SPE 134515

Novel Approach to Quantifying Deepwater Laminated Sequences Using

Integrated Evaluation of LWD Real-Time Shear, Porosity, Azimuthal Density
and High-Resolution Propagation Resistivity
Katerina Yared, SPE, Baker Hughes Inc.; Mauro Pelorosso, ENI US Operating Co.; Ismail Altintutar, SPE, Derek
Buster, SPE and Ela Manuel, Baker Hughes Inc.; James Doyle and Charles Russell, SPE, ENI US Operating Co.

Copyright 2010, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Deepwater Drilling and Completions Conference held in Galveston, Texas, USA, 5–6 October 2010.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Accurate reserve volume determination is crucial in the early stages of a project since planned subsurface capacity is
dependent on reserve expectations. The fundamental method of calculating reserves uses bulk formation resistivity and bulk
porosity to determine water saturation. This approach cannot accurately quantify reserves in laminated sand-shale sequences
where the sensor resolution is insufficient to characterize the fine laminae. A tensor petrophysical model can determine
laminar shale volume and laminar sand-fraction conductivities reducing the problem to a single dispersed shaly sand model.
Combining this with sand-fraction porosities can lead to accurate reserve quantification.

Identification and quantification of hydrocarbons within low-contrast, low-resistivity formations can be difficult when using
conventional log data. This is primarily due to the presence of laminar shale and the inherent vertical resolution of
measurements acquired by wireline and logging while drilling (LWD).

A Gulf of Mexico deepwater example is used to demonstrate this novel approach in quantifying hydrocarbons in laminated
sand-shale sequences. Real-time shear slowness is used in conjunction with LWD triple-combo data to identify potentially
productive low-contrast reservoirs. Then, advanced resistivity post processing extracts the vertical component of resistivity,
enabling calculation of sand-fraction resistivity. Sand-fraction resistivity, combined with normalized sand-fraction porosity,
yields sand-fraction water saturation. Shale volume, porosity and water saturation cut-offs determine the net hydrocarbon
volume. The LWD-calculated hydrocarbon volumes in place are then compared to results obtained from a wireline logging
suite.

This approach demonstrates that the use of conventional empirically derived bulk-volume porosity and saturation methods in
laminated sand-shale sequence formations results in underestimation of the reservoir producibility and hydrocarbon reserves.
Vertical resistivity, derived from LWD-acquired propagation resistivity and electrical anisotropy sensitivity, can be used to
quantify reserves in these environments.

Introduction
In-depth analysis and hydrocarbon estimates of reservoirs are often performed using wireline logging. However, wireline
acquisition becomes very challenging in highly deviated wellbores where wireline tools have increased risk of becoming
stuck. This paper illustrates the use of LWD-acquired data in high-angle wellbore environments in accurately quantifying
hydrocarbons that are present in the laminated sand-shale sequences. High-quality formation evaluation data acquired in
these environments directly impacts customers’ operating costs through savings in rig time.

The well being studied is a development well in a field of the US Gulf of Mexico (GoM) where water depths are
approximately 2500 ft. The field produces from several stratigraphic intervals ranging from lower Pliocene to upper Miocene
in age. This study well targeted two upper Miocene reservoirs that were either already producing or sanctioned for
development in other fault blocks and a third, deeper sand that had been penetrated by earlier appraisal wells. In this field,
2 SPE 134515

the trap is a faulted structural four-way closure with a stratigraphic component (lateral pinch-out of the levee-overbank
sands). Reservoirs were deposited as submarine channel and levee complexes in a slope setting (Fig. 1).

Fig. 1: Depositional environment; channel-levee complex (from Ruvo et al, 2008)

There are several well penetrations of the reservoirs in this field. This depositional model is supported by lag and core data..
Both massive sands, interpreted as channels, and laminated sand-shale and silt sequences, interpreted as levee deposits, were
penetrated in this well and in offset wells. Fig. 2 illustrates core samples from the laminated sand-shale sequences in the
nearby field. Conventional (full-diameter) cores acquired in two offset wells reveal that net-to-gross in the laminated
intervals ranges up to 40percent, and that the lamination is on a scale of several millimeters to several centimeters. Analysis
of the laminated levee sequence of the uppermost of the three sands in the subject well is the focus of this paper.

Fig. 2: Core photograph of laminated sand, silt and shale in an interpreted levee sequence from one of the upper Miocene reservoirs
in an offset well (from Ruvo et al, 2008).

The primary objectives of the LWD acquisition and interpretation are to (1) identify potential low-contrast hydrocarbon-
bearing intervals using real-time LWD acoustic compressional and quadrupole shear slowness measurements, together with
resistivity and porosity measurements (2) quantify the net-to-gross interval and define the formation structure (relative dip)
with LWD density images, (3) invert the LWD propagation measurements for horizontal and vertical resistivity and (4)
perform laminated sand-shale sequences analysis (LSSA) using tensor resistivities, along with Thomas-Stieber sand-fraction
normalized porosities to quantify these laminated sequences.
SPE 134515 3

Low-Contrast Low-Resistivity Reservoir Identification Using Real-Time LWD Shear

The velocities of acoustic waves traveling through a formation are dependent on different properties like lithology and fluid
types. Shear acoustic waves do not propagate through fluids, thus they only represent the structural frame of the rock.
Therefore they are less affected by the fluids in the pore space. Compressional acoustic waves on the other hand are overly
attenuated by fluids, especially by light hydrocarbons. The Acoustic Log Hydrocarbon Indicator (ALHI) proposed by
Williams 1990, contrasts water-saturated and hydrocarbon-saturated formation responses using the compressional wave (Vp)
to shear wave (Vs) velocity ratio, or Vp/Vs. A comparison of the near-linear relationship of Vp to Vs in water-saturated
mudrocks (Castagna et al. 1985) and a synthetically generated ALHI Vp/Vs ratio computed using only the real-time LWD
quadrupole shear arrival (Tang et al. 2003) highlights these hydrocarbon-saturated formations. The resulting ALHI Vp/Vs
ratio is plotted with the measured Vp/Vs ratio on the log and shading between their crossovers to indicate a possible presence
of hydrocarbons (Fig. 3). The real-time compressional acoustic wave response corrected for light hydrocarbon effects (called
“RT DT Comp. Corr”. in Fig. 3) using the ALHI method plotted against the uncorrected real-time compressional acoustic
wave also gives an indication of the possible presence of hydrocarbons in the thin bedded interval. The resistivity and density
neutron curves show the suppression of the hydrocarbon response of the low-contrast, low-resistivity thinly bedded reservoir:
the resistivities are low and the density/neutron profile is typical of shales.

Fig. 3: Real-Time Acoustic Log Hydrocarbon Indicator Plot. In track 5 possible hydrocarbon bearing zones are indicated by the
yellow shaded areas.

A crossplot (Fig. 4) of the measured Vp and Vs responses against the Castagna mudrock trend line (Vp/Vs = 1.16 · Vs +
1.36) and the Williams ALHI trend line (ALHI Vp/Vs = 1.13 · Vs + 1.116) is another way to indicate the presence of possible
light hydrocarbons when these points are between the ALHI Vp/Vs ratio and the dry mudrock line. Contrasts between the
computed and the measured Vp/Vs ratio highlight crossover events in the light hydrocarbon-saturated rocks similar to the
crossover of bulk density and neutron porosity but without radioactive chemical sources.
4 SPE 134515

Fig. 4: Real-time Vp vs. Vs crossplot indicating the presence of hydrocarbon in the low-contrast, low-resistivity reservoir (yellow
shaded area)

Borehole Imaging and Interpretation

Azimuthal LWD density image was acquired for this field to achieve an accurate structural description of the reservoir. The
thinly bedded nature of these sands is readily observed on the detailed borehole image below (Fig. 5). Full diameter cores
from a nearby offset well (Fig. 2) confirm the presence of sand-shale laminae in the analyzed section. The density images are
displayed in high-side and low-side borehole configurations so that the high-side is on the left and right of the image and
low-side is on the center of the image. Even though these beds are thinly bedded, the laminations in these sequences are
likely to have similar reservoir properties to the thick massive sands.
Structural interpretation in this study focuses on structural dip/boundary determination. Structural dip and azimuth of the
formation were manually calculated from the image based on the varying density readings of the boundaries between the
beds (Fig. 5). Low-density readings appear lighter on the image and higher density zones appear darker on the image. The
advantage of the manual dip technique is that each feature can be classified into a geological category (e.g. shale bedding,
sandstone bedding etc). Although the structural dips were calculated using LWD memory images, the real-time images
exhibited sufficient resolution enabling this interpretation to be made in real-time, especially at low rate of penetration (below
100ft/hr). The structural information (bed dip and its azimuth) calculated from the density image are then combined with
wellbore trajectory information (deviation and azimuth) to compute the relative dip angle, angle between the normal to the
bedding plane and the tool axis. The relative dip angle is used in the resistivity inversion process to determine horizontal
resistivities and anisotropy ratio leading to determination of vertical resistivities. Vertical resistivities are used to generate
sand-fraction resistivities, which in turn lead to quantification of hydrocarbons.
Another successful application made possible by the resolution and 100-percent wellbore coverage of the LWD density
image is net-to-gross analysis in these thinly bedded reservoirs. The absence of poor wellbore conditions and thinly cemented
beds yielded good-quality LWD density images, resulting in reasonable estimates of net-to-gross. The LWD density image
provided sand volume estimation in the absence of whole core data. The acquired wire line images have higher resolution of
while drilling images how expected but it was influenced by washout occurred after drilling.
Net-to-gross analysis from the LWD density image is performed through several steps. First, the average of 16-sectored
density image was calculated. The average density is then correlated with the image and other formation evaluation data to
determine density cut-offs to define three different lithologies, i.e., sandstone, siltstone, and shale. Based on this correlation,
the following cut-offs were used: sandstone <2.12 g/cc, shale >2.18 g/cc, with siltstone being represented by the values in
between. The sandstone and siltstone data were then integrated to produce an estimate of 221 ft of sand over a 510-ft interval,
corresponding to a final sand volume of 43.4 percent (Table 1). This approach, in combination with the petrophysical model
SPE 134515 5

described below reduces the uncertainty in net pay and minimizes the possibility of missing reservoirs not easily detected
using conventional techniques.

Fig. 5: Azimuthal Density Image, Formation Dips and Net-to-Gross Analysis from LWD Density Service
6 SPE 134515

Propagation Resistivity and Electrical Anisotropy

In anisotropic formations, the resistivity measurement made is a function of the direction of the measurement. Anisotropic
effects occur when the electromagnetic signal traverses bed boundaries, resulting in a measurement that is a combination of
the vertical resistivity, and the horizontal resistivity, (Hagiwara 1996, 1997). With LWD electromagnetic propagation
measurements, the magnitude of the observed anisotropy effect is influenced by the transmitting frequency of the
electromagnetic wave, the transmitter-to-receiver distance, the type of measurement (phase shift or attenuation), the relative
dip angle, and the anisotropic ratio ( / ) of the formation. Anisotropic effects cause curve separation with the
higher frequency, 2-MHz measurements reading higher than the lower frequency, 400-kHz (Hearn et al. 1997; Boonen et al.
2001). The higher frequency is affected more by anisotropy than the lower frequency.

The relative dip angle plays a key role on the exhibited effects of anisotropy. When the relative dip angle is small (less than
35°), the propagation sensors typically respond to the horizontal component of resistivity, as the currents induced by the
transmitters are in the horizontal plane. However, when the relative dip angle increases, the electromagnetic waves cross the
bed boundaries, resulting in measurements that now are sensitive to the vertical component of resistivity.

Fig. 6 plots the apparent propagation resistivity response as a function of the relative dip angle for a formation with an
anisotropic ratio ( / ) of 3 as an example for GoM data. It is clear that separation between phase and attenuation
measurements starts at approximately 35 of relative dip and becomes significantly large at higher relative dip angles.

Fig. 6: Propagation Resistivity as a Function of Relative Dip Angle

Advanced LWD Propagation Resistivity Processing

LWD electromagnetic multiple propagation instruments make numerous measurements based on the alteration of the
electromagnetic wave as it travels across a pair of receivers. The reduction of the amplitude and the change in velocity of this
wave are exploited and transformed into conductivities and converted into resistivities. The resulting multi-frequency and
multi-spacing resistivities are referred to as “true-resolution” resistivities and have varying depths of investigation with each
measurement having its own response to environmental effects (Hearn et al. 1998; Meyer 1997). Environmental effects
include adjacent bed effects, variations in formation anisotropy and dielectric permittivity. To interpret these LWD
resistivities properly, post processing is required to bring the various measurements to a common resolution and to account
for the contained effects. The results of the processing lead to accurate horizontal formation resistivities.

The Baker Hughes post-processing strategy (Meyer 1997) includes corrections for borehole, anisotropy, dielectrics and
nearby bed effects. Fig. 7 illustrates the steps involved for this unique post-processing scheme. Following borehole
SPE 134515 7

corrections and deconvolution to account for shoulder bed effects, simultaneous evaluation yields corrections for anisotropy
and dielectric effects. The inverted data are then linearly combined to produce fixed depth of investigation (DOI)
resistivities. The method has yielded excellent results for relative dip angles of less than 70˚.

Fig. 7: Advanced LWD Propagation Resistivity Processing Sequence

Extraction of the vertical component of resistivity has traditionally been made with wireline sensors that have mutually
orthogonal transmitter-receiver pairs that make 3-dimensional measurements. However, when LWD propagation tools
intersect laminated sand-shale sequences with relative dip angles above 35˚ (Fig. 6), the measurements are affected by
electrical anisotropy, and a correction is applied (simultaneously with dielectric corrections) to the affected measurements to
extract the horizontal component (Hearn et al 1998). This sensitivity to anisotropy is exploited to extract the vertical
component of resistivity, . Deep horizontal resistivity from the processing is represented by the 60-inch measurement,
60 (Fig. 8, track 5). With derivation of the horizontal component and knowledge of the anisotropy ratio, the vertical
component of resistivity can be derived:
60 (1)
/ (2)
Therefore,

60 / . (3)
Fig. 8 illustrates the results of the process described to extract the vertical component of resistivity, , for the bottom part of
the reservoir (bottom zone). Track 1 plots gamma ray, wellbore deviation and relative dip angle (calculated from bulk
density images). Note the subtle drop in the gamma ray (approximately 20 API) from the shales at the top towards the
bottom of the zone. As illustrated (Fig. 6), the relative dip angles (42°-58°), are sufficient to cause curve separation in the
propagation resistivities that are plotted in track 2 (electromagnetic waves are crossing boundaries). The true resolution
resistivities indicate a very small increase from approximately 1 Ω.m in the shales at the top, to approximately 2 Ω.m at the
8 SPE 134515

bottom of the zone. Both gamma ray and resistivity profiles are typical in laminated sand-shale sequences. Simultaneous
evaluation of dielectrics and anisotropy coefficients leads to the anisotropy ratio plotted in track 3, indicating possible
presence of hydrocarbons. Track 4 plots the processed horizontal resistivity, . (RES60 and calculated vertical resistivity,
. Vertical and horizontal resistivities are used in determining sand-fraction resistivities, .

Fig. 8: Vertical Resistivity Determination - Log of the bottom part of the reservoir (Zone 2) illustrating the anisotropy effect on true
resolution propagation resistivities in track 2. Determination of anisotropy ratio in this high relative dip angle environment enables
the determination of vertical resistivty plotted in the final track.

Macroscopic Anisotropy – Laminated Sand-Shale Sequences

Thinly bedded laminated sand-shale sequences are common in economically important Gulf of Mexico deepwater turbidite
reservoirs (Mollison et al. 2001). Electrical anisotropy is the macroscopic effect of thinly laminated sedimentary formations
in which the logging sensors have insufficient vertical resolution to properly quantify the properties of the individual layers.
The physical effect of the lack of vertical resolution in laminated sequences is “macroscopic anisotropy” (Hagiwara 1997),
where the measurements now exhibit a directional dependence with regards to resistivity. Macroscopic anisotropy is a direct
result of the vertical and horizontal resistivity differences created by the differing sand and shale resistivities. Therefore, the
traditional scalar model is transformed into a tensor resistivity model. There are no conventional logging instruments that
can resolve the individual lamina in thinly laminated sand-shale sequences. The measurements made parallel to the beds
(horizontal component) are greatly dominated by the conductive shales while the measurements made perpendicular to the
bedding (vertical component) are dominated by the resistive sand layers.
SPE 134515 9

Scenario 1: Isotropic Shale – Isotropic Sand

Consider the case where the individual shale layers and the sand layers are homogeneous and isotropic (Schoen et al. 1999).
Two equations can be written to address the horizontal and vertical component of resistivity. The horizontal component of
resistivity is represented by a parallel circuit (Eq. 4), while the vertical resistivity is represented by a series circuit (Eq. 5).
Eq. 6 represents the anisotropic ratio while Eq. 7 accounts for the volumetric balance.

1/ / 1 / (4)
1 (5)
/ (6)
1, 1 (7)

Using equations 1 and 2, we can solve for the sand-fraction resistivity, , and the laminar shale volume, .

(8)

and
(9)

Fig. 9 plots resistivity against laminar shale volume for an isotropic sand-isotropic shale formation with 10 Ω.m-sand and 1
Ω.m-shale layers with 50-perchent net-to-gross 1 ). With only 20-percent shale volume, the horizontal resistivity
reads 3.8 Ω.m. With 50-percent laminar shale volume, the resistivities calculated are 1.8 Ω.m, much less than the 10 Ω.m-
sands in which we are interested. Using these shale-conductivity dominated measurements in water saturation equations
results in significant underestimation of hydrocarbons in place.

Fig. 9: Resistivity vs. Laminar Shale Volume – Isotropic Sand / Isotropic Shale

Scenario 2: Anisotropic Shale – Isotropic Sand

The Gulf of Mexico shales are typically anisotropic and sands are isotropic (Hagiwara 1996, 1997). This makes horizontal
and vertical response equations for the shale portion of the laminar sequences necessary. The sand-fraction resistivity, ,
10 SPE 134515

and laminar shale content, for anisotropic shale-isotropic sand case can be calculated from horizontal resistivity, ,
vertical resistivity, , along with horizontal and vertical shale resistivities, , and , .

, , 1 ∆ (10)

where,
,
(11)
,

and,
, ,
∆ 4 (12)
,

The ∆ term represents the correction for the anisotropic shale. The laminar shale content is expressed as follows:

(13)
,

If the shale fraction is isotropic, then, , , , and the equations revert back to isotropic shale-isotropic sand scenario
and is given by Eq. 8.

Sand-Fraction Porosities
Porosity devices acquire measurements that represent bulk porosity of the formation. To use bulk-porosity measurements in
conjunction with fractional resistivities to determine saturations, the porosities must be converted to fractional quantities.
The laminar sand-fraction porosity is extracted from the Thomas-Stieber shale distribution model (Thomas et al. 1975) . The
graphical representation for the volumetrically balanced, sand-fraction total porosity is shown in Fig. 10. The triangle
represents the Thomas-Stieber volumetric shale distribution model based on the assumption that the maximum clean sand
porosity and shale porosity can be determined from the input data. The model plots total porosity against shale volume to
determine shale distribution (laminar and dispersed shale volumes) and laminar sand-fraction porosity. The main concept for
this methodology is in the removal of laminar shale effects using a total porosity system. While dispersed shale occupies and
alters the intergranular sand porosity, the laminar shale primarily changes the net-to-gross without occupying the pore spaces
and drastically reducing the porosity. Note the reduction in total porosity as we move from the 100-percent clean point (φmax)
towards the 100-percent dispersed point (φmax ·φshale). To determine the total sand-fraction porosity, the bulk laminar shale
porosity must be removed from the total porosity (as determined by bulk measuring devices) and the remaining porosity must
be normalized to laminar sand volume:

(14)

To determine the true effective sand-fraction porosity, dispersed shale volume must be removed from the total sand-fraction
porosity. Dispersed shale volume is also normalized to the laminar sand volume:
(15)

As an example of bulk method versus Thomas–Stieber method (Thomas et al. 1975), consider an interval of sand/shale
lamination with 50-percent net-to-gross where sand lamina porosity is 30-percent and shale lamina porosity is 15-percent
( 0.50, 1 0.50 . Conventional bulk-measuring devices would record an average of the two
porosities:

0.15 0.5 0.30 0.5 22.5% (16)

If one were to use effective (shale-corrected) porosity:

0.225 0.15 0.5 15% (17)

SPE 134515 11

Clearly, neither the average total porosity nor the effective bulk porosity reflects the actual sand-fraction porosity. The use of
total or effective bulk porosities results in significant underestimation of hydrocarbon saturation. True laminar sand total
porosity, corrected for laminar shale effects, is obtained through the volumetrically balanced (normalized to sand volume)
Thomas-Stieber method:

· / 1 0.225 0.15 0.5 /0.5 30% (18)

The graphical solution for the volumetrically balanced sand-fraction porosity is shown in Fig. 10.

Fig. 10: Thomas-Stieber Shale Distribution Model

Laminated Sand-Shale Analysis (LSSA)

The Baker Hughes LSSA (Fig. 11) is designed for evaluation of low-contrast low-resistivity sand-shale sequences using
vertical and horizontal resistivities, porosity and shale volume data (Mollison et al. 1999; Mollison et al. 2001). The routine
uses Thomas-Stieber volumetric total porosity - total shale distribution in combination with a macroscopic electrical
anisotropy model to extract true resistivity of the sand fraction, sand-fraction porosity and shale distribution. This in turn
leads to quantification of hydrocarbons within the sequences using Waxman–Smits, Dual Water or Archie total water-
saturation methods.

LSSA implements a petrophysical model that is internally self-consistent both petrophysically and mathematically. The
routine is limited to a two-component sand-shale lithology based on a two-component tensor resistivity model that is
electrically and volumetrically balanced.
12 SPE 134515

Fig. 11: Laminated Sand-Shale Analysis Processing Flow Diagram.

Results

Table 1 shows net-to-gross percentage values for the total zone of interest using LWD azimuthal imaging and LSSA. The
two independent methods are in good agreement and in line with core-determined net-to-gross in the nearby well.

Table 1: Net–Gross Comparison – Density Imaging vs. LSSA

Total Net‐to‐gross Estimation Comparison 
LWD Density Imaging  43.4% 

LSSA Using LWD Data  39.8% 

Fig. 12 illustrates the Thomas-Stieber shale distribution for the entire zone of interest with total bulk porosity plotted against
shale volume. Maximum porosity, shale porosity and 100-percent dispersed point defines the triangle. The majority of the
data falls on the laminated sand-shale sequence zone indicating the presence of laminated sand-shale sequences in the
reservoir. This laminated zone is primarily located at the bottom of the reservoir were the LSSA results indicate the thickest
net pay zone (Fig. 13). The top-left cluster of data points represents the water sand at the bottom of the reservoir.
SPE 134515 13

Fig. 12: Thomas-Stieber Triangle Plot over the Zone of Interest

Fig. 13 displays the results of the tensor petrophysical analysis over bottom zone. The log presentation is as follows:

Track 1: Gamma ray and shaliness indication

Track 2: Horizontal resistivity ( ), vertical resistivity ( ), and sand-fraction resistivity ( )
Track 3: Bulk density and neutron porosity
Track 4: Depth and anisotropy Ratio ( / )
Track 5: Thomas-Stieber sand-fraction porosity
Track 6: Sand-fraction total water saturation ) and net tensor pay flag (red bar)
Track 7: Bulk effective and total porosity (from density and neutron porosity), bulk water and hydrocarbon volume
Track 8: Bulk formation total water saturation and net conventional pay flag (red bar)
Track 9: Bulk formation volumetrics (sand, laminar and dispersed shale, bulk water, and hydrocarbons)

The LSSA net pay flag in track 6 is based on effective water sand-fraction saturation cut-off of 65-percent and effective sand-
fraction porosity of 12 percent. The conventional net pay flag in track 8 uses the same cut-offs, but is based on bulk
measurements. Over this bottom zone of interest, he conventional method yields no net pay while LSSA tensor method,
yields 54 ft of net pay.
14 SPE 134515

Fig. 13: Laminated Sand-shale Analysis (LSSA) Results of the bottom zone.

Table 2 lists the reservoir pay summary over the total zone of interest. LSSA tensor analysis, using sand fractional properties
derived from horizontal and vertical resistivities and Thomas-Stieber porosities, yields 203 ft of net pay with 29-percent
average effective sand porosity and 37-percent average effective sand-fraction porosity. Conventional bulk analysis, using
horizontal resistivities and average bulk porosities suggests only 6 ft of net pay with 15-percent average effective porosity
and 50-percent average effective water saturation. Interpretation based on the tensor analysis method recommends this zone
for completion, however, interpretation based on conventional bulk method would likely result in bypassing (overlooking)
this zone of interest. The use of heavily suppressed resistivities, due to conductive shale laminations, and average bulk
porosities, significantly underestimates hydrocarbon saturations in this interval of laminated sequences.

Table 2: LSSA Pay Summary

LSSA Tensor  Conventional Bulk 
   Petrophysiscs  Petrophysics 
Gross Interval           510 ft    510 ft   

Net Pay                   203 ft      6 ft 

Net‐to‐Gross             40%  1% 

Average Effective Porosity (pay)   28.9% (sand)  15.0% 

Average Effective Sw (pay)    36.9% (sand)  50.0% 

SPE 134515 15

Conclusions
LWD acquired real-time acoustic compressional and quadrupole shear slowness, along with propagation resistivity and
density/neutron porosity data were used to identify the presence of hydrocarbons in the thinly bedded sand-shale sequences.
The structural information from the processed LWD azimuthal density image was used to calculate the relative dip angles of
the formations. Post processing, using raw LWD propagation measurements, enabled the determination of horizontal
resistivity, Rh, and anisotropy ratio, (Rv/Rh). The anisotropy ratio determination was facilitated by the sensitivity of
propagation resistivities to the high relative dip angles (relative dip angles above 35° but less than 70°) across the zones of
interest. The presence of anisotropy enabled the extraction of the vertical resistivity, Rv.

Tensor petrophysical analysis was applied to determine sand-fraction resistivities in the laminated sand-shale sequences. The
sand-fraction porosities were determined using the Thomas-Stieber method. The resulting sand-fraction porosities were then
combined with the sand-fraction resistivities to compute sand-fraction saturations. Sand-fraction saturations were combined
with laminar shale volumes resulting in formation volumetrics. Conventional bulk method yielded only 6 ft of net pay across
the zone of interest while the tensor method with Thomas-Stieber porosities yielded 203 ft of net pay. The tensor/Thomas-
Stieber method, used in LSSA, enhances the reserves estimates by increasing the net pay thickness, increasing the porosity
and decreasing the water saturations in the sand laminations over the estimates provided by conventional methods. Relying
on conventional analysis would have significantly underestimated the volume of hydrocarbons over these laminated
sequences making this reservoir economically insufficient.

In addition to providing crucial structural and near-wellbore information, the LWD azimuthal density imaging provided net-
to-gross estimation that were in line with net-to-gross estimation provided by LWD post processing and tensor petrophysical
analysis (LSSA). The LSSA results, based on LWD-acquired data, were compared by the operator to the net-to-gross
analysis/interpretation based on wireline acquisition. Interpretation results were comparable and within the margin of error,
providing confidence on the reliability of the approach used by the LWD acquisition method. The enhancements of pay,
provided by tensor method over conventional, have warranted completion and have played a significant role in determining
the completion strategy for the stacked pays encountered by this well.

Nomenclature
Rh = Horizontal resistivity
Rv = Vertical resistivity
Rsd = Sand-fraction resistivity
Rsh = Shale- fraction resistivity
= Isotropic sand resistivity
, = Vertical shale resistivity component
, = Horizontal shale resistivity component
Vsd = Sand volume
Vsh = Shale volume
= Dispersed shale volume
= Laminated shale volume
 = Dispersed shale volume
λ2 = Anisotropy ratio
= Effective porosity
= Total porosity
= Sand porosity
= Laminated shale porosity
= Total sand porosity

Acknowledgments
We would like to thank ENI US Operating Co., Nexen Inc. and Baker Hughes Inc. for their support and permission to publish
this paper.
16 SPE 134515

References
Boonen, P., Haugland, S.M., and Laughlin, D.G. Analysis of LWD Propagation Resistivity Data in Anisotropic, Thinly-Bedded Formations
Identifies Significantly More Hydrocarbons. Paper SPE 95894. presented at the 2005 SPE Annual Technical Conference and
Exhibition, Dallas, Texas, October 2005.

Castagna, J. P., Batzle, M. L., Eastwood, R. L. 1985. Relationships between compressional-wave and shear-wave velocities in clastic
silicate rocks. Geophysics 50 (4): 571-581.

Hagiwara, T. 1996. EM Log Response to Anisotropic Resistivity in Thinly Laminated Formations with Emphasis on 2-MHz Resistivity
Devices. SPE Formation Evaluation Journal 11 (4): 211-2118.

Hagiwara, T. “Macroscopic Anisotropy” Approach to Analysis of Thinly Laminatted Sand/Shale Sequences: Sensitivity Analysis of Sand
Resitivity Estimate and Environmental Corrections. Paper SPE 38669. presented at the 1997 SPE Annual Technical Conference and
Exhibition in San Antonio, Texas, October 1997.

Hearn, F.P., Meyer, W.H. and Wisler, M.M. Advanced Processing Methods for a New Generation Propagation Resistivity Tool Improves
Interpretation Methodology. Paper SPE 49135. presented at the 1998 SPE Annual Technical Conference and Exhibition, New
Orleans, Louisiana, September 1998.

Meyer, H.W. 1997. Multi-Parameter Propagation Resistivity Interpretation. Paper SPWLA GG. presented at the SPWLA 38th Annual
Logging Symposium, Houston, USA, June 15-18.

Mollison, R.A., Schoen, J.S., Fanini, O.N., Kriegshäuser, B., Meyer, W.H., Gupta, P.K. 1999. A Model For Hydrocarbon Saturation
Determination From Tensor Resistivity Relationship in Thinly Laminated Anisotropic Reservoirs. Paper SPWLA OO. presented at the
40th Annual Symposium, Oslo, Norway, May 30 – June 3, 1999.

Mollison, R.A., Fanini, O.N., Kriegshäuser, B.F., Yu, L., Ugueto, G., van Popta, J. 2001. Impact of Multicomponent Induction Technology
on a Deepwater Turbidite Sand Hydrocarbon Saturation. Paper SPWLA T. presented at the 42nd Annual Logging Symposium,
Houston , USA, June 17-20.

Ruvo, L., Doyle, J., Cozzi, M., Riva, S., Scaglioni, P., Serafini, G. 2008. Multi-Scale Data Integration in Characterizing and Modeling a
deep-water Turbidite Reservoir. Paper SPE 115305 presented at the SPE Annual Technical Conference and Exhibition, Denver,
Colorado, USA, 21-24 September.

Schoen, J.H., Mollison, R.A., Georgi, D.T. Macroscopic Electrical Anisotropy of Laminated Reservoirs: A Tensor Resistivity Saturation
Model. Paper SPE 56509 presented at the 1999 SPE Annual Technical Conference and Exhibition in Houston, Texas, October 1999.

Tang, X. M., Dubinsky, V., Wang, T., Patterson, D. 2003. Shear Velocity Measurement in the Logging While Drilling Environment:
Modeling and Field Evaluations. Petrophysics 44 (2): 79-90.

Thomas, E. C., Stieber S. J. 1975. The Distribution of Shale in Sandstones and its effect Upon Porosities. Paper SPWLA T. presented at
the 16th Annual Logging Symposium, New Orleans, USA, June 4-7.

Williams, D.M 1990. The Acoustic Log Hydrocarbon Indicator. Paper presented at the SPWLA 31st Annual Logging Symposium, June 24-
27.