Content uploaded by Jan Henninges
Author content
All content in this area was uploaded by Jan Henninges on Feb 11, 2015
Content may be subject to copyright.
PROCEEDINGS, Thirty-Seventh Workshop on Geothermal Reservoir Engineering
Stanford University, Stanford, California, January 30 - February 1, 2012
SGP-TR-194
DOWNHOLE MONITORING DURING HYDRAULIC EXPERIMENTS AT THE IN-SITU
GEOTHERMAL LAB GROß SCHÖNEBECK
J. Henninges, W. Brandt, K. Erbas, I. Moeck, A. Saadat, T. Reinsch, and G. Zimmermann
GFZ German Research Centre for Geosciences
Telegrafenberg
14473 Potsdam, Germany
e-mail: janhen@gfz-potsdam.de
ABSTRACT
During the current project phase the geothermal
water loop at the Groß Schönebeck site is being set
up and short- to long-term hydraulic experiments are
performed in the well doublet in order to assess the
hydraulic properties and the sustainability of the
engineered reservoir. Downhole measurements are
carried out during production in order to observe the
performance of the stimulated zones and the naturally
occurring permeable intervals. The production string
has been constructed in order to allow for access to
the reservoir with logging tools during fluid
production using a new developed Y-tool to bypass
the submersible pump. In addition a novel hybrid
wireline production logging system for combined
measurements with electrical tools (pressure,
temperature, flow meter, gamma ray, casing collar
locator) and fiber-optic distributed temperature
sensing (DTS) has been developed. The results of the
first measurement campaigns show that valuable data
for the observation and understanding of reservoir
flow dynamics can be collected with the new system.
The observed hydraulic behavior is mainly controlled
by a variable contribution from a hydraulic fracture
zone, which appears to be influenced by the
production history and induced reservoir processes.
INTRODUCTION
The Groß Schönebeck site, located approximately 40
km North of Berlin, Germany, was selected in order
to set up an in-situ laboratory for the development of
technologies related to the production of energy from
Enhanced Geothermal Systems. The geological
setting is characteristic for many sedimentary basins
worldwide: temperatures necessary for geothermal
power generation are sufficient but permeabilities of
the sedimentary rocks are too low for a direct
exploitation [Holl et al., 2005].
The geothermal reservoir is at a depth of 4100–4300
m with a bottom-hole temperature of 150 °C. The
main targets are the permeable sandstones of the
Upper Rotliegend (Dethlingen Formation) and the
underlying volcanic rocks (andesites) of the Lower
Rotliegend, where permeability is mainly due to
connected fractures. The Dethlingen sandstones have
a connected porosity of 8–10 %, and an in situ
permeability of up to 16.5 mD [Trautwein and
Huenges, 2005].
The Gt GrSk4/05 well was drilled in 2006/2007
adjacent to the E GrSk3/90 well to a measured depth
of 4400 m. In the reservoir interval the well is
deviated in the direction of the minimum horizontal
stress with an inclination between 37 and 49°.
Hydraulic stimulation techniques have been applied
to enhance the permeability of the sandstones and the
underlying volcanic rocks [Zimmermann et al.,
2011]. In the low permeability volcanic rocks, a
cyclic water frac treatment was performed, where a
total of 13,170 m3 of water was injected. In the
overlying sandstones two gel-proppant treatments
were performed. Short term production tests show
that the productivity index (PI) was increased from
an initial value of 2.4 m3/(h MPa) to 10.1 m3/(h MPa)
after stimulation. In 2009, an acid matrix stimulation
was performed, and a short term production test
indicated a further increase of the PI to a value
between 13 and 15 m3/(h MPa). The sustainability of
these PI-values is the matter of an ongoing long-term
experiment performed in 2011 and 2012.
In the summer of 2010 an electrical submersible
pump (ESP) has been installed in the Gt GrSk4/05
well. The production string has been equipped with a
special Y-tool in order to allow access to the
reservoir with logging tools during production
(Figure 1). For this purpose a new hybrid wireline
production logging system for combined
measurements with electrical tools and fiber-optic
distributed temperature sensing (DTS) has been
developed. Here we report on the results of the first
application of this system during hydraulic
experiments which were carried out in September
2011. Until now, no long-term hydraulic test has
been performed yet, but the production data recorded
during the commissioning of the surface equipment
already indicated that the productivity would likely
be lower than after the previous tests. Moreover the
well exhibited different drawdown characteristics at
different times. The aim of the logging campaign was
therefore to record production profiles and to observe
the reservoir dynamics during the hydraulic tests in
order to find explanations for the observed behavior.
Figure 1: Installation of the production string with
the Y-tool to bypass the ESP with
downhole logging tools during
production.
THE HYBRID WIRELINE LOGGING SYSTEM
Classic PL includes downhole measurement of
pressure, temperature, and fluid velocities, e.g. with a
spinner flow meter, to estimate flow rates and phase
composition within a flowing well. Temperature logs
can also be used to locate fluid movement along a
well. Using DTS, profiles can be registered almost
instantaneously over long distances and changes over
time can be conveniently monitored once the sensor
cable is in place.
In the past ten years there is a rapidly growing
number of DTS applications in the petroleum
industry, e.g. with permanent downhole sensors for
production monitoring in wells which are not
accessible with wireline or coiled tubing, e.g. [Brown
et al., 2000]. Within an earlier study, a prototype
slickline DTS sensor cable has been used to assess
the effect of water frac stimulation treatments in the
E GrSk 3/90 well [Henninges et al., 2005].
The new 5,500 m hybrid wireline logging cable
contains both electrical conductors and steel tubes for
inclusion of the optical fibers. For the optical fibers, a
polyimide/carbon coating was selected to allow for
the required stability and resistance to chemical
degradation (e.g. hydrogen ingression) of the fibers at
elevated temperatures [Reinsch and Henninges,
2010]. The electrical tools include pressure,
temperature, and a spinner flow meter, as well as
gamma ray (GR) and casing collar locator (CCL) for
depth correlation.
In November 2010 the first baseline logging runs
with the hybrid logging system had been performed
under static conditions [Henninges et al., 2011]. For
an integration time of 30 minutes the DTS data
exhibited a temperature resolution of up to 0.06 °C,
which shows that even small temperature differences
can be resolved along this rather large profile of
5,500 m length.
PRODUCTION LOGGING RESULTS
Two logging campaigns during production tests were
performed on September 8 and 9, 2011, in the
GtGrSk04/05 well. The logging tool was placed at a
depth of about 4350 m, which is 5 m above the
beginning of the pre-perforated 5‖ liner and about 10
m above the maximum accessible depth of the well.
With this set-up, the pressure at bottom-hole and
DTS profile data over the reservoir interval can be
acquired simultaneously during the production phase.
Unfortunately, no bottom-hole pressure and electrical
wireline data could be recorded during the production
test on Sept. 8 due to a failure of the electrical data
transmission which occurred when the ESP was
started. After this problem had been fixed, both
bottom-hole pressure data and spinner logs during the
following day could be recorded.
In addition to the recorded logging data, surface
water flow rates and pressure data from a sensor
located below the ESP at a depth of 1202.49 m were
available for evaluation.
Day 1 (September 8, 2011)
The first discovery was that no indications for solids
deposition within the 5‖ liner across the reservoir
interval could be observed, based on the cable tension
and motion sensor data recorded during the first
decent into the well. Therefore a substantial blockage
of the well which could be responsible for the
reduced productivity could already be excluded.
The first production phase is characterized by a
decrease of well temperatures in the lower part of the
well (Figure 2), which is gradually progressing in
upward direction. In addition to this, the DTS profile
at 13:28 hrs displays a local increase of temperature
occurring at a depth between 4240 – 4220 m. This
indicates that flow is mainly occurring from the
volcanic rocks in the lowermost part of the well and
the perforations at the base of the Dethlingen
sandstones at this time. Shortly afterwards a sudden
temperature increase was noticeable at the position of
the perforations of the 1st gel/proppant frac located
just below 4200 m depth, which is visible in the
13:47 hrs profile.
Figure 2: DTS temperature data production test
Sept. 8, 2011. Acquisition times of DTS
profiles not displayed in log header: light
blue 12:29, gold: 13:02, orange: 13:28.
For a more detailed analysis of the temporal
evolution, DTS data of four depth intervals located
above and below the main inflow zones described
above, together with flowrate and pressure data is
displayed in Figure 3. The main production phase
was preceded by a very short production phase of
about six minutes duration. After this, a temperature
response at the lowermost observation point at 4335
m is already visible, which indicates that flow from
the slotted liner interval in the volcanic rocks had
already taken place. During the first part of the main
production phase, a similar decline in temperature at
the three upper observation levels is occurring with a
successive temporal delay corresponding to the
distance along the flow path.
After about 45 minutes, the temperature increase at
the 4250 m observation point located above the
perforation interval in the lower part of the
Dethlingen Formation sandstones is visible. This is
followed by the even more pronounced increase at
the 4195 m observation point above the 1st
gel/proppant frac occurring after about 60 minutes.
This time coincides with the beginning of a
noticeable phase with decreasing drawdown at the
ESP, which was characterized by a steep and almost
linear increase beforehand. Despite a short
fluctuation, the trend of the production rate remained
rather constant during this intermittent recovery
period. Afterwards, temperatures remained rather
constant for the rest of the production phase.
Figure 3: Surface water flow rate, drawdown at
pump (DD ESP), and DTS data for
different depths within the reservoir
interval, Sept. 8, 2011. In the temperature
data plot, the times for start of the
production phase 1 and 2, as well as the
beginning of the pressure increase are
marked.
Day 2 (September 9, 2011)
Before the start of the production test during the
second day, the borehole temperatures are still
slightly reduced compared to the preceding day
(Figure 4). After the start of pumping, temperatures
below the 1st gel/proppant frac are decreasing in a
similar fashion as during day 1 (Figure 5). But in
contrast to the first day, the temperature at the 4195
m observation point above the 1st gel/proppant frac is
remaining almost constant during the entire
observation phase. This indicates that a significant
amount of fluid is produced from the 1st gel/proppant
frac interval throughout the production phase.
Parallel to the acquisition of the DTS data the
bottom-hole pressure was recorded with the electrical
downhole tool (Figure 5). The recorded pressure
yields a drawdown curve with a gradually higher
slope than the curve derived from the sensor at the
ESP, which is most likely due to increasing
contribution of flow from the reservoir interval and
corresponding friction losses.
Figure 4: DTS temperature data production test
Sept. 9, 2011. Acquisition times of DTS
profiles not displayed in log header: light
blue 12:16, gold: 12:35.
Figure 5: Surface water flow rate, drawdown at
pump (DD ESP) and at downhole logging
tool (DD bottom-hole), and DTS data for
different depths within the reservoir
interval, Sept. 9, 2011. In the temperature
data plot, the time for start of the
production phase is marked.
Flowmeter logging
After the DTS monitoring period, one upward
spinner log with a cable speed of 10 m/min was
recorded (Figure 6). Due to temporary blocking of
the spinner, not a complete suite of down- and
upward runs with different logging speeds could be
recorded. Therefore it was decided to record
stationary spinner readings above the main inflow
zones which had been identified on the basis of the
DTS monitoring during the production test of the
previous day. A total of four stationary measurements
at depths of4300 m, 4228 m, 4200 m, and 4080 m
with a duration of about five minutes each were
performed (Figure 6).
Figure 6: 1st downward logging run (Run 1),
spinner log (Run 2), and stationary
spinner readings, Sept. 9, 2011. The main
inflow zone is located at the 1st
gel/proppant frac, with minor
contributions from the bottom of the well,
i.e. the water frac, and the perforations in
the lower section of the Dethlingen
sandstones. No inflow at the position of
the 2nd gel/proppant frac.
For the evaluation of the spinner data a calibration of
the spinner response was performed based on
recordings with different cable speeds during the first
downward run (Figure 7). The derived flow velocities
are summarized in Table 1. The readings at 4200 m
are disturbed by turbulences caused by the strong
inflow at the perforations of the 1st gel/proppant frac,
which is also visible in the data from the upward
logging run. Above the disturbed zone the spinner log
shows rather constant readings and no further
temperature anomalies are visible, which indicates
that no inflow into the well is occurring. It was
therefore assumed that the flow above the position of
the perforations of the 1st gel/proppant frac is equal to
the flow recorded at the top of the reservoir at 4080
m depth. The computed flow velocity at this depth of
64.74 m/min is very similar to the average flow
velocity for the 5‖ liner within the reservoir section
of 63.20 m/min, which can be calculated from the
average surface flow rate of 35.14 m³/h during the
stationary spinner readings.
Relative contributions of the individual reservoir
intervals to total flow are given in Table 1 and were
directly computed from the determined flow
velocities, as the volumetric flow rate is directly
proportional to it.
y = 5.914x - 25.90
R² = 0.998
-100
0
100
200
300
400
500
-20 020 40 60 80 100
Spinner response (CPS)
Cable speed (m/min)
calibration
4300 m
4228 m
4200 m
4080 m
Fit (calibration)
Figure 7: Average spinner readings during the
stationary measurements and calibration
curve derived from response at different
cable speeds during shut-in.
Table 1: Sept. 9 stationary spinner readings,
computed flow velocities, and
contributions to total flow of the
individual dept intervals. For comparison,
the results of the 2007 survey are given.
Depth
interval
Spinner
reading
Flow
velocity
Contri-
bution
(2011)
Contri-
bution
(2007)
m
cps
m/min
%
%
4080
357
64.74
0.00
14.08
4200
438
78.43
69.47
52.03
4228
91
19.77
5.75
6.85
4300
69
16.05
24.78
26.81
DISCUSSION
The recorded DTS data display favorable conditions
for flow profiling using temperature, because of the
remaining temperature anomaly within the volcanic
rocks at the bottom of the well resulting from the
injection of large amounts of cold water during the
water frac treatment in 2007. During production,
relatively colder water from this level is produced
and ascending to the shallower production intervals,
in which fluids with higher temperatures are
produced. Therefore inflow from these shallower
production intervals is obvious because of the
temperature difference of the produced fluids.
Two different inflow regimes can be derived from the
recorded temperature data: The first is characterized
by inflow from the water frac in the volcanic rocks at
the bottom of the well and the perforations at the
bottom of the Dethlingen Fm. Sandstones (first
production phase during Sept. 8), followed by a
sudden inflow at the 1st gel/proppant frac. The second
one shows a predominant contribution to flow from
the 1st gel/proppant frac with a minor contribution
from the water frac in the volcanics over the entire
production phase (Sept. 9 test).
The shape of the drawdown curve is significantly
different for the two inflow regimes. The first regime
displays a steep an almost linear drawdown at the
beginning, followed by an intermittent recovery
phase. The second regime exhibits a drawdown
which is steadily increasing with a significantly lower
slope. This is remarkable, since both the average
production rates during day one and day two of 45.3
m³/h and 40.45 m³/h, respectively, are comparably
high, and the initial pressures only differ by about 0.1
MPa. The different observed behavior must therefore
be related to the production history. Nevertheless,
despite of their different shapes, the drawdown
curves arrive at similar values after a production time
equal to the duration of the first test.
When comparing the flow contributions of the
individual reservoir intervals for the Sept. 9 test to
the results from the 2007 test data [Zimmermann et
al., 2010], the following observations can be made:
The relative amount of inflow from the volcanic
rocks is rather similar and the largest contribution is
coming from the 1st gel/proppant frac in both cases.
But during the 2011 test, a lower inflow is occurring
at the perforations at the bottom of the Dethlingen
sandstones, and the 2nd gel/proppant frac is showing
no production at all. At the 1st gel/proppant frac a
relatively higher inflow of approx. 69 % is occurring
in 2010, compared to a contribution of approx. 52 %
in 2007.
CONCLUSIONS
A new developed production string including a Y-
tool to bypass the submersible pump has been used
for access to the reservoir with logging tools during
fluid production. This system allows to record
transient reservoir performance data. First
preliminary conclusions are given here.
Based on the recorded logging data the following
aspects of the reservoir flow dynamics are observed:
Two different inflow regimes could be recognized
which are responsible for the variable drawdown
behavior. The difference is mainly based on a
variable inflow from the 1st gel/proppant frac, which
is probably depending on the production history and
processes in the reservoir system induced hereby.
The shortfall of flow from the second gel/proppant
frac located in the upper part of the Dethlingen
sandstones and reduced inflows from the perforations
in lower part of the Dethlingen sandstones can at
least explain a partial reduction in productivity with
respect to the test performed in 2007. Further
hydraulic testing with longer-term production phases
need to be performed in order to determine the
productivity.
The new hybrid wireline production logging system
was successfully deployed and valuable data for the
evaluation of reservoir hydraulics could be gathered.
Further effort will be put into improved configuration
of the DTS measurement and integrated evaluation of
the flow, pressure, and temperature data.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the support of
Jörg Schrötter, Mathias Poser and Christian Cunow
during the field measurements. This work was funded
by the GeoEn project of the German Federal Ministry
for Education and Research (BMBF), by the German
Federal Ministry for the Environment, Nature
Conservation and Nuclear Safety (BMU, grant no.
0325088), as well as through the project ―Modell-
Pilotvorhaben Geothermiekraftwerk Groß
Schönebeck‖ of the Ministry for Science, Research,
and Culture of the State of Brandenburg.
REFERENCES
Brown, G. A., B. Kennedy, and T. Meling (2000),
Using fibre-optic distributed temperature
measurements to provide real-time reservoir
surveillance data on Wytch Farm field horizontal
extended-reach wells, SPE 62952, paper
presented at SPE Annual Technical Conference
and Exhibition, Dallas, Texas, USA October 1-4.
Henninges, J., G. Zimmermann, G. Büttner, J.
Schrötter, K. Erbas, and E. Huenges (2005),
Wireline distributed temperature measurements
and permanent installations behind casing, in
Proceedings of the World Geothermal Congress
2005, Antalya, Turkey [CD-ROM], edited by R.
Horne and E. Okandan, p. paper 1021,
International Geothermal Association,
Reykjavik, Iceland.
Henninges, J., G. Baumann, W. Brandt, C. Cunow,
M. Poser, J. Schrötter, and E. Huenges (2011), A
novel hybrid wireline logging system for
downhole monitoring of fluid injection and
production in deep reservoirs, Conference
Proceedings, 73rd EAGE Conference &
Exhibition, Vienna, Austria (CD-ROM), C043.
Holl, H. G., I. Moeck, and H. Schandelmeier (2005),
Characterisation of the Tectono-Sedimentary
Evolution of a Geothermal Reservoir –
Implications for Exploitation (Southern Permian
Basin, NE Germany), in Proceedings of the
World Geothermal Congress 2005, Antalya,
Turkey [CD-ROM], edited by R. Horne and E.
Okandan, p. paper 0614, International
Geothermal Association, Reykjavik, Iceland.
Reinsch, T., and J. Henninges (2010), Temperature-
dependent characterization of optical fibres for
distributed temperature sensing in hot
geothermal wells, Measurement Science and
Technology, 21(9), 094022.
Trautwein, U., and E. Huenges (2005), Poroelastic
behaviour of physical properties in Rotliegend
sandstones under uniaxial strain, International
Journal of Rock Mechanics and Mining Sciences,
42(7–8), 924-932.
Zimmermann, G., I. Moeck, and G. Blöcher (2010),
Cyclic waterfrac stimulation to develop an
Enhanced Geothermal System (EGS)—
Conceptual design and experimental results,
Geothermics, 39, 59–69.
Zimmermann, G., G. Blöcher, A. Reinicke, and W.
Brandt (2011), Rock specific hydraulic
fracturing and matrix acidizing to enhance a
geothermal system — Concepts and field results,
Tectonophysics, 503(1–2), 146-154.
