PROJ00326

• Contract/PO #: C153498
• Contractor Name: U. S. Geological Survey
• Water District: 0
• County: Boulder
• Bill Number: SB 87-15
• Loan Projects - Doc Type: Contract Documents

part of the hole. The thinner mudcake in the upper part
of the hole could have enabled greater magnetite inva-
sion in this area than in the lower part of the hole, and
is consistent with the response of the free-fluid index
log, which indicates a smaller signal in the upper part
of the hole than in the lower part of the hole. Although
nuclear magnetism logging has been used successfully
in the deep wells of the petroleum industry for many
years, results of this work seem to indicate that nuclear
magnetism logging using magnetite doping might not
be applicable to relatively shallow wells in highly per-
meable materials, where drilling methods generally do
not produce a thick, impermeable mudcake.

Effective-Porosity Log

An effective-porosity log is produced as part of
the Cyberlook log-interpretation package. Schlum-
berger Well Services (1987, p. 120) defines effective
porosity as:
"Effective porosity is that porosity associ-
ated with the non-shale phase of the shaly
sand. It is the porosity that would exist if the
shale and the water bound to the clays were
removed, leaving only the clean sand phase."
The similarity between this definition and the
definition of specific yield implies that an effective-
porosity log also can provide a means for estimating
specific yield. Log values of effective porosity gener-
ally are larger than the corresponding specific yield in
well-sorted coarse-grained materials and are smaller
than the corresponding specific yield in clayey materi-
als. In coarse-grained materials, effective porosity is a
measure of the total pore space of the rock, whereas
specific yield is a measure of only the drainable part of
the total pore space. In clayey materials, effective
porosity is zero by definition, whereas specific yield is
generally slightly greater than zero.
An effective-porosity log is calculated by use of
the equation:

... ='" -1 ...
'I'e 'Yt c'fc'

where
$. = effective porosity,
$1 = total porosity,
Ie = clay index, and
$c = porosity of clay.

Crossplots of density and neutron-porosity logs
are used to define $1 and $c; the minimum shale index
log can be used to define Ie.

The feasibility of using effective porosity to
estimate specific yield was evaluated by use of a least-
squares linear regression analysis of specific yield from
holes CI and CIA on effective porosity from the log for
well USGS. The core data used for the regression anal-
ysis consisted of 145 specific-yield determinations
made on core samples from holes Cl and CIA. The
core-sample intervals were correlated from holes Cl
and CIA to well USGS, and selected samples were
excluded from consideration if they failed to meet the
following criteria:

I. Lithology and general geophysical log response at
the core-sample depth had to be similar
between holes Cl, CIA, and well USGS.

2. The log response in well USGS must be relatively
uniform at the correlated depth so a small error
in depth correlation would not make a large dif-
ference in log value.

3. The borehole of well USGS at the correlated depth
had to be of uniform diameter so hole rugosity
would not adversely affect log response.

(3)

Samples that meet all these criteria were well
suited for comparison to log response. The resulting
data set consisted of 64 samples-19 from the Dawson
aquifer, 20 from the Denver aquifer, and 25 from the
Arapahoe aquifer. Core data from the Laramie Forma-
tion were few and were excluded from consideration.

The regression equation of specific yield on
effective porosity (fig. 6) has a coefficient of
correlation of 0.84 and a standard error of estimate of
0.05 specific-yield units. This level of correlation indi-
cates that effective porosity can be a valid indicator of
specific yield, provided that specific yield is calculated
as a mean of several determinations. Mean specific-
yield values (table 3) calculated for each aquifer from
laboratory analyses of core are shown to be similar
to the mean specific yield calculated from the effective-
porosity regression. Sixty-eight percent of new
specific-yield estimates that are derived by use of the
regression equation can be expected to be within plus
or minus 0.05 specific-yield units of the laboratory
specific-yield value.

14 Techniques for Estimating Specific Yield snd Specific Retention from Graln.Slze Oats and Geophysical Logs from
Clastic Bedrock Aquifers