HIGH-RESOLUTION SEISMIC REFLECTION SURVEY AT DOVER AFB: A COMPARISON OF THREE SEISMIC SOURCES Steve Cardimona Boston College, PL/GPE, 29 Randolph Road, Hanscom AFB, MA 0 173 1 Katharine Kadinsky-Cade Earth Sciences Division, US Air Force Phillips Lab, 29 Randolph Road, Hanscom AFB, MA 0173 1 Richard Miller Kansas Geological Survey, 1930 Constant Ave., Lawrence, KS 66047 Jay J. Pulli and Walter Turpening Elohi Geophysics, Inc., 11606 Brook Meadows, Meadows, TX 77477 ABSTRACT In June of 1995, the Earth Sciences Division of the Air Force Phillips Lab, with survey equipment from the University of Delaware and assisted by the Kansas Geological Survey and Elohi Geophysics, conducted a geophysical site characterization of the SERDP-funded Groundwater Remediation Field Lab (GRFL) located at Dover AFB, Delaware and administered by Applied Research Associates for USAF Armstrong Lab. Seismic data were collected in order to 1) compare the results using three different compressional sources and 2) cover the field site well enough to characterize the seismic response of the shallow subsurface. This paper will focus primarily on the first of these two goals. Seismic data were collected along three north-south profiles set 10 meters apart, each profile with a different compressional source: a 5.5kg sledgehammer, a 12-gauge firing rod from Betsy Seisgun Inc. shooting 150 grain blanks, and a portable piezoelectrically driven vibrator, developed by Elohi Geophysics, operating with a 90Hz-450Hz sweep. An east-west cross line was collected using the sledgehammer source in order to tie the three profiles together. A laser theodolite provided station location and elevation control. The primary targets were the water table (that had been marked on maps at a depth of about 3 meters) and a sand-clay interface at about 15 meters depth. We collected 24-channel CMP data using a half meter spacing of both source and 1OOHz geophones. An end-on spread geometry was used, with a 1 meter offset between source and nearest geophone. Field QC after initial walkaway noise testing with each source did not show any one source to be outstanding A practical early result of the seismic survey showed the water table to be at just over 10 meters. We have associated the strongest reflection event with the water-table interface. Seismic data comparison in this study is based on spectral content, total energy and signal-to-noise ratios, as well as a discussion of coherency of the primary reflection event at the water table. The problem with the water table being deeper than expected is that the water table reflection may interfere with the other primary seismic target, the sand-clay interface. With a wavelength of about 4 meters at 100 Hz, interpretation of the data must take into account the possible interference of the two reflections in the seismic images. With a surface velocity of 400m/sec, the first Fresnel zone for 1 OOHz signals at 15 meters depth is about 5.5 meters under each seismic line, therefore overlapping between profiles. Thus, despite the separation of the three lines, they are sampling similar regions of the target area. Nevertheless, initial inspection of the seismic shot gather data showed that they are characterized by rapid variations in amplitude and phase across short distances. Both the firing rod and the vibratory source gave an initial look at the near surface during data acquisition via the use of augers necessary for deployment of these sources. The site had rapid lateral 171 Downloaded 07/07/14 to 129.237.143.20. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/