu2cos
overlap th
sp th sp
in which the velocity of the thermal front (t
) is 2 times PO in Fig. 10
(top). Substituting the relation between eciency (η
), thermal foot-
print (A
) and overlapping area:
=→=A AηπR
overlap th footpr overlap th thint
results in a relation between eciency, ow velocity and the thermal
πR Rtu
2cos *
th sp
For every ATES system with τ< 1 the eciency can be obtained
with this relation. When R
>u, the t
-term under the square root
contributes less than 1% to the obtained eciency. Under these con-
ditions, both right and left term of Eq. (12) depend on the ratio between
the traveled distance and the thermal radius. So for any constant
combination of u
over R
, the eciency is the same, which allows to
identify the eciency as a function of the R
/u-ratio for dierent sto-
rage periods, Fig. 10 (bottom). This can be used to identify minimum
desired thermal radius (i.e. maximum desired screen length for a given
storage volume) at a location with a given groundwater ow velocity to
meet a minimal eciency.
The derived relation is now used to assess the well design data with
respect to the local ambient groundwater ow velocity, hydraulic
conductivity and thickness of the aquifer. For each of the ATES systems
in the dataset the R
/uvalue was determined, the relation given in
Fig. 10 (bottom) is used to indicate lines of expected thermal eciency
only taking into account losses due to displacement caused by ambient
groundwater ow, Fig. 11.
Fig. 11 shows that about 20% of the systems have an expected ef-
ciency lower than 80% (R
/u < 1.1). For the ATES systems with an
expected eciency lower than 80% (Table 4) the average storage vo-
lume is relatively small and the average ow velocity relatively high at
36 m/y. Although minimizing screen length reduces heat losses due to
displacement, minimizing for conduction and dispersion losses require
an optimal screen length for a particular storage volume.
3.2.3. Conclusion analytical analysis
In optimizing the storage geometry of ATES systems the applied
length should be carefully considered. However, in both Figs. 6 and 9 it
can be seen that many ATES systems with varying storage volumes have
identical screen lengths, at various multiplications of 5 m. This likely
relates to the fact that screen sections are supplied in 5 m sections,
Fig. 8. The A/V values for dierent storage volumes and well screen
Fig. 9. L/R
-value relative to permit volume of ATES systems in
practice, combined with minimum (L/R
= 1), maximum (L/
= 4) and optimal (L/R
= 1.5) L/R
for conduction and disper-
sion losses.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
which can, but are not adjusted to a specically required length. The
wide range of storage volume per single screen length (e.g.
40,000420,000 m
for L= 20, Fig. 9) thus indicates that the screen
length design indicated in the permit application are generally not
based on an evaluation of storage volume and local geohydrological
conditions, Dutch design standards only consider the clogging potential
for ATES well design (NVOE, 2006). Particularly for smaller ATES
systems, the sensitivity of recovery eciency for screen length selection
is high, as these are most vulnerable for signicant losses as a con-
sequence of ambient groundwater ow and dispersion and conduction
(Figs. 8 and 10).
3.3. Numerical evaluation of energy losses
To assess the combined eect of conduction, dispersion and dis-
placement losses, the results of the performed numerical MODFLOW
simulations are discussed and compared with the straightforward and
simple analytical solutions presented in the previous section. The wide
range of ATES conditions for which the numerical simulations were
performed resulted in recovery eciencies between 10 and 70%.
(Fig. 12).
3.3.1. Contribution of displacement losses
The lowest eciencies are associated with the scenarios with high
ambient groundwater ow (> 50 m/year), together with relatively
small thermal radius, which results in a small thermal radius over
ambient groundwater ow (R
/u-ratio < 1 y). For both the numerical
and the analytical solution for the impact of ambient groundwater ow
on recovery eciency is very sensitive for low R
/u-values. However,
at higher R
/uvalues (> 1 y) the eciency becomes less dependent of
/u, as dispersion and conduction losses are dominant under such
conditions. In all cases the analytical solution overestimates the e-
ciency compared to numerical results, because the analytical solution
does not take account for conduction and dispersion losses. To estimate
Fig. 10. Top: schematic overview of calculating the overlapping sur-
face area of 2 identical thermal cylinders. Bottom: the derived ana-
lytical relation between losses and the thermal radius groundwater
ow velocity ratio.
Fig. 11. R
/u-values for ATES systems in the dataset with thresholds
for dierent thermal recovery eciencies.
Table 4
Results of analysis of screen length with respect to groundwater ow velocity.
average uaverage Vaverage R
[m/y] [m
/y] [m]
η> 80% 6 263,000 46
η< 80% 33 100,000 32
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
the eciency for the numerical simulations that would result under the
impact of displacement only, the obtained eciencies under no ow
conditions are used as a reference (following (Ŋ
) for u= 5 m/y; Ŋ
). These numerically derived estimates show a good re-
semblance with the analytical relation. This conrms that the analytical
approach is valid to determine displacement losses separately.
3.3.2. Contribution of conduction and dispersion losses
Simulated eciencies for the scenarios without ambient ground-
water ow were highest, up to 75%, and highly correlated with the
surface area over volume ratio A/V (Fig. 13), in contrast with the si-
mulations with the highest ambient groundwater ow (50 m/y). Also
the A/V ratios calculated for earlier simulation studies and experiments
without ambient groundwater ow (Caljé, 2010; Doughty et al., 1982;
Lopik et al., 2016) strongly correlate with the observed eciencies in
these studies. Like in this study, the results from Lopik et al. (2016) and
Doughty et al. (1982) consist of a series systematic changing boundary
conditions which allows for verication of the relations found in
Fig. 13. Results of both Lopik et al. (2016) and Doughty et al. (1982)
show a linear relation with similar slope between the surface area over
volume ratio (A/V) and eciency in the absence of ambient ground-
water ow. The excellent correlation eciency with the A/V ratio for
each study with no ambient groundwater ow, indicates that under
similar condition the eciency of ATES systems for a particular aquifer
system and operational mode can be interpolated based on A/V.
Although similar, the eciencies at a particular A/V ratio deviate
for these dierent modeling studies and are likely to be caused by small
dierences in parameters and model set-up. E.g.; both Doughty et al.
(1982) and Lopik et al. (2016) used an axisymmetric model and a ner
vertical spatial discretization compared to this study, resulting in dif-
ferences in numerical dispersion. Also, Doughty et al. (1982) uses no
dispersion, which explains why their simulations show the highest ef-
ciency. Lopik et al. (2016) uses shorter and less storage cycles as well
as a slightly smaller dispersion coecients compared to this study.
From these (small) dierences can be seen that at simulations with
higher dispersion, the A/V eciency relation becomes steeper, small
systems which have a larger A/V ratio then suer relatively more,
conrming the earlier observation from Fig. 7 that at larger storage
volumes conduction losses dominate.
3.3.3. Combined displacement and conduction & dispersion losses
As found by Doughty et al. (1982) the optimum for L/R
ratio for a
particular ATES storage volume is around 1.5 in the absence of ambient
groundwater ow. However this optimal ratio shifts to lower values
with increasing ambient groundwater ow velocity (Fig. 14). The op-
tima remains at for higher groundwater ow velocity, only for the
smallest system (12,000 m
) at the highest ambient groundwater ow
(50 m/y) tested, this is not the case within the simulated conditions.
To identify the optimal L/R
at dierent rates of groundwater ow
velocity, the L/R
value of the simulation series of each storage volume
and groundwater ow velocity with the highest eciency was selected
from the dierent L/R
scenarios simulated. To take into account the
at optima also the L/R
values with less than 5% deviation in e-
ciency were selected. For each of the simulated ambient groundwater
ow velocity, the average and the standard deviation of the optimal L/
values were calculated and plotted in Fig. 15. This empirical relation
shows how the well design for ATES wells can be optimized taking
account conduction, dispersion and displacement losses. It also shows
that at higher ambient groundwater ow, well design is more critical,
since the allowed deviation of the optimal solution becomes smaller.
Despite the limited number of simulations (120), the number and
spreading of dierent conditions is sucient to use this relation in
design practice.
Fig. 12. Relation between eciency and thermal radius over
groundwater ow velocity (R
/u) for numerical simulation results
and analytical solution (Equation (12)) for a 0.5 y storage period.
Fig. 13. Simulated eciencies relative to geometric property (A/V)
from this and other studies at u = 0 m/y and for u = 50 m/y from the
simulations done in this study. The Pearson correlation between A/V
and eciency is 0,99 for u = 0 m/y. and 0.58 for u = 50 m/y.
From the Lopik et al. (2016) study, only the data are used from the
simulations that excluded buoyancy ow.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
4. Discussion
4.1. Size and variation in seasonal storage volume
As shown in this research storage volume is an important parameter
aecting recovery eciency. In assessing this eciency it has been
assumed that the inltrated and extracted volume is equal for each
cycle. However, in practice the inltration and extraction volume from
wells are typically not equal due to variations in heating and cooling
demand. This can have a signicant inuence on the perceived recovery
eciency per cycle. Monitoring data indicates energy imbalances
varying between 22% and +15% (Willemsen, 2016). Because in
general ATES systems have to meet energy balance for a certain period,
in The Netherlands 35 years depending on provincial legislation, a
representative storage volume can be used to assess conduction and
displacement losses. Because the absolute losses increase with in-
creasing storage volumes, it is more benecial to optimize for maximum
storage volume. This is also reected in Eq. (7) where can be seen that
the A/V-value has a at optimum at larger storage volumes (Fig. 8), and
also in the relation identied by Doughty et al. (1982) and shown in
Fig. 14. Therefore, the permitted capacity data of the ATES wells in The
Netherlands were used to compare theoretical well design approaches
with eld data, Fig. 9. However, in practice ATES systems deviate from
their permit capacity to store heat because ATES operators request a
larger permit capacity to allow for exibility during operation; e.g.
building energy demand may be higher than expected, possible future
growth, change of building function and seasonal uctuations. This
inuences the shape and thus the losses of the heat storage. Operational
data of ATES systems from dierent databases have been used in
Fig. 14. Simulated eciencies for di erent groundwater ows (u)
and screen length over thermal Radius (L/R
) of various storage vo-
lumes. A. is at no/low ambient groundwater ow (Doughty applies).
B. is at high ambient groundwater ow.
Fig. 15. Optimal L/R
for dierent groundwater ows empirically derived from simu-
lation results.
Fig. 16. Volume in storage of warm well for dierent energy demand
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
regional and national studies and evaluations (CBS, 2005; Graaf et al.,
2016; SIKB, 2015b; Willemsen, 2016) all showing that ATES systems
yearly actually only use 4060% of their initially requested and per-
mitted capacity. The ranges of systems sizes presented in this study, e.g.
Figs. 5 and 6, are therefore much smaller in practice.
Also variations in seasons aect the total storage volume in the
ATES wells. In this study the common assumption was made, that the
average yearly volume is inltrated and extracted during the winter
and summer, with a storage period in between, resulting in a block-
scheme like inltration, storage and extraction pattern. However,
heating and cooling demand typically does not balance perfectly during
a year and seasonal variations may cause temporal imbalances, re-
sulting in a sometimes smaller and sometimes larger heat storage
compared to the yearly average storage. For example, heat may remain
in warm wells during a couple of warm winters until a colder winter
depletes the warm well. The eect of this aspect is illustrated by the
presentation of the cumulative volume stored in a well relative to the
average value for multiple years, Fig. 16. This pattern is derived from
the storage volume variation based on the monitored and projected
outside air temperature (20102020) of the weather station of De Bilt in
The Netherlands (KNMI, 2013). The energy demand pattern is de-
termined by deriving the energy demand for each day by scaling the
yearly average energy demand to the deviation of the daily temperature
from the average outside air temperature of the evaluation period. As a
result of this seasonal variations imbalances occur over the years, re-
sulting in varying stored volume in the wells. From Fig. 16 can be seen
that the maximum storage capacity occurring in practice is around
150% of the average yearly storage volume. This exercise was done for
dierent climatic datasets (monitored as wells as projections), all giving
the same outcome, that the maximum storage in the well is about 150%
of the average yearly storage.
The fact that well design can be best determined for maximum
storage volume, then leads to the conclusion that 150% of the expected
yearly average storage volume, which in turn is about 75% of the
permitted capacity (50% of permitted capacity is used in practice) must
be used as a basis for well design. Correcting the data of the permitted
volumes for these two aspects results in the ATES systems plotted in
Figs. 9 and 11 to respectively move up- and downwards.
4.2. Additional well design criteria in practice
The well design criteria required to assess and optimize the thermal
recovery eciency were considered in this study. However, in practice
additional aspects such as capacity, prevention of well clogging,
available aquifer thickness, mutual interaction and drilling and in-
stallation costs all play a role in determining the well design. In practice
the determination of screen length is mainly based on the maximum
desired pumping rate (NVOE, 2006). Together with minimizing drilling
costs this is a driver for screen lengths that are too short to achieve
optimal thermal eciency, which is clearly reected in Fig. 9. In the
Netherlands, a clear guideline or method available to take account for
losses as a result of ambient groundwater ow in well design is cur-
rently lacking (NVOE, 2006), which is reected in Fig. 11. The eect of
a partially penetrating well on the distribution and A/V-ratio of heat is
both not discussed in this study and not taken into account in current
practice. However, given the identied signicant eect of the A/V-
ratio on eciency, the eciency of a partially penetrating well may
deviate signicantly from a fully penetrating well with the same storage
volume and screen length. For partially penetrating wells the aquifer
anisotropy is also an important parameter to consider.
In this study is shown that suboptimal well design may have a large
inuence on well eciency, but can also be limited relatively easily. As
shown in Figs. 8 and 14, the dependency for both A/V and L/R
eciency has a at optimum beyond some threshold, which then al-
lows dealing with local aquifer thickness conditions and uncertainties
in storage volume now this threshold is known.
4.3. The impact of ambient groundwater ow on the eciency of ATES
High ambient groundwater ow aects the recovery eciency of
ATES systems signicantly. The missing framework to assess stored
heat losses due to groundwater ow is introduced in this paper. Also the
orientation of ATES wells with respect to the ambient ow direction
needs to be taken into account. Warm and cold wells need to be or-
iented perpendicular to the ow direction. For individual systems this
framework helps to improve well eciency, a drawback of the pre-
sented framework is, however, the resulting large thermal radii and
suboptimal use of aquifer thickness. In areas with many ATES systems
close together this may lead to scarcity of subsurface space for ATES. In
such busy areas with high ambient groundwater ow, planning stra-
tegies should work towards placement of same type of wells in the di-
rection of the groundwater ow, where then only the most upstream
wells will suer from losses due to groundwater ow, for which com-
pensation arrangements may be made. Multi doublet systems on the
other hand may better use the strategy to place well of the same type in
the direction of the ow and inltrate relatively more heat in the up-
stream and extract more from the downstream well to compensate for
the ambient groundwater ow losses, as was described by Groot (2013).
4.4. The eect of aquifer conditions
The shape of the stored heat was assumed to have a cylindrical
shape in this evaluation of well design. However, in a heterogeneous
aquifer the storage volume does not have the shape of a perfectcy-
linder, resulting in a varying thermal radius over the depth of the
screen. As a consequence of heterogeneity the A/V-ratio in practice is
higher compared to the expected value for a homogeneous aquifer.
Although they both use a single ATES conguration, Sommer et al.
(2013) and Caljé (2010) show that the net eect of heterogeneity on
eciency is limited over multiple storage cycles and its inuence is
much smaller compared to the eect of A/V and ambient groundwater
ow on the eciency. Only when gravel layers are present such het-
erogeneity may aect eciency signicantly, and should therefore best
be blinded (Caljé, 2010). Next to variations in hydraulic conductivity,
also variations in salinity may aect the shape of the storage volume
due to buoyancy ow due to density dierences. Such aspects will af-
fect the eciency dependencies derived for the homogeneous and
isotropic conditions evaluated in this study. Also the eciency de-
pendency for application of ATES in more challenging geohydrological
environments will require further study.
4.5. Combined wells and mutual interaction
This study focusses on optimizing the recovery eciency of a single
ATES systems and individual wells, ATES systems however cumulate in
urban areas (Bloemendal et al., 2014; Hoekstra et al., 2015) and reg-
ularly share subsurface space to store or extract heat. As a consequence,
additional considerations need to be taken into account, which might
lead to deviations from the design consideration presented in this re-
search. For example, planning of subsurface space occurs based on the
thermal footprint (Fig. 3) of an ATES well projected at surface level
(Arcadis and Bos, 2011; Li, 2014), which then promotes the use of
longer screens. From the at optima shown in Fig. 14 it can be seen that
the individual well eciency may not have to suer much from such
additional consideration. This will allow larger number of ATES sys-
tems to be accommodated in such areas and with that the overall CO
emission reduce (Jaxa-Rozen et al., 2015). Also, large ATES systems
often have multiple warm and cold wells which are placed together and
function as one single storage in the subsurface. The length of the
screens of such combined wells should therefore also be determined
based on the fact that they function as one storage volume in the sub-
surface, disregarding this aspect gives a suboptimal A/V and amplies
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
the eect of having a larger footprint, in areas where this must be
prevented. From this is concluded that combining wells, also requires a
well design for the individual wells based on storage capacity of both
wells together. However, in such busy aquifers best would be to pro-
mote the use of the full aquifer thickness for wells and use a full 3D
planning strategy.
5. Conclusion
In this study an evaluation of ATES characteristics from practice
together with analytical and numerical simulations were used to de-
velop the missing framework for ATES well design to achieve optimal
recovery eciency. This work includes the losses due to heat dis-
placement with ambient groundwater ow. The results show that two
main processes control thermal recovery eciencies of ATES systems.
These are due to the thermal energy losses that occur 1) across the
boundaries of the stored volume by mainly conduction and dispersion
only at smaller storage volumes and 2) due to the displacement of
stored volumes by ambient groundwater ow.
For the latter process, an analytical expression was deduced that
suitably describes thermal recovery eciency as a function of the ratio
of the thermal radius over ambient groundwater ow velocity (R
For the conditions tested, at R
/u < 1 the displacement losses were
dominant and thus would require minimization of the well screen
length or maximize the volume stored. Obviously, practical aspects,
such as required minimum well capacity or the availability of suitable
aquifers, may prevent the use of optimal screen lengths as is illustrated
for a large part (15%) of the evaluated Dutch ATES systems that in-
dicate an a eciency of less than 50%, due to ambient groundwater
ow (Fig. 11).
With respect to the dispersion and conduction losses it was shown
that conduction is dominating and for the numerical simulation results
of this and previous studies, thermal recovery eciency linearly in-
creases with decreasing surface area over volume ratios of the stored
volume (A/V) for a particular set of operational and geohydrological
conditions. With respect to the losses due to conduction and dispersion,
the optimal screen length has a at optimum, which allows to also take
account for other considerations in well design like neighboring sys-
tems and partially penetrating eects.
For the optimization of thermal recovery eciency with respect to
both main processes, the optimal value for the ratio of well screen
length over thermal radius (L/R
) decreases with increasing ambient
groundwater ow velocities as well as its sensitivity for eciency. With
the insights on the controls on thermal recovery eciency derived in
this study, the assessment of suitable storage volumes, as well as the
selection of suitable aquifer sections and well screen lengths, can be
supported to maximize the thermal recovery of future seasonal ATES
systems in sandy aquifers world-wide.
This research was supported by Climate-kic E-use (aq) and the
URSES research program funded by the Dutch organization for scien-
tic research (NWO) and Shell, grant number 408-13-030. We thank
two anonymous reviewers for their valuable comments on the manu-
Anderson, M.P., 2005. Heat as a ground water tracer. Ground Water 43, 951968.
Arcadis, T.T.E., Bos, W., 2011. Handreiking masterplannen bodemenergie. SKB, Gouda.
Bear, J., Jacobs, M., 1965. On the movement of water bodies injected into aquifers. J.
Hydrol. 3, 3757.
Bear, J., 1979. Hydraulics of Groundwater. Dover Publications Inc., Mineola, New York.
Bloemendal, M., Olsthoorn, T., Boons, F., 2014. How to achieve optimal and sustainable
use of the subsurface for Aquifer Thermal Energy Storage. Energy Policy 66,
Bloemendal, M., Olsthoorn, T., van de Ven, F., 2015. Combining climatic and geo-hy-
drological preconditions as a method to determine world potential for aquifer
thermal energy storage. Sci. Total Environ. 538, 621633.
Bonte, M., 2013. Imacts of Shallow Geothermal Energy on Groundwater Quality, Geo
Sciences. Vrije Universiteit Amsterdam, Amsterdam.
Bonte, M., Mesman, G., Kools, S., Meerkerk, M., Schriks, M., 2013a. Eecten en risico's
van gesloten bodemenergiesystemen. KWR Watercycle Research Institute,
Bonte, M., Van Breukelen, B.M., Stuyfzand, P.J., 2013b. Environmental impacts of aquifer
thermal energy storage investigated by eld and laboratory experiments. J. Water
Clim. Change 4, 77.
CBS, 2005. In: CBS (Ed.), Hernieuwbare energie in Nederland 2004. Central Authority for
Statistics in NL, Den Haag.
CBS, 2016a. In: CBS (Ed.), Issued Building Permits 19902015 Retrieved from Statline
Database. Central authority for statistics in NL, Den Haag.
CBS, 2016b. Voorraad woningen en niet-woningen; mutaties, gebruiksfunctie, regio.
Caljé, R., 2010. Future Use of Aquifer Thermal Energy Storage Inbelow the Historic
Centre of Amsterdam, Hydrology. Delft University of Technology, Delft.
Ceric, A., Haitjema, H., 2005. On using Simple Time-of-travel capture zone delineation
methods. Groundwater 43, 403412.
Doughty, C., Hellstrom, G., Tsang, C.F., 1982. A dimensionless approach to the Thermal
behaviour of an Aquifer Thermal Energy Storage System. Water Resour. Res. 18,
EIA, 2009. Residential Energy Consumption Survey. US Energy Information
EU, 2010. In: In: Union O.J.o.t.E (Ed.), Directive on the Energy Performance of Buildings
153. EU-Parliament, European Union, Strasbourg, pp. 1335.
Eugster, W.J., Sanner, B., 2007. Technological Status of Shallow Geothermal Energy in
Europe. European Geothermal Congress, Unterhaching, Germany.
Fry, V.A., 2009. Lessons from London: regulation of open-loop ground source heat pumps
in central London. Geol. Soc. Lond. 42, 325334.
Gelhar, L.W., Welty, C., Rehfeldt, K.R., 1992. A critical review of data on eld-scale
dispersion in aquifers. Water Resour. Res. 19551974.
Graaf, A.D., Heijer, R., Postma, S., 2016. Evaluatie Wijzigingsbesluit bodemenergiesys-
temen. Buro 38 in commision of ministry of Intrastructure and environment, Cothen.
Groot, J., 2013. Optimizing Energy Storage and Reproduction for Aquifer Thermal Energy
Storage Geosciences. University of Utercht.
Haehnlein, S., Bayer, P., Blum, P., 2010. International legal status of the use of shallow
geothermal energy. Renew. Sustain. Energy Rev. 14, 26112625.
Harbaugh, A.W., Banta, E.R., Hill, M.C., McDonald, M.G., 2000. In: USGS (Ed.), Modow-
2000, The U.S. Geological Survey Modular Ground-Water ModelUser Guide to
Modularization Concepts and the Ground-Water Flow Process. US Geological Survey,
Hartog, N., Drijver, B., Dinkla, I., Bonte, M., 2013. Field assessment of the impacts of
Aquifer Thermal Energy Storage (ATES) systems on chemical and microbial
groundwater composition. In: European Geothermal Conference. Pisa.
Hecht-Mendez, J., Molina-Giraldo, N., Blum, P., Bayer, P., 2010. Evaluating MT3DMS for
heat transport simulation of closed geothermal systems. Ground Water 48, 741756.
Hoekstra, N., Slenders, H., van de Mark, B., Smit, M., Bloemendal, M., Van de Ven, F.,
Andreu, A., Sani, D., Simmons, N., 2015. Europe-Wide Use of Sustainable Energy
from Aquifers, Complete Report. Climate-KIC.
Jaxa-Rozen, M., Kwakkel, J.H., Bloemendal, M., 2015. The Adoption and Diusion of
Common-Pool Resource-Dependent Technologies: The Case of Aquifer Thermal
Energy Storage Systems. PICMET, Portland.
KNMI, 2013. Hourly data 1980-2010, Climatic Projections KNMI W+ Scenario
20102045 for Weather Station 260, de Bilt.
Kasenow, M., 2002. Determination of Hydraulic Conductiviy from Grain Size Anaysis.
Water Resources Publications LLC, Highlands Ranch, Colorado.
Kim, J., Lee, Y., Yoon, W.S., Jeon, J.S., Koo, M.-H., Keehm, Y., 2010. Numerical modeling
of aquifer thermal energy storage system. Energy 35, 49554965.
LGR, 2012. In: Netherlands P.o.T (Ed.), Dutch Register/Database for Groundwater
Li, Q., 2014. Optimal Use of the Subsurface for ATES Systems in Busy Areas Hydrology.
Delft University of Technology.
Lopik, J.H.V., Hartog, N., Zaadnoordijk, W.J., 2016. The use of salinity contrast for
density dierence compensation to improve the thermal recovery eciency in high-
temperature aquifer thermal energy storage systems. Hydrogeol. J. 24 (5),
Ministry-of-Internal-Aairs, 2012. Bouwbesluit (Building Act). Ministry-of-Internal-
Aairs, Den Haag.
Molz, F.J., Warman, J.C., Jones, T.E., 1978. Aquifer storage of heated water.
Groundwater 16.
Molz, F.J., Melville, J.G., Parr, D.D., King, D.A., Hopf, M.T., 1983. Aquifer thermal energy
storage' a well doublet experiment at increased temperatures. Ground Water 19,
NVOE, 2006. Richtlijnen Ondergrondse Energieopslag, Design Guidelines of Dutch
Branche Association for Geothermal Energy Storage. Woerden.
Nagano, K., Mochida, T., Ochifuji, K., 2002. Inuence of natural convection on forced
horizontal ow in saturated porous media for aquifer thermal energy storage. Appl.
Therm. Eng. 22, 12991311.
Réveillère, A., Hamm, V., Lesueur, H., Cordier, E., Goblet, P., 2013. Geothermal con-
tribution to the energy mix of a heating network when using Aquifer Thermal Energy
Storage: modeling and application to the Paris basin. Geothermics 47, 6979.
RHC, 2013. Strategic Research and Innovation Agenda for Renewable Heating & Cooling.
Renewable Heating & Cooling, European Technology Platform, Brussels.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319
SER, 2013. Energie Akkoord. Social Economical Counsil.
SIKB, 2015a. BRL 2100 Mechnisch boren. SIKB, Gouda.
SIKB, 2015b. In: NL d.p.i (Ed.), Userdata of ATES Systems, Recieved from 4 provinces of
the Netherlands in 2015, Gouda.
Sommer, W., Valstar, J., van Gaans, P., Grotenhuis, T., Rijnaarts, H., 2013. The impact of
aquifer heterogeneity on the performance of aquifer thermal energy storage. Water
Resour. Res. 49, 81288138.
Sommer, W.T., Doornenbal, P.J., Drijver, B.C., van Gaans, P.F.M., Leusbrock, I.,
Grotenhuis, J.T.C., Rijnaarts, H.H.M., 2014. Thermal performance and heat transport
in aquifer thermal energy storage. Hydrogeol. J. 22, 263279.
Sommer, W., 2015. Modelling and Monitoring Aquifer Thermal Energy Storage.
Wageningen University, Wageningen.
TNO, 2002. Boringen uit Dinoloket, digitale grondwaterkaart, Utrecht.
TNO, 2002. REGIS, Utrecht.
Tsang, C.F., 1978. Aquifer thermal energy storage. Institute of Gas Technology
Symposium on Advanced Technologies for Storing Energy.
UN, 2015. Adoption of the Paris Agreement. United Nations, Framework Convention on
Climate Change, Paris.
Visser, P.W., Kooi, H., Stuyfzand, P.J., 2015. The thermal impact of aquifer thermal en-
ergy storage (ATES) systems: a case study in the Netherlands, combining monitoring
and modeling. Hydrogeol. J. 23, 507532.
Willemsen, N., 2016. Rapportage bodemenergiesystemen in Nederland. RVO/IF
Technology, Arnhem.
Zeghici, R.M., Oude Essink, G.H.P., Hartog, N., Sommer, W., 2015. Integrated assessment
of variable densityviscosity groundwater ow for a high temperature mono-well
aquifer thermal energy storage (HT-ATES) system in a geothermal reservoir.
Geothermics 55, 5868.
Zheng, C., Wang, P.P., 1999. MT3DMS: A Modular Three-Dimensional Multispecies
Transport Model for Simulation of Advection, Dispersion, and Chemical Reactions of
Contaminants in Groundwater Systems; Documentation and User's Guide.
Zuurbier, K.G., Hartog, N., Valstar, J., Post, V.E., van Breukelen, B.M., 2013. The impact
of low-temperature seasonal aquifer thermal energy storage (SATES) systems on
chlorinated solvent contaminated groundwater: modeling of spreading and de-
gradation. J. Contam. Hydrol. 147, 113.
M. Bloemendal, N. Hartog Geothermics 71 (2018) 306–319

Supplementary resources (2)

... Aquifer Thermal Energy Storage (ATES) is a reliable low-carbon technology for space heating and cooling of buildings. Their energy and environmental benefits are proven in various studies and applications (Bloemendal and Hartog, 2018, Schüppler et al., 2019. However, ATES is not a global widespread technology (Schüppler et al., 2019, Surface and Energy, 2018. ...
... During operation of an ATES system in winter, cooling capacity for next summer is stored, and during cooling in summer, heating capacity for next winter is stored. Hence, to sustainably exploit ATES systems, seasonal energy storage and recovery must more or less balance (Bloemendal and Hartog, 2018). Therefore, disproportionally warm or cold areas must be avoided as respectively cold or warm wells cannot be charged sufficiently to meet demand for the next season (Bloemendal and Hartog, 2018). ...
... Hence, to sustainably exploit ATES systems, seasonal energy storage and recovery must more or less balance (Bloemendal and Hartog, 2018). Therefore, disproportionally warm or cold areas must be avoided as respectively cold or warm wells cannot be charged sufficiently to meet demand for the next season (Bloemendal and Hartog, 2018). In some European countries, such as The Netherlands, this balance is regulated so that the ratio H/C (Heating and Cooling) must be lower than 15% in the first 5 years and lower than 10% in 10 years (Hendriks, 2010). ...
Full-text available
Aquifer Thermal Energy Storage (ATES) Systems is a technology to sustainably and economically provide space heating and cooling. However, it cannot be applied everywhere because successful application depends on the presence of a suitable aquifer and favorable climatic conditions. Despite some operational ATES systems, the Spanish ATES market is immature, and there are no regulations or guidelines developed. To foster ATES adoption in Spain, this paper introduces a potential study considering the resource potential, technical, economic, and environmental aspects. The GIS-based approach is focused on the geographical identification and assessment of the aquifer potential for ATES and climatic conditions. This allows to distinguish those areas considered more suitable for ATES systems for the residential and for the tertiary sector. The results show where in Spain potential for energy and GHG savings with ATES can be found. 38% of the aquifers in Spain show potential for ATES and 63% of large urban areas in Spain are located in such areas. Also, 50% of the population lives in areas where the residential sector seems to be suitable for ATES based on the climatic conditions. Energy and GHG savings can reach up to 91% and 68% respectively, derived from the use of ATES.
... The total simulation horizon is set to 5 years. Although this is shorter than the expected life span of ATES systems, it is sufficiently long to distinguish between performance under varying well placement policies [21][22]. ...
... For this study the thermal distribution and heat loss in the horizontal plane is of interest, while vertical loss is not. Vertical distribution and losses to confining layers is also expected to be relatively small compared to horizontal losses [21][22]. Therefore, using only 1 layer in the vertical direction with the height of the total aquifer thickness (26 m) will not affect the interaction effects between the wells of interest in this study while limiting the model complexity and required computational resources [6]. ...
... Aquifer properties are homogeneous because the effect of heterogeneity on ATES well recovery efficiency is shown to be insignificant [27][28][29] and may disturb the analysis of the subsurface interactions. Influence of buoyancy flow due to density difference that occur due to temperature differences are negligible for LT-ATES and is therefore not taken into account [21,[30][31]. Because hydraulic conductivity has negligible effects on thermal losses under homogeneous conditions [21], the horizontal and vertical hydraulic conductivity is set to a constant value of 30 m/d and 6 m/d respectively. ...
Full-text available
The primary energy use of ATES systems evaluated for high and low aquifer utilisation levels. • High aquifer utilisation levels reduce energy use of individual systems, as more wells can be placed. • The highest aquifer utilization level considered is 115% and resulted in 82% ATES adoption. • For aquifer utilization <80%, energy use of buildings is not affected by subsurface interactions. • For aquifer utilization >80%, interactions affect gas use +15% and electricity use +/-15%. A R T I C L E I N F O Keywords: Aquifer Thermal Energy Storage (ATES) Subsurface interaction between ATES systems Individual ATES system performance Optimal utilisation of subsurface space A B S T R A C T Low temperature (<25 • C) Aquifer Thermal Energy Storage (ATES) systems have a worldwide potential to provide low-carbon space heating and cooling for buildings by using heat pumps combined with the seasonal subsurface storage and recovery of heated and cooled groundwater. ATES systems increasingly utilize aquifer space, decreasing the overall primary energy use for heating and cooling for an urban area. However, subsurface interaction may negatively affect the energy performance of individual buildings with existing ATES systems. In this study, it is investigated how aquifer utilization levels, obtained by varying well placement policies, affect subsurface interaction between ATES systems and how this in turn affects individual primary energy use. To this end, a building climate installation model is developed and integrated with a MODFLOW-MT3DMS thermal groundwater model. For the spatial distribution and thermal requirements of 26 unique buildings as present in the city centre of Utrecht, the placement of ATES wells is varied using an agent-based modelling approach applying dense and spacious placement restrictions. Within these simulations ATES adoption order and well placement location is randomized. Well placement density is varied for 9 scenarios by changing the distance between wells of the same and the opposite type. The results of this study show that the applied dense well placement policies lead to a 30% increase of ATES adoption and hence overall GHG emission reduction improved with maximum 60% compared to conventional heating and cooling. The primary energy use of individual ATES systems is affected at varying well placement policies by two mechanisms. Firstly, at denser well placement, ATES systems are able to place more wells, which increases the capacity of their ATES system, thereby decreasing their electricity and gas use. Secondly, aquifer utilization increases with denser well placement policies and thus interaction between individual ATES increases. At subsurface utilization up to 80%, individual primary energy use does not change significantly due to subsurface interaction. At aquifer utilization level > 80%, both negative and positive interaction is observed. Negative interaction between wells of the opposite type leads to an increase of gas or electricity use up to 15% compared to spacious well placement. On the other side, buildings may experience a maximum decrease of 15% electricity use at dense well placement due to positive interaction between wells of the same type. Local conditions like building location, plot size, distance to other buildings and 2 heating/cooling demand determine the specific effect per building. The optimal well placement policy result from the aquifer utilisation levels discussed above. Maximum GHG emission reduction while maintaining individual ATES system performance, is achieved with well distances of 0.5-1 times the yearly average thermal radius for wells of the same type (cold-cold and warm-warm). Opposite well types (cold-warm) should be placed apart ~2 times the thermal radius to prevent negative subsurface interaction.
... Part of the injected energy is lost to the surroundings due to conduction, dispersion, and diffusion and cannot be recovered. Conduction losses generally dominate dispersion losses in ATES systems (e.g., Bloemendal and Hartog 2018). ...
... A small L/R th ratio means that the thermal radius is large compared to the length of the well screen so that the cylinder is short and wide. In practice, L/R th ratios vary roughly between 0.25 and 4 (e.g., Bloemendal and Hartog 2018). Energy losses occur mainly at the boundary of the thermal zone. ...
... Energy losses occur mainly at the boundary of the thermal zone. Minimization of the area A of the thermal zone compared to the volume V of the thermal zone leads to lower losses (e.g., Bloemendal and Hartog 2018). The lowest A/V ratios and highest efficiencies are obtained for systems with a large storage volume and an L/R th ratio of 2 (e.g., Bloemendal and Hartog 2018). ...
Full-text available
Aquifer thermal energy storage (ATES) is an energy efficient technique to provide heating and cooling to buildings by storage of warm and cold water in aquifers. In regions with large demand for ATES, ATES adoption has lead to congestion problems in aquifers. The recovery of thermal energy stored in aquifers can be increased by reducing the distance between wells of the same temperature while safeguarding individual system performance. Although this approach is implemented in practice, the understanding of how this affects both the recovery efficiency and the neeeded pumping energy is lacking. In this research the effect of well placement on the performance of individual systems is quantified, and guidelines for planning and design are developed. Results show an increase in thermal recovery efficiency of individual systems when the thermal zones of wells of the same temperature are combined, which is explained by reduced surface area of the thermal zone over which losses occur. The highest increase of the thermal recovery efficiency is found for systems with a small storage volume and long well screens. The relative increase of the thermal recovery efficiency is 12% for average-sized systems with a storage volume of 250,000 m3 /year, and 25% for small systems (50,000 m3 /year). The optimal distance between wells of the same temperature is 0.5 times the thermal radius, following the trade-off between an increase of the thermal recovery efficiency and the increase in pumping energy. The distance between wells of opposite temperature must be larger than 3 times the thermal radius to avoid negative interaction.
... In the summer, the waste and stored underground heat energy is re-used for the cooling type air conditioning of the buildings in winter period. The system is operated reversibly in cooling and heating periods for air conditioning purposes in residences, offices, etc. in cooling and heating periods to ensure sustainable air conditioning (Bloemendal and Hartog 2018). ...
... Hence hot water is stored at most 5°C above the average temperature of the aquifer (Dickinson et al. 2009;Lee 2010;Gao et al. 2017) at present study. In addition, buoyancy flow was ignored in this study because at the relative small temperature differences between the wells and ambient groundwater as applied for ATES, buoyancy effects are negligible (Bloemendal and Hartog 2018;Anderson 2005;Doughty et al. 1982). ...
... Researchers offer different opinions on thermal interaction distances between wells. Based on the results of numerical modeling studies, Kim et al. (2010) stated that the thermal recovery rate in ATES systems is not Rth Thermal radius (Rth) Fig. 2 Representation of thermal and hydraulic radius of an ATES well in homogeneous and isotropic aquifer conditions (Bloemendal 2018) significantly affected if the distance between abstraction and injection wells is more than one thermal radius. Sommer et al. (2015) stated that thermal recovery will decrease if the distance between wells is less than two thermal radii. ...
Full-text available
Low entropy shallow ground heat resources are gaining importance in recent years owing to their availability compared to difficult-to-reach geothermal energy sources. In the last decades, aquifer thermal energy storage (ATES) systems have begun to be utilized increasingly since they can provide one of the cleanest and most energy efficient heating and cooling system alternatives for buildings. One of the main problems in the design phase of ATES systems is the correct estimation of thermal interference distance between extraction and injection side of the system. In order to investigate this problem, an extensive modeling study was carried out to constitute a preliminary approach for the numerical model calibration of heat transport and storage in aquifers. For this purpose, actual site data available for a well-studied coastal aquifer in Mediterranean region of Turkey was used in the analysis and performance assessment of a conceptual open-loop ground source heating well doublet. Three-dimensional coupled numerical model of groundwater flow and heat transfer was produced to constitute an approach for the thermo-hydraulic model calibration by estimating the duration of thermal breakthrough between the abstraction and injection wells. Simulation results were compared with the analytical solution for doublet well breakthrough time with finite hydraulic gradient. The comparison of results indicated that the numerical model is able to represent the thermal behavior in the field. Therefore, calibration methodology established in this paper could be followed in the pre-feasibility study and the design phase of low-temperature aquifer thermal energy storage systems worldwide.
... A large body of literature on direct numerical simulations of the recovery efficiency exists, (e.g. Doughty et al., 1982;Sommer et al., 2013;Sommer et al., 2015;van Lopik et al., 2016;Bloemendal and Hartog, 2018;Pophillat et al., 2020a;Pophillat et al., 2020b). However, such numerical studies are computationally intensive and specific to certain combinations of parameter values and aquifer geometry, which makes their findings difficult to generalize. ...
... Expansions of S T are presented in Table 3. Concepts similar to S T have been used in prior studies to determine relative contributions of the dispersion processes in radial flow. Hoopes and Harleman (1967) obtained a weighted ratio, 4αv(R) 3Dm , which is (28) with d = 2 specifically, while Bloemendal and Hartog (2018) applied the unweighted ratio αv(R)/D m to a two-dimensional radial problem, which underestimates the contribution of mechanical dispersion. ...
... While A/V is identical for any combination of Q and T that yield identical QT, the recovery efficiency can significantly differ for different combinations of Q and T. Doughty et al (1982) found that when the total solute mass injected c 0 QT was kept constant, in a system with D = D m , the recovery efficiency was higher under large Q small T operation, than under small Q large T operation. Bloemendal and Hartog (2018) investigated heat storage in a system with a 2D flow field. They concluded from sensitivity analyses of Q that 1 − F r ∝A/V approximately when thermal diffusion dominates, which agrees superficially with our findings for 2D flow fields that 1 − F r ∝Q − 1 2 T 0 (Equation 23), because A/V∝(QT) − 1 2 . ...
Full-text available
For cyclic injection-extraction wells with various radial flow geometries, we study the transport and recovery of solute and heat. We derive analytical approximations for the recovery efficiency in closed-form elementary functions. The recovery efficiency increases as injection-extraction flow rates increase, dispersion decreases, and spatial dimensionality decreases. In most scenarios, recovery increases as cycle periods increase, but we show numerically and analytically that it varies non-monotonically with cycle period in three-dimensional flow fields, due to competing effects between diffusion and mechanical dispersion. This illustrates essential differences between the spreading mechanisms, and reveals that for a single well it may be impossible to optimize recovery of both solute and heat simultaneously. Whether retardation increases or decreases recovery thus depends on aquifer geometry and the dominant dispersion process. As the dominant dispersion process heavily determines the sensitivity of the recovery efficiency to other parameters, we introduce the dimensionless kinetic dispersion factor ST, to distinguish whether diffusion or mechanical dispersion dominates. We also introduce the geometric dispersion factor G, which is derived from our full solution for the recovery efficiency and improves upon the concept of the area-to-volume ratio (A/V), often used in analysing well problems. Unlike A/V, G accounts for spatio-temporal interactions between dispersion and flow field geometry, and can be applied to determine recovery efficiencies across a wider range of scenarios. It is found that A/V is a special case of G, describing the recovery efficiency only when mechanical dispersion with linear velocity dependence is the sole mechanism of spreading.
... No assumption of ergodicity is made, as in practice, for managed aquifer recharge applications such as ATES and ASR, the storage radius of the system may be even smaller than a single integral scale of heterogeneity (e.g., Sommer et al., 2013). For example, in the Netherlands, where most presently operating ATES systems worldwide are located, the storage radius of wells never exceeds 100m, with most being under 50 m (Bloemendal & Hartog, 2018). Real aquifers reported in many studies in the literature (e.g., Sommer et al., 2013;Vereecken et al., 2000;and Fernàndez-Garcia et al., 2005) have horizontal correlation lengths of between 2.5 and 100 m. ...
... In homogeneous aquifers, an increase in the injected volume necessarily leads to a decrease in the actual A/V (which is identical to ) for geometrical reasons. Many studies therefore suggest that an increased injected volume leads to an increase in the recovery efficiency (e.g., Bloemendal & Hartog, 2018;Forkel & Daniels., 1995;Novo et al., 2010;Schout et al., 2014;Sommer et al., 2015). In aquifers with heterogeneous hydraulic conductivity and scale-dependent macrodispersion, this rule might no longer apply, as increasing would necessarily decrease but possibly lead to an increase in the actual A/V, due to the large interfaces that may develop between fast and slow flow zones. ...
Full-text available
The recovery efficiency of aquifer storage systems with radial flow fields are studied for heterogeneous aquifers. Macrodispersion, arising from spatially heterogeneous hydraulic conductivity, is modeled as a scale‐dependent mechanical dispersion process. Approximate solutions for the recovery efficiency as a function of local dispersion and macrodispersion parameters, the injection‐extraction rate Q $Q$ and duration T $T$, and storage cycle count, are derived and validated against numerical simulations. If macrodispersion dominates and the macrodispersion coefficient scales linearly with distance, the recovery efficiency is independent of both Q,T $Q,T$. For sublinear and superlinear scalings, recovery increases and decreases respectively if Q,T $Q,T$ increases. However, if local dispersion dominates, increasing Q,T $Q,T$ always increases recovery. As macrodispersion becomes increasingly dominant with scale, the recovery efficiency may be a nonmonotonic function of Q,T $Q,T$, with a maximum. In homogeneous aquifers, nonmonotonicity does not occur for 1D and 2D radial flow, but occurs for 3D radial flow fields only as a function of T $T$, not Q $Q$. These methods may also be used for fitting local dispersion and macrodispersion parameters with push‐pull tests using recovery data, with advantages in scope of applicability and ease of data acquisition and interpretation, compared to existing push‐pull test methods, which fit to breakthrough curves and do not consider macrodispersion. Furthermore, characterizing macrodispersion with push‐pull tests may be advantageous over methods that use observation wells, as observation well placement may be challenging in highly heterogeneous aquifers. The results show that the macrodispersion parameters are not innate aquifer hydraulic properties, as their values vary with flow field geometry.
... Here, the groundwater flow and heat advection are very limited, and a global thermal balance for hot and cold injections would have been required before issuing the permitting certificate. On the contrary, in other hydrogeological conditions, if the groundwater flow and heat advection are very important, thermal energy cannot be stored efficiently locally as heat and cold plumes are transported far away from the wells [36,[44][45][46]. If the groundwater flow and heat advection are moderate, a detailed simulation of the groundwater flow and heat transport in the aquifer is particularly required to find out if the annual imbalance can be managed in relation to the specific local hydrogeological conditions. ...
Full-text available
A numerical model was built using FEFLOW® to simulate groundwater flow and heat transport in a confined aquifer in Brussels where two Aquifer Thermal Energy Storage (ATES) systems were installed. These systems are operating in adjacent buildings and exploit the same aquifer made up of mixed sandy and silty sublayers. The model was calibrated for groundwater flow and partially for heat transport. Several scenarios were considered to determine if the two ATES systems were interfering. The results showed that a significant imbalance between the injection of warm and cold water in the first installed ATES system led to the occurrence of a heat plume spreading more and more over the years. This plume eventually reached the cold wells of the same installation. The temperature, therefore, increased in warm and cold wells and the efficiency of the building’s cooling system decreased. When the second ATES system began to be operational, the simulated results showed that, even if the heat plumes of the two systems had come into contact, the influence of the second system on the first one was negligible during the first two years of joint operation. For a longer modeled period, simulated results pointed out that the joint operation of the two ATES systems was not adapted to balance, in the long term, the quantity of warm and cold water injected in the aquifer. The groundwater temperature would rise inexorably in the warm and cold wells of both systems. The heat plumes would spread more and more over the years at the expense of the efficiency of both systems, especially concerning building’s cooling with stored cold groundwater.
... The groundwater is re-injected with an upper limit approaching 25 C to preserve the quality of the water within the aquifer [27]. Storage efficiencies are typically high for ATES systems, recovering between 67.5 and 87% of stored heat and cold, with increasing storage volume serving to improve storage efficiencies [23,28,29]. ATES is commonly installed in universities, hospitals, large commercial buildings, and airports [26], with ventilation cooling during summer often providing the heat to be stored and subsequently used for heating throughout the winter. ...
As mitigating climate change becomes an increasing worldwide focus, it is vital to explore a diverse range of technologies for reducing emissions. Heating and cooling make up a significant proportion of energy demand, both domestically and in industry. An effective method of reducing this energy demand is the storage and use of waste heat through the application of seasonal thermal energy storage, used to address the mismatch between supply and demand and greatly increasing the efficiency of renewable resources. Four methods of sensible heat storage; Tank, pit, borehole, and aquifer thermal energy storage are at the time of writing at a more advanced stage of development when compared with other methods of thermal storage and are already being implemented within energy systems. This review aims to identify some of the barriers to development currently facing these methods of seasonal thermal energy storage, and subsequently some of the work being undertaken to address these barriers in order to facilitate wider levels of adoption throughout energy systems.
... Typically, in heating mode a heat pump is used whereas with cooling mode a socalled direct cooling loop without using the heat pump is often designed [2,4]. The vast majority of ATES systems are classified as low-temperature-ATES (LT-ATES) with maximum injection temperatures of below 25 • C and are usually using shallow groundwater of the upper few tens to hundreds of meters [3,10,11]. ...
Aquifer Thermal Energy Storage (ATES) is an open-loop geothermal system allowing long-term storage of thermal energy in groundwater. It is a promising technology for environmentally friendly energy generation that can reduce greenhouse gas (GHG) emissions. In the literature, there are few studies on the greenhouse gas emissions caused by ATES systems over their entire life cycle. Thus, this study presents a novel life cycle assessment (LCA) regression model that can be used for a wide range of ATES configurations due to its parametric structure. This model is a fast alternative to conventional time-consuming LCAs. Combined with a Monte Carlo simulation, it enables the analysis of the environmental impacts of a large variety of hypothetical ATES systems and therefore the evaluation of the technology as a whole. Compared to conventional heating systems based on heating oil and natural gas, the median value of the Monte Carlo simulation results in GHG savings of up to 74%. In comparison to cooling techniques using today's electricity mix, ATES can save up to about 59% of GHG emissions, while also being economically competitive. When considering a projected electricity mix for the year 2050, the GHG emission savings resulting from a second LCA regression model are as high as 97%. The findings of our sensitivity analysis show which ATES design parameters should be optimized when planning new systems. In particular, the most important design parameters operating time cooling and coefficient of performance (COP) of the heat pump should be carefully considered.
High-Temperature – Aquifer Thermal Energy Storage (HT-ATES) can significantly increase Renewable Energy Sources (RES) capacity and storage temperature levels compared to traditional ATES, while improving efficiency. Combined assessment of subsurface performance and surface District Heating Networks (DHN) is key, but poses challenges for dimensioning, energy flow matching, and techno-economic performance of the joint system. We present a novel methodology for dimensioning and techno-economic assessment of an HT-ATES system combining subsurface, DHN, operational CO 2 emissions, and economics. Subsurface thermo-hydraulic simulations consider aquifer properties (thickness, permeability, porosity, depth, dip, artesian conditions and groundwater hydraulic gradient) and operational parameters (well pattern and cut-off temperature). Subject to subsurface constraints, aquifer permeability and thickness are major control variables. Transmissivity ≥2.5 × 10 ⁻¹² m³ is required to keep the Levelised Cost Of Heat (LCOH) below 200 CHF/MWh and capacity ≥25 MW is needed for the HT-ATES system to compete with other large-scale DHN heat sources. Addition of Heat Pumps (HP) increases the LCOH, but also the nominal capacity of the system and yields higher cumulative avoided CO 2 emissions. The methodology presented exemplifies HT-ATES dimensioning and connection to DHN for planning purposes and opens-up the possibility for their fully-coupled assessment in site-specific assessments.
Full-text available
Results are presented of a comprehensive thermal impact study on an aquifer thermal energy storage (ATES) system in Bilthoven, the Netherlands. The study involved monitoring of the thermal impact and modeling of the three-dimensional temperature evolution of the storage aquifer and over-and underlying units. Special attention was paid to non-uniformity of the background temperature, which varies laterally and vertically in the aquifer. Two models were applied with different levels of detail regarding initial conditions and heterogeneity of hydraulic and thermal properties: a fine-scale hetero-geneity model which construed the lateral and vertical temperature distribution more realistically, and a simplified model which represented the aquifer system with only a limited number of homogeneous layers. Fine-scale heterogeneity was shown to be important to accurately model the ATES-impacted vertical temperature distribution and the maximum and minimum temperatures in the storage aquifer, and the spatial extent of the thermal plumes. The fine-scale hetero-geneity model resulted in larger thermally impacted areas and larger temperature anomalies than the simplified model. The models showed that scattered and scarce monitoring data of ATES-induced temperatures can be interpreted in a useful way by groundwater and heat transport modeling, resulting in a realistic assessment of the thermal impact.
Full-text available
The efficiency of heat recovery in high-temperature (>60 °C) aquifer thermal energy storage (HT-ATES) systems is limited due to the buoyancy of the injected hot water. This study investigates the potential to improve the efficiency through compensation of the density difference by increased salinity of the injected hot water for a single injection-recovery well scheme. The proposed method was tested through numerical modeling with SEAWATv4, considering seasonal HT-ATES with four consecutive injection-storage-recovery cycles. Recovery efficiencies for the consecutive cycles were investigated for six cases with three simulated scenarios: a) regular HT-ATES, b) HT-ATES with density difference compensation using saline water, and c) theoretical regular HT-ATES without free thermal convection. For the reference case, in which 80 °C water was injected into a high-permeability aquifer, regular HT-ATES had an efficiency of 0.40 after four consecutive recovery cycles. The density difference compensation method resulted in an efficiency of 0.69, approximating the theoretical case (0.76). Sensitivity analysis showed that the net efficiency increase by using the density difference compensation method instead of regular HT-ATES is greater for higher aquifer hydraulic conductivity, larger temperature difference between injection water and ambient groundwater, smaller injection volume, and larger aquifer thickness. This means that density difference compensation allows the application of HT-ATES in thicker, more permeable aquifers and with larger temperatures than would be considered for regular HT-ATES systems.
Conference Paper
Full-text available
This study illustrates that several processes contribute to groundwater quality effects of ATES systems, in which the effect of mixing is predominant for the lowtemperature ATES systems studied. Only for ATES systems in aquifers with relatively homogeneous groundwater quality, minor temperature effects could be observed. At higher temperatures (>30 C) groundwater quality is expected to be more significantly affected, due to the exponential temperature-dependence of both geochemical equilibria and rate constants. Overall, mixing was the predominant process affecting groundwater quality at the studied ATES sites, particularly, where groundwater gradients were strongest.
[1] Heterogeneity in hydraulic properties of the subsurface is not accounted for in current design calculations of aquifer thermal energy storage (ATES). However, the subsurface is heterogeneous and thus affects the heat distribution around ATES wells. In this paper, the influence of heterogeneity on the performance of a doublet well system is quantified using stochastic heat transport modeling. The results show that on average, thermal recovery decreases with increasing heterogeneity, expressed as the lognormal standard deviation of the hydraulic conductivity field around the doublet. Furthermore, heterogeneity at the scale of a doublet ATES system introduces an uncertainty in the amount of expected thermal interference between the warm and cold storage. This results in an uncertainty in thermal recovery that also increases with heterogeneity and decreases with increasing distance between ATES wells. The uncertainty in thermal balance due to heterogeneity can reach values near 50 percent points in case of regional groundwater flow in excess of 200 m/yr. To account for heterogeneity whilst using homogeneous models, an attempt was made to express the effect of heterogeneity by an apparent macrodispersivity. As expected, apparent macrodispersivity increases with increasing heterogeneity. However, it also depends on well-to-well distance and regional groundwater velocity. Again, the uncertainty in thermal recovery is reflected in a range in the apparent macrodispersivity values. Considering the increasing density of ATES systems, we conclude that thermal interference limits the number of ATES systems that can be implemented in a specific area, and the uncertainty in the hydraulic conductivity field related to heterogeneity should be accounted for when optimizing well-to-well distances.