1,765 and 1,259 ( 3,024 total) new
wells drilled and 2.09 and 2.07 tcf
( 4. 16 tcf total) of natural gas produced
from the field.
The Barnett shale study work-flow (Fig. 1) was developed to allow
iterative input from all disciplines,
ultimately leading to a rigorous production outlook model. This result
provides some level of confidence that
the model and its underlying assumptions are reasonable, at least in the
early years. A model update that will
include all wells drilled through 2012
and their production through mid-
2013 will be run in 2014.
We chose a study area to encompass
the extent of previous drilling within known geologic boundaries of the
field. A total of 8,000 sq miles was included, although only 4,172 sq miles
had been tested by drilling through
2010. The Barnett thins to the west
and south, and the boundaries of the
study area were extended to include
the conceivable developable extent
We performed a log-based assessment of the Barnett. 2 Raster logs from
16,730 wells were available, 2,010 of
which were deep enough to view the full Barnett section,
and only 420 had both density and neutron-porosity logs.
Of these, 146 wells with good log quality and caliper
(measure of hole quality) and good spatial distribution were
selected. Major stratigraphic tops were picked in all wells,
and the study focused on the lower Barnett, the target interval of the vast majority of production.
To the west the complete Barnett section is included in
the pay zone, but to the east the Barnett is subdivided into
upper and lower members by the Forrestburg limestone formation. The lower Barnett thickens near its east edge with
the addition of lower-quality porosity zones and interspersed
lime mudstones that could inhibit fracture growth during
In this area, only the lower part of the lower Barnett is
included in the pay zone (Fig. 3).
Density-log porosities across the pay zone were analyzed
so that, when aggregated, they yielded a distribution average of 11.6% porosity (Fig. 4). This average was compared
with crushed-core porosity from 796 samples in four wells, 3
yielding an average porosity of 6.0%. The study used the
ratio of log and core porosities to create an adjustment fac-
Two cross sections of the Barnett shale show upper Barnett, Forrestburg limestone,
upper part of lower Barnett (nonpay zone), and lower part of lower Barnett (pay zone).
Shaded areas of density-log porosity (DPHI) curve represent DPHI values <5%. Anomalously thick Barnett shale in a well in cross-section B-B’ is interpreted to have deposited
in karst sinkhole or collapsed cave. Pay zone is highlighted in yellow. Locations of cross
sections are shown in Fig. 2. Source: Reference 2. (Fig. 3)
tor of 52% to adjust density-log porosity to core-equivalent
Maps were developed of adjusted porosity across the
field, in addition to maps of net pay-zone thickness. The two
maps were combined into a PhiH map indicating reservoir
volume. The porosity-thickness map was found to provide
the best correlation to well productivity, a key element in
predicting future field production from undeveloped areas of
the field. PhiH is clearly driven by the influence of pay-zone
thickness, rather than by more gradually varying porosity
We recognize that the geologic description could have
been enhanced greatly by analysis of seismic data to identify faulting, karsting, and other anomalies across the field.
Seismic analysis was beyond the scope of the study, however, and we think that PhiH mapping is adequate to explain
the first-level geologic drivers of production and to predict
productivity across the field. The 3D seismic would be more
crucial at an individual-prospect level.
Production decline; economic analysis
The study conducted individual decline analysis on all
15,144 wells drilled through 2010 to determine their indi-