Kansas Geological Survey, Open-File Rept. 93-1
Annual Report, FY92--Page 3 of 20
Hand-contoured maps of the top of the following stratigraphic units were produced: Carlile Shale, Greenhorn Limestone, Graneros Shale, Dakota Formation, Kiowa Formation, Longford Member (Kiowa Formation) or Cheyenne Sandstone, Morrison formation, Permian System, Cedar Hills Sandstone, Salt Plains Formation, and Stone Corral Formation. Using the formation tops, thicknesses were computed and mapped for the following geologic units: Carlile Shale, Greenhorn Limestone, Graneros Shale, Dakota Formation, Kiowa Formation undifferentiated marine shale, Longford Member (Kiowa Formation) or Cheyenne Sandstone, Morrison formation and other undifferentiated Jurassic and Triassic deposits, Permian strata above the Cedar Hills Sandstone, the Cedar Hills Sandstone, Salt Plains Formation, and Harper Sandstone.
The hand-drawing of the isoelevation and isopach contours has been completed, and the contours and the points are in the process of being digitized into the ARC/INFO geographic information system, where they will be used to create coverages for map production. The finished maps will be assembled into a geologic atlas of southwestern Kansas, which should be ready for publication during the early part of FY94. The maps will also be used later in the flow modeling of the shallow part of the regional system to delineate regional hydrostratigraphic units, the aquifer, and aquitard units.
Work in the future will be directed toward using the digital gamma-ray data to generate sandstone/shale ratios in the stratigraphic units that constitute the upper and lower Dakota aquifer units across southwestern Kansas. The ratios will be contoured to produce isolith maps for each sequence-stratigraphic subdivision to identify the dominant sandstone-body trends in the Dakota and to map three-dimensionally the distribution of sandstone. This is an important first step in the process of defining the complicated plumbing system of interconnected sandstone bodies in the Dakota.
Figure 5--Reduced color image of gamma-ray log intensity of the Permian to Upper Cretaceous strata in western Kansas along T. 16 S. from R. 31 W. to R. 9 W. A larger version of this figure is available.
Geophysical Log Analysis of the Gray County Site
Spectral gamma-ray, lithodensity-neutron, acoustic velocity, and resistivity logs were run
in the Gray County well drilled in March 1993 in sec. 26, T. 27 S., R. 28 W. The
stratigraphy differs from other wells of the Dakota program series in that it has an interval
of the Ogallala Formation above the Cretaceous section. A major point of interest in this
logging run was whether the Ogallala can be distinguished from the Dakota sandstones on
the basis of geophysical logs alone. If so, then what are the petrophysical properties that
make this distinction possible? Some conclusions are discussed in this summary with
reference to the resistivity (Figure 6), lithodensity-neutron (Figure 7), acoustic velocity
(Figure 8), and spectral gamma-ray logs (Figure 9) from this well.
Figure 6--Resistivity logs from KGS Gray County Feed Yard 1, SENE sec 26, T. 27 S., R. 28 W., Gray County, Kansas.
Some geophysical log comparisons between the Ogallala and the Dakota were easier than others because most of the Ogallala section was above the water level in the well. As a result, the pore space in the Ogallala sands and gravels was only partially saturated with water. The high air content of the pore space can be seen readily on several logs, particularly those from tools with relatively shallow depth of investigation beyond the borehole wall. The relatively high-density porosity and low neutron porosity readings in the Ogallala (Figure 7) would be recognized immediately by most petroleum geologists as a gas effect. In this instance the gas is not a hydrocarbon but natural air. However, the effect is the same because of the low density and low hydrogen content of air compared with water. The effect of the air phase is also evident in the acoustic velocity log. Above the water level there is a marked increase in transit time (Figure 8), because the speed of sound is much slower in gas than in aqueous media.
Figure 7--Lithodensity-neutron logs from KGS Gray County Feed Yard 1, SENE sec 26, T. 27 S., R. 28 W., Gray County, Kansas.
Figure 8--Acoustic velocity log from KGS Gray County Feed Yard 1, SENE sec 26, T. 27 S., R. 28 W., Gray County, Kansas.
The spectral gamma-ray logs (Figure 9) proved to be a valuable aid in the interpretation of rock types and units from drill cuttings. In particular, the feldspar content of the Ogallala gives a useful signature on the potassium curve of the spectral log, differentiating the Ogallala from the Dakota rocks. The discrimination is best made by the computation of a Th/K ratio log (Figure 10), so that the effects of potassium-feldspar (high potassium and moderate thorium) can be accentuated relative to those of illitic shales (high potassium and high thorium). Note that the distinction of the Ogallala from the Dakota in this well is quite clear-cut when using the Th/K ratio as a criterion.
Figure 9--Spectral gamma-ray logs from KGS Gray County Feed Yard 1, SENE sec 26, T. 27 S., R. 28 W., Gray County, Kansas.
The Th/U ratio values of the Ogallala and the Dakota are similar (Figure 10), and their range indicates fairly neutral redox conditions in their formation. However, both units differ drastically from the intervening Graneros Shale and Greenhorn Limestone, where a consistently low Th/U ratio reflects enhanced uranium concentrations. The uranium was probably fixed by organic matter by the reducing conditions that prevailed during the marine deposition of these units.
Figure 10--Gamma-ray spectral ratio logs from KGS Gray County Feed Yard 1, SENE sec 26, T. 27 S., R. 28 W., Gray County, Kansas.
Collectively, the gamma-ray spectral ratio signatures are valid and useful measures for distinguishing the Ogallala and Dakota Formations and for recognizing the Graneros Shale and Greenhorn Limestone. The discrimination criteria are also likely to be of regional rather than localized significance. The Miocene Ogallala Formation represents clastic deposits derived from the uplift of the Rocky Mountains to the west, with feldspar supplied from fresh granitic material. These sediments contrast with the more mature clastics of the Dakota Formation, which originated from more distant sources to the east. The relative increase of uranium concentration in the Graneros Shale is well known across Kansas and is a typical phenomenon in regional marine shales of this type, where the seafloor was under reducing conditions. Enhanced uranium contents are less common and much more localized in the fluvial and deltaic deposits of the Dakota and Ogallala Formations, where neutral or mildly oxidizing conditions appear to have been the rule in either the original deposition or subsequent diagenesis.
Spectral gamma-ray logs are more expensive to run than conventional gamma-ray logs.
However, their use can be justified on occasions where the aquifer stratigraphy is poorly
understood and where distinctions between the Ogallala and Dakota Formations are needed.
Alternatively, they could be run as reference logs (i.e., as a stratigraphic standard) at
selected locations for the correlation of conventional gamma-ray logs run in wells in the
surrounding area.
Geostatistical Analysis of Sandstone and Shale Distribution in Hodgeman
County
Approximately 360 gamma-ray logs of the Dakota aquifer have been digitized in a nine-
township area of Hodgeman County (Figure 11). These logs provide an extremely high
density of control compared with the average state coverage of the Dakota (one well per
township). At each well location stratigraphic tops have been picked as lithostratigraphic
subdivisions of the Lower Cretaceous section. These tops and their codes are shown in
Figure 12, where they are matched with their corresponding Dakota aquifer subdivisions.
The data set will be analyzed with respect to vertical, lateral, and directional change in the
distribution of sandstones and shales. The initial phase of the project should be completed
by the fall of 1993. The database construction is almost complete, and the analytical
strategy has been determined. The data will be analyzed by Ling Bian (Geography
Department, University of Kansas) using geostatistical methods to determine the degree of
lateral continuity and orientation in the Dakota aquifer units. The results will have
important implications regarding the design and performance of the dynamic models that
will be applied to simulate flow in the Dakota.
Figure 11--Distribution of digitized gamma-ray logs in the Hodgeman County intensive study area.
Figure 12--Formation elevations in Hodgeman County data set referenced with lithostratigraphy and aquifer subdivisions.
