SPE 130261

Fluid Contact Determination in Tight Gas Reservoirs; Formation “B” Case

Study, Sultanate of Oman
Ibrahim, Abd_Elmoula, Said Hasani, Sultan, Jahwari, SPE, PDO

Copyright 2010, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Deep Gas Conference and Exhibition held in Manama, Bahrain, 24–26 January 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
The traditional methods for fluid contacts determination in deep seated tight gas reservoirs are very difficult. This is
a result of the down hole environment toughness and the complex nature of the rock. In addition to that, most of
the tight gas reservoirs have either no, or poorly defined water table.
This paper presents a systematic methodology to determine the proper fluid contacts and to define the movable
and unmovable fluids in tight gas reservoirs. It is a comprehensive workflow which integrates core measured
Mercury Injection capillary pressure (MICP) and log data measurements.
MICP measurements have been used to detect the irreducible water saturation (Swirr), which was then integrated
with core porosity to create the Porosity-Swirr relationships. These relationships have been transformed into bound
water volume log profile (Bvwirr) using the log porosity measurements. The gas water contact (GWC) has been
located by following the irreducible water saturation (Swirr) and the calculated water saturation (Sw) log profile, and
the separation between the bound water (Bvwirr) and the total water bulk volume (Bvw). The GWC is also defined
using a synthetic resistivity log, R0 and the relative resistivity discriminator curve (RRD) derived from the simple
Archie’s equation and following the R0 and true formation resistivity log separation.
These techniques have been successfully applied on the “B” reservoir section of a major gas field to detect the true
GWC and to prove the presence of movable fluids. The results were successfully calibrated by well test and
production logging results.
As demonstrated in the field case discussed in this paper, the findings of this study have important implications for
hydrocarbon maturation assessments, field reserves evaluation and consequently the field development strategy.

Introduction
An accurate description of the fluid distributions in a tight reservoir is a key to reduce uncertainty in reserve
estimate and hydrocarbon assessment, and consequently the field development strategy. Fluid contacts
identification in tight reservoirs is extremely difficult to establish by using the traditional methods. It owes strong
influences from capillary forces. The complexity of the rock properties and rock tightness leads to drastic changes
in the fluid contacts, mostly have either no, or poorly defined water table. Additionally, the measured resistivity is
high. It has either no or a slight resistivity log separation between deep and shallow resistivity log profiles, thus
revealing the rock tightness and the rock nature.

In “B” reservoir, the water saturation derived from resistivity and porosity logs are commonly used to estimate the
productive pay zones and define the completion and stimulation intervals by using an empirical, field-specific cut-
offs. It is in the middle range of the water saturation calculation – 40 to 70 % values, often prove ambiguous, and
on occasion leadings wrong decisions and unwanted water or no gas production.
A key element of this paper is how well log data, core measurements and Mercury Injection capillary pressure
(MICP) are integrated to address this challenge. It is an attempt to identify the true fluid contacts and define the
movable and immovable fluids in the “B” tight gas reservoir as case study, by two techniques, a) examining the
Resistivity_ porosity log derived saturation and Core derived Mercury Injection capillary pressure (MICP) data and
b) Computing the log-based R0 and relative resistivity discriminator (RRD) reference curves by using the simple
Archie formula and by following the R0 and true formation resistivity (Rt) log separation.
2 SPE 130261

These two methods were applied successfully on 155 wells drilled in one of the major gas producer fields in central
Oman. It was used to detect the true GWC and to identify the presence of movable fluids. The results were
successfully corroborated by well test and production logging results.
Four well examples are presented in this paper illustrating the application of the two techniques and the
corroboration with production results.

Geology and Rock Properties

Formation “B” is a deep steated Cambro-Ordovician age clastic, dry gas; reservoir rock, approximately ~ 210 m
thick, belonging to the Mahatta Humaid Group, which overlies the Precambrian to Cambrian age, Huqf Supergroup
carbonates, and underlies the Ordovician to Silurian age Safiq Group clastics in central part of Sultanate of Oman,
(Millson et al 1996).
Formation “B” facies are dominated by sandy to muddy sabkha, interbedded with better reservoir quality Aeolian
sands in the lower - middle interval. In the upper half the sandy to muddy sabkha is interbedded with fine-grained
reservoir-quality sheet-flood sands, deposited from current and wave action
The majority of the Formation “B” contains a pronounced horizontal stratification from pervasive mm-scale sub-
horizontal fabrics, to extensive m-scale sheet sandstones that dominate reservoir layers that are themselves
laterally extensive (between multiple with km-scale spacing). Evaluation of core analysis data suggests that
reservoir quality is highly variable within and between litho types, facies associations, wells and reservoir layers.
The primary depositional fabric does appear to exert some influence on reservoir quality. Coarser grained and
‘cleaner’ sandstone litho types and facies (e.g. weakly confined alluvial) typically have better reservoir quality than
finer grained and/or argillaceous litho types, and facies (e.g. sheet flood-dominated alluvial). However, many of the
‘clean’ sandstones appear preferentially cemented, notably at bed bases, and have poor reservoir quality. Bed
thickness and/or the degree of vertical amalgamation may be a significant influence on the reservoir potential of
these ‘clean’ sandstones.
Formation “B” Porosity ranges between 3 to 10 % and Permeability varies between 0.02 to 1 mD. It is subdivided
into two major units, upper (100 m thick) and lower (100 m thick). Subsequently, each unit is subdivided into 6
reservoir subunits, figure-1.
The upper reservoir units are better quality compared with the Lower reservoir units. The lower reservoir units are
very poor quality with 3-5 % porosity and ~0.02 – 0.05 mD permeability ranges. The Lower Section is capped by a
regional mudstone marker. The overlying Upper formation “B” siltstones, some 100m thick, is unconformably on
lapped by the Mid-Late Cambrian thick Shale bed
Gamma Ray log signature is always high (110-150 GAPI) due to the presences of K40 (feldespathic sandstone
most likely lacustrine in origin), figure-1. Shale beds discriminated using Density / Neutron log separation as:
RPD = Øn – Ød / Ød (Ratio)
Where RPD = Relative Porosity Difference, ratio, (modified after Robbert Niewenhuijs, 2007); Øn = Neutron log
Reading (LPU) and ØD = Density Porosity in Fraction (based upon the limestone matrix Rhom=2.71 g/cc).
It is found that Sandstone RPD value is < 0.5, figure-2.

Procedures and Techniques

In “B” reservoir, the traditional concept for GWC identification was used. The Water saturation derived from Archie
formula is commonly used to estimate the productive pay zones and to define the completion and stimulation
intervals, using an empirical, field-specific cut-offs. It is in the middle range of the water saturation calculation 40 to
70 % values, can often prove to be ambiguous, leading to wrong decisions and unwanted water or none-
insignificant gas production. In order to address this complex problem of identifying GWC, two consistent
techniques were evolved successfully on “B” reservoir,

1- From the Resistivity_ porosity log derived saturation and core derived Mercury Injection capillary pressure
(MICP) irreducible water saturation. Core porosity and Swirr relationship was created. This relationship was
transformed into log-based synthetic reference curve for irreducible water saturation (Swirr) and bound water
volume (Bvwirr). The GWC and pay intervals is determined by following the irreducible water saturation (Swirr) and
the calculated water saturation (Sw) log profile, and the separation between the bound water (Bvwirr) and the total
water bulk volume (Bvw).
Mercury injection (MICP) pressure curves for 135 samples were measured at high pressure of almost 10,000 psi.
Data have been converted into Gas /Brine system at in-situ condition using lab measurements, table-1. The
Capillary pressure was converted into height above free water level (HAFWL). Swirr was figured out at the
maximum reservoir Height ~ 210 m for each individual sample, fugure-3. Then Swirr _Porosity empirical
relationships were derived using porosity versus core Swirr crossplot techniques, figure-4, as:

Swirr = -0.9379744* log10 (Ø) – 0.76423 (R2=0.82)

SPE 130261 3

Log-Based Swirr and Bvwirr reference curves were computed every 6 inches using the log derived porosity and the
associated Bvw for the appropriate porosity range.

The GWC and pay intervals were determined by observing the irreducible water saturation reference curve (Swirr)
and the calculated water saturation (Sw) log profile computed from simple archie formula, and the separation
between the bound water (Bvwirr) and the total water bulk volume (Bvw).
It is found that, whenever the formation is at irreducible water or produces 100 % gas, the Sw log reading is (± 5
units) equal to the Swirr reference curve. When the formation has potential water production, the Sw log reading
will be significantly greater than the Swirr reference curve, figure-5.

2- The Second method is to compute a synthetic resistivity log-based R0 and relative resistivity discriminator,
RRD, using simple Archie formula, and following the R0 and true formation resistivity, Rt log separation. The
separation between R0 and Rt (or the deep resistivity) log curve could be expressed as relative resistivity index,
RRD (ratio), which is used as index curve for, fluid contacts identification, pay footage and to define the
completion and best stimulatable intervals.
R0 = Rwa / Ø ^ m (Archie) “ohm-m”
Where
Rwa = Formation water resistivity “ohm-m” derived from Pickett plot (~ 0.025 ohm-m ± 0.002),
figure-6. Rwa” (apparent formation water Resistivity derived from Pickett plot at 100 % tested water
zones) is mostly grater than “Rw” (Formation water Resistivity) due to the effect of rock tightness on
the deep resistivity log and presences of significant amount of residual hydrocarbon volume varies
between 30 – 50 %
Ø = log porosity “fraction”, (derived from Density log);
m = Cementation Exponent (Spcail core analysis), = 1.95 ± 0.05, figure-7.
The water zones or tight sections (immovable fluids) are identified whenever the R0 = Rt. While
the movable gas section is defined whenever the Rt is significantly grater than > R0 reference curve,
figure-8.
The log separation between R0 and Rt is presented as Relative Resistivity discriminator “RRD,
ratio”. Where
RRD = (Rt – R0) / Rt
It is found that, the Movable hydrocarbon bearing section is defined whenever the RRD reference curve
greater than 40 % with a large separation between deep resistivity (Rt) log profile and R0 reference curve as seen
in the histogram figure-9.

Background and Well Examples

A complete suite of logging tools have been run in the most of the drilled wells in the study area, including
GR/Calipers/Dual Density (900 apart)/Neutron/ Resistivity/Sonic/Formation Scanner Image tool and Dual Packer
Formation tester. Dual Density tools (900 apart) were being used to minimize the porosity uncertainty and effect of
hole rugosity occurred due to the hole breakout. Dual packer formation tester is being used to acquire the
formation pressure and increasing the pressure acquirements success ratio. However, couldn’t differentiate
between gas and water using pressure gradients due to formation pressure depletion (deplete reservoir system).
Unsuccessful NMR logging has been tried in several wells, due to high down hole temperature (> 140 0C), low
signal noise ratio (Salt Saturated mud), and gas effect.

Example-1: Gas well

Density/Neutron log separation reveals sandstone matrix, couldn’t differentiate between water zone and
gas leg. The deep resistivity is reading ~ 6 ohm-m in lower section and 10 ohm-m in upper section (very
close). Slight hydrocarbon log separation was noticed between the resistivity log profiles in the upper
section; however it is doubtful, figure-10.
9 R0/Rt log separation and RRD reference curve suggested gas down to level (GDT) and water leg below
this depth.
9 The computed Sw is coincide with Swirr reference curve (± 5 %) and Bvw is equal Bvwirr above the gas
down to level proved R0 curve results, suggested gas leg. The computed Sw is grater than the Swirr in
water leg below water up to level.
9 Subsequently,
¾ The lower interval was perforated and tested formation water confirmed the results obtained
from the two techniques.
4 SPE 130261

¾ The Upper section was perforated and hydraulically fraced confirmed a presence of a
significant gas production as proved by the production log data results, as seen in, figure-10.

Example-2: Dry well

9 This well is drilled as appraisal well toward the flank of the main field structure. The Density/Neutron log
separation reveals sandstone matrix, couldn’t differentiate between water zone and gas leg. The deep
resistivity varies between ~ 6 ohm-m - 8 ohm-m in upper section, with no separation between resistivity log
profiles, figure-11.
9 No Separation between R0 and Deep Resistivity log (RRD reads < 40 %), and Sw is greater than the Swirr
reference curve for all the reservoir section, suggested dry well log signatures and water bearing section,
figure11.
9 Subsequently the upper most part of the section was perforated, hydraulically fraced and tested with
nitrogen with minor amount of formation water volume production, confirmed the evaluation techniques
results.

Example-3: Stacked Contact

9 Gas bearing sandstone bed of 4 m thick was noticed by following the R0/Rt (RRD) and Sw/Swirr
(Bvw/Bvwirr) synthetic log profiles at the middle part of Formation “B” unit-3, embedded between two shale
beds (section AA), figure-12.
9 The upper (unit-2) and lower (unit-4) of Unit-3 reveal water bearing log signature, suggested stacked
contacts.
9 Subsequently
¾ Dual Packer Formation tester with fluid analyzer was decided to confirm the evaluated results
and collect PVT samples for lab measurements. Three bottles of gas samples were collected
from the AA sandstone bed (4 m thick) of unit-3 confirmed by the fluid analyzer reveal a typical
dry gas signature, as seen in figure-12. Another two water samples were collected from unit-2
and unit-4 confirmed the interpreted data results.
¾ The well evaluation results have been confirmed by tested the log data results and production
logging results as shown in figure-12.

Example-4: Unexpected Gas-Water Contact

9 This well was drilled at the far flank close to the boundary fault
9 Gas bearing sandstone bed and unexpected shallower Gas/water contact was observed at the upper most
part of the reservoir, interpreted by following the R0/Rt, (RRD) and Sw/Swirr, (Bvw/Bvwirr) synthetic log
profiles, figure-13.
9 The upper interval was hydraulically fraced.
9 Gas and water were produced; production log was run to check the detected fluid levels, fracture
propagation and well performance.
9 Increase in temperature log and dense fluid of 1.07 g/cc and capacitance log reading were detected at the
bottom part of the perforated interval, supporting water production contributed from the bottom part of the
section confirmed the log data results. It is also indicated that the hydraulic fractures have been
propagated down into the water leg.

Conclusions

9 The determination of the fluid contacts and movable fluid in low permeability gas reservoirs are extremely
difficult using traditional methods.

9 Core measurements and Mercury Injection capillary pressure (MICP) together with open hole log data
have been successfully used to create irreducible water saturation and irreducible bulk water volume
reference curves. The log separation between log based synthetic Resistivity “R0“ created from simple
Archie formula and true formation resistivity was successfully used to detect fluid contacts and movable
hydrocarbons in tight gas reservoirs.

9 The gas zone was determined when the log derived water saturation was equal to the irreducible water
saturation reference curve (± 5 units). The water bearing section was determined when the water
saturation computed from log was greater than the irreducible water saturation reference curve. Gas was
SPE 130261 5

interpreted when the formation resistivity log profile is considerably greater than the R0 reference curve.
The separation between true formation resistivity log and R0 reference curve is expressed as Relative
Resistivity discriminator (RRD) log reference. The gas water contact (GWC) in formation “B” was located at
the point where the RRD exceeds 40 %.

Acknowledgement
The authors thank the Ministry of Oil & Gas and the management of PDO for permission to publish this work. The
authors would like to express his thanks to PDO gas and Saih Rawl asset team for the valuable input. The authors
wish to show appreciation to Karavadi, Rao and Stoltz, Roland for constructive support.

Nomenclatures
MICP = Mercury Injection Capillary pressure
Sw = Water Saturation
Swirr = Irreducible Water Saturation
Bvw = Bulk Volume of Water
Bvwirr = Bulk Volume of Bound Water
R0 = Synthetic Resistivity
RRD = Relative Resistivity Decimator
RPD = Relative Porosity Difference
Ød = Density Porosity
Øn = Neutron Porosity
Rhom = Matrix Density
HAFWL = Height above Free Water Level
Ø = Porosity
Rt = True Formation Resistivity
Rwa = Apparent Formation Water Resistivity
Rw = Formation Water Resistivity
m = Cementation Exponent
GR = Gamma Ray
NMR = Magnetic Resonance Tool

References
Scott Jacobsen1, Dale May, Jeff Grant, Jeff Little1 Schlumberger Oilfield ‘Producibility Prediction In Gas Sands Through Effective
Integration Of NMR, Resistivity, and Porosity Log Data.’ SPWLA 47 Annual Logging Symposium, June 4-7, 2006.
Richard Merkel, SI International. “Integrated Petrophysical models in tight Gas Sands, SPWLA 47th Annual Logging Symposium, June 4-7,
2006.
N. Al-Marhoon, SPE, G. Clark and S. Babajan, SPE, Petreoleum Development Oman” Integration of Surface and Subsurface Models of
Commingled Production of Different,Reservoir Fluids - Case Study: Gas Condensate Field, Oman, SPE 101290, abu Dhabi, 2006
Millson et al 1996, The Lower Palaeozoic of Oman and its context in the evolution of a Gondwanan continental margin. Journal of the
Geological Society, London, 153, 213-230
Buckley, R.C., 1997, “The sedimentary evolution of the Cambrian of east-central Oman: an outcrop study of the Haima Supergroup, Ph.D.
Thesis
Vroon Ten Hove, M. - A Sedimentological, Diagenetic and Geochemical Review of the Barik Sandstone Reservoir in the Barik, Saih Rawl,
and Saih Nihayda Fields. PDO Exploration Report 367 1996
Robbert Nieuwenhuijs, “A new methodology for modelling facies in fluvial reservoirs”, Central Oman, 2007-7003_06

Metric
1m = 3.281 ft
6 SPE 130261

Figure 1 - A typical Formation “B” log signatures Table 1 - Liquid loading gas wells in Oman Deep, TVD ~4800 m

GR/Calliper Perme Resistivity Density/Neutron/ Porosi Hg/Air System

ability Sonic/Pef ty σ 480 dyne/cm
θ 140 °
σcos(θ) 367.7

Gas/Brine System
σ 55.34 dyne/cm
θ 0 °
σcos(θ) 55.3
Conversion 0.45 Pc Hg/Air * conversion = h o/w
2.22 Pc Hg/Air / conversion = h o/w

10m Top FM “B” Upper Density

Gas 0.33 g/cm3
Brine 1.1 g/cm3
Perm Phi_n
Figure 3 - Irreducable Water Saturation at Maximum Reservoir Height

Formation “B”
GR RD
F orm ation “B ” T otal T hickness 2 00 – 2 10 m

Mercury Injection Capillary Pressure

HAFWL Vs Sw
Pef
RS Rhob Maximum Reservoir Height ~ 210 m

Cali
(m)

Top FM “B” Lower

HAFWL

Dts Phi

Dt Swirr variation at the Maximum Reservoir Height

Sw (v/v)

Top Formation C

Figure 2 – Relative Porosity Difference, RPD, Multi_well Histogram

Differentiate between sandstone and shale
Figure 4 – Core Porosity Vs Irreducible water Saturation crossplot
Formation “B”
Relative Porosity Difference, RPD Formation “B”
Multi _ well Histogram Mercury Injection Capillary Pressure
Swirr Vs Core Porosity
F Sandstone
r
e Swirr = -0.938 * log10 (Ф, Fraction) – 0.7624
0.50 R2 = 0.82
q
u
e
Shale
n
c
i
(v/v)

e
s
Swirr

RPD, Relative Porosity Difference, Fraction

Core Porosity, stress corrected (fraction)

SPE 130261 7

Figure 5 – Typical well example exhibits Gas and Water zones Figure 7 – Formation “B” cementation Exponent, m
detected from Sw/Bvw and Swirr/Bvwirr reference log separation
1000.00
Den/Neu
m = 1.95
GR/ Perm Resistivity Porosity Water Saturation (Sw)
Cali /Pef Bvw/Bvwirr Irreducible water Sat (Swirr)

Formation Resistivity, Factor

100.00

10.00

Gas Zone
1.00
Sw = Swirr ± 5 %
1 10 100
Bvw=Bvwirr
Porosity %
10m

Figure 8 – Typical well example exhibits Gas and Water zones

Bvw detected from R0/Rt log separation and RRD reference curve
Gas Down To
GR Perm Den/Neu/Pef Resistivity Rt / R0 RRD Poro SW
Cali sity

Water Up To
Sw
Swirr

Movable Hydrocarbon
Water Zone Rt > R0 and RRD > 40 %
Bvwirr Sw > Swirr
Bvw > Bvirr

Figure 6 – Multi-well Pickett plot for 100 % tested water zone,

Rwa=0.025 ohm-m ± 0.005 10m
Rt

R0
RRD Sw

Sw = 100%
Gas-Water Contact

Shoulder
effect

Water/tight Zone
Overlapped Rt and R0 and RRD < 40 %
8 SPE 130261

Figure 9 – RRD, Muti-Well Histgram, difrentiated between movble Figure 11 – Well example-2, Rt ~= R0 and Sw > Swirr
hydrocarbon and tight/water zones RRD < 0.4 (40%), Supported Dry well, confirmed by test

GR Perm Rt and Ro Sw and Swirr RRD

RRD ~ 0.4 Cut-off value

(~40 %)
Movable Hydrocarbon

Tight or water bearing Rt ~= Ro SW > Swirr

10m
Frequencies

Perforated interval: N2 assisted Bvw Swirr

minor amount of water flow
after Hydraulic frac job

Bvwirr Sw
RRD

Relative Resistivity Discriminator, RRD, Ratio

Porosity

Figure 10 – Well example-1, Gas and water leg perfectly matched

with R0/Rt and Sw/Swirr log separation

Open Hole Log Production Log Figure 12 – Well example-3, Stacked contact, Gas bearing sandstone
GR Per Rt/R0 Den/Neu Porosity Sw/Swirr Den/Temp Q gas Spinner bed of 4 m thick was noticed by following the R0/Rt (RRD) and
Cali m Pef Cap Sw/Swirr (Bvw/Bvwirr) synthetic log profiles in the middle part of
Formation “B” unit-3, embedded between two shale beds

Rt / R0 Sw Swirr RRD
pef/sonic
Den/Neu

Q _Fluid
Spinner
Perm eability
GR

Tem p/Den
M DT Fluid analyser

Bvw/Bvwirr

Production log

Rt > Ro SW ~= Swirr
Sw
Density
D ensity

25m
T e m pe ra tu re

Swirr
Rt

10m

Q Gas
Tem perature

Spinner Water sample

Q-water
R0

Gas entry RRD

Swirr
Perforated intervals

Gas Down To Increasing in temperature

Q -G as

Dry Gas signature due to water production

Water Up To
Sw
Rt

R0 Temperature and density

Tested Interval log showed Gas Production
Formation water
Rt = Ro SW > Swirr
SPE 130261 9

Figure 13 – Well example-4, unexpected Gas-water contact and

unexpected water production due to hydraulic frac propagation down
into water leg

G Rt/R0 Sw
Production log
R Swirr

Q-Water
D e n sity

Q-Gas

Gas Production
Spinner

Perforated interval
R0 Rt
Water entrance

Gas-Water contact

Swirr

Temperature

10m
Sw

Capacitance

RPD