SPE 164881

Improved Petrophysical Analysis in Horizontal Wells: From Log Modeling

Through Formation Evaluation to Reducing Model Uncertainty – A Case
Study

A. Valdisturlo, M. Mele, Eni e&p. D. Maggs SPE, S. Lattuada, R. Griffiths SPE, Schlumberger.

Copyright 2013, Society of Petroleum Engineers

This paper was prepared for presentation at the EAGE Annual Conference & Exhibition incorporating SPE Europec held in London, United Kingdom, 10–13 June 2013.

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

This paper presents the results of using a new workflow to correct and validate logging-while-drilling measurements (LWD)
from horizontal wells, and the impact of the results in the petrophysical answers derived from the measurements. The
workflow involves building a layered geological model and modeling the log responses in a model-compare-update loop to
obtain the log properties of each layer. While similar methodology has been available in the past, the process was laborious
and time consuming, and therefore it was not well applied in the industry. The paper demonstrates how the new process
addresses the most common effects in horizontal wells in a timely and efficient manner, allowing it to form a part of
petrophysical analysis in high angle and horizontal wells.
In high angle and horizontal wells it is often difficult to apply the traditional petrophysical interpretation techniques normally
used in vertical wells, due to geometric effects on the data in particular the resistivity logs. These effects include local layering
or resistivity anisotropy, and boundary effects such as proximity and polarization horns on the resistivity measurements. Other
effects complicating the borehole environment include asymmetric invasion profiles, the presence of cuttings beds and drilling
mud segregation. This means that the data is challenging to interpret and the petrophysical answers from horizontal wells are
not always fully used in static reservoir models. The inclusion of the corrected petrophysical properties from this processing
into the static reservoir model reduces uncertainty and improves model accuracy.
The workflow was applied on wells in a development field in North America. The reservoir consists of a few tens of feet thick
silty sand and siltstone layers deposited in a shelfal environment. The extended reach wells used in the development of the
field have long lateral sections (from 5,000 to 10,000 ft). Due to the geological complexity of the area, the wells often cross
multiple layers and faults and are actively steered to optimize reservoir contact. The geological environment from static
reservoir model was efficiently confirmed and refined, log responses corrected and verified before being used in the
petrophysical analysis. The comparison of log responses between vertical and deviated wells was helpful both for quality
control and in the well log modeling phase to assess the correct record of petrophysical properties for the input logs and for the
modeling results. The update of the log measurements and resulting improvement in petrophysical answers is presented. The
workflow requires strong integration between the Reservoir Geology, Drilling and Petrophysics teams.
The paper presents a case study of the application of a new workflow to improve petrophysical answers from logging while
drilling measurements in high angle and horizontal wells. The study demonstrates how log modeling in high angle and
horizontal wells can be used to improve formation evaluation. The improved formation evaluation and updated geological
model reduces uncertainty and adds detail.

Introduction

The first commercial success of a horizontal well was in the Rospo Mare Field in the Adriatic Sea (Italy). In this early
application, heavy oil production from the horizontal well was twenty times that expected from a standard vertical well
producer, while the cost was only double (Bosio, 1986). Since then the oil industry has continuously enhanced the potential of
2 SPE 164881

horizontal wells and the demand for technology for supporting this activity has been for the most part satisfied. Drilling
technology, reliable geosteering and LWD tools, have allowed high angle and horizontal (HaHz) wells to become common
practice, leading to increases in recoverable hydrocarbon reserves and optimization of hydrocarbon production. The economic
development of unconventional reservoirs, such has gas and oil shales, and tight sandstones, would not be possible without the
extensive use of horizontal wells.

Horizontal wells are often defined as having a deviation from vertical exceeding 80 degrees (Passey et al., 2005). Under these
conditions several technologies have been developed or updated to allow standard petrophysical measurements to be acquired.
These include methods such as coil tubing and drill pipe assisted logging to convey conventional logging strings downhole,
and the development of LWD tools and assemblies. The latter has become the primary technology for acquiring logging data
in HaHz wells for several reasons: the real-time measurements available while drilling aid geosteering operations; the time
required for logging operations is minimized, and complex and risky drill pipe or coiled tubing operations are avoided.

However, LWD data acquisition strategy in horizontal wells is often designed primarily for geosteering purposes and can be
unsuitable for petrophysical evaluation, for example when the data acquired is limited to GR and resistivity. Even when full
LWD logging suites are available including GR, resistivity, density and neutron porosity measurements, in HaHz wells it is
often difficult to apply the traditional petrophysical interpretation techniques normally used in vertical wells. This is primarily
due to geometric effects on the data, in particular, but not limited to, the resistivity logs. These effects include bed boundary
effects and proximity to uncrossed layers on all measurements, and local layering or resistivity anisotropy, and polarization
horns on the resistivity measurements (Griffiths et al., 2012). Other effects complicating the borehole environment include
asymmetric invasion profiles, the presence of cuttings beds and drilling mud segregation (Passey et al. 2005). Today our
capability of landing and navigating HaHz wells inside a target in the desired stratigraphic position, has contributed to the
successful use of these wells in many development projects. However, our ability to interpret unambiguously the acquired logs
in order to obtain a correct petrophysical evaluation of the HaHz well data is still far behind our capability to drill the wells.
This case study will demonstrate how a new high angle well evaluation package and associated workflow was applied on two
complex horizontal wells (Well A and Well B) to obtain improved determination of the formation properties for use in a
standard petrophysical evaluation.

Geological background

The case study focuses on the application of the workflow to two horizontal wells drilled in North America. The stratigraphic
section of interest is part of a large-scale terrigenous prograding complex of shallow marine to marginal marine facies. In this
context, the reservoir facies represent stacked distal prodelta to shelfal lobe sandy sediments, overlying marine shales. The
successive environmental evolution indicates the progressive onset of coastal conditions, according to a general regressive
trend.

The reservoir basal part is characterized by the rapid transition from moderately bioturbated shaly siltstones and siltstones,
representing low-energy outer shelf sediments to thin-bedded (less than 1.0 ft.) very fine/fine silty sand laminated beds
alternating with silty mudstone and bioturbated siltstone, representing prodelta and distal shelfal lobe deposits. This basal
interval is by relative high-energy thicker (up to some feet) stacked beds of fine grained silty sandstone, with low-angle and
hummocky laminations and a basal shelly lag, corresponding to proximal shelf lobes. The topmost section is dominated by
strongly bioturbated siltstones with thin sand beds, indicating a shift into offshore transition and inner shelf environmental
conditions. In summary, the reservoir interval (some 40 ft thick), is organized in four intervals (from top to bottom):

L4 – siltstone to thin beds: outer shelf/prodelta to distal lobe transition

L3 – main shelf lobe consisting of various amalgamated sand beds
L2 – mostly fine shelf sediments, indicating a temporary deactivation of system
L1 – mostly shelf lobe facies as L3, rapidly grading at the top to inner shelf fine sediments

The main 3 lobes (L4, L3, and L1) are separated by silty layers and/or alignments of tight nodular carbonate concretions of
diagenetic origin. The stacking pattern of these sand bodies is very similar and well correlated at the field scale. Minor
thickness variation of individual cycles\beds can be linked to lobes compensation geometries. These objects are interpreted as
shallow-water, turbidite-like sandstones deposited by high-density hyperpycnal flows deriving from flow-dominated fluvial
systems (Mutti et al., 2007).

The trap is structural and stratigraphic; the structure is a gently dipping monocline interrupted by a set of high-angle (near-
vertical) normal faults, and the up-dip seal is guaranteed by the shaling out of the sand bodies (see Fig 1 below). Due to the
thinness of the reservoir interval, its lateral continuity is often interrupted by the faults, although no significant
compartmentalization is observed at the field scale.
SPE 164881 3

Fig 1: Geological section of Well B showing the target layers and faulting. LWD density image and interpreted dips are shown in the
horizontal log display at the top of the high angle formation evaluation module interface. Note the extreme vertical exaggeration – 50ft
TVDSS (true vertical depth sub-sea) vs. 6000ft THL (true horizontal length)

The development plan for the field calls for horizontal producers to maximize well performance. After landing, the wells are
required to undulate within the reservoir, targeting the best quality intervals and maximizing reservoir contact. Given the
thickness of the section and the structural complexity, this requires a proper LWD log acquisition program to suit both the
active geosteering required and the formation evaluation strategy.

In such conditions, well logs are affected by a number of artifacts caused by the complex interplay of borehole trajectory and
geologic geometries. The log response depends on the physics and volume of investigation of the measurement, the contrast of
petrophysical properties between the layers, and the incidence angle between the borehole and the layer interfaces (Griffith R.
et al., 2012). The consequence of this is that:

 shoulder effects are amplified

 the log responses are affected by the proximity of non-penetrated layers adjacent to the target intervals
 the formation resistivity measurements are affected by polarization

Workflow description

The case history starts with a field geological model and a field petrophysical model. The HaHz interpretation package
integrates these two sources of information to facilitate the following workflow (illustrated in Fig 2).

The workflow is centered on a newly developed high angle well evaluation module, which at the time of writing is in its pre-
commercial testing stage. The workflow involves building a layered geological model and modeling the log responses in a
model-compare-update loop to obtain corrected physical properties for each layer which take into account the influence of
nearby layers on the log response in the layer of interest. In this version only GR and resistivity modeling codes were
available. The paper demonstrates how the new workflow addresses the most common effects in horizontal wells in a timely
and efficient manner, allowing it to form a part of petrophysical analysis in high angle and horizontal wells.

 LWD logs are checked and compared with existing vertical wells to assess data reliability.
 Environmental corrections are not applied on LWD logs as they are routinely applied by the acquisition engineer in
the field and the industry standard corrections are not always suitable for HaHz wells.
 Sections from the geological model are used to build the formation model in the proximity of the well.
 If available, LWD image logs are used to calculate formation dip and azimuth along the trajectory.
 The formation model is verified based on the LWD logs and image log analysis, refined and updated. The formation
model required for the HaHz workflow is normally more detailed than the geological sections.
 Each layer of the formation model crossed by the well is semi automatically populated with formation properties (i.e.
4 SPE 164881

natural radioactivity, horizontal and vertical resistivity, density, etc.), based on selected well logs. In layers not
crossed by the well the properties are manually entered based on offset well response.
 If necessary, lateral property boundaries are inserted in the layers allowing for property variations to exist in a layer
which the well intersects more than once.
 Forward model simulations of the (FM) logs are computed based on the model (the layer properties and the geometric
relationship between the layers and the wellbore trajectory).
 The FM logs are then compared with the measured logs.
 The layered model and layer properties are manually refined until the agreement between simulated and acquired logs
is acceptable.
 The validated layer properties from the workflow (saved as square logs along the well path) form the basis for the
new petrophysical evaluation.
 The new petrophysical evaluation obtained with layer properties is compared with the original evaluation performed
using the raw acquired data. Substantial differences are evaluated and understood, before any validation.
 The final petrophysical evaluation and structural model are used to update the existing geological model.
 A final Hydrocarbon-In-Place (HIP) calculation update is performed.

Fig 2: HaHz workflow showing the iterative loop in which the geometry and formation property model is refined until an acceptable
match between the simulated (forward modeled) and measured logs is achieved.

Presentation of Data and Results

The horizontal sections from Well A and Well B reviewed in this paper were drilled with a 8 ¾” bit and a 6 ¾” nominal
diameter bottom hole assembly containing GR, resistivity, azimuthal density and neutron porosity LWD measurements. After
the wells were landed in the top of the reservoir, the laterals were actively geosteered in the reservoir, by comparing predictive
models created from the offset wells with the real time LWD measurements.
The cored offset well used as a reference both for the geosteering and for the quality control of the LWD logs is shown in Fig
3. The reservoir facies consist of mostly very fine to fine grained silty sand with a little clay, characterized by high matrix
density (average 2.72 g/cm3), due to the proliferation of lithics (carbonates, metamorphic & volcanic rocks), and shale clasts in
the sand. The petrophysical analysis was performed using a deterministic approach as this was considered suitable for the logs
SPE 164881 5

available. The shale volume (VSH) was defined using a density-neutron cross plot method, as the GR log is affected in some
intervals by non-shale radioactive elements. The VSH can be considered an expression of heavy, very fine to fine-grained
siltstone and a minor amount of clay. Effective and total porosity was computed using the density neutron cross plot technique
and tuned to the core total porosity.

Fig 3: Offset well (near-vertical) logs in the cored reservoir section. Core data displayed includes particle size (track5), RCA porosity
(track6) & grain density (track7), in addition to the petrophysical interpretation.

The water saturation was computed using the Indonesia equation (Poupon & Leveaux, 1971), using the electrical parameter
values derived by special core analyses. For the vertical or near-vertical wells the water saturation was calculated using the
recorded deep resistivity response. In the horizontal wells the standard interpretation was performed using the most appropriate
phase shift resistivity at 2MHz frequency. The new workflow saturation was performed with the vertical resistivity layer
property square log. As the software version used for these examples was still under development, only resistivity and GR
layer properties were available (see discussion section for further details) for the interpretation. Therefore the volume of shale
and porosity were computed using the measured density and neutron logs and combined with the resistivity layer property for
the final saturation answer.

Calcite nodule effect on interpretation

In addition to the normal bed boundary issues associated with horizontal well interpretation, the presence of tight nodular
carbonate concretions of diagenetic origin in various layers of the reservoir causes artifacts, and deserves to be mentioned. The
availability of azimuthal density images from the LWD tools run in the horizontal wells provides formation dip and
geosteering information. Quadrant density measurements around the wellbore help quantify the true formation porosity even
though it may vary around the borehole (Griffiths R, 2009), such as the case where nodules are intersected. The images also
demonstrate the impact of the carbonate concretions and help explain the impact they have on the petrophysical analysis.
Figure 4 shows an interval where lighter patches on the density image indicate that discontinuous concretions have been
intersected by the borehole. The schematic at the top of figure 4 displays an interpreted distribution of concretions consistent
with the density image and azimuthal density log data.
6 SPE 164881

Fig 4: Image and quadrant density across an interval in the high angle section of Well A where concretions were intersected. Light
color on the image corresponds to high density. The upper schematic displays an interpretation of the distribution of the concretions
around the borehole which is consistent with the image and azimuthal density data, (not to scale).

Comparison of the HaHz LWD density-neutron data with wireline density-neutron response from offset vertical wells clearly
showed that the LWD bottom quadrant density (normally associated with best formation contact) may not be the appropriate
expression of the average rock property around the wellbore in the presence of calcite nodules. As demonstrated in Figure 4
the nodules are often intersected on only one side of the borehole, so comparison of an azimuthal density to a non-azimuthal
neutron measurement introduces scatter when plotted on a density-neutron cross plot (Fig. 5, plot a). Using the maximum of
the four quadrant readings (Fig. 5, plot b) was found to provide the best correspondence with the WL response measured in the
vertical wells (Fig. 5, plot c), and this was the curve chosen for the petrophysical interpretation.

Fig 5: Density-neutron cross plots (GR color scale) demonstrating the effect of concretions on density-neutron interpretation. a) Well
A – bottom quadrant bulk density vs neutron porosity showing scatter caused by comparison of an azimuthally focused density to an
azimuthal average neutron porosity in a zone with concretions. b) Well A – using the maximum of 4 azimuthal bulk density readings
provides a cross plot showing similar characteristics to that seen in the vertical wells. c) Density-neutron data from the offset vertical
wells.
SPE 164881 7

An overview of the results from the standard saturation evaluation and that from the new workflow in Well A are shown in
Fig. 6. Due to the extended reach nature of the wells, the lateral section in well A is 11,500 ft MD, it is difficult to clearly
identify differences and the impact of the workflow. Therefore, three shorter sections have been selected for more detailed
review.

Fig 6: Petrophysical analysis comparison of the standard and new workflow in Well A

The detailed examples below show specific cases of the differences between the standard petrophysical evaluation and the
results of the new workflow. The data displayed in subsequent figures is arranged from top to bottom as:
1. Density image and formation dips. Lighter color indicates higher density.
2. Volumetric petrophysical interpretation (0 to 100%). White with dash = shale, yellow = sand, green = oil, white = water.
3. Effective water saturation (0 to 100%). Black line = standard workflow saturation, red line = HaHz workflow saturation.
4. Acquired and forward modeled resistivity (0.2 to 2000 Ohm.m)
5. Acquired resistivity and layer properties (0.2 to 2000 Ohm.m). Blue line = horizontal resistivity, red line = vertical
resistivity.
6. Acquired resistivity (0.2 to 2000 Ohm.m). Blue dots = shallow 16” phase resistivity, black line = deep 40” attenuation
resistivity.
7. Curtain Section showing the layer model and trajectory in black, colored by resistivity
8 SPE 164881

Detailed Example 1

Two effects are seen in the interval from Well A displayed in Fig. 7. On the left the well crosses three thin layers at a high
angle (highlighted by the left blue ellipse). The extended shoulder bed effects and resistivity measurements responding to the
multiple layers within their volume of investigation cause the measured resistivities to read higher than they would in a single
layer. Consequently the standard interpretation delivers a water saturation which is too low. By accounting for these shoulder
bed effects the HaHz interpretation delivers a higher and more precise water saturation in the thin shaly beds, avoiding
undeliverable hydrocarbon volume in a substantial non-reservoir section. Also in this interval the effect of the calcite cemented
nodules is clearly seen on the density image. On the middle of the interval (highlighted by the right blue ellipse) the well path
maintains an undulating sub horizontal trajectory in the L4 close to the boundary of a shalier and more conductive sub layer
below. The effect of the shaly bed in the volume of investigation of the resistivity, but not touching the wellbore, is seen to
lower the measured resistivity and increase the water saturation calculated in the L4 reservoir layer using the standard
interpretation. The HaHz workflow corrects the layer properties for these proximate layer effects, so when the true layer
resistivity is used in the interpretation the correct water saturation of the layer is determined. In this case the water saturation is
lower and consistent with the saturation observed in the L4 layer when the trajectory is further from the boundary.
This example demonstrates how, depending on the geometry and property distribution, application of the HaHz workflow to
determine the true layer properties in the petrophysical model can either increase or decrease water saturation. In both cases
the result better reflects the true saturation of the layer(s) penetrated by the wellbore.

Fig 7: Well A example displaying the effect of thin beds (highlighted by the left blue ellipse) and proximity effect (highlighted by the
right blue ellipse). In this case, correction for the thin beds results resulted in a water saturation increase, while correction for the
proximate conductive shale layer resulted in a decrease in the water saturation. In both cases petrophysical accuracy and HIP
determination is improved.
SPE 164881 9

Detailed Example 2

Two effects are also seen in the interval from Well B shown in Fig 8. The well trajectory is rather undulating but for long
sections (highlighted by green rectangles in the saturation track) remains sufficiently far from any boundaries that their effect
is negligible. The lack of nearby layers means that acquired logs reflect the true layer properties without the need for
geometrical correction. Consequently, in this condition, the saturation derived from the standard workflow is in good
agreement with the saturation derived from the HaHz workflow.

At the lowest point of the trajectory in this interval (highlighted by the red ellipse on the curtain section) the well crosses, first
down and then up, through a series of reservoir layers (L2 and L1) with different properties. At the point of maximum TVD,
the well is very close, but not contacting, the conductive shale basement (labeled Bottom Series on the curtain section). The
combined effect of extended shoulder bed effects, mixed resistivities and proximity to the conductive shale basement result in
a lower resistivity than expected in this interval. The HaHz workflow delivered a higher resistivity and consequently increased
the calculated oil in place in L1 reservoir layer. In this example the effect of the calcite cemented nodules is clearly seen on the
image.

Fig 8: Well B example showing the effect of thin stacked layers (highlighted in the red ellipse). The green rectangles in the saturation
track highlight intervals where the well is sufficiently far from any boundaries that the measurements are unaffected by nearby layers.
Consequently the geometry-corrected saturation (red line, saturation track) and uncorrected saturation (black line, saturation track)
agree.
10 SPE 164881

Detailed Example 3

A further two effects are highlighted in the interval from Well B shown in Fig 9. The interval highlighted by the right red
ellipse shows clear proximity effect. The standard petrophysical analysis delivers a water saturation which is too low as the
input geometrically uncorrected resistivity is affected by a combination of the nearby thin bed (L3) which is shaly and
conductive, and the underlying high resistivity reservoir layer (L2). In this case the density image suggests that the wellbore
just touches the L3, as seen by the lighter colored patch in the middle of the image, which corresponds to the bottom of the
hole. The resistivity response is effected by a combination of L4+L3+L2, all of which lie within the resistivity volume of
investigation. The contributions from each can only be determined from the tool-specific response equation made available
with the HaHz interpretation package. The fast resistivity forward model provides the simulated log response in these
conditions allowing comparison to the measured response. The overall result indicates that the apparent resistivity is slightly
higher than it would be recorded if only the L4 were in the resistivity measurement volume of investigation. The new water
saturation calculated with the geometry-corrected resistivity is slightly higher in comparison with the original water saturation,
because it is based on the real resistivity properties of the L1 layer. In this example the effect of the calcite cemented nodules
is clearly seen on the image. The low porosity associated with these nodules causes the computed water saturation to spike to
100% at the location of the concretion.

In the interval highlighted by the left red ellipse, the effect of extended bed boundaries and associated blurring of the resistivity
response is evident. The well trajectory exits the top of the L4 reservoir layer, and enters the Top Series shale. The well path
was steered down to renter in the L4 reservoir section. The recorded resistivity is affected simultaneously by extended bed
boundaries on exit and reentry to the L4, and adjacent bed influence. The use of the layer properties and a fine geological
model sharpens the bed boundaries and takes them and the surrounding layer resistivities into account when the resistivity fast
forward model is computed. Use of the geometry-corrected layer resistivity in the petrophysical analysis indicates water
saturation of almost 100% in the shale (as expected in this area) and reveals higher hydrocarbon saturation in the L4 layer near
the exit and entry points. The final result is improved water saturation determination in both the L4 and the Top Series shale,
which is substantially a non reservoir section.

Fig 9: Well B example showing bed boundary effects where the well drills up into the overlaying shale and then back down into the L4
reservoir (left red ellipse). Proximity effects can be seen where the trajectory gets close to the conductive L3 and resistive L4 layers.
SPE 164881 11

Discussion and Future Recommendations

The examples shown were developed using an early version of the high angle well evaluation software package. In this version
only GR and resistivity modeling codes were available, and the petrophysical comparisons were made using only the
resistivity properties of the layer model combined with the measured density and neutron porosity logs. The results clearly
demonstrate the impact of the high angle effects on the resistivity curves and how these results propagate through to the
saturation answer. The current version of the package includes density response fast forward modeling. Neutron porosity fast
modeling will be included in the near future.

Fig 10 demonstrates the impact of using all the layer properties in the petrophysical evaluation over a 1,300 ft MD section of
Well B (a similar interval to that shown in Detailed Example 2, Fig 8.). The GR, resistivity and density logs were modeled and
used to iteratively refine the geometry and layer properties. The neutron porosity logs were squared and adjusted based on
interpreter experience and local knowledge.

Fig 10: Petrophysical comparison using layer properties GR, resistivity, density and neutron porosity, all corrected for geometry
effects

The top track compares the various water saturation answers; computed directly from measured logs (black solid line),
geometry corrected resistivity with measured nuclear logs as in the previous examples (red dashed line), and the fully
geometry-corrected set of GR, resistivity, density and neutron porosity logs (solid blue line).

The modeled density image is particularly useful for confirming the geometry of the model, refining the local dip and
identifying where additional features such as thin layers need to be added. The inclusion of the density and neutron porosity
layer properties in the petrophysics improves the shale volume and porosity computation and this feeds through to the
saturation – no other parameters have been changed. The final saturation curve simplifies and further improves the evaluation:
 In the thin layer crossed twice at the bottom of the well at 3705 ft TVDSS, the saturation is better defined.
 Calcite concretions which are only touched on one side of the borehole can be identified and appropriately excluded
from the interpretation as they have negligible impact on the resistivity response (as they are discontinuous and have
higher resistivity than the surrounding formation).
 Where the tight layers cross the wellbore at high angle, such as at 9,840 ft THL, they are modeled using property
boundaries.
 The light colored triangle at 9,700 ft THL is an attempt to model the shape of a concretion which partially crosses the
well bore. The unusual feature highlights the limitation of using property boundaries for this application. A future
12 SPE 164881

development of the high angle well evaluation software will address this through the addition of local bedding
features.

One consideration for future development is the extension of fast forward modeling to include other measurements such as
acoustics, NMR, sigma and spectroscopy. Environmental corrections may also need to be revisited as existing corrections
assume homogeneous formation, which is clearly not the case for the majority of HaHz well data.

Conclusion

The data from high angle and horizontal wells is often not fully used in calculations and reservoir models as it is difficult to
interpret in an unambiguous way. Evaluation is complicated by geometric and unusual near borehole effects which influence
log responses. This means that traditional or standard petrophysics interpretation techniques are difficult to apply directly to
the high angle and horizontal well logs, without taking into consideration all of these effects.

The application of a model-compare-update iterative workflow using fast forward models of commercial log responses allows
a formation geometry and property model to be built which is consistent with the log responses actually measured in the sub-
surface. Use of the formation properties derived in this way provides the inputs for improved formation evaluation as the
influence of the geometry effects has been removed.

The examples shown demonstrate some of the typical cases observed in HaHz well evaluation such as bed boundary and
proximate bed effects. Though the module was only in field test at the time of writing and hence did not have the full suite of
fast forward models available, the application of just the propagation resistivity forward model was able to improve the
interpretation of these real examples. Successful application of the HaHz interpretation workflow requires integration between
Geological and Petrophysical teams to provide a mutually consistent picture of the reservoir. Improved geometry definition
and hydrocarbon in place calculations increase confidence in reservoir definition and reduce reservoir risk.

Acknowledgements

The authors would like to thank the management of Eni e&p and Eni US operating Co. for their support and permission to
publish this study, and in particular Paola Giaj-Via (Eni e&p) who made available the sedimentological interpretation of the
cores.

References

Bosio, J. 1986, Horizontal Wells are Now Used for Industrial Development. SPE Paper 14614

Bigelow, E. L., 1992, A New Frontier: Log Interpretation in Horizontal Wells, SPWLA 33rd Annual Logging Symposium,
Paper OO.

Griffiths R., Morris C., Ito K., Rasmus J. & Maggs D., 2012, Formation Evaluation in High Angle and horizontal wells – A
New and Practical Workflow. Paper FF presented at SPWLA 53rd Annual Logging Symposium, June 16-20, 2012

Griffiths, R., 2009, Well Placement Fundamentals. Schlumberger ed.

Mendoza, A. et Al. 2010, Quantitative formation evaluation in high angle and horizontal wells using LWD measurements:
field application of integrated log modeling workflow. SPE Paper 135732

Mutti E., Tinterri R., Muzzi Magalhaes P. and Basta G.. Deep-Water Turbidites and Their Equally Important Shallower Water
Cousins. Extended Abstract for AAPG Annual Convention, Long Beach, California, April 1-4, 2007.

Passey Q.R., Yin H., Rendero C.M., & Fitz D.E. 2005, Overview of High-Angle and Horizontal Well Formation Evaluation:
Issues, Learning and Future Directions. Paper A, SPWLA 46th Annual Logging Symposium, June 26-29, 2005

Poupon A. and Leveaux J. 1971, Evaluation of Water Saturations in Shaly Formations, Paper O, Transactions SPWLA 12th
Annual Logging Symposium, 1971.

Singer, J.M. 1992, An example of log intepretation in horizontal wells. The Log Analist. March-April, p. 85-95
Rendero C., Passey Q, & Yin H. 2005. The Conundrum of Formation Evaluation in High-Angle/Horizontal Wells:
Observations and Recommendations. SPE Paper 96898.
SPE 164881 13

Worthington, P.F. 2008. Formation Evaluation in Horizontal Wells - the Pivot Role of Anisotropy PETROPHYSICS Vol.49
no.4.

About the Authors

Antonio Valdisturlo is Senior Petrophysicist in Eni e&p HQ, Milan (Italy). He joined the company in 1988 and has worked in
integrated field and exploration projects with multi-disciplinary teams from a wide variety of locations (Europe, Middle and
Far East, North America and Australia), experiencing petrophysical and geological technical roles. He is presently focal point
for reservoir characterization techniques. Antonio holds a Degree in Geology from the University of Milan.

Maurizio Mele is Senior Petrophysicist in Eni e&p HQ, Milan (Italy). He joined the company in 1988 as well site geologist.
He has worked in integrated field and exploration projects with multi-disciplinary team from a wide variety of locations
experiencing petrophysical technical roles. Presently he is involved in exploration and development projects on Far East
activities. He is team leader of the R&D project on LWD and Horizontal wells. Maurizio holds a Degree in Geology from the
University of Parma, Italy.

Silvia Lattuada is currently in charge of Techlog Business Development, based in Milan, Italy. She joined the company in
1997 working for a data management project for AGIP Corporate DB. For several years Silvia has covered a technical position
providing software support and training to clients about Schlumberger software packages. For 5 years she had a service
delivery and management position. Silvia holds a Master Degree in Geology from the University of Milan, Italy. 

Roger Griffiths is the Petrophysics Domain Head for the Drilling & Measurements segment of Schlumberger. He joined the
company in 1987 and has worked in a wide variety of geographical locations including the Far and Middle East, Europe,
Africa and North America, in a variety of field, management, engineering and technical roles. He is a technical Advisor in
Petrophysics and Well Placement, having written 2 books and numerous technical papers on these subjects. Roger holds an
honors degree in Mechanical Engineering from the University of Melbourne, Australia.

David Maggs is currently LWD Petrophysics Product Champion for Techlog, based in Grabels, France. He joined the
company in 1988 and worked as a Wireline Field Engineer for 8 years in South America and the North Sea. He has since
worked in Management and Technical positions in a variety of geographical locations including North America, Europe, South
America, the Far and Middle East, mainly supporting LWD measurements and Well Placement Operations. David holds a
Masters Degree in Mechanical Engineering from the University of Southampton, England.