Click to view article in PDF format.
A Petrophysical Method to Evaluate Irregularly Gas Saturated Tight Sands Which Have Variable Matrix Properties and Uncertain Water Salinities*
Michael Holmes1, Dominic Holmes1, and Antony Holmes1
Search and Discovery Article #40673 (2011)
Posted January 11, 2011
*Adapted from oral presentation at AAPG International Conference and Exhibition, Calgary, Alberta, Canada, September 12-15, 2010
1Digital Formation, Denver, Colorado ([email protected])
A problem in many Rocky Mountain tight gas sandstones is a sequence that is only partially gas saturated, with changing matrix properties combined with variable (and often unpredictable) water salinities. Often it is difficult to distinguish between high resistivity fresh water wet sands, and high resistivity, gas-bearing sands. A standard approach is to make a qualitative judgment based on
density
/
neutron
response – the gas “cross-over” effect. However, if matrix properties are variable, this approach can be misleading, and is at best a qualitative judgment.
The methodology presented here is a quantitative assessment of gas saturation by comparing matrix specific
density
and
neutron
responses with
porosity
, calculated such that gas effects are minimized. Cross plot
porosity
from
density
/
neutron
combination is only minimally affected by gas and by changing matrix properties.
Three sets of calculations are made assuming sandstone (bulk density 2.65 gm/cc), calcareous sandstone (bulk density 2.68 gm/cc) and heavily cemented calcareous sandstone, or limestone (bulk density 2.71 gm/cc). Quantified estimates of gas saturation, as “seen” by each
log
, are available for each assumed rock type. Pressure effects on
porosity
log
responses are included in the calculations.
Four sets of saturation profiles are now available, one from standard resistivity
log
analysis and three from
porosity
log
analysis assuming different matrix properties. Comparisons among the 4 sets of saturation profiles can be combined with other data, such as mud
log
shows and (if available) core measured matrix densities. Using such comparisons, it is often relatively simple to distinguish between wet intervals and gas-bearing intervals. With increasing assumed grain density, gas saturations calculated from the
porosity
logs increase. If gas saturations so defined are unrealistically high, it is an indication that actual grain density is less than assumed grain density.
Additionally, if matrix properties are well-defined, it is possible to verify Rw input for resistivity
log
interpretation, and adjust as necessary. It is important to recognize that the
porosity
logs, and particularly the density
log
, investigate close to the wellbore, and may well be influenced entirely by the flushed zone. Examples are presented from Rocky Mountain reservoirs, in sequences where the problem of irregular gas saturation in systems with variable Rw is particularly severe.
Copyright � AAPG. Serial rights given by author. For all other rights contact author directly.
|
Shaley Formation Resistivity Analysis
The interpretation is a standard shaley formation analysis, involving calculations of total
The second set of analysis involves the calculation of effective
Gas Saturation from
Comparison of
Standard resistivity interpretation from a Piceance Basin well, NW Colorado, is shown in Figure 2. Interpretation of the
Rw = 0.1 Water resistivity a = 1.0 Cementation constant m = 2.0 Cementation exponent n = 1.7 Saturation exponent
A summary of the interpretation is as follows:
Figure 4 shows the same interval as Figure 2, with the addition of interpretations of gas saturation from
|

