Professional Documents
Culture Documents
Abstract
This paper describes the utilization of Real-Time geochemical analysis to support geosteering of a smart multi-lateral well,
located in one of the highest flow potential areas in Kuwait. The Burgan reservoir consists of vertically stacked channel sands
along with a fault network connected to the aquifer and contains highly viscous reservoir fluid. This drastically enhances the
water mobility, and results in severe premature water breakthrough. Hence, leaves zones of by-passed oil. For optimum
reservoir characterization, it was essential to integrate all reservoir-related data from macro to micro scale. X-ray
Fluorescence elemental data collected from offset cores were used to predict key rock attributes and calibrated with standard
petrophysical logs.The scope was constructing predictive models for the following properties: 1) lithological variations which
cannot be captured by other LWD tools 2) detailed mineralogy to determine the diagenetic overprint 3) depositional
environment of different Burgan sand facies. XRF elemental analysis while drilling was used to improve borehole
positioning, and identify faults in correlation with Image logs. Nature of the fractures/faults, contributing to porosity and
communicating with the aquifer, was inferred from XRF-obtained elemental markers. The integrated approach has resulted in
successful geosteering and placing the well with maximum reservoir contact. Moreover, XRF elemental markers have been
utilized for isolation of faulted and lower reservoir quality zones, splitting up of horizontal sections and optimization nozzle
sizes of the ICDs and hence an optimized Smart completion design. X-ray fluorescence analysis on cuttings in Real-Time
provides lithological information otherwise not available while drilling. It gives proxies contributing to the identification of
faults and reservoir intervals in an otherwise homogeneous sequence. It helps designing the completion string, isolating
sections of low quality or potentially producing water.
1. Introduction
The Minagish Field in Kuwait was discovered in 1959 and is located in the southwestern part of Kuwait. It contains several
reservoir intervals in its stratigraphic column varying from early Jurassic to late Cretaceous. The Field is situated 12 km north
west from West Umm Gudair Field. The field has been penetrated by more than 180 wells, not only for the middle and lower
Minagish reservoirs of lower Cretaceous age, but also for other shallow reservoirs such as Mishrif-Rumaila carbonates and
Wara-Burgan sandstone. The Minagish Field structure of Wara and Burgan Formation is a closed elongated asymmetrical
anticline oriented in North-South direction. The top of the Burgan structure is located at about 5500 feet TVDSS. (Figure 1)
IPTC 16617
Figure 1: Minagish Field is located in south west corner of Kuwait. The Burgan is of Albian age (Lower Cretaceous), and structure is
asymmetrical elongated north-south anticline.
IPTC 16617
Figu
Figu
Figure 2: Porosity vs. Permeability plot showing the difference in reservoir properties of channel sands in the upper Burgan compared with
other sands (within, above and below). Figure 3: north-south correlation line flattened on top Mauddud limestone marker shows the Burgan
sand sequences below and Wara shlaes/ or sands overlie the top Mauddud.
IPTC 16617
Figure 4: Seismic Coherency map along a horizontal section of a Burgan producer is used to locate areas of faulting. The coherent areas in
beige color in the cross section show no faulting. All other colors (in-coherent areas) show faulted areas starting from dark blue, yellow,
orange and red colors (less faulting to more faulting). Figure 5: The production log shows that in-coherent areas on seismic having very high
potential for water coning. This zone is interpreted as a fault zone connected with the aquifer. The water coning is mainly due to very high
mobility ratio (~40) and Oil wet reservoir characteristics.
2. Pre-Job Modeling
Pre job modeling is an important aspect of geosteering as this not only enables us to understand the reservoir challenges but
also helps to develop possible solutions to mitigate them. In the current work scope the pre job modeling consisted of a
geochemical model based on XRF analysis of core chips from offset wells. This is a crucial step in the elemental analysis
case because it enables to generate a geochemical model where all the main markers are recognized and concentration
patterns are registered.
2.1 Geochemical Pre-job Model:
Pilot Study: Prior to building the geochemical model based on offset well, a pilot geochemical study was done on 3 WaraBurgan wells. The high resolution geochemical analysis based on major, trace, REE and ratios of elements defined 10
chemostratigraphic packages and 15 chemostratigraphic units for the Wara-Burgan reservoir of Minagish Field (Figure 6).
The possible diagnostic geochemical elements are provided in Table 1.
MN-A
390 feet sampled
MNB
352 feet sampled
MN403 feet sampled
IPTC 16617
Figure.6 High resolution correlation panel showing main Chemostratigraphic packages defined in the studied wells.
The pilot geochemical study of the upper Burgan reservoir shows an increase in heavy mineral contents (key-ratio U/Mg),
particularly in the middle part of the main clean sand channel, then an increase in dolomite content, followed by Fe maximum
elevated values in lower part. This followed by increasing U/Th and K/Th ratios respectively at the lowest part indicating
dirty sand and shaly sections (Figure 7). The majority of calcium came from calcite or dolomite after Diagenesis. Whereas,
the main source of Fe is Glauconite, which is either detrital or early diagenetic. K is derived mainly from feldspars with little
amount from detrital glauconite. Main Ti source is Anatase which is authigenic precipitated from acidic Ti-rich fluids. Zr is
derived from Zircon mineral deposited in a higher flow regime. U/Mg: key-ratio - U sourced from heavy mineral vs. Mg
from dolomite.
Figure 7: Proposed geochemical model for Upper Burgan (MN-A) vertical well with very good correlation with (Well-B) horizontal well.
The petrographic analysis of Burgan sand shows the presence of heavy minerals like Anatase and Zircon. Anatase is an
authigenic mineral common in sediments with titanium rich fluids. Titanium is closly related to iron and the origin of it can
be ores or iron rich minerals like amphibole, pyroxene or Glauconite (green), which are not stable in acidic fluids. A possible
source for the iron and titan sulpher in the pore fluid could be the dissolution of the Glauconite by the organic acids which
have been generated along with the hydrocarbon generation. Authigenic hematite, Fe2O3 crystals in black at the centre are
IPTC 16617
formed due to same diagenetic processes. A result of the feldspars dissolution is secondary porosity and higher concentration
of SIO2 in the pore fluid. Bottom center right side shows the consequence of the precipitation of Quatz cement in pores.
Figure
Figure
Figure 8 shows Fine grained Quartz Sandstone with 23% porosity. Authigenic Anatase (TiO2) and relicts of Glauconite are still visible in the
center following early diagenesis. Figure 9 shows lower quality sandstone with 21%.porosity.The image in the center shows a dolomite
cement which has been affected by diagenetic leaching. The dolomite cement itself corrodes the touching detrital Quartz grains. The dissolved
SiO2 from Quartz, feldspars and clays in the fluid is able to generate quartz cement. The Glauconite grains have been almost leached and
transformed into Fe- chlorite
Figure 10 shows dolomite nodules associated with a coastal sabkha environment at upper part of Burgan, while Dolomite cements associated
with shallow marine conditions at lower part of Burgan.
IPTC 16617
Figure 11: Well X Core Description (Upper and Lower Burgan Sand Sequences). To the right, the first track shows the GR log response in
front of core facies of upper and lower Burgan sand sequences. The depositional environment evolution in the Burgan sand sequence
showed a transgressive trend above the Lower Burgan fluvial sand with clear geochemical evidence.
The elemental pattern obtained from the XRF analysis of 34 samples covering the upper and lower Burgan reservoirs is
presented in Figure 12 and Figure 13. Based on this analysis, key elemental markers are being defined to be used as a guide
while geosteering the lower and upper laterals (LAT-0& LAT-1).
Figure 12: Well X elemental pattern obtaine from XRF analysis for the upper Burgan sand reservoir.
IPTC 16617
Figure 13: Well X elemental pattern obtaine from XRF analysis for upper and lower Burgan sand reservoirs.
The well locations for smart multilateral wells are optimized by integrating data from multiple disciplines starting from the
macro (seismic) to micro (petrography) scale. Further the data from seismic, geology, petrophysics, reservoir engineering,
and well surveillance were incorporated into the pre-drill characterization program. The lower lateral, LAT-0 (mother bore)
of smart multilateral wells is placed in the lower Burgan, consisting of a braided river system with stacked sand bodies. The
sediment ranges from fine, medium to coarse grain sizes with the porosity ranging from 20%-30% and having permeability
values in the order of few Darcy. The bottom part of massive thick sand bodies is directly connected to the bottom aquifer.
The upper lateral, LAT-1, of smart multilateral well is targeted in the upper Burgan reservoir consisting of silt to medium
sands. The porosity is relatively low (between 15 - 18 %) and the permeability values are in the order of hundred mD as the
reservoir still retaining the fluvial sand character. There are shaly sediments between lower Burgan and upper Burgan, which
can act as permeability barrier or baffle for vertical migration of fluids (Figure 3 & Figure 11). This complex channel
geometry makes these reservoirs most challenging for implementation of smart multilateral wells. The well trajectories were
optimized to encounter best reservoir sections by minimizing the exposure of fault networks as interpreted from high
resolution seismic (Figure 14).
IPTC 16617
Figure 14: Coherency attribbute seismic section and map view showing MN-A muliti-lateral well location. The blue, green, yellow and red
colors on seismic section refer to the faulted areas with variable degree of intensity.
1.
The most relevant aspect in designing the smart multilateral well is the selection of appropriate multilateral junction4 by
considering the well completion requirements, intelligent completion requirements, downhole equipment specification,
formation exposed across the junction area, formation fluids and the well production conditions. After a complete screening
of various multilateral junctions, the level-4 junction is selected for smart multilateral wells. During the well trajectory
optimization the multilateral junction depth was selected across excellent compact shale at the base of Wara layers to ensure
junction stability during drilling, well completion and also to have junction integrity during long term production. The high
resolution XRF analysis shows a high argillaceous siliclastic zone based on chemical behavior of elements. (Figure 15)
10
IPTC 16617
61/8hole(LAT1)
8hole(LAT0)
Figure 15 The junction depth was selected across excellent compact shale at the base of Wara layers based on the high resolution XRF
elemental data analysis
2.
The geochemical model built utilizing offset well data provided a good understanding of the geochemical imprints of the
reservoir units and thus guided us to chemosteer the well in integration with other information. In the absence of resistivity
data, XRF analysis data infact identified a fault @ 7340 ft which was not observed by near Bit Gamma (Figure 16).
Elementary changes, supplemented with lithological changes, confirms the Faults encountered in 7490ft, 7950ft, 8450ft &
8700ft (in the first & last fault-Sand to Sand- the main indication is Cl, while in the F2 & F3 the other elements are also
indicating the Fault). Cl could be used as a good marker before encountering the Fault. The findings of XRF analysis are
further confirmed while drilling from interpretation of azimuthal Density Images. Chemosteering thus helped to change the
well path based on elemental analysis and maximize the reservoir contact. (Figure 17)
IPTC 16617
11
Identification of Faults/Fractures: Azimuthal litho-density Image was interpreted while drilling not only to understand the
formation dip but most importantly to identify cluster of fractures/faults and are complemented with XRF analysis.
The Figure 18 below shows 4 fault zones identified through the LAT-0 section of Well A and the same were quite
evident from XRF Analysis as shown in Table 2. Though the real time density images were not of high confidence still were
able to provide information about formation dip and guiding the Geosteering process in upper and lower Burgan reservoirs.
Figure 18: Litho-density Image showing zones of faults (as a cluster of fractures)
12
IPTC 16617
Table 2: XRF analysis complementing the findings from litho-density Image Interpretation
During geosteering the LAT-0 of well A in the lower Burgan reservoir unit, increment of K, Al, Ti, Zr is observed
(Figure 19) indicating dirty sands. The target is to keep these elements at minimum to make sure we are in good sand,
there by having a higher value of Si. XRF Geochemical Analysis showed a clean Sand/Sandstone over the drilled interval
with Si values between 18-35% and minimum concentration of Al, Fe, K, Ti, S, Zr, CO and AS elements.
MNAL0Wellpath
Upper Burgan
Lower Burgan
Transition zone
Clean sand
Figure 19: XRF elemental alanysis in near-real-time mode for the Lower Burgan target
IPTC 16617
3.
13
During landing the well at the top of Upper Burgan sand, XRF geochemical analysis based on elemental signature showed a
sharp increment of silica (Si) and heavy minerals (Titanium,Ti & Zirconium,Zr). Al (Aluminum), Fe (Iron), and K
(Potassium) had a downward trend. There are also other elements like Cl, Mo, Cr, Ag, Co and Sn going to zero values at the
top of Burgan sand. These proxies are considered as excellent markers and had very good correlation with the offset well
MN-X. The Silica (Si) values decreased after displacing the oil base mud (OBM) with FAZEPRO calcium carbonate mud
due to the contamination and high percentage of calcium. The correlation became difficult with MN-X and other elements
have been selected for correlation. There are 10 proxies with elemental signature in clean sand lobes of upper Burgan. While
geosteering in the sweet zone an increase in heavy mineral contents (Ti and Zr), particularly in the middle part of the main
clean sand channel, then an increase in Mg content has been noticed while cutting down structure, followed by Fe and Al
maximum elevated values while cutting further down at the lowest part of Burgan indicating dirty sand and shaly sections.
(Figure 20) The evidence of faulting in the upper lateral (LAT-1) showed elementary changes, supplemented with
lithological changes very similar to the faults encountered in the Lower Burgan lateral section (LAT-0). See (Figure 17). The
pilot XRF model and offset well MN-X are used with high confidence to steer the well based on clear geochemical finger
print and markers associated with Burgan sub-layers. This has resulted in a successful chemosteering in the clean sand lobe
with less than 10 feet thickness and maximized the reservoir contact even with structural complexity associated with faults
and significant dip changes.
MNX(offsetwell)
MNA(LAT1)Multilateral
Figure 20: XRF elemental alanysis in near-real-time mode for the Upper Burgan target
14
IPTC 16617
In summary, the current work scope was a success, solely due to merger of independent data sets from different analysis
through a collaborative approach, extracting the right information at the right time. The first data to be gathered are seismic
data, providing the vital reference framework, such as: identifying the most likely faulted sections and the type of dislocation
they can cause. The seismic image below (Figure 21) shows the horizontal well section, superimposed with abundance of
main elemental marker obtained through XRF analysis. The markers chosen are amongst those which provided the better
contrast during the pre-well study: Al, Zr, Ti, K. Clear changes in the abundance of these elements are associated with the
main features of the well section, i.e. the landing point approximately @ 7300 ft, the faulted sections @ 7940 ft and 8450 ft
in particular. Al also provided an early indication of the approaching fault as it started showing an increase at the start of the
disturbed section at 7836 ft.
Similarly, elemental markers are also matched with the MWD azimuthal tool response (Figure 22). Al started to increase with
the first minor fault, and peaked at 7940ft. It remained high till 8100 ft where it started to lower values. Ti had similar
behavior with the difference that it only appeared in earnest when the main fault was encountered at 7940 ft. The indication
from Aluminum is particularly important because it came as an early sign of the fault. The azimuthal information arrived
later and confirmed that a fault had dislocated the well trajectory which was corrected as a consequence.
Figure 22: Elemental markers superimposed with main events: landing and main faults
However the utilization of chemical elements measured in real-time with high density points is not the only utilization of this
technology. In fact, the same data can be utilized during the production phase to isolate the faulted interval. In our case Al
IPTC 16617
15
concentration, which had a significant increase across the faulted sections was used as a proxy along with permeability
measurements to mark and isolate these zones of potential trouble (Figure 23) which would be very detrimental for water
production.
Figure 23: Isolated zones along the well path L0, Well A.
6. Conlcusions
The paper demonstrated key merits of an integrated approach for geo-steering of smart multilateral well and its successful
outcome in a highly challenging Burgan reservoir. It is a belief that this workflow will benefit the industry while developing
reservoirs of similar nature.
Whenever a new dataset is entered in the Formation Evaluation process, it is crucial to make sure that it becomes part of the
work flow loop, otherwise the advantages and new information it can provide may be underutilized or ignored. This is a vital
part of the integrated approach.
It is imperative that a collaborative workflow from planning till completion involving seismic, geochemical analysis along
with petrophysical evaluation is necessary to achieve the objective. This was clearly evident in our case where:
i) Seismic data has enabled to prepare the background platform,
ii) Geochemical analysis of offset data was utilized to make a pre-job geochemical model providing the reference indications
for the key marker elements,
iii) In real time geosteering utilizing all the available data sets to place the well in the zone of interest and maximize the
drainage area.
The quick reaction to changes based on available data sets has granted a longer reservoir section to be exposed in the
horizontal drain holes, subsequently led to modification of the production plans, isolating troublesome or less productive
sections. In summary, this work scope was successfully utilized to drill the first Smart Multi-lateral Well in Kuwait as well
as in Middle East Region.
16
IPTC 16617
Pre drill: Reinterpret seismic using a more aggressive approach involving more small faults.
2.
3.
Use experience gained in current job to plan and execute the next job with a focus on consequences of hard turns in
a narrow zone.
4.
Chemosteering should be an integral part of geosteering while dealing with such kind of complex reservoirs.
Acknowledgement
The Authors would like to thank Kuwait Oil Company and the Ministry of Oil for permission to publish this paper. They are
grateful to Hassan Bunain (Manager, Field development, West Kuwait) and Khalaf Al Anezi (Team Leader of Minagish,
west Kuwait) for their encouragement and support.
References
1.
Khalaf Al-Enezi, Om P. Das, Muhammad Aslam, Khaled Ziyab and Taher El-Gezeeri; Kuwait Oil Company:
Successful Case Histories of Smart Multilateral Well with Inflow Control Device and Inflow Control Valve for
Life-cycle Proactive Reservoir Management in High Mobility Reservoir, Minagish Field West Kuwait SPE 161632
2.
OM. Das, Khalaf Al-Enezi, , Muhammad Aslam, Taher El-Gezeeri and Khaled Ziyab and; Kuwait Oil Company:
Novel Design and Implementation of Kuwait's First Smart Multilateral Well with Inflow Control Device and
Inflow Control Valve for Life-cycle Reservoir Management in High Mobility Reservoir, West Kuwait SPE
159261
3.
Taher El Gezeery, Abdul Aziz Khaled Ismael and Khalaf Al Anezi, Kuwait Oil Company Horizontal Wells
Optimize Production in a Super K Sandstone Reservoir, Minagish Field, West Kuwait paper SPE 684529,
presented at GEO 2010, the 9th Middle East Geosciences Conference and Exhibition held in Bahrain at Bahrain
International Exhibition Centre, 7 to 10 March 2010.
4.
K. Al-Enezi, O.P. Das, M. Aslam, R. Bahuguna, and A. Latif, Kuwait Oil Company: Water Coning Model for
Horizontal Wells in High Mobility Reservoir, West Kuwait paper SPE 130302, presented at the CPS/SPE
International Oil & Gas Conference and Exhibition held in Beijing, China, 810 June 2010.
5.
Taher El Gezeery, Abdul Aziz Khaled Ismael, Khalaf Al Anezi; Kuwait Oil Company and Christian Scheibe
Chemostratigraphic Differentiation Between Fluvial and Shore-Face Sands as a Real-Time Geosteering Tool
Within the Albian Upper Burgan Formation, Minagish Field, West Kuwait" paper SPE 684527, GEO 2010, the 9th
Middle East Geosciences Conference and Exhibition held in Bahrain at Bahrain International Exhibition Centre, 7 to
10 March 2010.
6.
Taher El Gezeery, Fawaz Al Saqran; Kuwait Oil Company and Ekpo Ita Archibong, and Somaya Al-Radhi,
Schlumberger Utilizing Real Time Logging While Drilling Resistivity Imaging to Identify Fracture Corridors in a
highly fractured Carbonate Reservoir, Kuwait SPE 118152, Abu Dhabi International Petroleum Exhibition and
Conference held in Abu Dhabi, UAE, 36 November 2008.