Outcrop-Scale Ground-Penetrating Radar and Seismic Imaging of Unsaturated Lacustrine Delta Sediments

Publication Date


Type of Culminating Activity


Degree Title

Master of Science in Geophysics



Major Advisor

Spencer H. Wood


Lacustrine delta depositional environments are known as major targets for hydrocarbon reservoir and groundwater aquifer explorations. Outcrop-scale geophysical imaging in complex structural and stratigraphical lacustrine environments is also difficult due to extreme contrasts in physical property values, poor source and receiver coupling, and low penetration and resolution. Knowledge of geophysical imaging capabilities and outcrop-scale studies for such environments, therefore, are very important to identify structures and strata of the sediment deposits.

The purposes of this thesis are to map geology of a complex stratigraphic outcrop containing poorly-consolidated unsaturated lacustrine delta sediments, to obtain subsurface ground-penetrating radar (GPR) and seismic images of a well-exposed section of this outcrop, and to examine methods to improve acquisition designs and processing parameters of GPR and seismic data to increase the resolution and image quality of the data.

The well-exposed outcrop is located on the Sommercamp Road, Marsing, Idaho. The site lies on the southwest side of the Western Snake River Plain, a Neogene-aged intracontinental rift basin, in southwest Idaho. The outcrop contains a 10 m exposure of an unsaturated lacustrine delta sequence with topset beds tectonically tilted 10° , underlain by 25° dipping foreset sand beds with mud drapes, and cut by a near-vertical fault. Beneath the foresets are lacustrine clay sediments over an irregular rhyolite bedrock surface.

I collected data using GPR and seismic reflection methods. For GPR data acquisition, I used 50, 100, and 200 MHz frequencies. I processed the GPR data using the following processing sequence: dewow, first-arrival time alignment, automatic gain control, bandpass filter, eigenvector filter, and migration. The processed 100 MHz GPR data show the best reliably images of the dipping subsurface structures of the sediments with up to the depth of 6 m using a constant velocity of 0.12 m/ns in dry sands. The 200 MHz data show the dipping subsurface structures up to only 3 m deep using the same velocity. The 50 MHz data show the deepest signal penetration up to more than 20 m but they poorly resolve the dipping structures of the subsurface.

I collected the seismic reflection data using two sources: a .223 rifle gun and a 25-lb sliding-dropped hammer. Seismic shot gathers from both sources show that ground roll is prominent in the shot gathers with velocities ranging from 150-300 m/s. Refraction analysis from the shot gathers shows that reflection velocities range from 350-1200 m/s. Reflections from subsurface interfaces may appear in the shot gathers with low velocities and low frequencies thus making it very difficult to distinguish reflections from the ground roll. Frequency spectra from the shot gathers of both sources show that the hammer source may contains lower reflection frequencies than the rifle sources. Due to the difficulty in satisfactorily separating the ground roll from the reflection events, I could not produce good subsurface images from the seismic data in this complex geology area.

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