THE CRETACEOUS LIMESTONE PRODUCING AREAS OF THE MARA A N D

MARACAIBO DISTRICTS, VENEZUELA

RESERVOIR A N D PRODUCTION ENGINEERING *
RY

M. S T E P H E N S O N **

Synopsis Introduction
An outline is given of the reservoir characteristics The four Cretaceous Limestone fields which arc
and produckion engineering of the important new situated short distances to the west and north of the
Cretaceous Limestone fields of the Mara and Mara- city of Maracaibo now contribute very largely to
caibo Districts in Western Venezuela. Venezuela’s total output of oil. Production was only
These fields are characterised by large reservoir discovered here in the Cretaceous in 1944 but
rock volumes and thick oil columns, but extremely development in the post-war years has been rapid.
IOW matrix porosities. Their productivity is due to At the end of September, 1950, just over six years
the presence of widely distributed fissure systems, after the initial discovery, in La Paz, a total of about
and many of the wells have high Potentials. Develop- 260 million barrels had been produced, and the out-
i i i m t has been rapid and considerable care has been put had reached approximately 260,000 barrels per
taltcn to produce the reservoirs soundly, conserving day. 83 wells had been drilled to depths between
their natural energy and avoiding waste. 4,500’and 13,500’, and 58 of these were producing.
TWOof the fields, La Paz and Mara, had been ap-
Résumé praised by 31 and 50 wells up to 12,000’ deep,
respectively, and one well had been completed in each
L’exposé donne un aperçu des caractéristiques de of the more recently discovered La Concepcion and
réservoir et de la mise et production des gisements Sibucara accumulations.
importants de calcaire crétacé dans les districts cle L a Paz and Mara, although separated by non-pro-
Mara et de Maracaïbo à l’ouest du Venezuela. ductive limestone, form a geological unit, a sharply
Ces champs sont caractérisés par de grandes masses folded anticline with an overall length of some 50
de roches-réservoirs et d’épaisses colonnes d‘huile, kilometers (30 miles) and a breadth of up to 8 kilo-
mais aussi par des porosités de très faible indice. meters ( 5 miles). L a Concepcion and Sibucara are
Leur productivité est due à la présence de réseaux different structures *. Smith has described how the
de fissures très étendus et un grand nombre des productivity of the Cretaceous Limestone “is essen-
pui ts présentent une production potentielle impor- tially due to fracturing and jointing, which is a
tante. result of the tectonical deformation of the area”. The
Le développement en a été rapide et on a pris reservoir rock has a normal thickness of about ~,h’,
grand soin d’exploiter les “réservoirs” de façon adé- although its vertical thickness may be very much
quate, en conservant leur énergie naturelle et en évi- greater due to steep dips and fault repetitions. At the
tant les pertes. tops of the structures it is about 4,000’ deep in La
J. E. Smith a décrit, dans un exposé joint à celui-ci, Paz, 5,000’ in Mara, 10,000’in La Concepcion and
la géologie des champs, les caractéristiques du calcaire 12,000’ in Sibucara. The very marked subsurface
et les méthodes de carottage (voir section I). relief is hardly reflected a t all in the relatively flat
surface topography, and the highest well is less than
* The geology of the fields, the characteristics of the 400’ above sza level.
limestone, and logging methods, have been described by J. E. I n a general way these fields are very similar to
Smith in a companion paper (see section I).
** Reservoir Engineer, Shell Caribbean Petroleum Com-
pany, Maracaibo, Venezuela, * See general.map of the area appended to this paper.
666 PROCEEDINGS S I I I R U \VOKLD PETIIVLLUM LUNC~XIL>S->OL1iVN il

the betterknown Persian Asmari Limestone fields. advantages of unit operation. I n Mara, however, the
The main basic differences are that here the lime- Creole Petroleum Corp., Mene Grande Oil Co., and
stone matrix has a very much lower average porosity Texas Company, as well as Shell Caribbean, have
-and permeability-than in Persia, water en- concessions, and an interesting form of unit develop-
croachment appears to be far less important, the ment has been evolved. I n view of the probable
tempo of development has been relatively more rapid, character of the reservoirs, working arrangements
and in certain aspects operating practices have been were made in order to attain fair and efficient re-
different. As regards reservoir control, conservation covery and to minimise capital investments, and the
of energy, and care to attain maxiinum-efficient re- progress of development is regularly reviewed by
covery, the differences lie mainly in matters of detail. joint engineering committees. The arrangements
The La Paz, La Concepcion and Sibucara con- cover spacing, rate of drilling, and production oper-
cessions are held by the Venezuelan Oil Concessions ating practices. Shell acts as production operator for
Ltd. (Shell), so that these fields already had the the partners.

RESERVOIR CHARACTERISTICS AND PERFORMANCE

Reservoir fluids small cuts, about 0.1"/o or less, of salt-free watrr,
Oil and Gas. The crudes obtained from these which are produced by all wells irrespective of their
reservoirs, though differing somewhat from field to position. The edgewater is very saline, containing
field, are all of good quality and relatively free from around 50,000 milligrams of solids per litre, mostly
sulphur troubles. The Sibucara-La Concepcion crudes sodium chloride. Water lost while drilling generally
are paraffinic, according to the U.S. Bureau of picks up these salts in the limestone. As an example,
Mines' Classification, and L a Paz and Mara are both water which was lost in one Mara well picked up
paraf finic intermediate, the Mara crude being the some 9000 mgllitre before being produced in an-
most naphthenic. The specific gravities of the tank other well, 3400 feet away, which till then had only
oils range from about 0.84 (37" A P I ) to 0.88 (29O shown very small salt-free cuts.
API). Viscosities are low, around 1.6 to 3.0" Engler
at 86' F.,and the production can be handled with- Initial reservoir pressures
out special difficulties. The initial reservoir pressures in L a Paz and Mara
The solution gas-oil ratios are fairly high, ranging near the original oil-water contacts are closely in line
from 550 cu.ft.lbb1. in Mara, to 1600 cu.ft.jbb1. in with what would be expected from a hydrostatic head
Sibucara. The gases are entirely free from hydrogen to the surface, i.e. around 4900-5000 pcig at 11,000'
sulphide, and are generally suitable for use as fuel. subsea. A similar generalisation may apply to the
The shrinkages of the crudes, from subsurface to other oil accumulations though the depths of the
tank conditions, are in line with the GOR'S, being water tables are not yet known. The original pres-
least in Mara, where the shrinkage is 26 %, and sures at the tops of the structures were, however,
greatest in Sibucara, about 47 %, Viscosities of the relatively high, e.g. in La Paz at 4000' S.S., the initial
crudes under subsurface conditions are very low, and pressure was 2800 psig, corresponding to a gradient
range from I centipoise in Mara, down to 0.3 centi- of some 0.7 psilfoot.
poise in La Concepcion. Undoubtedly it is the low The original pressure gradients in the oil colun~ns
viscosity which largely contributes to the prolific of the reservoirs range from 0.319 psilfoot in Mara
nature of the production. down to 0.257 psilfoot in Sibucara. It is interesting
Details of the properties of the reservoir crudes that these gradients, in any one field, apparently
and the produced oils, as measured in the regional decrease slightly with depth, because the thermal ex-
laboratories, are set out in Table I. pansion rather more than of fsets the compression.
The compositions of the original reservoir crudes Apart from this the original pressures-and fluid
within each of the two fields which have been exten- distributions-give every indication of having been
sively drilled, viz. L a Paz and Mara, appear to b- in static equilibrium.
remarkably uniform, although there are distinct dif-
ferences between the crudes of the different fields. Reservoir temperatures
Water. Water production can be classed under The temperature distribution in these f ields, shown
three heads, edgewater, so far obtained only in La in outline on Figure I , is on the fullowing general
Paz and Mara, water lost while drilling or acidicing lines; I ) a mean surface teniperature of about 80-85"
and produced from the same or other wells, and I;,2 ) a geothermic gradient in the Younger Tertiary
 M. STEPHENSON-THE MARA AND MARACAIRO D l S T R I C T S 667

TABLE I
Characteristics of reservoir crudes and tank oils

Mara La Paz La Concepcion Sibucara

Original reservoir crude

I) Saturation pressure (psig) ..... 3050 2900 4350 5730

a t 215" F a t 190" F a t 238" F at 270° F
2) Change in saturation pressure
with temperature (psi/"F) ..... 4 4 2-5 N.A.
3) Thermal expansion coefficient of
saturated * reservoir crude (per
O F ) ......................... 0.00052 0.00052 0.00069 0.0007
4) Compressibility of saturated *
reservoir crude (per psi) ...... 0.0000105 0.000012 0.000019 * 0.00002g
5) Specificgravity of saturated * re-
servoir crude ................ 0.734 0.703 0.635 0.593
6) Viscosity of saturated * reservoir
crude (centipoises) . . . . . . . . . . . 0.9 0.6 0.3 0.5
7) Gas-oil ratio (CU. ft,íbbl.) . . . . . . 550 640 I 060 1590,
8) Shrinkage factor from saturation
to tank conditions' .......... 0.74 0.72 0.62 0.53
9) Fluid analysis, mol % N, . . . . . - 0.4% 1.9% 2 %
CO, .... - - 0.7% 1.1%
CH, 39.2% 38.6% 48.0% 58.5%
C,H, ... 9.4% 8.1% 6.5% 5.0%
C,H, ... 4.8% 6.0% 5.1% 4.0%
i - C,H,, .. 1.0% 1.5% 1.0% 1.4%
n - C4H,, . . 3.1% 3.2% 3.3% 3.7%
i - C5H,, .. 1.4% 1.3% 1.2% 1.0yo
n - C5H,, . . 1.6% 2.2% 1.9% 1.5%
C6H14 .. 2.4% 3.7% 2.7% 2.3%
c, plus . 37.1% 35.0% 27.6% 19.6%
Tank Oil /

IO) Density, Sp. Gr. ............. 0.878 0.855 0.837 0.847

(.886) ** (366) ** (-842) * *
"A.P.I. ............. 30" 34" 38" 36"
(28') ** (32") * * (37") **
Viscosity Engler a t 30" C .....
II)

12)
a t 5 0 ° C .....
Distillation I.B.P. (typical ap-
3.0'
2.0"
2.0"
1.5" 1.3" I 1.4"
at 40°C

. prox.) ...................... 78" C 66" C 60° C 65" C

10% at 10% at 10% a t 10% a t
168" C 162" C 135" c 135" c
36% a t 41% a t 48% at 48% at
300" C 300" C 300" C 300" C
13) Sulphur yo . . . . . . . . . . . . . . . . . . 2% 1.4% 0.8% 0.8%

* temperatures given under I). ** Wellhead samples.

and Eocene shales and sands of about 18O F per dient of about 7O F per 1000' would result in a uni-
1000', and 3) in the Cretaceous limestone markedly form subsurface density.
reduced gradients, of the order of IO to 1 6 O F~IOOO'. The details of the present temperature distribution
The mean temperature gradient in the limestone are complicated, as a consequence of production from
appears to be least where the overall permeability is various depths and fissure systems.
greatest, possibly due to freer circulation of the
reservoir fluid. It may be calculated as an example Reservoir rock and distribution of oil
from the properties of the reservoir fluid in La Paz Although on the whole these fields show good pres-
that, if its composition is uniform, a temperature gra- sure communication over coqsiderable parts of each
668 PROCEEDINGS T H I R D WORLD PETROLEUM CONGRESS-SECTION 11

300'F-

S i BUC I , RA
O

250.F -

CONCEP<:iON
Ø /O
p'
a./

Eocene
6, GLASARE 'NELLS 4

o 1
 M. STEPHENSON-THE MARA A N D MARACAIBO D I S T R I C T S 669
structure, and the limestone as a whole is considered permeability and of 'the amount of oil in place, have
to form (from the pr oduction point of view) one been difficult, particularly because of the irregular
Icontinuous oil bearing zone-as is usual in fissured distribution of the important secondary voids-which
i,limestones,
. . . important . . Permeability gradations and are largely of a fissure nature- and the very low
barriers, both vertical and horizontal, have also been overall porosity and permeability of the matrix. The
found to exist. In Mara 3 main reservoirs are problems are being tackled both by detailed
recognised (see map), each separated by intervening examination of cores, particularly diamond cores, ancl
areas of almost completely tight formation. These by studies of reservoir behaviour. One outstanding
main reservoirs can be still further subdivided. In- Plroblem, to which only very approx imate answers
deed, a few wells have tapped permeable zones which h ave been given, is the determinatic)n of the pro-
appear so limited as to have inappreciable connection P ortion of the oil in place that is loca ted in voids of
with any other well. In La Paz no clear barriers have .___
low L:l:L-.
~CIIIIC~UIIIL
.
..1. - .. _ _
wriere
~,
... .-
gas segregation
L.
is negligible,
1leen recognised, and the overall permeability change' S and from which the oil could only be recovered at a
ippear to be more graduai, though the effects of thes e practical rate by solution gasdrive depletion.
Iiermeability gradations are important. The geological report, by J. E. Smith (see section
Thc first 5 wells drilled to appraise the CretaL- I ) has already described the fissure systems in some
LLUUJ l:--ctn*- :* T D,, :,,,,,t:,,t-A
I I I I I L J L U I I L 111 La I aL i i i v L a L i 5 a L L - U
t L n nnFF:h:l:t
~VJJIUIIILY
LIIL detail, and indicated their distribution.
that there might have been distinctive productive As regards the core investigations, somewhat over
horizons. These wells were completed over relatively 5000 feet of diamond cores of the limestone, and
Iiarrow intervals at various stratigraphic depth:Y . several hundreds of feet of ordinary cores, have been
1Each found the same crude, and the original reser taken and analysed. Visual analysis shows that only
7ioir pressures, when reduced to a datum level, wer e very limited volumes of the matrix are obviously oil
1 ,'
iaenricai. 1 - 7 . -,
i n i s eviaence nas Deen DacKea up ,DY all
4 , , 1 1
stained, and the stained portions are generally only
:jubsequent data in La Paz, and also independently i n associated with fractures. On the other hand, labora-
iMara. tory results point to the presence of considerable
The problems of determining the net permeabilitY amounts of oil in the matrix samples, though analysis
1,, eIrecrive
ana rile cc > ' . porosity 01r me
.1
oil-Dearing 1 ' > ,, reservoir is hampered by the practical dif ficulties of measuring
rocks, and investigating the distribution of the small quantities, and by the occurrence of solid

TABLE II
Analyses o f matrix samples from cores and ditch cuttings

4-I yo Porosity Fluid saturation of Cores

Field Formation

I- Cores
I
I
Ditch
Cuttings

-
La Paz Colon 3.4 (24) 1
La Luna 3.3 (99) ~
-
Upper Cogollo 2.1 (48) 6.0 (15)
Middle Cogollo 3.6 (69) 4.3 (18)
Lower Cogollo

I Overall Average I 3.0 (273) 1 5.0 (35)

Mara Colon 2.8 (29) -

La Luna 1.4 (354) 3.2 (33)
'
Upper Cogollo 1.4 (351) 3.1 (32)
Middle Cogollo 2.5 (500) 2.1 (62)
Lower Cogollo 1.2 (49) 1.4 ( 8 )

N.R. (I) PERMEABILITY - almost all samples show less (2) ( ) = number of sample; analysed.
thail 0.1 mD. (3) - = not measured.
 670 PROCEEDINGS T H I R D \?rORLD PETROI.EUM CONGRESS-SECTION 11

Fig. 2. Limestone staining tests. Above: original samples of limestone matrix from Mara with no visible oilstain,
showing sawed and broken faces. Below: similar samples, but after having been evacuated, covered with crude oil
and subjected to reservoir pressures for a few hours, then sawed and broken, showing heavily oilstained interiors.

hydrocarbons, which occupy up to 35 % of the gross Attempts have been made to estimate the fissure
volume of some samples. An interesting experiment porosity by direct measurement of the sizes of the
was carried out on the staining of the matrix. Mara openings found in cores. The results are subject to
core samples of the limestone, as obtained and with- a large personal factor, but total figures of say % %
out visual oil stain, were placed in a cell which was to I % have been quoted for all fissures, including
evacuated for 2 hours. Crude oil was then admitted, haircracks.
and compressed to reservoir pressures for a few The reservoir behaviour studies have given
hours. The samples were removed, partly sawed, and estimates of the effective oilbearing porosities from
then broken. The oil had penetrated them com- material balance calculations, and of effective
pletely, and stained them throughout. Fig. 2 shows permeabilities from studies of well behaviour-espec-
photographs of original samples, and of samples ially from the rates of pressure build-up after wells
treated as described. are shut in. The material balances, equating sub-
The average gross porosity of the matrix, as shown surface expansions etc. to the production, are
in Table II, is of the general order of z % in Mara modified from the standard forms to take account of
and 3 "/o in L a Paz, but the water content may aver- the thick oil columns, the widely differing degrees to
. age over 5 0 %. The average permeability of the which gas liberation has progressed, and the gas
samples is very small. Indeed, only exceptional pieces segregation. Reasonably consistent results may be
of the matrix show permeabilities over 0.1mD. Direct obtained a ) from studies of the initial productions
laboratory measurements on representative long per psi pressure drop, b) from the inanner in which
cores, including fissures, have been planned for the the production per psi changes with cumulative pro-
'future. duction, and c) from the total productions obtained
M. STEPHENSON-THE MARA n N D M A R A C A I R O DTSTRTCTC 671
672 PROCEEDINGS THIRD WORLD PETROLEUM CONGRESS-SECTION ii

between various pressure stages. These results range gascap in the limestone, have been considered rather
from say 5 % to 2 % net oil porosity, in the Mara anomalous.
reservoirs and in La Paz. The original oillwater contact in L a Paz and Mara
The rates of build-up of bottom-hole pressure, is considered to have been about 11,000’ subsea. I t
after shutting in producing wells, vary widely from is difficult to establish its position exactly in either
well to well, and these build-ups may be analysed on field, largely owing to the heterogeneity of the lime-
analagous lines to those used in studies of flow be- stone and the very low porosity and permeability of
haviour in homogeneous reservoirs. T h e results the matrix. T o date it has only been possible to
permit the calculation of overall quasi-permeabilities, complete one satisfactory oiljwater observation well
and also the appraisal of the efficiencies of well with good connection both to oil and water zones.
completion methods. Figure 3 shows typical build-up This well is at the southern end of La Paz.
curves, and indicates the methods of analysis. The
quasi-permeabilities cover an extremely wide range, Development summary
from prac?ically impermeable up to permeabilities too The numbers, depths and productions of the lime-
high to be measured. stone wells in these fields, at the end of September,
I t is now considered that the greater part of the 1950, are summarized as given in the table below.
oil originally in place was contained in the finer void The first three La Paz Cretaceous wells were
spaces rather than in the very coarse fissures. The drilled some 400 to 600 metres apart, on the top
effective matrix porosity may all be closely associated
with the fissures and the finer cracks.
of the structure. Outsteps were then located at much
wider intervals, the steps normally being of the order
,
I t is of importance in this connection to note that of I to 2 kilometers. When the character of the
the apparently impermeable saddle between La Paz reservoir was more clear, it was decided to drill the
and Mara, though extending far above the water wells which were primarily intended as producers to
table, forms a barrier between crudes of markedly meet the top of the limestone about 9000’ subsea,
different qualities. locating them with a view to maximum productive life
and adequate drainage of the oil in place. Other wells
are still being drilled to investigate the downflank
Original gascaps and edgewater
conditions, and the edgewater, in more detail. The
I n La Paz there appears to have been a small present overall spacing in the proven area per com-
origina’l ‘gascap, which was tapped by one of the pleted well is roughly 800 acres, or 1200 acres per
early appraisal wells, P-73. However, due to the producer.
thickness of the oil column, which was at least 7000’, In Mara, the spacing is rather closer.
the bulk of the oil was undersaturated. I n Sibucara
also the upper part of the limestone was gasbearing. Weil production performances
There was no original gascap in Mara, and it is un- The production performances of the wells differ
likely that there could have been one in La Concep- very widely, but they may be grouped as follows:
cion. The presence in L a Paz of active oil seeps along Firstly, there are the completely tight and unpro-
the cresta1 fault (though probably connected with ductive wells, to which reference has already been
the shallower Eocene and Guasare reservoirs) and a made. Secondly, there are the poor producers, having

Field La Paz

Wells completed ............... 31 50 I I

Wells drilling ................. 6 5 - I
Wells producing .............. 20 35 I I
Observation wells for gascap and
edgewater ................... 5 I - -
Average depth of producers. ..... 8,800’ 9,000’ 12,080’ 13,450‘
Average daily production per pro-
ducing well in barrels. . . . . . . 8,100 2,400 1,500 9,600
Average cumulative production per
completed well in barrels, to
Sept. 1950 .................. 5,800,000 I ,600,000 1,180,000 4,800,000
 M. STEPHENSON-THE MABA A MARACAIBO DISTRICTS
~ D 673
no obvious pressure communication to any other well, used for observation purposes, especially in L a Paz.
and with low initial productions, say of the order of These wells include producers to which the gascaps
300 bblslday. They may flow, or be gaslifted or have extended and which have not been worked over
pumped. Their performance resembles that of wells to exclude the gas, outsteps drilled into the edgewater,
in depletion-type low-permeability sand fields. and some wells whose potential is very low due to
Thirdly come the wells with good initial productions, the tightness of the formation. R'Ieasurements of gas-,
say 5000 bblslday, but which apparently tap very oil-, and water-zone pressures, and of the levels of
limited sub-reservoirs and have no apparent con- gasloil and oillwater interfaces are made in the ob-
nection with any other wells. They flow strongly, at servation wells, usually monthly, but often fort-
least initially, and in their performance they resemble nightly also. Tjhe degree of reliability of the wells
the next group, except that their productive lives may for observation purposes, especially for observing
be short, with heavy declines in pressures and oil fluid interfaces, has to be established, by investigating
production, and early increases in gas-oil-ratio. the location of zones of fluid entry. This is partic-
Finally, there are the wells which clearly tap the main ularly important in these fields, where it may only be
reservoirs. They all flow naturally, but vary con- possible to follow say, the gasjoil contact satisfactorily
siderably in capacity, from say 1,000 to over 20,000 in any one well for a few months, while the contact
bblslday, and their degree of communication varies lies between two or more good permeable zones.
from nearly perfect to poor, so that the equality of Shut-in subsurface pressure measurements are
bottom hole pressure over the reservoirs is in some also taken, as needed, in producers. At present these
areas relatively independent of the distribution of wells are closed-in on the average about half-yearly,
offtake, and in other areas is markedly dependent on in rotation. As has been mentioned, and shown on
this distribution. The initial pressures and pro- Figure 3, the rate of pressure build-up from the
ductions of these wells depend very much on the flowing pressures can be analysed to give useful
amount of drainage which has already taken place information on the overall permeability of the lime-
from other wells. The completion of additional wells stone. The analysis is frequently used for extra-
in this group serves to accelerate the production, but polating to fully built-up pressures.
not necessarily to increase the field ultimate, except Pressure, temperature, and rate-of-fluid-flow
insofar as affected by structural relief, etc. surveys are made in producing wells, in order to
Of the wells drilled to date in Mara, 16 % are locate zones of fluid entry for workovers, future
non-producers, 14 % are poor producers in the estimates, and observation purposes, e:c., to
second category, some 4 "/o are classed as in- determine the relations between pressure drawdownc
dependent, though initially good wells, and the re- and rates of production, and to investigate optimum
maining 66 % tap the main reservoirs. In L a Paz the flow conditions and the usefulness of artificial lift
respective percentages are 6 70, IO %, zero and -especially gaslif t. Figure 5 shows typical relations
84 %-reflecting the greater overall permeability. between bottom hole flowing pressures and rates of
Of the final category in the two fields, about half production.
to two thirds may be designated a s prolific, and it is
the performance of these prolific wells which is most Reservoir performance
important from the production viewpoint. The performance of La Paz can be taken as more
Figure 4 shows in outline the performances of 3 or less typical f o r these reservoirs in general. The
typical wells, from the main group. One was a salient features of its behaviour are outlined in the
prolific producer, characterised by rapid pressure following, and the trends of pressures, gas-oil con-
build-up on shutting-in, which has now gone com- tact levels, and gas-oil ratios are shown on Figure 6,
pletely to gas due to the growth of a gascap. The together with t,he production rates.
second was a moderate producer with very slow pres- I-The rate of field offtake has been raised to
s h e build-up, which has almost gone to gas. The 163,000bblslday, in some IO stages, the more clearly
third was another prolific producer, which after pro- to observe the reservoir behaviour. These stages of
ducing for just over 2 years, had to be worked over steady production have lasted for times ranging from
to exclude gas. about 4 to 12 months.
2-The mean cumulative gas-oil ratio to October
Reservoir observation 1950 is no more than 5 % above the original solution
I n order to observe, control and predict the be- ratio, and although this year the excess is slightly
haviour of these fields, frequent subsurface surveys higher, it may for practical purposes be said that pro-
are made both in producing and shut-in wells. Wells duction is around solution ratio.
which are currently unsuitable for production are 3-Water production has remained negligibly
Proceedings 3rd W.P.C., Section II 43
Fig. 4. Examples of individual well erf for mar ice curves.
 M. STEPHENSON-THE MARA ANI> MARACAIBO DISTRICTS
675

LJUU

<
4
n
-I
3
e
W
I-
z
z
W
n
O
'8 2000
n
O
I-

.;
Y
O
c
U
Li
U
C
5
I
W
a
7-
RATI?

2 1500
y>
W
U
n
o
;
3
U
W
u
Y
U
3
v)
m
2
<o
1000

io000 i5000
iTE OF O I L PRODUCTION, BARRELS PER DAY
l-
U
7
4
t
o
c>
-1
Y
Lo
d
o
y
O.

r
I-
n
W
O

10
w
a
3
(o
w
l
i
a
0
7
w
(o

15000C

>.
cl
'y:
m
m
5 100000
l-
Y

a
Z
0
U
3
$ 50000
n

0 S.&W.CU
i%

O
CUMULATIVE OIL PRODUCTION
Fig. 6. La Paz reservoir performance curve in outline,
 STEPHENSON-THE MARA A N D MARACAIBO DISTRICTS 677
he oil-water contact ob- an interval of 130 feet, and no significant trend is
edgewater has been yet apparent. The variations are apparently due to
local rearrangements. However, other outstep wells
: prolific wells, with the .are now drilling to help clarify the picture.
South sector, follow a I t is apparent that production to date has been ob-
,bows somewhat higher tained by liquid expansion, and solution gas drive,
)wing to imperfect com- with gravity segregation of the liberated gas, and the
)ment. Pressures in the formation of secondary gascaps. If there has been
erred to as categories I any net rrdvance of edgewater it is clear that its effect
g. The general trend in on pressures has to date been very limited.
e by far the most im-
standpoint) has been a Reservoir control
essure decline. The problem of determining the optimum pro-
e most permeable zones duction rate for each well, or indeed of criteria by
has been dropping a t a which to judge the optimum, had to be tackled at an
i relatively regular, al- early date. There was little doubt that in the early
s much heavier. Slightly stages h e r e should be reasonable conservation of
rded in the tighter wells. reservoir energy, and consequently that the overall
shut down the field to gas oil ratios should be kept to a practical minimum.
Id for 7 days. As was to The first criterion which suggested itself, as in the
in the major fissure Middle East, was to produce with bottom hole
shut down, this being a flowing pressures above the saturation pressures. It
fluids from the tighter was soon apparent that this would limit offtake to
lagged. In the gasloil an unworkable extent, and trials were carried out to
8, which had been shut find the effect of producing with heavier drawdowns.
oil pressures increased The results were gratifying-most of the wells could
,i over IO days, reaching be flowed with bottom hole pressures far below the
ifter production recom- bubble points without causing any increase in the
was hardly affected. gas-oil ratios. Indeed, rather the converse happened,
ming production, at an f o r many of the prolific wells showed decreases in
in before, the pressures GOR‘S,with time, corresponding a t least to the losses
ue almost exactly their of gas from the reservoir fluid to the gas caps. In
e first direct indication due course, as the gas caps have grown, some of these
ow of oil and gas from trends have reversed. Some of the other wells . _

D f the limestone. -especially those characterised by low productivity

the main fissure system indices-showed the usual features of local depletion,
pping with rather sur- with increasing GOR’S. One of the examples
ch are completed in less appended (Figure 4), i.e. DM-I, shows how a pro-
eratic results. One well, duction of just over 6ooo hblslday was satisfactorily
:t above the main lpasloil maintained by beaning up, with a steadily declining
contact, and which had contained gas only f o r m a n y . GOR, almost until the gas-oil contact actually reached
months, suddenly in April I950 showed an oil column the top productive interval, even though throughout
400 feet deep, which gradually disappeared over the this time the bottom-hole drawdown was of the order
next two months. I t is obvious that though the overall of 550 psi, and the flowing pressure at the depth
picture may be simple, the details of the movements of the productive zone was about the same amount
of oil and gas are very complex. below the corresponding saturation pressure. As an:
7-As regards :he oiljwater contact, owing to the other example, it has been found possible to produce
relatively great depth of the edgewater this was only the La Concepcion well at a bottom-hole flowing
reached in an outstep well in 1949, about 11,000 feet pressure some 3000 psi below the saturation pressure
subsea. The movements of the edgewater have been already for over 2 years, with a constant gas-oil-ratio
observed in this well continuously over 6 months, equal to the solution ratio. Further details are shown
after a series of initial tests which were intended to in Figure 5.
clean the hole and adjoining limestone thoroughly. It is therefore clear why the criterion of limiting
Although the pressures in the well are dropping, the the bottom-hole flowing pressure has been abandoned
movements of the oiliwater interface are eratic, over in favour of direct observation of GOR trends.
678 PROCEEDINGS T H I R D LVOKLD PETROLEUM CONGKESS-SECTION II

As regards the production of edgewater, it has of shutting-in, or opening-up, a neighbouring pro-

been evident that this should be kept to a realistic ducer.
minimum, and to date it has been negligible. The practice now followed in L a Paz is to make
The next problem was the distribution of the off- regular shut-in pressure surveys in a suitable number
take over each field. The basic principle accepted was of wells, and adjust the regional offtake so far as
that production should be distributed so that pressure f easible, in the direction of equalizing the pressure
depletion in each reservoir is as uniform as practic- distribution.
able, in order to minimize possible edgewater trouble, The third problem was the maximum efficient rate
or irregular gascap development. I t is mainly a matter of production for each field as a whole. The criterion
of simplifying practical operations. I n this connection taken as a working rule is that the gross offtake
several tests have been made to investigate inter- shall not be so high as to cause any serious and pro-
ference between wells. I t became obvious that there longed steepening of the decline in mean reservoir
is very good communication in certain areas, as may pressure, when this is plotted against the cumulative
be seen from the general decline of pressures, and production. Provided there is no such steepening of
the initial pressures of new wells. For example, the decline, the main effect in these reservoirs of in-
initial shut-in pressure of the La Paz well P-84, creases in, the rate of offtake is only a shortening
6,600 feet from the nearest producer at that time, of the time factor, especially in view of the limited
P-62, was within IO psi of its pressure when cor- water advance-if any, and there is no harm to the
rected for depth, though this was 190 psi below the fields from the viewpoint of the conservation of
original reservoir pressure. However, at the same reservoir energy, or the attainment of maximum
time there has been no clear evidence of short-term ultimate recoveries. This criterion may indeed be
interference between any single well and another, over-cautious.
such as a sharp rise, or fall, in pressure as a result

PRODUCTION ENGINEERING
Producing conditions Acidizing
J. E. Smith has discussed drilling and completion Acidisation with hydrochloric acid is usual, soon
practice briefly in his report in section I. The after drilling. The quantities of acid which are used
sizes of the final casing strings, set to the top of the vary from about 1,000 to 20,000 gallons per job, at
limestone, are usually 7” or 9%”. All wells are tubed, concentrations between 8 % and 20 %. Commercial
to facilitate killing, swabbing and acidising, and in inhibitors are used, though it is found that if the
order to be able to produce from low down in the tubing etc. is already oil-wet the corrosion is much
open section if this should be required. The success- reduced. The problem of inhibition is difficult at
ful completions all produce by natural flow, either the higher temperatures, approaching 300’ F. De-
through the tubing or annulus-usually both, depend- pending on the circumstances, the acid treatments
ing on the capacity of the well and its position with may be of an overall nature, or selective, using
respect to a gas cap or edgewater. Standard Xmas packers or the “acid pilot” technique. The general
trees are used, sometimes modified by the addition of practice is, first to wash with say 1000 gallons, then
an extra casing spool to provide additional outlets for 0 give a squeeze treatment of about 5000 gallons,
the production and so minimize friction losses. As a followed later if needed by larger jobs. Surface
safety measure, and in case of need, most wells are tension reducers, and also retarders have been used,
hooked up by high pressure lines to central mud but their effect has not been very clear. Spectacular
plants. Some need for gaslift is foreseen, at least to production increases of up to some hundreds of
boost some wells, but of the ultimate cumulative pro- percent are frequently obtained, usually from the
duction, the bulk is expected to flow naturally. first treatment. Follow-up treatments are normally
Optimum flowstring sizes are estimated both from much less successful, although there are exceptions.
empirical pressure-rate-depth-data obtained in the One well in Mara gave a remarkable increase in
field, and from the analysis of the available energy potential, from 60 b/d to 8800 b/d, as a result of a
of lift in the reservoir crudes and of the various treatment with 10,000gallons after an unsuccessful
sources of energy loss. However, owing to the widely initial treatment with 3,000 gallons.
varying capacities of the wells, and the difficulty Shooting, and formation-f racturing by pressure,
of foreseeing these capacities, these studies can only have been tried but without success to date.
be applied in a general way.
 M. STEPHENSON-THE MARA AND MARACAIBO DISTRICTS
679

Fig. 7. Self recording subsurface flowmeter, as constructed for prolific wells. L e f t : assenibled instrument,
but without centralisers; top part is modified recording section of Amerada IIPG3 gauge ; overall length 55”.
Centre: instrument without housing of recording section ; 3-hour clock removed. Right: disassembled impeller
section, showing itator, impeller, impeller shaft, magnets for coupling through stainless steel wall, reduction
gearing I : 3600, and stylus shaft.

Exclusion of gascaps at considerable rates. For example, in La Paz the gas

One of the major operating problems is the ex- oil contact in the fissure system fell by about 48 feet
clusion of gascaps from the producers, in order to during September 1950. The exclusion of excessive
conserve reservoir energy. These gascaps are growing gas from a well is generally accomplished by :
 a ) choking the well back to minimize coning, miniature separators. The lowest pressure separators,
b) lowering tubing, and producing on tubing only, which are of the horizontal type, and may handle up
c) setting a gas shut-off liner. to 60,000 bblslday each, are elevated to a height of
If these methods are unsuccessful the well is shut some 50 feet on towers, in order to be able to collect
in. The procedure adopted depends on the circum- the rich gases, and at the same time permit simple
stances. The third method is preferred if a good zone gravity flow into the fiscalizing tanks. This system
’ of oil entry is present at adequate depth. The perme- has worked very well.
able zones are located by subsurface rate-of-flow and Many of the wells produce through their own first
pressure and temperature surveys, and by a study of and second stage separators. Others produce through
drilling fluid losses. An lmpeller type of self-record- large common horizontal separators, but are switched
ing subsurface flowmeter was specially developed through individual separators for testing. The gas
for this work three years ago. Its overall diameter is obtained from each separator is metered daily, in
1213”. I t may be run on piano wire through the order to keep a close check on all gas-oil ratios.
tubing, and is now in regular use. Fig. 7 shows the The small quantities of saline water, which are
instrument. The recording section is part of a produced in one o r two wells only, are separated
standard Amerada RPG3 subsurface pressure gauge. by high pressure water separators located before the
Injection-type surveys have been made frequently choke, and chemical treatment, the chemical being in-
using the Dowel1 Spinner. However the results often jected at the well heads.
differ markedly from those obtained in the
producing surveys.
Gasoline recovery
I n order to ensure maximum recovery of the lighter
Oil and gas handling hydrocarbon fractions, and avoid losses in shipment
With few exceptions the limestone wells in La Paz -although at considerable cost--high-pressure oil
and Mara flow directly to central separating (de- absorption plants were planned, to extract the butanes
gassing) and oil gathering stations. The exceptions and gasoline from the gases, and then ship the pro-
are those weak or outlying wells which produce to ducts independently. The main La Paz plant, which
local pickup stations, or other wells where the first is now operating, has a design capacity of about 1500
stage of gas separation is carried out at the wellhead, bblslday of natural gasoline, and about the same
in order to eliminate most of the gas from the quantity of mixed butanes. In addition there is a
production and thereby increase the carrying capacity smaller plant with a present output of about 230
of the flowlines, or reduce the pressure drop along barrels of gasoline per day. The Mara plant, which is
them. The well flowlines normally range in length similar to the main La Paz plant, is under con-
from 5 kilometer up to 7 kilometers, but lines as st ruct ion.
long as IO kilometers have been used temporarily.
These lines are usually 6” diameter, welded, and laid Shipping
on the ground or on blocks. as straipht o
as possible to I t has been convenient that all of these Cretaceous
the nearest flowstation. N o expan cion bends are fields are relatively close to the shore of Lake Mara-
needed, and in v iew of the rather flat terrain and caibo. All the production is piped to a common
fairly high pressu leen worthwhile shallow-water shipping terminal at Palmarejo, which
to carry out any @ d U l I l g W L IIIKJ. I wo or more extra is jointly owned by four Companies, although Shell
lines are laid as necessary for the bigger wells. have the major share.
Two central flowstations are in use in La Paz. The La Paz crude is handled by 10’712” lines,
They are situated 6 kilometers apart and at present following two routes. The one route, 40 kilometers
handle some IOO,OOO and 60,000 bblslday respectively. long, passes Mara. The other, which is similar in
In Mara there are three principal flowstations, each length, goes through La Concepcion. The Mara main
with capacities of some 30,000-50,ooo bblslday, some crude lines, also 10”,l12”, eventually join the first-
6 to 7 km. from each other. The Concepcion and Si- mentioned L a Paz route, and have a maximum length
bucara wells thave independent stations. of about 27 kiloineters. The Concepcion crude is
So long as well pressures are high enough 3-stage pumped via a 6” line to Sibucara, 20 kilometers away,
separation of the oil and gas at about 300, 50 and 5 whence the Concepcion/Sibucara mixture (there is no
psig is employed. The conditions of separation and practical dif fercnce between the oils) flows by
the numbers of stages were selected as a practical op- gravity, from an elevation of I ~ o ’for
, 15 kilometers
timum on a basis of direct laboratory tests on re- through a 12” line to Palmarejo.
combined samples of the reservoir crudes, using These main lines are all laid in the same general
 M. STEPHENSON-THE MARA A N D MARACAIBO DICTlìICTS 681

well flowlines. They are welded, laid nominal capacity of 230-240,000 barrels per day. In
he surface, and are straight, as there addition to the Lake crossing, 5 miles long, there will
expansion bends. be a 14 mile crossing of the shallow Gulf of Coro.
g of crude from the fields is done by Pumping facilities will be located at the Palmarejo
pumps, with capacities of the order Terminal and will consist of three centrifugal pumps,
/day each at operating pressures, viz. each driven by a dual fuel gas engine through a speed
Il the pumps are driven by independent increaser. The pumps will be in series, two being in
xtrification being planned for the near continuous operation, the third acting as a spare and
available for use to boost as and when required. Each
I and gasoline from the gas plants in pump will have a capacity of 12,000barrels per hour,
iped separately through buried 3”/4” at about 300 psi head, and its prime mover is rated
o Palmarejo. The Mara products will at 1700 B.H.P.
the same way. In all, forty-seven thousand tons of 20 inch and
ejo lake-shore terminal now handles a 30 inch pipe will be used. For the Lake crossing,
1260,000 bblslday. It has twelve crude special 3/4” walled pipe is being used, primarily to
with a capacity of 1% million barrels, give added weight to prevent floating, and to provide
. hortonspheres for butanes, and two additional protection against corrosion. This pipe will
:s for gasolines. be laid in a dredged ditch, laying operations following
ingle shipping jetty, 1200 meters long, just behind the dredge. The 30” pipe has a wall
iormally used by shallow-draft tankers thickness of %”, and the greater length of it will
10tons. The gross loading capacity is be laid on the ground. The line will be straight and
ur. The butanes are shipped under without expansion bends.
e arrangements
Acknowledgements
zted oi! line to Paragaana The author wishes to thank the management of the
e capacities of these light-oil fields are Shell Caribbean Petroleum Company for permission
)praised, work has been commenced on to publish this paper and is glad to acknowledge the
ie project, in order to take the bulk assistance freely given him in its preparation by other
oil from Palmarejo directly to the members of the staff of this Company. I n particular
ry and deepwater shipping points on he would like to mention the advice of the Manage-
. peninsula. This will bypass the bar ment and the help received in the Exploitation Engi-
,aibo, and Considerably reduce the need neering Department in Maracaibo.
aft Lake tankers. T h e information and experience which have been
distance between Palmarejo and Car- briefly summarised have been accumulated by the
;ulf of Venezuela, is slightly more than work of many, too numerous to name individually, in
. the line will be approximately 150 the Fields, the Laboratories, and the Maracaibo
is designed as a 30 inch welded con- Office of the Shell Caribbean, as well as in the head
uith two 20 inch submarine lines in office of the Bataafsche Petroleum Mij. in The
ig Lake Maracaibo, and it will have a Hague.

DISCUSSION
B. KEIMERING,rapporteur (N.V. De be interesting to learn what estimate has been made
2troleum Maatschappij, The Hague, of the minimum bottom hole working pressure neces-
remarked that as regards the fluid sary to keep the w e h flowing naturally.
:ores mentioned in Table II it would Mr. L. C. STEVENS (Petroleum Engineer, The
to know what drilling fluid was used Hague, Netherl’ands) remarkcd that the problem of
he cares and to what extent the per- the determination of the proportion of the total oil in
fluenced thereby. place present in the primary and the secondary poro-
oximate figure be given on the relative sity is most important but very difficult t3 solve, par-
the oil in place contained in the finer ticularly the ultimate recovery of oil from the primary
that in the very coarse fissures? porosity. No reliable statistical analysis of the fluid
d that most of the ultimate cumulative content of the matrix is considered possible due !.:
’
1 be obtained by natural flow. I t may the low core recovery by conventional coring methods,
 Recently diamond coring has been more successful, problems are being faced in other limestone areas
but only applied on a limited scale. The oil saturation and any new approach would be very helpful.
value being determined as a difference of two labor- Mr. J. T. WARDE (N.V. De Bataafsche Petroleum
atory determinations, namely porosity and water Maatschappij, The Hague, Netherlands) replied in
saturation, is very liable to error. Moreover, part of writing as follows: W e have used a great variety of
the primary porosity is occupied by solid hydro- drilling fluids from weighted drilling emulsion to
carbons. water and more recently oil. Comparisons in a small
The method of estimation of fissure porosity by way made between the fluid contents of cores taken
direct measurement, apart from the large personal in water and cores taken in oil show that there is very
factor, is not considered representative of average little difference between them. This applied only to the
conditions. as it has only been applied in one or two matrix. Flushing along cracks certainly takes place.
diamond cored wells on the flank in Mara. Visual I am afraid I cannot give even an approximation as
examination of the diamond cores would seem to to the proportion of oil in piace in the coarse fissures
indicate that the oil has migrated into the structure to that in place in the finer void space. Future pro-
along the major fissure planes and thence into the duction policy has to take into account that there is
finer cracks and has also saturated a band of the a strong possibility that the major part of the oil is
matrix formation on either side of the fissures ex- contained in the very fine cracks.
tending backwards to a depth of only 1-2cms with It has been estimated that when the oil has all been
a visible, rapid reduction of oil content away from recovered the reservoir pressure will be about 1000
the crack. Hence production of oil from any ap- psi and with a pressure of this order most of the
preciable distance from a fissure system would seem wells will flow.
unlikely. Nevertheless, so-called dry holes whicli Mr. Stevens summarizes the difficulties involved
showed no water loss when tested considerably above in evaluating a reservoir of this type and I can fully
hydrostatic pressure, showed some small production agree with his remarks. As regards the question of
of oil after the well had been standing under a the future production policy to obtain the greatest
water head. recovery I would repeat that not only the matrix is
The solution of this problem is very important involved in this problem but also all the low per-
from the point of view of estimating ultimate re- meability cracks which exist in great numbers and
covery and of future production policy, i.e. whether which might contribute considerably to the volume
to drop the reservoir pressure rapidly or not in order of oil in place in the reservoir.
to achieve maximum recovery. Presumably similar