SPE-188887-MS

Investigating Low Resistivity-Low Contrast Resistivity Pay in a

Permo-Carboniferous Reservoir, Central Saudi Arabia

C. B. Ayadiuno, S. Khan, A. A. AlAbbad, and F. AlMohsen, Saudi Aramco Oil Company

Copyright 2017, Society of Petroleum Engineers

This paper was prepared for presentation at the Abu Dhabi International Petroleum Exhibition & Conference held in Abu Dhabi, UAE, 13-16 November 2017.

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
The Permo-Carboniferous reservoirs of central Saudi Arabia comprise important accumulations of natural
gas and light oils. Some of the reservoirs exhibit low resistivity and low contrast resistivity (LRLC)
phenomena. Low resistivity pay reservoirs often produce gas/oil with little or no water at very low
resistivities. Low contrast resistivity pay zones, on the other hand, produce hydrocarbon at minimum
resistivity contrast between hydrocarbon-bearing intervals and adjacent water-wet or shaley zones.
Evaluating these types of reservoirs poses a major challenge to petrophysicists and petroleum engineers
due to the difficulty in recognizing them on logs and quantifying their hydrocarbon potential when using
simple resistivity-based petrophysical models. As a result, potential pay zones can be incorrectly evaluated
or bypassed. The objectives of this study, therefore, are: 1) to understand the causes of LRLC pay in the
study reservoirs through a detailed assessment of the textural and mineralogical composition of the rock, and
2) to validate a simple porosity-based methodology for recognizing LRLC pay in both old and new wells.
To achieve the set objectives, 38 core samples and 107 thin sections were selected from two
wells exhibiting low resistivity (Well-1) and low contrast resistivity (Well-2) phenomena, for detailed
petrographic and mineralogical studies (SEM/XRD). Furthermore, a water salinity map was created to
provide general salinity trends in the area. Well data combined with results of mineralogical and petrographic
studies suggest that low resistivity and low contrast resistivity pay in the study reservoirs is the product of a
complex mix of: 1) clay mineral types and its mode of distribution, 2) thin reservoirs below resistivity tool
resolution, 3) significant grain size variation and microporosity, and 4) variable formation water salinities.
This paper highlights the causes of low resistivity and low contrast resistivity phenomena in reservoirs
in central Saudi Arabia. Furthermore, the applicability of a porosity-based methodology for recognizing
LRLC phenomena was verified with promising results.

Introduction
Low resistivity (LR) and low contrast (LC) resistivity pay reservoirs pose a major challenge to the industry
due to the difficulty in recognizing their presence and quantifying their unlocked hydrocarbon potential
when using simple resistivity-based methods. Examples of LRLC pay abound from the Gulf of Mexico
to the Niger Delta. This is because resistivity-based saturation equation may fail to clearly identify pay
2 SPE-188887-MS

when deep resistivity (Rt), is very low or when there is little or no contrast between deep resistivity (Rt) in
hydrocarbon-bearing zones and resistivity of the formation when fully saturated by water (Ro). LRLC pay
zones can occur in clastics as well as in carbonate formations and would require a different approach to be
recognized and to quantify the hydrocarbon present in such reservoirs.
There have been several efforts in the past to address this challenge in the industry. Many of the results
however, are either empirical (not portable to other geological settings outside where they were developed)
or require expensive modern logs to address the issues, thereby preventing any look-back efforts towards
developing petrophysical workflows for legacy wells. Additionally, some of the solutions are, quite frankly,
too complicated to be implemented routinely with consistent results. We note, that the most important part
of LRLC challenge workflow, is recognizing such pay from wireline logs. This is because if they can be
identified, necessary steps can be taken to fairly characterize the reservoir properties and fluids.
The difficulty in identifying LRLC pay is in part, related to the dependence of saturation equation on
resistivity. It is noteworthy that Archie's (1942) experiments produced the empirical models that gave rise to
the water saturation equation (Eqns. 1 and 2). He found that for most rocks there is an empirical relationship
between the ratio of the bulk water-saturated rock resistivity, Ro, and the resistivity of saturating water, Rw.
It is known as formation factor, F, which is also inversely related to the volume fraction porosity, ϕ. The
relationships are given as:
(1)
and
(2)
where Rw is resistivity of water saturating the pore space, a is tortuosity constant, m is cementation factor,
Cw is the conductivity of saturating water and Co is the bulk conductivity of water saturated rock. Archie's
(1942) original conclusion that formation factor is a constant has been found to fail in shaley sands due to
excess conductivity introduced by shale/clays. Combining Equations 1 and 2 above, and by replacing Ro
with Rt, it follows that Archie's empirical power law function can be written as:

(3)

where Sw = water saturation, n = saturation exponent and Rt = true resistivity of the formation when it is
partially saturated with hydrocarbon. This is arguably the industry's most popular fluid saturation equation.
A review of Equation 3, may reveal one important reason for the failure of Archie's equation in LRLC pay
reservoirs. It is heavily dependent on resistivity for calculating fluid saturation in a reservoir. However,
this important input, is inversely related to water saturation. Therefore, a decrease in resistivity will lead
to an increase in water saturation which in turn results in lower hydrocarbon saturation, given as (1 – Sw),
assuming that all other input parameters are accurate. This shortcoming of Archie's principle underscores
the critical need for a non-resistivity dependent method for recognizing LRLC pay zones from wireline logs
and quantifying their hydrocarbon potential. In this paper, we examined a non-resistivity based approach
for identifying LRLC challenge in a siliciclastic reservoir which will be discussed in section 4.

Geological setting
The early to middle Permian (Sakmarian to Kungurian) Nuayyim Formation of Central Saudi Arabia
is a complex sand-prone interval that contains important accumulations of natural gas and light oils.
Sediments are represented by a predominantly continental succession (red-beds) that accumulated within
large extensional, intra-cratonic basins (Franks and Carrigan, 2006). These sediments record deposition
primarily within aeolian dune, interdune, wadi and playa-lake settings. Hydrocarbons typically occur within
SPE-188887-MS 3

porous, commonly brick-red, structureless to cross-laminated very fine to coarse grained sandstones located
towards the top of the Nuayyim succession. Rapid changes in thickness, absence of biostratigraphic markers
and lateral changes in facies have led to poor understanding of the formation in central Saudi Arabia.

Figure 1—Depositional model of Central Arabia sandstones (Shammari et al, 2011)

Petrography
One hundred and seven thin sections were selected across the two reservoirs and analyzed using standard
petrographic techniques. Textural properties, grain/matrix components, porosity/pore types and cements/
replacements were noted with volume percentages derived from 300 point counts. A subset of samples
was analyzed under the SEM to investigate the type, volume and distribution of clay minerals. Samples
were also selected for XRD analysis (whole rock and clay fraction) to further aid mineral identification.
Lower and middle sections of the reservoir comprise predominantly clean, quartz-rich porous sandstones
with argillaceous sandstones dominating the upper intervals. The clean sandstones are very fine to coarse-
grained, typically bimodally sorted and commonly displayed cross-laminated, graded and mottled fabrics.
Grain contacts range from tangential to straight, suggesting only weak to moderate levels of compaction. The
sandstones are dominated by well-rounded to sub-angular grains of monocrystalline quartz. Polycrystalline
quartz, feldspar, chert, micas, rock fragments, clay clasts and heavy minerals occurre in subordinate
amounts. Intergranular macropores dominate the pore networks and are commonly augmented by secondary
pores after leached labile grains. Micropores are also common. The argillaceous sandstones range from very
fine to coarse grained, are typically laminated and comprise a significant volume of undifferentiated clays
(possibly a mixture of both detrital and authigenic clays). Detrital grains are dominated by monocrystalline
quartz with smaller volumes of polycrystalline quartz, feldspar, chert, sedimentary lithic grains, micas, clay
clasts/mudclasts and heavy minerals. Detrital clays fill the pores or occur along discrete laminae.
4 SPE-188887-MS

Figure 2—SEM showing clay minerals in Well-1|(top set) and Well-2 (bottom set).

In the clean sandstones, authigenic cements are dominated by dispersed clay minerals with common iron-
oxide coatings, cements and replacements and minor quartz, carbonate and sulphate cements. Kaolinite is
volumetrically the most abundant clay type and occur throughout the reservoir interval. Illitic clays are
also present but occur in smaller volumes. Kaolinite occurs in two main forms: (i) as thick, dark brown
tangential coatings and pore-fillings displaying geopetal fabrics. These clays are very fine, poorly structured
and closely associated with palaeosols (infiltrated); (ii) as a replacement of labile grains (mainly feldspars
and possibly other labile grains). This type of kaolinite typically occurs as patchy irregular lumps and
dense masses that occludes pores. SEM images show well-structured, microporous ‘books’ and ‘verms’
commonly occupying the sites of secondary pores. Illitic clays (illite, illite-smectite) formed thin continuous
to discontinuous coatings on detrital grain surfaces. The clays typically occur in minor volumes displaying
both tangential and radial habits. SEM images show thin coatings of sheet and web-like clays carpeting
most detrital grains. The clays commonly show filamentous terminations. Subordinate cements include
syntaxial quartz overgrowths and fine carbonate and sulphate nodules. The argillaceous sandstones show
abundant grain coating and pore-filling micaceous/illitic clays with only minor discernible kaolinite. Quartz
overgrowths and anhydrite nodules are locally common.

Petrophysical Rock-types
Based on electrical properties, reservoir rocks can be broadly classified into two main petrophysical rock-
types – Archie and non-Archie rocks (Figure 3a). According to Worthington (2004), Archie rocks are: 1)
single, clean homogenous rock-types (no clay minerals, no silts), 2) water wet and electrolyte is high-saline
brine (over 20k ppm), 3) have uni-modal pore system with intergranular porosity, and 4) have no electronic
SPE-188887-MS 5

conduction (no pyrite, glauconite, mica). This implies that for Archie-type rocks (A) as shown in Figure 3a,
formation factor, F or resistivity index, RI, do not change with salinity. On the other hand, non-Archie rocks
(B) do not satisfy the above conditions, therefore, F or RI changes with salinity (Figure 3a). A schematic
crossplot of Co and Cw for clean Archie sand with zero intercept, with Co=Cw/F; and non-Archie sand with
X intercept and Co=Cw/F* + X/F* to account for excess conductivity from shales, silts or other conductive
minerals in the reservoir is shown in Figure 3b.

Figure 3—Shows a) the relationship between Rw and F in Archie and non-Archie rocks, b) Schematic crossplot
of Co vs Cw showing the effect of extra conductivity, X, in non-Archie rocks (modified after Worthington, 2004).

The complex mix of clays (types and distribution), thin beds, bimodal grainsize distribution,
microporosity and wide spectrum of salinity ranges (<10,000 ppm – over 130,000 ppm) often observed in
the Permo-Carboniferous sandstones in central Arabia increases the conductivity of the rock and qualifies
it as a non-Archie rock type. Two wells that aptly illustrate low resistivity and low contrast resistivity pay
phenomena were selected for this study. Both wells have a decent suite of logs, petrophysical analysis,
conventional and special cores analyses, as well as well-test results. Well-1 is an example of low resistivity
(LR) pay with an average resistivity of 1.0 ohm-m across the pay zone. Well-2, on the other hand, is an
example of low contrast resistivity (LC) pay with an average resistivity of 10 Ohm-m across pay and non-
pay intervals (Figure 4).
6 SPE-188887-MS

Figure 4—Well logs with photomicrographs illustrating low resistivity and low contrast resistivity pay in two study wells.

Methods/Procedures
Thirty-eight core plug samples and one hundred and seven thin sections selected from two wells that
exhibit LRLC phenomena were analyzed using a standard petrological microscope. Each thin section was
impregnated with blue dye resin to highlight porosity and stained with Alizarin Red S to help differentiate
carbonate cements. Some thin sections were also stained for K-feldspar. Grain types, matrix components,
porosity/pore types and cements/replacements were recorded. Volume percentages were derived from 300
point counts where possible, otherwise percentages were visually estimated when thin section quality was
poor. A subset of samples was also analyzed under the SEM to investigate the type, volume and distribution
of clay minerals in the studied reservoirs. Samples were also selected for XRD analysis (whole rock and
clay fraction) to aid mineral identification. These would later serve as input for multi-mineral petrophysical
model to determine volume of shale, porosity and fluid saturation.
SPE-188887-MS 7

Figure 5—Whole rock and clay XRD Analysis for Well-1

Figure 6—Whole rock and clay XRD Analysis for Well-2.

8 SPE-188887-MS

Identifying low resistivity low contrast pay zones from porosity logs
Sonic, density and neutron logs are used to supplement resistivity measurements for formation evaluation.
Each of these logs is sensitive to formation porosity, but each is also sensitive to the presence of
hydrocarbons and clay (Tixier et al., 1968). In clean, completely water-saturated zones, with no fractures,
porosities calculated from all three logs are expected to be equal. Shales/clays, microporosity and fractures,
if present in the near well-bore region (i.e. within the depth of investigation of the porosity devices) influence
porosity tools readings. Studies (Tixier et al., 1968; Etnyre and Mullarkey, 1996), have shown that in the
presence of hydrocarbons, acoustic and density logs over-estimate formation porosity, while neutron logs
under-estimate porosity in a clean formation. There are many legacy wells that have only basic wireline
data, and it is suspected that a number of these wells may have potential pay zones that exhibit LRLC pay
phenomena. If this assumption is true, then in the absence of modern sophisticated logs like NMR, dielectric
and Formation Testers, a non-resistivity based approach is needed to identify LRLC pay in these wells.

Estimating formation porosity from sonic, neutron and density logs

In their paper, entitled "Low Resistivity Contrast Gas Bearing Formation Identification from Conventional
Logs in Tight Gas Sandstones", Liu et al. (2013), presented a number of non-resistivity based methods for
identifying LRLC pay in a tight reservoir rock. These methods include 1) the difference of three porosities
(B), 2) the ratio of three porosities (C), and 3) porosity overlap, for identifying gas/light oil bearing zones.
We plan to extend and verify the non-resitivity workflows in the LRLC reservoir in central Arabia. But first,
we estimate density, sonic and neutron porosities which are required as inputs for subsequent workflows
using the following relations:
Sonic porosity:
(4)
Density porosity:
(5)
Neutron porosity:
(6)
DT is the slowness measured by acoustic logging tool, DTma is the slowness of rock matrix (quartz or
limestone, dolomite, etc), DTf is the slowness of pore fluid (this is assumed to be mud due to possible
invasion), ρb is the measured formation bulk density, ρma is the density of the matrix, ρf is the density of pore
fluid (assumed to be mud) and φn is the measured formation neutron porosity. PHIS, PHID and PHIN are
computed sonic, density and neutron porosities respectively.

Three Porosities Difference Method (B)

The underlining assumption for the three relationships as suggested by Liu et al. (2013) to work, is that in
a fully water-saturated reservoir all the three porosities are equivalent in value, which is not the case when
the reservoir is gas saturated. The three porosities difference method equation is given by:
(7)
Liu et al. (2013) proposed that in a gas zone, B is higher than zero, while in a water-wet layer, it is zero.
The necessary conditions for distinguishing a gas zone from a water-wet interval are given in Table 1.
SPE-188887-MS 9

Table 1—Parameter cut-off values for distinguishing gas zone from water-wet interval as proposed by Liu et al. (2013).

B C Ø-XPLOT

Gas B is greater than 0.0 C is greater than 1.0 Øt is greater than Øw

Water B is equal to 0.0 C is equal to 1.0 Øt is equal to Øw

Three Porosities Ratio Method (C)

The three porosities ratio, C, is another parameter for identifying gas bearing formations as suggested by
Liu et al. (2013). It is formulated as the multiplication of density and acoustic porosities divided by the
square of neutron porosity:
(8)
They concluded that in a gas-bearing formation, C is higher than 1.0, while in a water zone, it is equal
to 1.0, thus, making C useful for identifying gas pay.

Total Porosity vs Water-filled Porosity Overlap Method

It is possible that when information from porosity logs is insufficient for quantitative interpretation, it may
be sufficient for qualitative identification of gas or light oil. This is most readily accomplished with an
overlay of comparably scaled porosity logs (Tixier et al., 1968). The third approach, called the porosity
overlap method, is based on Archie's (1942) equation relating resistivity to porosity as shown in Equation
2. Re-writing the equation for a water-filled rock with bulk resistivity Ro, we obtain:

(9)

Since Ro = Rt and water saturation, Sw = 1.0, in a 100% water saturated rock, then replacing Ro with Rt,
an apparent water-filled porosity, φw is given as:

(10)

It follows from Equation 10, therefore, that in a hydrocarbon saturated reservoir, true resistivity, Rt is
expected to be higher than Ro and the calculated porosity, φw, will be smaller than that of same reservoir
filled with 100% water.

Verification of the three porosity-dependent methodologies: Case studies

Porosity-based workflows for identifying gas, as suggested by Liu et al. (2013), were extended to, and tested
in light hydrocarbon-bearing LRLC reservoirs of the two study wells. The common underlining assumption
of these methods is that in a clean, fully water-saturated sandstone reservoir, all three porosities (sonic,
density and neutron) are equivalent in value, which is not the case when there is gas in the near well-bore
region. In their paper, Equations 7 and 8 were formulated and boundary conditions set for identifying gas
pay and non-gas pay zones. Additionally, they suggested crossplotting water-filled porosity against total
porosity in the same reservoir. Both porosities are expected to be equal in a water-wet reservoir and different
if the interval is hydrocarbon saturated. Given the simplicity of the earlier discussed methodologies and
their use of minimum logs, they seem attractive, as a necessary tool for identifying LRLC pay zones in both
old and new wells. The methods were verified using two light hydrocarbon-bearing LRLC wells and the
results are discussed below.
10 SPE-188887-MS

Well-1 Example. Well-1 is an example of low resistivity (LR) pay with an average resistivity of 1.0 ohm-
m across the pay zone. Wireline logs show a thick, relatively blocky succession of clean to argillaceous
sandstones. Computed B and C values based on Equations 7 and 8 satisfy necessary conditions for
hydrocarbon bearing zone as set by Liu et al. (2013) in Table 1. Additionally, crossplotting total porosity
POR_MM (shale-corrected) and water-wet porosity, POR_WT (track 7), resulted in porosity cross-over
(shaded blue) in the hydrocarbon zone (figure 7). There is no water contact in this well. From these results,
the proposed workflows were useful in identifying the presence of hydrocarbon in Well-1.

Figure 7—Well-1 is a low resistivity (LR) pay example. The B (three porosities difference) value is plotted in track
4. It is generally higher than zero. In track 5 is plotted C (three porosities ratio), which is generally more than one.
Additionally, total porosity, POR_MM (magenta) and water-filled porosity, POR_WT (blue) are plotted in track 7. All three
parameters met the boundary conditions for identifying hydrocarbon saturated zone as suggested by Liu et al. (2013).

Well-2 Example. Well-2, on the other hand, is an example of low contrast resistivity (LC) pay with a water
contact. It is predominantly a blocky, thick succession of porous, clean to locally argillaceous sandstone
with light hydrocarbons at the top of the reservoir and water in the lower part. Computed B (three porosities
difference) was found to be higher than zero in both hydrocarbon and water-bearing sections of the reservoir.
Also, we found C (three porosities ratio) to be higher than 1.0 in both water-wet and hydrocarbon zones.
These results did not satisfy the set limits and fail to clearly distinguish between hydrocarbon and water-
filled intervals. However, the porosity overlap method (track 7) clearly discriminates between pay and non-
pay zones in the reservoir (Figure 8). From these results, therefore, two out of the three methodologies failed
to clearly identify hydrocarbon zones in Well-2.
SPE-188887-MS 11

Figure 8—Well-2 is a low resistivity contrast (LC) example. The B (three porosities difference) value is plotted in track 4. It is
generally higher than zero both in hydrocarbon and water-filled zones. In track 5 is plotted C (three porosities ratio), which
is also generally more than one in both pay and non-pay zones. In contrary, however, total porosity, POR_MM (magenta) and
water-filled porosity, POR_WT (black) are plotted in track 7. It clearly identified pay zone from non-pay interval. Only one of the
three parameters met the boundary conditions for identifying hydrocarbon saturated zone as suggested by Liu et al. (2013).

It is possible that all three methodologies work in strictly gas-bearing reservoirs, but not in formations
with liquid hydrocarbons. It may also be that the cut-off values as currently suggested are fluid type-
dependent and should be changed. These hypotheses were not tested in this paper.

Observations & Results

Low resistivity in a reservoir interval results from excess conductivity in the rock (Rider and Kennedy,
2011). Conductive minerals such as pyrite and clays are generally well known for introducing excess
conductivity in the rock. Textural properties such as fine grain size and the abundance of micro-pores
(if saline water is present) can also reduce resistivity values. On the other hand, fresher connate water
may cause low resistivity contrast in rocks. The low resistivity pay formations in central Arabia can be
classified petrophysically as non-Archie rocks, due to the presence of excess conductivity according to
the Worthington classification shown earlier in Figure 3 above. Based on petrographic and mineralogical
analysis of the study reservoirs, there are a number of reasons for LRLC phenomena in the study reservoirs.
Details of the findings are discussed below.

Bed thickness
Many of the pay zones occur at millimeter scale and are simply too thin to be resolved by the electrical
logging tools (Figure 9a). Laminated intervals are deposited as distinct layers of different proportions
of sand, silt and clay. Each layer can be highly variable in thickness, however, the allogenic clay layers
are generally thinner than the framework grain layers (Darling and Sneider, 1993). Thin beds may cause
significant reduction in resistivity tool readings. These effects may occur whenever there is a large contrast
in resistivity between a resistive interval and adjacent conductive shale beds which may provide a conduit
for current flow away from the tool.
12 SPE-188887-MS

Figure 9—Well-1 core showing: a) thin bed laminae and b) bi-modal grainsize/clay distribution

Grain size and microporosity

The reservoir is characterized by bimodal grain size distribution as shown in Figure 9b below. Zones with
abundant finer grains tend to have larger surface area, which increases the amount of irreducible water in the
reservoir (Darling and Sneider, 1993). High capillary-bound water, if saline, will certainly reduce formation
resistivity, especially if the grains are bimodally sorted and laminar as observed in Well-1. This reduction
in resistivity may result in computation of high water saturation, Sw. This suggests that it is possible to
produce hydrocarbon from such a reservoir even at high water saturations with little or no water production.

Clay types and distribution

Clay minerals are considered to be one of the primary causes of low resistivity in hydrocarbon reservoirs
(Boyd et al. 1995). Their effect on resistivity readings will depend on the type, volume and distribution
within the reservoir. Clay minerals are emplaced in the reservoirs in three ways: 1) laminar shales - vertically
unresolvable by most logging tools, if sufficiently thin; 2) dispersed clays - grain coats, filling pores, and 3)
as structural clays - grains or nodules. All three modes of clay mineral distribution are observed in the study
reservoirs (Figure 10), which further complicates the challenge. Substitution of ions in the clay mineral
structures results in a negative charge on the clay mineral surface. This surface charge will attract cations
such as Na+ and K+ when the clay is dry. When the clay comes into contact with water the cations are
released increasing the water conductivity. The ability of a clay mineral to release cations is known as the
cation exchange capacity (CEC). The higher the CEC of the clay mineral the greater the effect it has on
lowering resistivity. Therefore, smectitic clay minerals which have a high CEC will have a greater impact
on lowering resistivity than the kaolin group of clays (e.g. kaolinite) which have much lower CEC (Darling
and Sneider, 1993).
SPE-188887-MS 13

Figure 10—Photomicrographs showing bimodal grain size distribution and clay minerals emplacement in both wells.

Formation water salinity

High interstitial water salinity may significantly decrease formation resistivity. Low contrast pay, on the
other hand, occurs mainly when formation water is fresher or have lower salinity. Saline water acts as
an electrolyte and its conductivity can be related to formation resistivity factor F, the cementation factor
m, porosity, and saturation exponent, n (Li and Wang, 1995), as shown earlier in Equations. 1 to 3. It
was observed that salinity values are quite different in the two wells. Well-1 has fairly high salinity value
while that of Well-2 is about 6 times less (Table 2). Water salinity in a reservoir can be affected by several
factors such as fresh water charge, diagenesis, clay minerals, mineral dissolution (fluid-rock interaction)
and pressure. Dissolution of K-feldspars for example increases the concentration of potassium and thus
affects the overall salinity of formation water (Xie et al., 2003). Water geochemistry data with minimal
contamination was collected from the study reservoirs and used for creating regional salinity map which
provided an important general salinity trend of the area (Figure 11). There is an ongoing effort now to
expand the current salinity map to cover a larger area in central Saudi Arabia.

Table 2—Summary of observations in the study wells.

Well-1 (Low Resistivity Pay) Well-2 (Low Resistivity Contrast Pay)

Thin bed/Lamination Yes Yes

Grain size bimodality Yes Yes

Clay content Approx. 9% average Approx. 6.5% average

Kaolinite (replacive) Dispersed Dispersed

kaolinite (pedogenic) Dispersed Dispersed

Illite/Smectite Dispersed Dispersed

Illite/Mica Laminated Laminated

Chlorite Structural? Structural?

Feldspar Yes Yes

Deep Resistivity, Rt Low (average 1 Ohm-m) Medium (average 10 Ohm-m)

Water salinity (NaCl) Approx. 50,000 ppm Approx. 8,000 ppm

Hydrocarbon type Light Oil Light Oil

Water contact? No (ODT) Yes (OWC)

14 SPE-188887-MS

Figure 11—A map showing the general trend of water salinity of the study area.

Furthermore, observed differences between the study wells are summarized in Table 2 below.

Verification of porosity dependent methodologies

Three porosity-based methodologies for identifying gas (Liu et al., 2013), were extended to and verified
in a light hydrocarbon-bearing LRLC reservoir. The common underlining assumption of the methods is
that the presence of hydrocarbons in the near well-bore region affects all three porosity logs. By analyzing
the effects of hydrocarbon on the porosity logs, one can distinguish pay from non-pay zones. And since
shaly clastic reservoir rocks require extensive core sampling for electrical properties, which is tedious, time
consuming, and expensive (Ruhovets and Fertl., 1982), the simplicity of the three-porosity methodologies,
and the minimum data requirement, is quite attractive as a solution for LRLC pay identification. However,
the major challenge lies in its ability to consistently distinguish pay from non-pay. All three methods clearly
identified the hydrocarbon zone in Well-1. However, only the porosity-overlap method distinguished pay
from non-pay zone in Well-2. The other two methods (B and C parameters) failed to differentiate between
hydrocarbon and water-wet zones in Well-2.

Conclusion
Based on petrographic and mineralogical analysis of the sandstone reservoirs of the study wells, we infer
the following:

• The most common clay minerals in the study wells are kaolinite and illite-smectite mixed layered.
Where clay minerals are present and spatially continuous, they affect rock electrical properties.
When micropores present within the clays are filled by trapped connate (saline) water, they can
be electrically conductive. Although kaolinite does not have high cation exchange capacity its
abundance and distribution is an important factor in this case.
SPE-188887-MS 15

• The reservoir has variable grain size distribution, ranging from clay-silt to coarse grains. The effect
of this is large surface area and microporosity, which may lead to very high irreducible water
saturation that are both clay- and capillary-bound and immovable.
• The beds are thinly laminated in an alternating sand-shale sequence, making them too thin to be
resolved by resistivity logging tools. The shale laminations provide a conductive path for current
flow away from the resistivity tool, thereby resulting in a lower resistivity measurement.
• Formation water salinity significantly affects formation resistivity. Interaction between conductive
clay minerals and interstitial brine may lead to low resistivity as seen in well_1, while fresh connate
water with minimal clay conductivity can cause low resistivity contrast as seen in Well-2.
• Conductive minerals such as pyrite and other clays such as smectite, chamosite and glauconite are
either absent or occur in only trace amounts. These minerals can therefore, be excluded as causes
of low resistivity, low contrast resistivity phenomena in the interval of interest.
• Of the three porosity-based methodologies suggested by Liu et al. (2013), we find the overlap
porosity approach to produce a more consistent result for identifying LRLC pay phenomena in
study wells. The other two methods appear inconsistent especially at lower porosities and higher
clay content.
In conclusion, we propose that the complex mix of clay minerals (type, volume, and distribution), thin
beds (below tool resolution), variable grainsize, microporosity and variable formation water salinity are the
primary causes of low resistivity and low resistivity contrast pays in the Permo-Carboniferous sandstone
reservoirs of Well-1 and Well-2. Furthermore, the porosity overlap method has shown promise, ahead of
other related methodologies tested, in identifying LRLC pay in the study wells in central Saudi Arabia.

Recommendations
Based on the observations and conclusions, and the need to improve the petrophysical analysis outcome of
legacy wells, it is recommended that:

• Detailed electrical property measurements of cores selected from LRLC reservoirs must be
conducted and utilized for petrophysical analysis. Electrical properties such as cation exchange
capacity (CEC), formation water resistivity (Rw), cementation factor (m) and saturation exponent
(n) are critical for computing accurate water saturation (Sw) values.
• Systematic and detailed field-wide petrographical and mineralogical (SEM/XRD) study of clay-
types, mode of distribution and volumes should be carried out and utilized in all petrophysical
analysis for improved lithology, porosity and saturation results.
• Representative databases of water resistivity (Rw), cementation factor (m) and saturation exponent
(n) should be created from existing Special Core Analysis and Conventional Core Analysis studies
already completed in the fields. This should be utilized for water saturation (Sw) evaluation and
for creating a comprehensive salinity map/Rw catalogue across central Saudi Arabia.
• Detailed analysis of the chemistry of formation brine and drilling fluids should be conducted.
This will allow for better understanding of the interactions that occur between these fluids and the
clay minerals and how these interactions affect log measurements in the near well-bore region.
Additionally, this will improve our decision-making on drilling mud weight and chemistry.
• More work is required to improve the promising results of the porosity-based methods for
identifying LRLC pay. This may involve applying different types of corrections to the input logs-
shale correction, deep invasion correction, setting different boundary conditions from the proposed
ones, etc.
16 SPE-188887-MS

Future work
The next phase of this study will focus on:

• Creating the necessary database of electrical properties for field-wide property maps.

• Developing and testing LRLC petrophysical workflows in both legacy and new wells utilizing
acquired electrical data.
• Improve the current porosity-based methodologies for identifying gas pay zones to make them
applicable to light hydrocarbons in both new and legacy wells.

Acknowledgements
Authors wish to thank Maher Marhoon, Abdullah Zahrani, Khalid Mhmoud, Omar Kharouf and Taha
Okasha for their support and contributions to this work. Appreciation is equally extended to the Central
Arabia Exploration team for the opportunity to collaborate.

References
1. Archie, G.E. (1942), The Electrical Resistivity Log as an Aid in Determining Some Reservoir
Characteristics. Trans. of AIME 146 (1): 54–62.
2. Boyd, A., Darling, H., Tabanou, J., Davis, B., Lyon, B., Flaum, C., Sneider, R.M., Sibbit, A. and
Singer, J. (1995), The lowdown on low-resistivity pay. Oilfield Review, 7(3), 4–18.
3. Darling, H. L., and Sneider, R. M. (1993), Productive low resistivity well logs of the offshore
Gulf of Mexico: Causes and analysis. Productive low resistivity well logs of the offshore Gulf of
Mexico: New Orleans Geological Society and the Houston Geological Society, 23–36.
4. Etnyre, L. M., and Mullarkey, J. C. (1996), Low Contrast, Low Resistivity Reservoirs - Causes,
Pitfalls, Environments of Deposition. Rocky Mountain Association of Geologists, 11–16.
5. Franks, S. G., and Carrigan, W. J. (2006), Depositional and Diagenetic Controls of Unayzah
Reservoir Quality; GTT-2006-03 Volume 1: Discussion and Interpretations
6. Li, K., and Wang, J. (1995), The effect of water salinity on electrical properties on sandstones,
First Joint Symposium of the CSPG and the Canadian Well Logging Society (CWLS): Program
and Abstracts, (225-226).
7. Liu, X., Hu, X., and Zhang, X. (2013), Low Resistivity Contrast Gas Bearing Formation
Identification from Conventional Logs in Tight Gas Sandstones. Canadian Research &
Development Center of Sciences and Cultures Vol. 5, No. 1, 2013, pp. 51–54.
8. Rider, M., and Kennedy, M. (2011), The Geological Interpretation of Well Logs. Rider-French
Consulting.
9. Ruhovets, N., and Fertl, W. H. (1982), Volumes, Types, and Distribution of Clay Minerals in
Reservoir Rocks Based on Well Logs. Society of Petroleum Engineers. DOI:10.2118/10796-MS
10. Shammeri, S., Franks, S., and Soliman, O. (2011), Depositional and Facies Controls on
Infiltrated/Inherited Clay Coatings: Unayzah Sandstones, Saudi Arabia. AAPG Annual
Convention and Exhibition, Houston.
11. Tixier, M. P., Morris, R. L., and Connell, J. G. (1968), log evaluation of Low-resistivity pay sands
In the gulf Coast. Ninth Annual SPWLA Logging Symposium, June 23-26, 1968 in New Orleans,
Louisiana.
12. Worthington, P.F. (2004, December), Petrophysical Evaluation of Shaly Sands; Gaffney, Cline &
Associates Saudi Aramco training manual.
SPE-188887-MS 17

13. Xie, X., Jiu, J. J., Li, S., and Cheng, J. (2003), Salinity variation of formation water and
diagenesis reaction in abnormal pressure environments. Science in China Series D: Earth
Sciences, 46(3), 269–284.