Seismic Refraction and Resistivity Imaging of Shallow Sedimentary/Volcanic Interfaces Beneath the Western Snake River Plain Near Orchard, Idaho

Publication Date


Type of Culminating Activity


Degree Title

Master of Science in Geophysics



Major Advisor

Paul R. Donaldson


Geophysical investigations of subsurface lithologic units in regions of thick, unsaturated, and poorly consolidated sediments have always been a problem in exploration geophysics. An experimental high-resolution seismic refraction profile and four geoelectric surveys were acquired in a problem area of this nature near Orchard, Idaho. The objective was to test the viability of using these techniques to do detailed subsurface mapping of sedimentary/volcanic interfaces underlying this portion of the western Snake River Plain. Although only standard methods were employed, the refraction and geoelectric surveys produced results which are internally consistent and consistent with information obtained from nearby boreholes.

The refraction survey was conducted at a location locally referred to as Orchard Ranch. Eight 24-channel geophone spreads, with a 5-m station interval, were used to acquire the data resulting in 920 m of subsurface coverage. A buried explosive energy source was utilized at inline offsets of 100 m and 0 m from both ends of each receiver spread.

Data analysis was conducted using two independent processing techniques including the reciprocal method (RM) and a refraction statics software package developed by Green Mountain Geophysical Corporation. The RM technique computes refractor depths as the product of a time-depth and depth conversion factor for every receiver. Refractor depths were computed from north to south with forced depth ties between spreads at stations of overlap. The Green Mountain technique uses a ray theory model of seismic waves in the near surface layers of the earth to estimate layer geometries and velocities. Refractor depths are derived from time-terms computed for each receiver, and time differences between each shot/receiver pair. This method uses constant layer velocities in the computation.

The refractor depth models generated from the RM and the Green Mountain analyses reveal very similar features: a slightly irregular refractor surface at depths of 30-45 m with an abrupt discontinuity in the vicinity of Station 87. This discontinuity has been interpreted as representing either a down-to-the-south normal fault or a flow edge of a younger basalt overlying a deeper, and older, basalt unit.

A second study conducted at the Orchard Ranch site involved the acquisition, processing, and interpretation of resistivity data collected from four separate geoelectric soundings to map stratigraphic boundaries in the subsurface, both laterally and with depth. These included a single point Schlumberger vertical sounding, 20-m and 60-m dipole-dipole profiles, and a controlled source audiomagnetotelluric (CSAMT) survey.

The first set of resistivity data was acquired using a symmetric expanding Schlumberger electrode configuration running parallel to a seismic refraction line acquired in a previous study (Howarth, 1989, Line 1), and centered about the midpoint (Station 100) between the PVC and the Orchard Ranch wells. One-dimensional inversion of the data using an automatic interpretation program (Zohdy, 1989) resulted in a 16-layer resistivity model. By imposing constraints based on the available geologic information, an equivalent four-layer model was established using an iterative process of inversion and forward modeling. The resultant model reveals three layers of differing resistivities within the unconsolidated sediments in the near surface, underlain by a highly resistive basalt at a depth of 30 m.

The second set of resistivity data was acquired using a dipole-dipole electrode configuration. Two separate surveys, using 20-m and 60-m electrode separations (a-spacings), were conducted parallel to Line 1 beginning at the PVC well and ending near the Orchard Ranch well, providing resistivity data both laterally and vertically. Measurements were made at each field setup corresponding to n-spacings of 1, 2, 3, and 4. Pseudo-sections of apparent resistivity versus n-spacing allowed for qualitative interpretations of the data. In general, the near-surface sediments contain a shallow layer of high apparent resistivity, overlying a section of relatively low apparent resistivity. The absence of an abrupt resistivity increase in the deepest resistivity samples (60-m dipole pseudo-section) may indicate that the basalt was not reached.

The last survey was a CSAMT profile conducted along the entire length of the Line 1 refraction survey. A transmitting current electrode separation of 1525 m, located 1825 m due east of, and parallel to, refraction Line 1, was used in conjunction with a potential electrode spacing of 20 m and a magnetic coil to record orthogonal horizontal components of the electric and magnetic fields at 47 surface locations. This current and potential electrode separation distance was in excess of the 3 skin depth (far field) minimum separation necessary to ensure plane wave behavior for the resistivity and phase measurements.

The CSAMT resistivity and phase data were processed with Zonge Engineering DATPRO software. A 2-D smooth model inversion algorithm generated an earth resistivity model for a depth range of 50-400 m. This range likely falls entirely within the massive basalt unit as suggested from well logs. Two anomalous features stand out in the model. A region of extremely high resistivity in excess of 6000 Ω-m (as opposed to an average basalt resistivity of 200 Ω-m) appears near the vicinity of the Orchard Ranch well. This is most likely due to recording in a scrap metal dumping site near a steel cased well which has significantly contaminated the data and caused an unreliable inversion. A second notable feature is the presence of elevated resistivity vertical zonation at various locations along the profile. This has been interpreted as representing air-filled porosity within the basalt, possibly related to fractures or large void spaces.

Files over 30MB may be slow to open. For best results, right-click and select "save as..."