Evaluation of Geophysical Methods for the
Detection of Subsurface Tetrachloroethylene in
Controlled Spill Experiments

Aldo Mazzella, U.S. EPA, ORD, NERL, ESD, CMB, Las Vegas, NV, and
Ernest Majer, Lawrence Berkeley National Laboratory, LBNL, Berkeley, CA

M/Yf i/"9 Tetrachloroethylene (PCE), a dense non-aqueous phase liquid (DNAPL), is a predominant contaminant of drinking water
aquifers. Effective remediation requires the location of the non-aqueous phase PCE in the subsurface.

Geophysical methods have the potential to detect subsurface DNAPLs in a safe, non-invasive, cost-effective manner.
^ number of controlled spill experiments were conducted in which measurements with 10 different geophysical methods
were made before, during, and after the injection of PCE into the subsurface. This approach clearly identifies any geophysical
anomaly associated with the PCE.

The Experiment

A number of sand and clay/sand layers
were constructed in a fiberglass tank at
LBNL.

Geological formation
cross-section.

E 50

¦B 100

CL
V

o 150

3 % clay

6% clay

4 sand

In May 2004,85 liters of PCE was injected into the
subsurface over a period of 26 hours.

Eleven scientists from the U.S. EPA, LBNL and the
U.S.Geological Survey (USGS) obtained measurements
with 10 different geophysical methods.

100	150	200

Diameter (cm)

s'k

The geophysical methods evaluated were surface ground penetrating radar (GPR), cross
borehole GPR, directional borehole GPR, borehole dielectric tool, high frequency and very
early time electromagnetic systems, cross borehole seismic, complex resistivity, self potential
and borehole video.

Results Preliminary results indicate that most of the methods observed some anomaly associated with the PCE. Partial data from
the surface GPR, seismic and complex resistivity methods are shown below taken at prespill, spill and post spill periods.
These three methods respond to different intrinsic physical properties of the formations.

The seismic and complex resistivity data shown below were taken at the top of the 3%
clay/sand layer, 50 cm depth. Both methods show a large anomaly during the spill period,
returning almost to background after the spill stopped.

Prespill ¦ 2 hours

Horizontal distance (cm)

50 100 150 200 cm

GPR data show a
large anomaly
during spill at 50
cm depth, post spill
data show reduction
in the anomaly.

Borehole video
indicated that 4 hours
after the spill started, a
4 cm thick layer of PCE
had accumulated and
spread across the 3%
clay/sand layer at 50 cm
depth

Seismic Cross Borehole

Complex Resistivity Cross Borehole
Dipole • DipoPe 10 Hz data at 50 cm depM

+9 hours after spill started

Summary

+68 hours after spill started
(+42 hours after spill stopped)

120 -60 -40 0 40 80 120 160
Tlron (hour*)

1)

resistivity h
hps^II start*-sJc>j»

-phase

Anomalies were observed with multiple geophysical methods indicating changes occurred in different physical
properties of the formations with the presence of the PCE. A monitoring approach utilizing multiple, different geophysical
methods and scientific expertise could provide unique detection and identification of subsurface PCE.

Notice: Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy. Mention of trade n

>r commercial products does not constitute endorsement or recommendation by EPA for u:

ho

o
o
cn

Collaborative Science
for Environmental Solutions

-------