SPE 163010

Sanding in Oil Well Reservoir Completions

Vincent O. Nwabueze, SPE, Abiola S. Onikoyi, SPE, Felix O. Okoro, SPE, Shell Nigeria, Prof. J .A. Ajienka,
SPE, Uniport

Copyright 2012, Society of Petroleum Engineers

This paper was prepared for presentation at the Nigeria Annual International Conference and Exhibition held in Abuja, Nigeria, 68 August 2012.

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.

combination of both. It is also widely believed that

Abstract formation burial depth can be used as a consolidation
Sand production in oil wells impairs full reservoir parameter to decide whether to include sand control in
production capability, erodes sand face completions, oil completion design or not.
down-hole tubular and surface equipment. The
debilitating effects of sand production on surface This paper seeks to share the results of the review of a
production equipment are manifested in the plugging of large population of wells located and completed in
flow lines, production manifolds and separators, different Depo-Belts in the Niger Delta with a view to
leading to significant deferment in production due to helping operators streamline their decision-making
downtime of facilities for sand clean out and component process to include or not to include sand control
repair and replacement. systems in their oil wells for efficient production
performance at less deferment due to sand production
2000 oil wells in the Niger Delta area have been and lower completion and operating cost.
reviewed to understand the sanding tendencies of the
oil well completions and establish the completion
strategy and practices that have successfully reduced Introduction
sand production and its impact. It is observed that over The Niger Delta Basin is one of the largest hydrocarbon
100 Mbopd oil is locked in as a result of produced provinces in the world. Known oil and gas resources of
sand. A plethora of sand control mechanisms such as the Niger Delta rank the province as the twelfth largest
Internal Gravel Packs, External Gravel Packs, Stand- in the world. To date, 34.5 billion barrels of recoverable
Alone Screens, Premium Screens and Sand oil and 93.8 trillion cubic feet of recoverable gas have
Consolidation Chemicals have been installed to reduce been discovered. The area of the Niger Delta can be
sand production in oil wells to acceptable rates but divided into a series of Depo-belts, separated by major
several cases of failures have been observed reviewing syn-sedimentary faults zones. These Depo-belts can be
the past history of the oil wells. thought of as transient basinal areas suceeeding one
another in space and time as the delta prograded
While several operators have developed guidelines to
southward. Thus each depo-belt forms the structurally
judge when sand control is required and how to operate
and depositionally most active portion of the delta at
the oil wells safely, there are still grey areas to be
each stage of its development. In fact, they are families
explored to understand the variation of formation
of genetically and temporally related growth fault
consolidation indices from one Depo-Belt to another. trends or macrostructures. The Niger Delta spans across
The sand production performance of 2000 wells have five major depo-belts viz Northern Delta, Greater
been reviewed to examine whether the tendency for
Ughelli, Central Swamp, Coastal Swamp and Offshore
sanding can be attributed to oil well completion
Depo-belts. Each depo-belt may be defined in terms of
techniques or in-situ formation consolidation or a
the age of its paralic sequence and the age of the
2 SPE 163010
alluvial sands that cap the paralic sequence1. Generally,
the depo-belts are younger southwards and thus the
Northern Delta is the most consolidated depobelt whilst
the Offshore depo-belt is the least consolidated.
However, specific core samples show variations that do
not precisely conform to this general prognosis. An
estimated 4,493 wells have been drilled in 246 fields till
date spanning across the five major depo-belts2. A large
majority of the wells are concentrated in the Coastal
and Central Swamp depo-belts. In this paper, 2127
wells were analysed. 1728 wells were completed with
sand control while 399 wells were completed without
sand control. The focus of this study is to evaluate the
effect of completion depth/consolidation in each depo-
belt on sand production from completed wells in the
Niger Delta with the aim of providing some guidelines
on the inclusion or exclusion of sand control systems in
oil wells in the Niger Delta.
Theoretical background
Sand Production in Niger Delta
Oil accumulations in Nigeria are generally of small
areal extent, usually highly faulted and characterized by
thin oil columns often with overlying gas caps. Most
reservoirs have at least a partial water drive which,
since in many instances the sand thickness exceeds that
of the oil column, is often manifested as a bottom water
drive. The accumulations occur in Miocene sands that
to a depth of about 10,000 feet are only loosely
consolidated. The sands are mostly fine grained and
well sorted with porosities of 25 to 35 percent;
permeabilities range from 0.5 to 5 darcies5. Typically,
sand control is not installed in formations that are at
10,000ftss or deeper. Sand control is installed in
shallower completions because it is believed that they
are in transition or poorly consolidated state.
Depobelts
The Niger Delta sedimentary wedge consists of series
of discrete depocentres or depo-belts each characterized
by individual proximal distal facies trend within the
Agbada Formation. This depo-belts are formed as the
delta progrades southwards and represent important
stages in the history of delta progradation1. The
magnitude of throw and growth faults bounding belts,
resulted in the paralic sequence of the down thrown
side being younger than that of the up thrown side. The
depo-belts are successively younger from north to the
south.
The sedimentary wedge in the Niger Delta has been laid
down in five depo-belts with the oldest lying furthest
inland and the youngest located offshore. Each of the
depo-belts ranges from 30 to 60km in width. The
sedimentary sequence in the depo-belts gives rise to the
stacking patterns of marine interbedded silt sands that is
repetitive and vertical. This is usually followed by
lower to upper shore face sands of various types and
also by coastal plain deposits that are sometimes absent
due to erosion or non-deposition. Each of the stacks is
marked at the top by the next transgressive event. Knox
and Omatsola (1989) described the sedimentation
mechanism in the depo-belts as escalator regression2.
Each depo-belt contains one or more paleontologically
transgressive shale horizons, which represents
interruptions in the overall regressive sequences that are
probably related to sea level rise.
The interplay of subsidence and sediment supply rates
resulted in deposition of discrete depo-belts. As further
crustal subsidence of the basin could no longer be
accommodated, thus shifting the direction of sediment
deposition sea ward, forming a new depo-belt1. Each
depo-belt is a separate unit that corresponds to a break
in regional dip of the delta and is bounded landward by
growth faults and seaward by large counter-regional
faults or the growth fault of the next seaward belt3,1.
Characteristically, the paralic sequence in each depo-
belt differs in age from that of its counterpart and
becomes younger towards the sea.
Rock Mechanics
Rock mechanics is the theoretical and applied science
of the mechanical behaviour of rock and rock masses;
also compared to the geology, it is that branch of
mechanics concerned with the response of rock and
rock masses to the force fields of their physical
environment. Rock mechanics itself forms part of the
broader subject of geomechanics which is concerned
with the mechanical responses of all geological
materials, including soils. Rock mechanics, as applied
in mining, petroleum, and civil engineering practice, is
concerned with the application of the principles of
engineering mechanics to the design of the rock
structures generated by mining, drilling, reservoir
production, or civil construction activity4.
Rock exhibits permanent, plastic deformation when
excessive levels of loading are applied. Plastic
deformation initiates as the stress exceeds the yield
stress and progressively develops as stress further
increases. When the failure stress is reached, the rock
loses load bearing capacity, fractures develop and the
stress drops to a residual level. The stress beyond which
plastic deformation occurs is described in terms of three
strength criteria viz; the Mohr-Coulomb criterion, the
Griffith criterion and the Cap criterion; all of which
provide an adequate description of intact rock strength.
The rock mechanical parameters required for petroleum
engineering applications are acquired both direct, by
carrying out rock mechanical tests on core material, and
indirect, by applying correlations of the mechanical
properties with other core and log parameters.
Direct laboratory testing of core material provides
superior quality data and is generally recommended.
However, the number of tests and the core coverage
across relevant wells and intervals is limited. Hence,
SPE 163010 3

correlations are indispensable for generating a vertical Where E is the Young Modulus and ! is porosity
and lateral distribution of the rock mechanical
parameters. Also for sandstone,
Rock deformation is characterised by a limited number
of rock mechanical parameters also referred to as E = 0.2 * BHN (8)
mechanical index properties. The elastic and plastic
mechanical index properties are determined in standard Combining Equations (7) and (8),
rock mechanical tests such as;
Confined Compression: BHN = 315*10-0.063 !
(9)

Note that Brinell hardness number correlation is limited

!m = " (#a #r) (1)
by:
where !m is maximum shear stress #a is axial stress and 1. Usage of only macroscopic properties such as
#r is radial stress. porosity while ignoring microscopic
parameters such as grain size, grain shape,
Uniaxial Strain: cementation and surface area.
$%a = $%v = Cm# * $#a (2) 2. Data sets from fields other that the Niger
and Delta.

$#r = k# * $#a (3) Sonic Transit Time Correlation8

where Cm# is the uniaxial compressibility under stress
increase and k# is the uniaxial radial to axial stress ratio
under stress increase.
Brazilian Indirect Tension:
As a strength indicator, the sonic transit time, "t, has
been used to characterize rock strength with sufficient
accuracy. Estimates and equations of "t have been
obtained from the Depth versus "t for oil reservoirs
established for clean sands in the Niger Delta. The
equation for oil wells is given as;

#t = 2F / &Dt (4) Z = 2.56"t2 620.95"t +44563.11 (10)

where D is the sample diameter and t is the sample
In order to calculate "t values given a depth, the
thickness. equation is re-arranged as follows:
Fracture toughness:
"t = (620.95 (((620.95)2 10.24*(44563.11
Z)))/5.12 (11)
Klc = F *Cg (5)
Where Cg is a geometry factor.
NB: This equation is valid for 6865ftss and deeper.
The Brinell Hardness Number is an index property
Interpretation of Sonic Transit Time:
which is defined as:
< 90 Consolidated
BHN = F / & ' d (6) 90 110 Intermediate
> 110 Unconsolidated
where F is the load, ' is the displacement and d is the
diameter of the steel indentor (traditionally expressed in Depositional Environment
kgf/mm2). Channel Sands
These are generally uniformly sorted grains, as such the
It ranges from smaller than 2kgf/mm2 for sands are usually clean and have good reservoir
unconsolidated, extremely weak formation material to properties. The relatively large grains translate to larger
larger than 100kgf/mm2 in case of hard rock7. pore throats which is directly proportional to
permeability. On the Gamma Ray Log, the Channel
BHN versus Youngs Modulus versus Porosity sand is usually identified by its characteristic sharp
Van Kooten and Vlaardingerbroek trended sandstone erosive base, a barrel shape signature (with grain size
behaviour and related sandstone uniaxial remaining the same along the vertical profile) or Bell
compressibility to porosity. Assuming a default Shaped signature, an increase in Gamma Ray reading
Poisson ratio of 0.25, from the base to the top of the sand (fining upward
sequence; upward reduction in grain size).
E = 63*10-0.063 !
(7)
4 SPE 163010
Figure 1: Log showing Channel Sands
Upper Shoreface
This is the more land-ward part of the Shoreface sands
and as such consists of better quality sand than the
Lower Shoreface (basin ward part of the Shoreface).
The upper Shoreface sand generally has good reservoir
properties due to their being well sorted. The relatively
large grains translate to larger pore throats which is
directly proportional to permeability. On the Gamma
Ray log the Upper Shoreface sands have a Funnel
Shaped signature. This indicates a coarsening upward
sequence of the sand grains.
Figure 2: Log showing Upper Shoreface Sands
Lower Shoreface
This is the more basin-ward part of the Shoreface
Sands. It is of poorer quality because of the
intercalation of Shale within the sands. The presence of
the shale greatly affects the sorting and as a
consequence negatively affects the reservoir properties
(porosity, Permeability. e.t.c.) of the Lower Shoreface
Sands. On the Gamma Ray log the Lower Shoreface
sands also have a Funnel Shaped signature, indicating
a coarsening upward sequence of the sand grains.
However it is distinguishable from the Upper Shoreface
Sands by the presence of shale intercalations.
Figure 3: Log showing Lower Shoreface Sands
Heterolith
This is a zone of sand with thin Shale layers within.
Due to the shale intercalations, the vertical
interconnectivity of the reservoir is greatly impaired.
On the Gamma Ray Log, the heterolith has a similar
signature to the Channel sand with the characteristic
sharp erosive base, a barrel shape signature, with
shale intervals within the barrel shape or Bell Shaped
signature with shale intervals within. This signature
occurs as a result of an increase in Gamma Ray reading
from the base to the top of the sand (fining upward
sequence; upward reduction in grain size). The shale
layers within are usually as a result of repetitive
transgressive (depositing the shale) and regressive
(depositing the sands) episodes.
Figure 4: Log showing Heteroliths
Data Analysis
Sand Production Analysis
Of the 2127 SPDC wells analysed, 602 wells
representing 28% of the total wells sampled are in the
Central Swamp depobelt, 735 wells representing 35%
in the Coastal Swamp depobelt, 656 wells representing
31% in the Greater Ughelli depobelt, 111 wells
representing 5% in the Northern Delta and 23 wells
representing 1% in the Offshore depobelt (Fig 5). An
average sand production of 5 pounds of sand produced
per thousand barrels of oil over the producing days of
the conduit was used as the criterion to determine Sand
producers and Non-Sand producers. Upon classification
accordingly, 1362 wells representing 64% of the total
wells sampled were sand producers while 765 wells
representing 36% of the total wells sampled were Non-
Sand producers (Fig 6 and Fig 7). Various sand control
types such as Gravel Pack (Milled Casing Under-
SPE 163010
reamed, Internal and External), Sand Consolidation and
Sand Screen (Expandable, Stratapac and Reslink) have
been deployed in the past based on sieve-analysis,
empirical results and field observations. The
availability of sand control was used to further
categorize the sand producer and non-sand producers.
1180 conduits with 538 (the largest group) of them
situated in the Greater Ughelli depobelt are sand
producers with sand control installed (Fig 8). 182
conduits with 62 (the largest group) of them situated in
the Greater Ughelli depobelt are sand producers with no
sand control (Fig 9). On the other hand, 548 conduits
with 236 (the largest group) of them situated in the
Central Swamp depobelt are non-sand producers with
sand control (Fig 10) while 218 conduits with 100 (the
largest group) of them situated in the Central Swamp
are Non-Sand producers with no sand control (Fig 11).
Depth of Sand Control Installation Analysis
Typically, formations at depths greater than or equal to
ten thousand feet subsea (10,000ftss) are thought to be
consolidated. This hypothesis was tested with the
available data. The 2127 wells under analysis were
divided based on the following criteria; 0-8000ftss,
8000-10000ftss and >10000ftss representing Prone,
Intermediate and Consolidated respectively. 14% of the
total number of wells analysed (293) reside in the
consolidated band otherwise known as deep wells while
1834 wells are shallow (Fig 12). 126 of the 293
conduits representing 43% are sand producers while
167 conduits corresponding to 57% are non-sand
producers (Fig 13 and Fig 14). Of the 126 sand
producing conduits, 40 conduits representing 32% have
sand control while 86 conduits representing 68% do not
have sand control (Fig 15 and 16). On the other hand,
the 167 non-sand producing conduits have 26 conduits
(16%) with sand control and 141 conduits (84%) with
no sand control (Fig 17 and 18).
About 90% of the Non-sand producers have sand
control installed and they are typically shallow
(<10,000ftss) implying that the sand control was
efficient. 70% of the Non-Sand Producers without sand
control are typically deep (>10,000ftss). Over 90% of
sand producing wells with sand control installed are
shallow (<10,000ftss) which implies compliance with
the Depth Criteria (Fig 15). This implies that the sand
control devices are not efficient at excluding sand in-
situ. This borders around screen slection and sizing. In
Fig 16, 50% of sand producing conduits with No Sand
Control installed are deep (>10,000ftss).
Sanding in Greater Ughelli Depobelt
The Greater Ughelli depobelt has the largest number of
sand producers with sand control installed (46%) as
well as the largest number of sand producers (34%)
with no sand control installed (Fig 8 and Fig 9). This
made it the most suitable candidate for a more detailed
5
study. A further analysis of the depositional
environment of the Greater Ughelli depobelt was
undertaken. The analysis involved the study of the
petrophysical data logs of each completion (at the
perforated interval) in the depobelt and a categorization
of the sandface completion into Channel, Upper
Shoreface, Lower Shoreface and Heteroliths. 174
conduits corresponding to 50% of the wells analysed
had completions on channel sands, 23% were Lower
Shoreface, 19% were on Upper Shoreface and 8% on
Heteroliths (Fig 19). The Brinell Hardness Number was
evaluated for each of the reservoir sands and the
average Brinell hardness number was 6 for the
Heteroliths, 8 for the Channels, 10 for the upper
shoreface and 11 for the lower shoreface; indicating the
least consolidated to the most consolidated respectively
(Fig 20).
Low Mid High
Channel 4 6 20
Heterolith 4 5 11
Lower Shoreface 4 8 27
Upper Shoreface 2 6 27
Table 1: Brinell Hardness Number distribution across
Depositional Environments
A distribution of the Brinell hardness number with
depth showed that consolidation (depicted by increasing
BHN) increased with depth in all depositional
environments and this observation was corroborated by
the analysis of the sonic transit time versus depth
distribution (Fig 21 and 22). This behaviour was clearly
shown by the Channel sands and was also seen in other
depositional environments. The reservoirs/formations
also showed a gentle decline in sonic transit time with
increasing depth indicating an increase in consolidation
across all depositional environments. The anomalous
behaviour (high BHN, Low porosity, shallow depth)
exhibited by some shallow shoreface sands is as a result
of shale intercalations lodged between the sand, causing
diminished porosity at the evaluated shallow depths.
Conclusions
This study has clearly shown that sand production in oil
wells in the Niger Delta is influenced by the depobelt,
the depositional environment as well as the depth of
burial of the reservoir.
The coastal Swamp is the most drilled depo-belt in the
Niger Delta whereas the Greater Ughelli depo-belt is
the most prone to sand production. Most of the oil
completions are in channel sands, which is the most
dominant depositional environment in the depo-belt.
6 SPE 163010

Field observations reveal that the depth criterion, as an depositional environment is a very important
indication of consolidation, was largely violated in this consideration in the decision to install sand control.
depo-belt since most shallow completions at below However, it has been observed that depth criterion as an
10000 ftss completed with sand control devices in place
indication of consolidation can be used in specific
still produced sand and the deeper oil completions also
produced sand with sand control in place depositional environment. Field observation showed
that 40% of the deeper completions without sand
The Greater Ughelli depobelt (consisting of OML 16, control actually produced sand after initial completions
21, 26, 27, 30, 34, 41 and 42) stands out as the most because of the generalised depth criteria used.
prolific sand producing depobelt amongst the depobelts
of the Niger Delta with over 90% of its reservoirs
producing sand over the 5pptb threshold, regardless of
sand control installation and depth of formation.
Acknowledgments
Based on this study, the Central Swamp depobelt
(consisting of OML 11, 17, 22, 28 and 35) emerged the The authors wish to thank the management of SPDC
most relatively sand-free depobelt in the Niger Delta Asset Development and Petroleum Engineering
with 55% of its reservoirs producing sand below the Discipline Teams for granting the permission to publish
5pptb threshold. this paper. Special thanks to Daniel Amogu, Alphonsus
Oyibo, Olusegun Akinrolabu, Tohira Shakioye,
Upon closer analysis of the Greater Ughelli sand Olakunle Ayoola, Olalekan Otubu, Anthony Bassey,
producing reservoirs, it was observed that the Channels Asmau Nayagawa and Isimbabi Onoze.
and Shorefaces (Upper and Lower) comprise 92 % of
the population, with the Channels dominating the sand
producers in 1 out of 2 reservoirs. This implies that
sanding in oil well reservoirs in the Greater Ughelli Nomenclature
Depobelt occurred largely in Channel sands as shown ! shear stress
by the data analysed. Further work is required to test # directional stress
this observation in other depobelts to further qualify % strain
this conclusion. C compressibility
k# uniaxial radial to axial stress ratio under stress
Considering the depth criterion, it is interesting to note increase
that even though sand control equipment are installed in F load or force
92% of all shallow wells and 55% of all deep wells, D sample diameter
97% of all 1180 shallow wells with sand control still t sample thickness
produced sand. Clearly, the sand control equipments are Cg geometry factor
not performing optimally across the depo-belts. Further Klc Fracture toughness
work is required in categorizing the various sand ' displacement
control equipment and the causes of their under- d diameter of steel indentor
performance. Z Depth
"t Sonic Transit Time
The Brinell Hardness Number Correlation and the
Sonic Transit Time Equation are excellent correlators
Subscript
of the consolidation observable in the Niger Delta as
m maximum
evidenced in their prediction of the expected trend of
a axial
consolidation in the Greater Ughelli depobelt. Their r radial
usefulness was more clearly depicted when anomalous
behaviours in some reservoirs were highlighted as
outliers in the BHN and STT trends. They are however
References
limited by their over dependence on porosity and depth
1. Doust, H and Omatsola, E., (1990), Niger Delta.
data respectively which are subsets of the universal sets
AAPG Memoir v. 48, 201-238, 1989.
of parameters that influence consolidation in rock 2. Knox, G.J., and Omatsola, E.M., (1989),
strata. Development of the Cenozoic Niger Delta in
terms of the Escalator Regression model and
Finally, the depth criterion for judging whether or not impact on hydrocarbon distribution. Proceedings
formation rock is consolidated cannot be generalised. KNGMG Symposium, Coastal Lowlands 1989:
181-202 Kluwer Academic Publishers, Dordrcht.
The knowledge of the areal geology and its dominant
SPE 163010 7

3. Evamy, B. D., Haremboure, J., W/kanerling P.,

Knaap, W.A., Lolloy F.A., and Rowlands, P. H.,
1978. Hydrocarbon habitat of the Tertiary Niger
Delta. American Association Petroleum Geologist
Bulletin. V. 62, p. 1-39.
4. Department of Petroleum Resources (2003),
Business Opportunities in Nigeria and the E&P
Environment, paper presented at the International
Oil and Gas Business Days Conference held in
Oslo, Norway, 26-28 August 2003.
5. Du Mee J. A. et al, (1963), Sand Consolidation
Experiences in Nigeria, paper presented 26 June
1963.
Figure 5: Well distribution according to depobelts
6. Brady, B.H.G., Brown, E.T. (1999), Rock
Mechanics For Underground Mining, Kluwer
Academic Publishers.
7. Veeken, C. A. M., (1994), Rock Mechanics
Manual, Part1- Introduction and Basics, Shell
Research, Rijswijk.
8. Kadana, R.E.I., (1995), Sand Failure Prediction
Study, Shell Petroleum Development Company,
Nigeria.

Figure 6: Sand producers according to depobelts

Figure 7: Non-Sand producers according to depobelts

Figure 8: Sand producers with Sand Control according to depobelts

8 SPE 163010

Figure 9: Sand producers with No Sand Control according to Figure 13: Sand Producers according to depth and depobelts
depobelts

Figure 10: Non-Sand producers with Sand Control according to Figure 14: Non Sand Producers according to depth and depobelts
depobelts

Figure 11: Non-Sand producers with No Sand Control according to

depobelts Figure 15: Sand Producers with sand control according to depth and
depobelts

Figure 12: Well Distribution according to depth and depobelts

Figure 16: Sand Producers with No Sand control according to depth
and depobelts
SPE 163010 9

Figure 17: Non Sand Producers with Sand control according to depth
and depobelts Figure 20: Distribution of Average Brinell Hardness Number
according to depositional environment in Greater Ughelli.

Figure 18: Non Sand Producers with No Sand control according to

depth and depobelts

Figure 21: Distribution of Brinell Hardness Number versus Depth

according to depositional environment in Greater Ughelli.

Figure 19: Distribution of wells according to depositional

environment in Greater Ughelli.

Figure 22: Distribution of Sonic Transit Time versus Depth according

to depositional environment in Greater Ughelli.
10 SPE 163010

Figure 23: Distribution of Depobelts across the OMLs in the Niger

Delta.