TIGHT GAS BASICS
Most of us are familiar with traditional tight gas reservoirs – clean,
low porosity sandstones or siltstones that look unattractive
on log analysis, at least by the conventional wisdom of the
1960’s. Porosity averages from 5 to 10% with permeability
between 0.01 and 5 mD. The best know play of this type is
the Alberta Deep Basin, developed in the 1970's and 1980's,
and still an area of interest today.
This
particular tight gas play is called a basin-centered gas
accumulation - the trapping mechanism is not structural or
stratigraphic, but a "water block" above the gas caused by low
relative permeability. There is considerable exploration and
development effort being expended on such plays today, especially in
the USA, Europe, and the Middle East.
With
the steady improvement in massive fracturing jobs, pioneered in the
Deep Basin, and in horizontal well placement, even tighter, lower
porosity gas zones are now economically feasible. These have
porosities that may average 3 to 6% and permeabilities from below a
microDarcy to a few milliDarcies. The Doig and Montney in Alberta and
northeast BC are examples. Both are radioactive due to uranium and
have often been called shales, even though the average grain size is
above 4 microns and there is little clay mineral or clay sized
particles. The organic content is fairly low (1 to 3% TOC) so there
is little adsorbed gas. They do not qualify as "shale gas" until 67%
of the particles are less than 4 microns.
Some tight gas
plays have a significant liquids component. Such wells are highly
desirable due to the price differential between gas and oil.
Unfortunately, log analysis cannot partition gas from oil at these
low porosities so tight rock core analysis and petrographic analysis
are important tools. Mineralogy and clay content are highly variable
so XRD analysis is also vital.
Alberta Deep Basin strat chart, showing gas producing reservoirs,
circa 1981. Note that Montney was not among the productive category
at that time.
The
petrophysical model uses conventional log measurements with
conventional shale corrected density neutron complex lithology
porosity model to handle quartz plus heavy minerals. A shale
corrected water saturation equation, such as the Simandoux or Dual
Water models are used.
Most zones in a tight gas
environment produce little water except at the updip edge, so RW is
actually derived from pre-determined water saturation values found
by capillary pressure measurements. A table of RW values and a
stratigraphic chart for the Deep Basin play were published in "Log Evaluation Results in the Deep Basin Area of Alberta",
by E. R. Crain, Transactions: 8th Formation Evaluation Symposium,
CWLS, Calgary, September 1981.
Saturation exponents are often
default values (A = 1.0, M = N = 2.0) because not very many
electrical properties measurements have been reported. Lower values
of M = N = 1.7 or 1.8 may be needed to force calculated water
saturation to match core analysis or capillary pressure minimum
water saturation.



“Tight Gas” example showing core porosity (black dots), core oil saturation (red dots).
core water saturation (blue dots), and permeability (red dots). Note
excellent agreement between log analysis and core data. Separation between red dots and blue
water saturation curve indicates significant moveable gas. The core
analysis shows considerable residual oil - some of this may be
moveable, in addition to any condensate carried in the gas. This is
a relatively high grade example with porosity between 5 and 6% and
permeability between 0.1 and 0.8 mD. Grid lines are 1 meter spacing.
NOTE the high uranium content (left hand curve in Track 1) is in the
middle of the best pay.
Without the Thorium curve, this interval would look shalier and it would
be difficult to match
core porosity using the total gamma ray as a shale indicator.
RUN THE SPECTRAL GAMMA RAY LOG TO ELIMINATE THIS PROBLEM.
GEOLOGY OF TIGHT GAS
The geology of most tight gas plays, whether old
fashioned like the Alberta Deep Basin, or the new "really-tight"
plays is slightly more complicated than conventional gas plays,
although the two can combine or grade into each other. One of the
better descriptions is in "Tight
Gas Reservoirs – An Unconventional Natural Energy Source for the
Future", by G. C.Naik, SPE 2003.
Naik
describes four criteria that define basin-centered gas accumulations
like the Alberta Deep Basin, including low permeability,
abnormal pressure, gas
saturated reservoirs and no down dip water leg.
Later he mentions the updip water block that traps the gas, as
proposed by Masters and Gray in the late 1970's.But not all tight
gas is a basin-centered accumulation, so different concepts apply in
different areas.
The
illustration at the right attempts to show the difference between
conventional and unconventional basin centered gas accumulations.
The red interval acts as an updip water block, with gas below it
in unconventional traps, with conventional traps above the seal.
Master and Gray wrote that the water block is transitional over a 5
to 10 mile wide band and is not related to a structural or
stratigraphic boundary.
Masters
and Gray explained this phenomenon by showing where the water block
and gas reservoirs fit on the relative permeability curve for the
gas - water system, as shown on the left. Little water is produced
in the gas zones, regardless of water saturation, due to the low
relative permeability of the water phase.
The
concept of a water block merely means that irreducible water
saturation is very high. On log analysis depth plots, this looks
like "water over gas", which cannot happen in a conventional
reservoir. The result looks like an upside-down transition zone.
Basin
margin gas traps, as in the Cretaceous of south eastern Alberta, are
relatively low porosity, low permeability plays, but are not now
considered to be unconventional. They do not have a water block as a
seal and occur in normal stratigraphic traps. The gas is biogenic
(formed from kerogen in situ). Gas does not migrate far from its
source.
Their
behaviour is sometimes a little unusual, depending again on the
relative permeability curves.
According to Naik's paper "In
a traditional reservoir, there is relative permeability in excess of
2% to one or both fluid phases across a wide range of water
saturation. Further, in traditional reservoir, critical water
saturation and irreducible water saturation occur at similar values
of water saturation. Under these conditions, the absence of
widespread water production commonly implies that a reservoir system
is at, or near, irreducible water saturation. In low-permeability
reservoirs, however, irreducible water saturation and critical water
saturation can be dramatically different. In traditional reservoir,
there is a wide range of water saturations at which both water and
gas can flow. In low-permeability reservoirs, there is a broad range
of water saturations in which neither gas nor water can flow. In
some very low-permeability reservoir, there is virtually no mobile
water phase even at very high water saturations."

Comparison of capillary pressure and relative permeability curves
for conventional gas (left) and tight gas (right) showing a large
water saturation range in tight gas reservoirs over which no gas or
water will flow.
The Montney distal shelf (‘tight gas’) play has become one of the
hottest natural gas resource plays in the WCSB. Horizontal drilling
and multi-stage frac technology have been the key to unlocking the
resources and placing the Montney in the top three most economic
resource plays in North America. Industry analysts estimate upward
of 5,000 horizontal wells will be drilled in the upcoming decades,
with a capital outlay approaching $30 billion.
To
illustrate petrophysical analysis of tight gas sands, we will use
the Montney as the classic example of the problems and solutions.
Some of those problems are radioactivity from uranium associated
with kerogen, highly variable mineralogy, very fine grained texture,
and several hydrocarbon types that are difficult to segregate.
Most tight gas sands have a wide variety of rock textures and
mineral compositions vertically in the wellbore as well as laterally
between wells or along the track of horizontal wells. Trying to find
"sweet spots" and steering along them is a challenging task. The
illustration below shows microphotos of four distinct facies in the
Montney from west east across west central Alberta. Porosity, grain
size, saturation, and permeability vary considerably.

Photo micrographs of Montney facies. The stratigraphic cross section
is also complicated so correlation across long distances is somewhat
difficult.

SHALE VOLUME CALCULATIONS IN TIGHT GAS
Many tight sands are radioactive due mainly to uranium
associated with phosphates or kerogen. This can be identified with a spectral gamma ray log and it
should always be run when penetrating radioactive sands. Sadly, it
is often not requested, even though the service is cheap and costs
no extra rig time.
Spectral gamma ray log shows
Uranium (U), Potassium (K), Thorium (Th), and standard gamma ray (GR).
Red vertical line is TH0, the clean line for the Thorium curve, and
the black vertical line is GR0, the clean line for the GR curve.
The Thorium curve is best for shale volume calculations. The SP is
flat and useless, Density neutron separation is mostly due to
dolomite and other heavy minerals so it cannot be used. The gamma ray can be used in the
absence of the Thorium curve by assuming Uranium content is
constant.
1: VSHth = (TH - TH0) / (TH100 - TH0)
2: VSHgr = (GR - GR0) / (GR100 - GR0)
The Clavier correction to the gamma ray result is often used to
smooth out minor variations in uranium content that make the gamma
ray look "noisy":
3: VSHclavier = 1.7 - (3.38 - (VSHgr + 0.7) ^ 2) ^ 0.5
Choose VSHth in preference to VSHgr or VSHclavier when the thorium
curve is available. This becomes Vsh for all future calculations.
The clean lines TH0 and GR0 are
easy to pick (red and black lines on the illustration). Shale lines
are harder as they are often off-scale to the right or buried under
a plethora of backup curves. In the absence of a good pick from the
log, use:
4: TH100 = TH0 + 25
5: GR100 = GR0 + 150
Adjust the constants to suit your
local knowledge.
Calculation of porosity is very
sensitive to the shale volume in tight gas sands, so it is critical
to calibrate Vsh from logs with clay volume from bulk XRD data sets
or tables of petrographic thin section point counts. A difference of
a few percent clay can mean the difference between NO PAY and ALL
PAY.
IMPORTANT: Remember that all log analysis models for TOC are
calibrated to standard geochemistry lab data that often do not
discriminate between kerogen and pyrobitumen. Either or both may be
present. Both have variable but fortunately similar physical
propertiees so converting log derived TOC to "kerogen" may actually
be a conversion to pyrobitumen or a mixture of the two components.
In the following material, you may want to substitute the words
"Organic Matter" for "Kerogen" to be more general.
POROSITY CALCULATIONS IN TIGHT GAS
Even though most tight gas sands are a complex mixture of quartz,
dolomite, calcite, and sometimes pyrite, with a little clay, the
standard density neutron complex lithology crossplot model works
well most of the time. However a correction for kerogen needs to be
made to the data if any is present:
6: PHIdc = PHID
– (Vsh * PHIDSH)
– (Vker * PHIDKER)
7: PHInc = PHIN
– (Vsh * PHINSH)
– (Vker * PHINKER)
8: PHIe = (PHInc + PHIdc) / 2
A
more complete description of the porosity method and the conversion
of TOC weight fraction to kerogen volume fracture are given in the
Gas Shale
Chapter.
This step requires careful calibration. For example, if Vker
> PHIek, there is something seriously wrong in the calculation
of PHIek or TOC.
Methods that
avoid the neutron log are also useful, including sonic-only, density-only, or sonic
density crossplot. Each method should be shale corrected and
calibrated to core. Matrix and fluid properties are needed for this,
possibly on a zone by zone basis. Some samples are shown below.

Crossplots of sonic (left) and density (right) versus
core porosity. Best fit lines give DTCmatrix = 182 with
DTCfluid
= 500 usec/m and DENSmatrix = 2710 with DENSfluid = 1050 kg/m3. The red
line on the density crossplot shows a relationship with DENSfluid = 400 kg/m3. Such a relationship has received some support
in the industry but clearly does not fit the core data available on
this project.
The equations for solving the
sonic and density models are as follows:
9: PHIdc = PHID
– (Vsh * PHIDSH)
– (Vker * PHIDKER)
10: PHIsc = PHIS
– (Vsh * PHISSH)
– (Vker * PHISKER)
11: PHIe = (PHInc + PHIdc) / 2
The matrix
values that lead to PHID and PHIS may need some juggling to
calibrate to core porosity. Values in the quartz + heavy mineral
range usually work. An example is shown later on this page.
More sophisticated multi-mineral and statistical methods are
definitely desirable, but these are not always available
quickly.
If these methods agree with each
other, then the regression worked well. If they are in general
agreement with the density neutron crossplot, then it should be used
because it has slightly better compensation for mineral variations.
However, if it is considerably higher than sonic and density results
(or core data), then abnormal neutron absorber minerals should be suspected and the
density neutron method should be discarded.
To reduce rough hole and sonic
skips, taking an average of 3 or 4 methods may be used.
Nuclear magnetic resonance logs
have become popular in tight gas, but they require special
attention. They generally show near zero effective porosity (BVI +
BVM) but the NMR total porosity (CBW + BVI + BVM) is close to the
effective porosity from the methods discussed above. This suggests
that the NMR cannot tell the difference between clay bound water,
capillary bound water, or gas in these low porosities.
WATER SATURATION CALCULATIONS IN TIGHT GAS
There may be enough clay that the Archie model should not be be used.
It costs nothing extra to use a shale corrected saturation equation
such as Simandoux or Dual Water model:
12: C = (1 - Vsh) * A * (RW@FT) / (PHIe ^ M)
13: D = C * Vsh / (2 * RSH)
14: E = C / RESD
15: Sw = ((D ^ 2 + E) ^ 0.5 - D) ^ (2 / N)
Electrical properties variations
between facies and with depth or diagenesis are not published. This
lab work is worth the effort, as considerable increases in gas in
place are possible with small reductions in M and N values.
Tight gas reservoirs are not "average" sandstones, so the electrical properties must be varied from
world average values in common use (A = 1, M = N = 2.0). To get
log analysis Sw to match lab data, much lower values are needed. Typically, A =
1.0 with M = N = 1.5 to 1.8. Unless lab derived properties are
available, vary M and N to obtain a good match to core Sw. If core
Sw is not available, the recommended default is M = N = 1.7.
Water recoveries are usually
negligible and fairly fresh as the water is condensed from entrained
water vapour. The minimum water saturation on capillary pressure
curves might be used to work backwards to find an RW that will give
a reasonable water saturation. Core analysis water saturation, at
least in more modern wells, seems to be a good guide and RW can be
adjusted accordingly.
PERMEABILITY CALCULATIONS IN TIGHT GAS
Permeability
may show a reasonable relationship to core porosity. In the example
at right, there is a strong correlation. Many older core analyses do
not record permeability below 0.01 mD so are quite useless. Modern
tight rock methods can give permeability in nanoDarcies. The
equation of the line in this example is Perm = 10^(20.0 * PHIe -
2.75). A few high perm data points are fractured samples.
Lithology CALCULATIONS IN TIGHT GAS
How
do we know which minerals to use in the petrophysical log analysis?
Detailed sample descriptions are a good start. Both X-Ray diffraction data and thin section point counts can be
used. Both methods are considered semi-quantitative and come from
tiny samples compared to the volume measured by logs. So we don't
get too excited about obtaining a close numerical match .
Standard 3-mineral models using PE, density, and neutron data are
used with appropriate parameters for the selected minerals, provided
the neutron log is not shifted to the left due to iron or other
neutron absorbers. If this happens, we can calculate a matrix
density from the sonic density crossplot porosity and run a
2-mineral model:
22: DENSma = (DENS - PHIe * DENSfluid - (1 - Vsh)
* DENSSH) / (1 - PHIe - Vsh).
If Vsh > 0.85, we set
DENSma = DENS. If matrix density is too high compared to known
lithology, it means porosity is too high. Adding shale or
eliminating bad data will reduce this problem - the calculation is a
good quality-control step.
Multi-mineral solvers can be used if spectral gamma ray data is
available. In this case, shale volume would be derived also.
Elemental capture spectroscopy logs are also used. These solve for
the minerals and kerogen from the chemical element composition of
the rock discovered by the logging tool, but it may not do justice to the porosity,
which should be checked by conventional methods.
As in all things
petrophysical, there is not much accuracy in the calculation of
small volumes of minerals or porosity.
PYRITE CORRECTIONS
Dispersed pyrite is described in most XRD reports on at least the
Doig and Montney, but not in the Cretaceous Deep Basin reservoirs.
However, unlike the Bakken tight oil case, it seems to be
unconnected to the porosity and has no impact on the high
resistivity values seen in these zones. So far, no corrections have
been needed in the tight case plays.
TIGHT GAS EXAMPLE
This example shows a tight gas sand in the Montney play that behaves
like a normal sandstone as far as log analysis is concerned. The
porosity is halfway between the density and neutron porosity curves.
The gamma ray is quite high due to uranium, but the XRD bulk clay is
low (0.00 to 0.10 fractional, 0 to 10% by weight). These values
don’t change much when converted to volume fraction. Resistivity is
reasonably high and RW is moderately low, so water saturation is
low. Only the raw log curves are shown here to illustrate the fact
that visual analysis is quite straight forward provided you get over
the “hang-up” of the high gamma ray values.


Example showing raw logs for a
typical Montney tight gas sand.
Zone is clean and core
porosity is halfway between the density and neutron porosity. Zone
is radioactive, quartz + dolomite.
This next example
shows the effect of abnormal neutron absorbers on the porosity
results, and the use of sonic and density data to avoid giving too
high a porosity. Some wells show larger neutron offsets; some show
no abnormal effects. Calibrating each curve to core and then taking
an average tends to remove variations in lithology that would
otherwise distort the porosity result.



This illustration of a portion of a lower Montney shows the use of
multiple pay flags (porosity of 3% = red, 4% = green) beside GR
track that emphasize the laminated nature of the porosity. The
"multi-porosity" track to the right of the resistive has (from left
to right) PHIN_SS, PE, PHIxndc (grey), PHIS (blue), Final Porosity
(solid black), PHIDcustom (green), and PHID_SS.The final porosity is
a weighted average of the other curves, after shale corrections,
because there appear to be abnormal neutron absorbers that have
shifted the neutron to the left, causing the density neutron complex
lithology model to read a bit too high. The matrix density track
shows reasonable values that agree with XRD and core grain densities
in other wells. The porosity track shows the final porosity (shaded
red) with crossplot porosity just a little bit to the left.
Permeability is from regression of core data. Lithology on the far
right is a 2-mineral model using the matrix density derived from
porosity and density data, using quartz+feldspar and dolomite+heavy
as two "generic" minerals.
Grid lines are 1 meter spacing.
|