xvEPA
          United States
          Environmental Protection
          Agency
            Office of Research and
            Development
            Washington DC 20460
EPA/540/R-95/520
August 1995
Site Characterization
Analysis Penetrometer
System (SCAPS)

Innovative Technology
Evaluation Report
                 SUPERFUND INNOVATIVE
                 TECHNOLOGY EVALUATION

-------
                                           CONTACT
Laiy Jack and Steve Rock are the EPA contacts for this report. Lary Jack is presently with the new
Characterization Research Division (formerly the Environmental Monitoring Systems Laboratory) in Las
Vegas, NV, which is under the direction of the National Exposure Research Laboratory with headquarters in
Research Triangle Park, NC.

 Steve Rock is presently with the new Land Remediation and Pollution Control Division (formerly the Risk
Reduction Engineering Laboratory) in the newly organized National Risk Management Research Laboratory
in Cincinnati, OH.

-------
                                EPA/540/R-95/520
                                August 1995
     SITE CHARACTERIZATION ANALYSIS
       PENETROMETER SYSTEM (SCAPS)

INNOVATIVE TECHNOLOGY EVALUATION REPORT
 NATIONAL RISK MANAGEMENT RESEARCH LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION AGENCY
              CINCINNATI, OHIO 45268
     NATIONAL EXPOSURE RESEARCH LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION AGENCY
             LAS VEGAS, NEVADA 89193
                                       Printed on Recycled Paper

-------
                                                Notice

The information in this document has been funded wholly or in part by the U.S. Environmental Protection Agency (EPA)
in partial fulfillment of Contract No. 68-CO-0047, Work Assignment NO. 0-40, to PRC Environmental Management, Inc.
It has been subject to the Agency's peer ad administrative reivew, and it has been approved for publication as an EPA
document. The opinions, findings, and conclusions expressed herein are those of the contractor and not necessarily those
of the EPA or other cooperating agencies. Mention of company or product names is not to be construed as an endorsement
by the agency.

-------
                                               Foreword
    The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air, and water
resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions leading
to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this
mandate, EPA's research program is providing data and technical support for solving environmental problems today and
building a science knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect
our health, and prevent or reduce environmental risks in the future.

    The National Risk Management Research Laboratory is the Agency's center for investigation of technological and
management approaches for reducing risks from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air, land, water and subsurface resources;
protection of water quality in public water systems ; remediation of contaminated sites and ground water, and prevention
and control of indoor air pollution. The goal of this research effort is to catalyze development and implementation of
innovative, cost-effective environmental technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy  decisions; and provide technical support  and information transfer to ensure effective
implementation of environmental regulations and strategies.

    This publication has been produced as part of the Laboratory's strategic long-term research plan. It is published and
made available by EPA's Office of Research and Development to assist the user community and to link researchers with their
clients.
                                          E. Timothy Oppelt, Director
                                          National Risk Management Research Laboratory

-------
                                                 Abstract

 In August 1994, a  demonstration of cone penetrometer-mounted sensor technologies took place  to evaluate their
 effectiveness in sampling and analyzing the physical and chemical characteristics of subsurface soil at hazardous waste sites.
 The effectiveness of each technology was evaluated by comparing each technology's results to the results obtained using
 conventional reference methods. The demonstration was developed under the Environmental Protection Agency's Superfund
 Innovative Technology Evaluation Program.

 Three technologies were evaluated: the Site Characterization and Analysis Penetrometer System (SCAPS) Laser Induced
 Fluorescence (LIF) sensor developed by the Tri-Services (Army, Navy, and Air Force), the Rapid Optical Screening Tool
 (ROST™) developed by Loral Corporation and Dakota Technologies, Inc., and the conductivity sensor developed by
 Geoprobe® Systems.  These technologies were designed to provide rapid sampling and real-time, relatively low cost analysis
 of the  physical and chemical  characteristics  of subsurface soil to  quickly distinguish  contaminated, areas  from
 noncontaminated areas. Results for the ROST™ and Geoprobe® technologies are presented in separate reports similar to
 this one.

 Three sites were selected for the demonstration, each contained varying concentrations of coal tar waste and petroleum fuels,
 and wide ranges in soil texture.

 This demonstration found that the SCAPS technology produced screening level data. Specifically, the qualitative assessment
 showed that the stratigraphic and chemical cross sections from SCAPS technology were comparable to the reference
methods. The technology's identification of the relative magnitude of contamination generally matched the reference data.
The quantitative assessment found that the SCAPS data was most closely correlated to the total petroleum hydrocarbons and
volatile petroleum hydrocarbons data. Based on this study, the SCAPS technology appears to be capable of rapidly and
reliably mapping the relative magnitude of the vertical and horizontal extent of subsurface contamination, when that
contamination is fluorescent. This type of contamination includes petroleum fuels and polynuclear aromatic hydrocarbons.
                                                     IV

-------
                                     Table of Contents
Section                                                                                  Page

Notice  	  ii
Foreword  	Hi
Abstract	iv
List of Figures	vii
List of Tables	vii
List of Abbreviations and Acronyms	viii
Acknowledgements	  x

1   Executive Summary	  1

2   Introduction	   3
       Demonstration Background, Purpose, and Objectives 	   3
       Demonstration Design	   4
               Qualitative Evaluation	   4
               Quantitative Evaluation  	6
       Deviations from the Approved Demonstration Plan	7
       Site Descriptions  	   7

3   Reference Method Results	9
       Reference  Laboratory Procedures	9
               Sample Holding Times  	9
               Sample Preparation	9
               Initial and Continuing Calibrations	 10
               Sample Analysis	 10
               Detection Limits 	11
               Quality Control Procedures	 11
               Confirmation of Analytical Results	12
               Data Reporting	12
       Quality Assessment of Reference Laboratory Data	 12
               Accuracy	 12
               Precision	 12
               Completeness 	12
        Use of Qualified Data for Statistical Analysis	12
        Chemical Cross Sections 	13
               Atlantic Site	13
               York Site	13
               Fort Riiey Site	17
        Quality Assessment of Geotechnical Laboratory Data	17
               Geotechnical Laboratory	 17
               Borehole  Logging	17
               Sampling Depth Control 	17
                                               v

-------
                             Table of Contents (Continued)
 Section
                                                                                      Page
        Stratigraphic Cross Sections	17
               Atlantic Site	
-------
                                       List of Figures
2-1    Typical Transect Sampling Line and Stratified Random Sampling Grid	6
3-1    TPH Reference Method Chemical Cross Section — Atlantic Site	14
3-2    PAH Reference Method Chemical Cross Section — Atlantic Site  	14
3-3    TPH Reference Method Chemical Cross Section — York Site	15
3-4    PAH Reference Method Chemical Cross Section —York Site	15
3-5    TPH Reference Method Chemical Cross Section — Fort Riley Site	 16
3-6    PAH Reference Method Chemical Cross Section — Fort Riley Site	 16
3-7    Reference Method Stratigraphic Cross Section — Atlantic Site	 18
3-8    Reference Method Stratigraphic Cross Section — Yorlc Site	 19
3-9    Reference Method Stratigraphic Cross Section — Fort: Riley Site  	 19
4-1    Tri-Services SCAPS	24
4-2    SCAPS Panel Plot — Node 4 Atlantic Site	24
4-3    SCAPS Chemical Cross Section — Atlantic Site			29
4-4    SCAPS Chemical Cross Section —York Site 	29
4-5    SCAPS Chemical Cross Section — Fort Riley Site	30
4-6    SCAPS Stratigraphic Cross Section — Atlantic Site	30
4-7    SCAPS Stratigraphic Cross Section — York Site	33
4-8    SCAPS Stratigraphic Cross Section — Fort Riley Site	33
4-9    Fluorescence Intensity vs. Wavelength — Node 4 Atlantic Site  	34
5-1    Normalized LIF and Qualitative Reference Data — Atlantic Site	38
5-2    Normalized LIF and Qualitative Reference Data — Yoirk Site	38
5-3    Normalized LIF and Qualitative Reference Data — Fort Riley Site	39
                                       List of Tables
Table

2-1
3-1
4-1
4-2
4-3
5-1
5-2

5-3
                                                                                  Page
Criteria for Data Quality Characterization Site	
Comparison of Geologist Data and Geotechnical Laboratory Data — All Sites ...
Quantitative Evaluation Data for the Atlantic Site	
Quantitative Evaluation Data for the York Site	
Quantitative Evaluation Data for the Fort Riley Site	
Regression Analysis Results for SCAPS and the Reference Methods — All Sites
Regression Analysis Results for the Average of Both SCAPS Pushes
       and the Reference Methods — All Sites	
Data for Mean SCAPS, TPH, and VPH — All Sites	
.  5
20
28
28
28
42

44
45
                                              VII

-------
            List of Abbreviations and Acronyms
AEG          Army Environmental Center
ASTM        American Society for Testing Materials
bgs           below ground surface
BTEX        benzene, toluene, ethylbenzene, and xylene
CCAL        continuing calibrations
cm           centimeter
cm/s          centimeters per second
CP           cone penetrometer
DQO          data quality objective
EPA          Environmental Protection Agency
ERA          Environmental Resource Associates
ETS          Environmental Technical Services
FID           flameionization detector
FMGP        former manufactured gas plant
GC           gas chromatograph
Geoprobe®    Geoprobe®Systems
HPLC        high performance liquid chromatography
Hz           pulses per second (hertz)
ICAL          initial calibrations
ITER          innovative technology evaluation report
LCS          laboratory control samples
LIF           laser induced fluorescence
MDL          method detection limit
Method OA-1   University of Iowa Hygienics Laboratory Method OA-1
              micrograms per kilogram
              micrograms per liter
              micrometer
mg/kg        milligrams per kilogram
mg/mL        milligrams per milliliter
Mj            megajoule
mL           milliliter
mm           millimeter
MMTP        Monitoring and Measurement Technologies Program
MS           matrix spike
MSD          matrix spike duplicate
NRMRL       National Risk Management Research Laboratory
NERL-CRD    National Exposure Research Laboratory-Characterization Research Division
nm           nanometer
OMA          optical multichannel analyzer
%D           percent difference
%RSD        percent relative standard deviation
PAH          polynuclear aromatic hydrocarbon
y^g/L
                                  viii

-------
        List of Abbreviations and Acronyms (Continued)
PARC         Princeton Applied Research Company
PDA          photodiode array
PE            performance evaluation
PID           photoionization detector
ppm          parts per million
PRC          PRC Environmental Management, Inc.
PRL          PACE reporting limit
PTI           Photon Technology, Inc.
QA/QC        quality assurance/quality control
QAPP         quality assurance project plan
ROST™       Rapid Optical Screening Tool
RPD          relative percent difference
RSD          relative standard deviation
SCAPS Site Characterization and Analysis Penetrometer System
SITE          Superfund Innovative Technology Evaluation
TER          technology evaluation record
TOC          total organic carbon
TPH          total petroleum hydrocarbon
USCS         Unified Soil Classification System
USDA         United States Department of Agriculture
VOC          volatile organic compound
VPH          volatile petroleum hydrocarbon
WES          Waterways Experiment Station
                                  IX

-------
                                        Acknowledgements

We wish to acknowledge the support of all those who helped plan and conduct this demonstration, interpret data, and prepare
this evaluation report In particular, for demonstration site access and relevant background information, Dean Harger (Iowa
Electric Company), Ron Buhrman (Burlington Northern Railroad), Abdul Al-Assi (U.S. Army Directorate of Engineering
and Housing); for turn-key implementation of this demonstration, Eric Hess, Darrell Hamilton, Harry Ellis (PRC
Environmental Management, Inc.) (913) 281-2277); for editorial and publication support, Suzanne Ladish and Frank
Douglas (PRC); for technical and peer review Dr. T. Vo-Dinh (Oak Ridge National Laboratory), Robert Knowlton (Sandia
National Laboratories), and JeffKelley (Nebraska Department of Environmental Quality); and for EPA project management,
Lary Jack (National Exposure Research Laboratory-Characterization Research Division) (702)798-2373). In addition, we
gratefully acknowledge the participation of the developers, the Tri-Services Site Characterization Analysis and Penetrometer
System (SCAPS) group (410)612-6836).

-------
                                               Section 1
                                        Executive Summary
    Recent  changes hi environmental site character-
ization  have  resulted  hi  the  application  of cone
penetrometer (CP) technologies to site characterization.
With a variety of in situ physical and chemical sensors,
this technology is seeing an increased frequency of use
in environmental site characterization. CP technologies
employ a wide array of sampling tools and  produce
limited investigation-derived waste.

    The  EPA's  Monitoring   and  Measurement
Technologies  Program  (MMTP)   at  the  National
Exposure Research Laboratory, Las Vegas,  Nevada,
selected CP sensors as a technology class to be evaluated
under the Superfund Innovative Technology Evaluation
(SITE) Program. In August 1994, a demonstration of
CP-mounted sensor technologies took place to evaluate
how effective they were in analyzing the physical and
chemical characteristics of subsurface soil at hazardous
waste sites.  Prior to  this demonstration, two separate
predemonstration sampling efforts were  conducted  to
provide the developers with site-specific samples. These
samples were  intended to provide data for site-specific
calibration of the technologies and matrix interferences.

    The main objective of this  demonstration was  to
examine technology performance by comparing each
technology's results  relative to physical and  chemical
characterization techniques obtained using conventional
reference methods.    The  primary focus  of the
demonstration was to evaluate the  ability of the
technologies  to  detect  the  relative magnitude   of
fluorescing subsurface  contaminants.  This evaluation is
described in this report as the qualitative evaluation.  A
subordinate  focus  was  to  evaluate  the  possible
correlations  or   comparability  of  the  technologies
chemical  data  with   reference  method  data.  This
evaluation is described hi this report as the quantitative
evaluation.  All of the technologies were designed and
marketed to produce  only qualitative screening data.
The  reference methods for evaluating  the  physical
characterization  capabilities  were  stratigraphic logs
created, by a geologist from soil samples collected by a
drill rig  equipped with hollow stem augers, and soil
samples analyzed by a geotechnical laboratory.  The
reference  methods   for   evaluating  the  chemical
characterization  capabilities   were   EPA  Method
418.1  and SW-846 Methods  8310  and  8020, and
University of Iowa Hygienics Laboratory Method OA-1.
In addition, the effect of total organic carbon (TOC) on
technology performance was evaluated.

    Three  technologies  were  evaluated:   the Site
Characterization and  Analysis Penetrometer  System
(SCAPS)  laser induced fluorescence  (LIF) and CP
sensors developed by the Tri-Services (Army, Navy, and
Air Force), the Rapid Optical Screening Tool (ROST™)
developed  by    Loral  Corporation  and  Dakota
Technologies,   Inc.,   and  the  conductivity   sensor
developed by Geoprobe® Systems.  These technologies
were designed to provide real-tune, relatively low cost
analysis of the physical and chemical characteristics
(primarily petroleum fuels and coal tars) of subsurface
soil to quickly distinguish  contaminated  areas from
noncontaminated  areas.   The SCAPS  technology  is
designed and operated to produce screening level data.
Results  of  the demonstration  are  summarized  by
technology and by data type (chemical or  physical)  hi
individual innovative   technology  evaluation  reports
(ITER).  In addition to the three  technology-specific
ITER's,   a  general   ITER  that  examines   cone
penetrometry, geoprobes, and hollow stem auger drilling
hi greater detail has been prepared.

    The  purpose of this ITER is  to  chronicle the
development of the SCAPS technology, its  capabilities,
associated equipment,  and accessories.  The document
concludes with an evaluation of how closely the  results
obtained  using the technology compare to the  results
obtained using conventional reference methods.

    One hazardous waste site each was selected hi Iowa,
Nebraska, and Kansas to demonstrate the technologies.

-------
 The  sites were  selected because of  thek varying
 concentrations of coal tar waste and petroleum fuels, and
 because of thek ranges hi soil textures.

    This   demonstration  found  that  the  SCAPS
 technology produces screening level data.  Specifically,
 the qualitative assessment showed that the stratigraphic
 and the chemical cross sections were comparable to the
 reference methods. The SCAPS sensors did not requke
 sample collection,  and thus,  avoided  the  sampling
 difficulties encountered by the reference methods during
 this  demonstration.   The relatively continuous data
 output  from the  LIF  sensor  eliminated  the  data
 interpolation requked by the  reference method.  This
 also increased the apparent resolution of the sensor's
 data.

    The  SCAPS  LIF   operator  also  qualitatively
 identified changes in contaminant type by  detecting
 significant changes hi peak emission wavelength.  The
 gross soil classifications identified by the technology
 generally matched the reference method classifications.
 The chemical cross sections for the LIF sensor showed
 close agreement to the reference method cross sections
 in  identifying  low,  medium,  and high zones  of
 contamination.  Generally, the relative LIF intensity was
 positively related to the concentration of total petroleum
 hydrocarbons   and   total   polynuclear   aromatic
 hydrocarbons.     In  only  one  case   during   this
 demonstration did the SCAPS LIF sensor not identify
 fluorescence above background for zones sampled that
 indicated  contamination.  Reference method sampling
 indicated  contamination hi the 100 's of the  parts per
 million (ppm) range at Node 5  at the York site.  The
 failure of the SCAPS LIF sensor to identify this  zone
 may have been a result of the  horizontal separation
 between the SCAPS and reference method  sampling
 points,  and inherent  matrix  heterogeneity.   The
 quantitative assessment found that the SCAPS LIF data
 was most closely  correlated to the TPH and volatile
 petroleum hydrocarbons  (VPH) data.  Due to matrix
heterogeneity up to 50 percent of die original data set
used in the quantitative  evaluation was  eliminated as
 outliers. This greatly reduced the predictive value of the
 regression models,  however, the remaining data was still
used to identify trends. The quantitative data assessment
 also  produced  a  first approximation  of a  detection
threshold  for the SCAPS LIF sensor.  For TPH and
VPH, based on thek regression models, the fluorescence
intensity (background corrected) at 0  milligram per
kilogram was 157 and 336, respectively. In addition, the
lowest concentrations of TPH and VPH detected during
the quantitative  assessment were 60 and  19 mg/kg,
respectively.   Both of these low concentrations  had
fluorescence intensity  readings  near the  thresholds
(157 and 336) discussed above.

    Based on the continuous data output for both the
chemical  and physical properties of soil, the SCAPS
sensors (physical and chemical) appear to be valuable
tools for qualitative site characterization.  The lack of
better correlation for the quantitative evaluation cannot
be solely  attributed to the technology. It may also be
due to the combined effect of matrix heterogeneity, lack
of instrument calibration,  uncertainties regarding the
exact contaminants being measured,  and the age  and
constituents hi the waste.  Based on the data from this
demonstration, it is not possible to conclude that the
technology  can  or  cannot  be  quantitative  hi  the
configuration used during this demonstration. Based on
the effects listed above, potential users should not expect
the SCAPS LIF sensor to produce data which shows a
high  degree of correlation  when  comparisons  with
conventional data are made on a point-by-point basis.
Verification of this technology's performance should be
done only on a qualitative level.  Even though it cannot
quantify actual levels  of  contamination or identify
individual   compounds,   it   can  produce relative
contaminant  distribution   data  very   similar   to
corresponding data produced  by conventional methods,
such as drilling and laboratory sample analysis, and it
can monitor changes hi emission wavelength to identify
possible changes  hi  contaminant constituent.    The
general magnitude of the LIF sensor  data directly
correlated to the general magnitude of contamination
detected by the  reference  method.    The SCAPS
performance during this demonstration showed that it
could  generate  this  data  faster than the  reference
methods and with little to no waste generation relative to
the reference methods.  The cost associated with using
this  technology to produce  the qualitative  data  was
approximately $42,000, as compared to the $55,000 used
to produce the reference cross sections,  hi this case, the
SCAPS LIF and CP  sensors cost less than reference
methods,  produced almost  1,200 more  data  points
(continuously) than the  conventional  approach,  and
provided data in a real-tune fashion.  It should be noted
that the technology's data is screening level, while the
reference  method  approach produced  definitive data.
The question that this demonstration cannot  answer is
whether or not it is better to have few data points at the
highest data quality level or many more at a lower data
quality level. Issues such as matrix heterogeneity  may
greatly  reduce the need for definitive level data hi an
initial site characterization. Critical samples will always
requke definitive analysis.

-------
                                                Section 2
                                              Introduction
     The purpose of this ITER is to present information on
 the demonstration of the SGAPS LIF and CP sensors, a
 system designed to provide screening type data on the
 physical and chemical characteristics of subsurface soil.
 This  system uses  laser  light to  cause  fluorescing
 contaminants  hi soils to fluoresce  and measures  the
 resulting fluorescence. Currently, this technology is being
 used most commonly to detect PAH compounds associated
 with petroleum fuel.

     More detailed information regarding aspects of this
 report can  be found hi the  January 1995  technology
 evaluation record (TER) for this demonstration. The TER
 contains all of the raw data and is not intended for general
 circulation, however, portions of the TER can be accessed
 by contacting the EPA technical project manager.

    The SCAPS sensors were demonstrated in conjunction
 with  two other  sensor technologies:  (1)  the  ROST™
 developed by Loral Corporation and Dakota Technologies,
 Inc.,  and  (2) the  conductivity • sensor developed  by
 Geoprobe®.   The results of the demonstration of these
 other two technologies are presented in individual ITERs
 similar to this document. An additional general ITER was
 prepared which discusses the history, sampling, and other
 capabilities of cone penetrometry, Geoprobe®,  and hollow
 stem   auger  drilling.     Complete  details  of  the
 demonstration,   descriptions  of the  sites,  and  the
 experimental design are provided hi the August 1994 final
 demonstration plan  for geoprobe-  and  CP-mounted
 sensors.  This information is briefly summarized hi this
 document.

    This section summarizes general information about
the demonstration such as the purpose, objectives, and
design.  Section 3 presents and discusses the validity of
the data produced by the reference  methods used hi the
evaluation of two SCAPS sensors:  the LIF sensor and the
CP sensor. Section 4  discusses the SCAPS sensors, their
capabilities,  and  equipment and accessories.  Section 5
evaluates how closely the results  obtained using the
SCAPS sensors compare to the results obtained using the
reference  methods.   Section 6  discusses the potential
 applications    of    the    technology.        Section
 7  presents  developer  comments,  EPA response  to
 developer comments,  and developer update  on  the
 technology.

 Demonstration Background, Purpose,
 and Objectives

     The  demonstration  was  developed  under  the
 Measuring  and   Monitoring   Technologies  Program
 (MMTP), a component of the EPA's SITE Program. The
 goal of the MMTP is to identify and demonstrate new,
 viable technologies that can identify, quantify, or monitor
 changes hi contaminants at hazardous waste sites or that
 can be used to characterize a site cheaper, better, faster,
 and safer than conventional technologies.
     The SCAPS LIF sensor uses LIF  to  detect the
subsurface presence or absence of fluorescing compounds,
such as petroleum fuels  and coal tar wastes.   This
technology is attached to and advanced into the soil with
a  conventional CP sensor.   The SCAPS  LIF and CP
sensors, were designed to provide rapid, continuous, hi situ
real-time, relatively low cost analysis of the physical and
chemical  characteristics   of   subsurface  soil.    The
identification  of subsurface   chemical characteristics
involves quickly identifying the presence or absence of
contamination, and relative  concentrations.    These
capabilities would allow  investigation and remediation
decisions  to be  made more efficiently  and quickly,
reducing overall  project  costs such  as the number of
samples that need to be submitted for  costly and time
consuming confirmatory analyses, and  costs associated
with multiple mobilizations.

    The primary focus of the  demonstration was  to
evaluate the ability of the technologies  to  detect the
relative magnitude of fluorescing subsurface contaminants.
This evaluation is described in this report as the qualitative
evaluation.  A subordinate focus was  to  evaluate the
possible correlations or comparability of the technologies
chemicid data with reference method data. This evaluation
is  described hi this report as the quantitative evaluation.

-------
evaluate the possible correlations or comparability of the
technologies chemical data with reference method data.
This  evaluation  is  described in this  report as the
quantitative evaluation.  All of the technologies were
designed  and  marketed to  produce only  qualitative
screening data.

    There were  three objectives for  the  qualitative
evaluations,  and  one objective for the quantitative
evaluation conducted during this demonstration.   The
first qualitative objective evaluated the SCAPS  LIF
sensor for its ability to vertically delineate  subsurface
soil contamination.   Cross sections  of  subsurface
contaminant plumes produced by the technology were
visually  compared  to  corresponding  cross  sections
produced by the reference  methods.    The second
qualitative objective evaluated the SCAPS CP sensor for
its  ability  to characterize  physical  properties  of
subsurface soils.  The third qualitative objective was to
evaluate  the  SCAPS  sensors for  their  reliability,
ruggedness, cost, and range of application. The SCAPS
LIF sensor was quantitatively evaluated on how its data
compared to the data from the reference methods, and an
attempt was made to identify the technology's threshold
detection limits.

Demonstration Design

    The experimental design of this demonstration was
created to meet the specific qualitative and quantitative
objectives described in Section 3.   The experimental
design was approved by all demonstration participants
prior to the start of the demonstration.  This experi-
mental design is detailed in the final demonstration plan
(PRC 1994).

    Sample  results  from  the SCAPS  sensors  were
compared to results from the reference methods.  For
this demonstration,  the reference methods  included
standard  SW-846 methods  for measuring  petroleum
hydrocarbons and  PAHs,  and  borehole logging and
sampling  by  a geologist  using hollow stem auger
drilling.   These  comparisons  are called intramethod
comparisons.    These  comparisons  were  used to
determine the  quality  of  data produced  by  the
technology.  Two data quality levels were  considered
during this evaluation:  definitive and screening data.
These data quality levels are described in Data Quality
Objectives Process for  Superfund  -  Interim Final
Guidance (EPA 1993;.

    Definitive  data  are  generated using  rigorous
analytical methods, such as  approved EPA reference
methods.  Data are analyte-specific with confirmation of
analyte identity and concentration.  Methods produce
tangible raw data (e.g., chromatograms,  spectra, digital
values)  in   the   form   of   paper  printouts   or
computer-generated  electronic files.    Data may  be
generated at the site or at an off-site location, as long as
the   quality   assurance/quality   control   (QA/QC)
requirements are satisfied.  For the data to be definitive,
either analytical or total measurement  error must be
determined.

    Screening data are generated by rapid, less precise
methods  of  analysis  with  less  rigorous  sample
preparation. Sample preparation steps may be restricted
to simple procedures,  such as dilution  with a solvent
instead of  elaborate  extraction/digestion and cleanup.
Screening  data provide  analyte identification  and
quantification,  although the  quantification may  be
relatively  imprecise.   At  least 10  percent of the
screening data are confirmed using analytical methods
and  QA/QC procedures and  criteria associated with
definitive  data.   Screening data without  associated
confirmation data  are not considered to be of known
quality.

    Since  this  technology  is new  and  Innovative,
approved EPA methods  for in  situ  laser induced
fluorescence  do not exist.   For the purpose of diis
demonstration, the lack of approved EPA methods did
not preclude the  technology  from being  considered
definitive.  The evaluation of this technology as to its
quantitative capabilities was included to provide potential
users a complete picture of die technology's capabilities.
However,  the  developer   never claimed that  the
technology  was quantitative.  The main criteria for data
quality level assignment was based on the comparability
of the technology's data to the  data produced by the
reference methods.  Table  2-1  defines the statistical
parameters used  to define  die data  quality  levels
produced by SCAPS. These criteria were defined in the
approved demonstration plan, and  accepted by the
developers.  These  are based on past  SITE  demon-
strations of monitoring  technologies.

    The sampling  and analysis methods  used to collect
the baseline data for this demonstration are currently
accepted by EPA for providing legally defensible data.
This data is defined as definitive level data by Superfund
guidance.    Therefore,  for  the  purpose of  this
demonstration, these technologies  and analytical metiiods
were considered reference metiiods.

Qualitative Evaluation

    Qualitative  evaluations  were   made  through
observations  and  by   comparing  stratigraphic  and
chemical cross sections from  the technology to cross
sections produced from the reference methods.  The
reference method for the stratigraphic cross sections was

-------
TABLE 2-1. CRITERIA FOR DATA QUALITY CHARACTERIZATION SITE
Data
Quality
Level

Definitive
Screening
                                        Statistical Parameter

      r2 = 0.80 to 1.0, and the slope3 and y-intercept are statistically similar to 1.0 and 0.0, respectively, the
      precision (RSD) is less than or equal to 20 percent and inferential statistics indicate the two data sets
      are statistically similar.

      r2 < 0.80, the precision (RSD) is greater than 20 percent, and the technology meets its developer's
      performance specifications, inferential statistics indicate the two data sets are statistically not similar;
      or in the case where the regression analysis indicates the data is of definitive quality, but the
      inferential statistics indicate the data sets are statistically different.
Notes:
r2
RSD
Since the SCAPS technology did not produce data in equivalent units to the reference method, the slope cannot
be used to assess accuracy, however, comparability can still be evaluated.
Coefficient of determination.
Relative standard deviation.
continuous sampling with a hollow stem auger advanced
by  a drill rig and  the  corresponding borehole logs
created by a geologist.  In addition, the technology's
ability to determine  subsurface soil texture at discrete
intervals was further compared to data produced by an
off-site geotechnical laboratory.   Soil samples were
analyzed  for  total organic carbon  (TOC)  using the
90-3 Walkley-Black Method;  and soil texture analysis
was performed  by American  Society  of Testing
Materials (ASTM) Method D-422.

    The  reference  methods  for the  chemical cross
sections were subsurface sampling using a drill rig and
off-site sample  analysis by EPA Method 418.1  and
SW-846 Method 8310.  EPA Method 418.1 produces
data on TPH concentration. EPA Method 8310 produces
data on PAH concentrations.  These reference methods
were selected for the qualitative evaluation based on
recommendations made by the developer, consideration
of the types of fluorescing  target compounds, and the
project objectives.

    To qualitatively  assess the ability of the SCAPS LIF
sensor  to  identify the  presence  or  absence  of
contamination  and  produce  contaminant distribution
cross   sections,  the  technology  was required  to
continuously  sample at five points located along a
transect line at each of the demonstration sites (Figure
2-1).  These  points were called sample nodes.   The
transect line  was placed across an  area  of known
subsurface     contamination    identified     during
predemonstration sampling  activities and  previous
investi-gative sampling at these sites.  A 6-foot by 6-foot
area was marked around each sample node.  This area
was  subdivided into nine sections  of equal,  size,
identified as Sections A through I. At least one sample
                                              node per site was placed outside the area of contami-
                                              nation.

                                                   Once  each 6-foot by  6-foot area  was marked,
                                              sampling points for each technology and the reference
                                              methods were  assigned randomly at each node.  This
                                              produced a stratified random sampling design. Sections
                                              for sampling at each node were only used once.

                                                   The potential effect of organic matter was evaluated
                                              qualitatively by TOC  analysis of soil samples.  This
                                              evaluation was intended to examine potential trends
                                              between TOC content,  and how data analyzed using the
                                              technology compared to that obtained by the reference
                                              methods.

                                                   The chemical and geotechnical data generated by the
                                              technology was used to  produce  qualitative   data
                                              regarding contaminant and  stratigraphic cross sections
                                              along each transect line.   These cross sections  were
                                              compared to cross sections  generated by the reference
                                              method results from soil samples collected with a drill
                                              rig.   The comparison of  contaminant cross  sections
                                              involved  visual  comparisons  with  overlays,   and
                                              minimum and maximum depths of contamination at each
                                              location along a transect line.

                                                   Other factors that underwent qualitative evaluation
                                              were technology costs, ease of operation, ruggedness,
                                              instrument reliability, environmental sampling capability,
                                              and production rates.  PRC assigned a person to  work
                                              with the developer to become knowledgeable in the use
                                              and application of the SCAPS sensors.  Through this
                                              work, PRC was able to assess the operational factors for
                                              the technology.

-------
 FIGURE 2-1. TYPICAL TRANSECT SAMPLING LINE AND STRATIFIED RANDOM SAMPLING GRID
         STRATIFIED RANDOM SAUPLINR GRID
                                                        TYPICAL TRANSECT SAUPI INC I IMF
                                                LEGEND

                                                  1  NUU8ERED SAMPLING NOOC
                                                 «£,. CONTINUOUS VERTICAL MEASURING POINT FOR
                                                    QUALITATIVE ASSESSMENT

                                                 ^> TARGETED 6 INCH DEPTH INTERVAL FOR
                                                 ^^ OUANTITAIIVE ASSESSMENT
    During the demonstration, a total of 78 soil samples
were collected and analyzed by the reference methods,
and used in the  qualitative data evaluations.  These
samples  were  distributed  as  follows:  28 from  the
Atlantic site, 26 from the York site, and 24 from the
Fort Riley site.  Only sample data reported as positive
values were used in the evaluation. As described hi the
approved demonstration plan, sample data reported as
"not detected" were not used. As stated hi the approved
demonstration plan, the  elimination of these points are
not expected to  have a lesser  or  similar  effect  as
assigning arbitrary values to non-detects.

Quantitative Evaluation

    The SCAPS LJF sensor was evaluated quantitatively
on its ability to chemically characterize subsurface soil
contamination relative to classes of contaminants and
specific contaminants.   This evaluation consisted  of
comparing data generated using the technology to data
obtained using the reference methods  over a wide range
of concentrations.   The reference  method for  the
chemical cross sections soil sampling was hollow stem
auger drilling. University of Iowa Hygienics Laboratory
Method OA-1 (VPH), SW-846 Method 8020 benzene,
toluene,  ethylbenzene,  and  total xylenes  (BTEX),
SW-846  Method  8310  (PAH),  and  EPA  Method
418.1 (TPH) were used as  the  reference analytical
methods. This demonstration attempted to determine if
the results  from the SCAPS LIF sensor could be
correlated to results from the reference methods, and if
the technology was able to differentiate between different
types of contamination, such as PAHs, BTEX, coal tars,
and petroleum fuels.  In addition, PRC attempted to
determine the detection thresholds for  these classes of
contaminants.

    To quantitatively assess the comparability of the data
produced by the  SCAPS LIF sensor to the reference
methods" data, the demonstration plan required  the
technology to conduct its most accurate  and precise
measurements at discrete depths at each sampling node.
These depths represented zones of initial  contaminant
detection, medium, and high fluorescence. However, at
the start of the  demonstration, the developer of the
SCAPS sensors informed PRC that the data produced
during standard  dynamic  push mode was the most
accurate  data that could be produced.  Therefore, the
SCAPS LIF  data for quantitative evaluation was the
same as that used hi the qualitative evaluations.

    The locations for the reference method sampling for
the quantitative evaluation were selected after reviewing
the SCAPS and ROST* data for a site. Sample intervals
that showed similar data from both technologies were
selected  as  reference  method  sampling  intervals.
Reference method sampling intervals represented zones
of initial contaminant  detection,  medium, and high
                                                   6

-------
fluorescence.   The  data  produced at these  intervals
was used  to  quantify contamination,  identify con-
taminants,   establish  precision  control  limits,  and
establish contamination detection thresholds.  Reference
method data was used to assess the comparability of the
data produced by the SCAPS LIP sensor to reference
method chemical analysis.

    For the quantitative evaluation, data produced by the
SCAPS LIF sensor was averaged over a  12-inch push
interval corresponding to intervals sampled for reference
method analysis.  This data was used to determine a
mean fluorescence over that interval.  This data was
compared to corresponding mean reference method con-
centrations  for any given interval. To create these mean
reference method  concentrations,  PRC collected and
homogenized five  replicate samples from the 12-inch
depth intervals identified as reference method sampling
intervals which were chosen based on the SCAPS and
ROST™ data. Each replicate sample was collected from
a randomly assigned section at each sample node.

    The data developed by the SCAPS LIF sensor was
compared to reference method data for the following
compounds or classes of compounds: TPH, total BTEX,
VPH, total PAH,  and individual compounds (BTEX,
naphthalene, 1-methylnaphthalene, 2-methylnaphthalene,
acenaphthene,  fluoranthene, pyrene,  benzo-a-pyrene,
and anthracene). These comparisons were described in
the August 1994 demonstration plan.

    Method precision  also was  examined during the
demonstration. The SCAPS LIF sensor was required to
produce 10 replicate readings or measurements at given
depths without  moving the sensor between  readings.
From these 10  measurements  at each discrete depth,
precision control limits were established.

    For the quantitative evaluation, a total of 103 soil
samples were collected and analyzed by the reference
methods.   The distribution of these  samples was as
follows:    8 replicate sampling  intervals  producing
38 samples at the Atlantic site, 7 replicate sampling
intervals producing 35 samples at the York site, and
30 samples from 6 replicate sampling intervals at the
Fort Riley  site.  Only sample data reported as positive
values were  used hi  the  evaluation.  Sample data
reported as "not detected" was not used.

Deviations from  the Approved Demonstra-
tion Plan

    The   primary  deviation   from   the   approved
demonstration plan dealt with the statistical analysis for
the quantitative evaluation.
    Since the SCAPS sensors did not produce data
directly representing the concentration of contaminants,
or  data in the same units  as the reference  method
analysis, the Wilcoxon Rank  Sum Test could not be
used, .and the comparison of the technology's data to
reference method 99 percent confidence intervals was
not made.  In addition, the effect of soil moisture was
not examined due  to  the fact that the  bulk of the
contaminated  zones at each site were  at  or near
saturation.  Finally,  the approved  demonstration plan
identified a hydraulic probe sampler as the reference
method for collecting the soil  samples  used hi the
quantitative evaluations. However, due to sample matrix
affects, the hydraulic probe samples could not meet the
soil sampling objectives regarding sample volume. The
inability of this method to produce full sample recovery
was caused by the saturated  fine sands encountered at
many of the target sampling depths.  To allow for
adequate sample volume, PRC changed the reference
method for this soil sampling to hollow stem augering
and split spoon sampling.

Site Descriptions

    The demonstration took place at three sites within
EPA   Region  7.     The   three  sites  are  the
(1) Atlantic-Poplar  Street Former  Manufactured Gas
Plant (FMGP) site (Atlantic site), (2) York FMGP site
(York site), and (3) the Fort Riley Building 1245 site
(Fort Riley  site).   Brief summaries for each site are
given below.  Complete details are located in the August
1994 final demonstration plan.

    The Atlantic site is located hi Atlantic, Iowa. The
site is surrounded by gas stations, grain elevators, a seed
supply  company,  and  a railroad  right-of-way.   All
structures associated with  the FMGP   have  been
demolished.  A gas station now operates on the location
of the  FMGP.   The  Atlantic  Coal  Gas Company
operated the FMGP from  1905 to  1925.   During that
time, an unknown quantity of coal tar was disposed of on
site.  In addition to the coal  tar waste,  more recent
releases of petroleum from two nearby gas stations also
have occurred.  An investigation conducted at the site
from  1990 to 1992  identified the  following  primary
contaminants:  BTEX and PAHs.  The local ground-
water contains free petroleum product and pure coal tar.

    The York site is located in York, Nebraska.  The
site encompasses  nearly a half acre hi an industrial
section, of the city. The site is surrounded by a former
railroad right-of-way,  a concrete  company,  a  seed
company, and a farm^supply store.  The site is nearly
level, and several buildings occupied by the FMGP are
still present.   The York Gas and  Electric Company
operated the FMGP from 1899 to 1930.  Coal tar waste

-------
was disposed of at the site.  Current information on the
site suggests that coal tar waste and its constituents
should be the only waste encountered.

    The Fort Riley site is located at Building 1245 on
the east side of the Camp Funston area at Fort Riley,
Kansas.   Between 1942 and 1990,  five 12,000-gallon
steel underground storage tanks were located at this site.
The tanks were used to store leaded  and unleaded
gasoline, diesel  fuel, and military operations gasoline.
Soil at the site is contaminated with gasoline and diesel
fuel believed to be the result  of past petroleum fuel
releases from the underground storage tanks.

-------
                                            Section 3
                                  Reference Method Results
    All soil samples collected during this demonstration
were submitted to PACE, Inc. (PACE), for chemical
and geotechnical analysis.   The PACE  laboratory in
Lenexa, Kansas, performed the 418.1  and Methods
8020 and OA-1 analyses, while the PACE laboratory hi
St.  Paul, Minnesota, performed the Method 8310 an-
alysis.   PACE   subsequently   subcontracted  the
geotechnical  analyses to   Environmental  Technical
Services (ETS), Petaluma, California.   The chemical
data  supplied  by  the   reference  laboratory,  the
geotechnical  data  supplied  by  the  geotechnical
laboratory, and  the  data produced by the  on-site
professional geologist is discussed in this  section.

Reference Laboratory  Procedures

    Samples collected  during this demonstration were
homogenized and split for the following analyses:

    •   TPH by EPA Method 418.1 (EPA 1986)

    •   PAH by  EPA SW-846 Method 8310 (EPA
        1986)

    •   BTEX by EPA  SW-846 Method 8020 (EPA
        1986)

    •   Total VPH as gasoline by University of Iowa
        Hygienic Laboratory Method OA-1 (University
        Hygienic Laboratory 1991)

    •   Soil texture  and TOC  by  the 90-3  Walk-
        ley-Black Method (Page 1982)

    The results of these analyses  are summarized hi
Appendix  A.  The results are reported as wet weight
values as required hi the approved demonstration plan
(PRC 1994). The data is grouped by analytical method,
site, and whether the data is  intended  for qualitative or
quantitative data evaluation.
    The  data from  the reference" laboratory  was
internally reviewed by PACE personnel before the data
was delivered to PRC. PRC personnel conducted a data
review on the results provided by PACE following EPA
guidelines (EPA 1991). PRC reviewed the raw data and
checked the calculated sample values.

    The   following  subsections   discuss  specific
procedures used to identify and quantitate TPHs, VPHs,
PAHs, BTEX, and TOC.  Most of these  procedures
involved requirements that were mandatory to guarantee
the quality of the data generated.

Sample Holding Times

    The required holding tunes from the date of sample
receipt for each analytical method used to analyze the
soil samples  were as follows:  University of Iowa
Hygienics Laboratory Method OA-1 (Method OA-1),
14 days  for  extraction  and  analysis;  EPA SW-846
Method 8020  (BTEX),  14  days  for  extraction and
analysis;  EPA Method  418.1  (TPH), 14 days for
extraction and 40 days  for  analysis;  EPA SW-846
Method 8310 (PAH), 14 days for extraction and 40 days
for analysis; and  90-3 Walkley-Black Method (TOC),
28 days for extraction and analysis.

    All holding tunes for the samples were  met during
this demonstration.

Sample Preparation

    Preparation of soils for TPH analysis was performed
following EPA Method 418.1.  This  method uses a
Soxhlet  extraction as  stated  in   SW-846 Method
9071.  The soil sample extracts were  analyzed for TPH
using SW-846 Method 418.1.

-------
    Extracts for VPH analysis were prepared following
Method OA-1. The BTEX sample preparation require-
ments were carried out as specified in that method.

    The preparation of soil samples  for TOC analysis
were carried out as specified in the 90-3 Walkley-Black
Method.

    Sonication extraction, SW-846 Method 3550, was
used for the preparation of soil samples for SW-846
Method 8310 analysis.  The preparation of samples for
PAH analysis by SW-846 Method 8310 were carried out
according to the method requirements.

Initial and Continuing Calibrations

    Initial calibrations (ICAL) were  performed before
sample analysis began.  ICALs for  SW-846 Methods
8020, 8310, and 418.1 consisted of the  analysis of five
concentrations of standards. Method OA-1 required the
analysis of three  concentrations of  standards for the
ICAL.  Linearity for these ICALs was  evaluated by
calculating  the percent  relative  standard  deviation
(%RSD) of the calibration factors.  The %RSD QC limit
for SW-846  Methods  8020 and  8310  and Method
OA-1 was 20 percent.  The calibration  factors  were
calculated by dividing the response (measured as the area
under  the peak  or peak  height)  by the amount  of
compound injected on the gas chromatograph (GC)
column.  The 90-3 Walkley-Black Method for TOC
required a  daily  calibration  to  a  reference  sulfate
solution. This ICAL was performed in duplicate.  All
initial   calibrations  met  the  respective  method
requirements.

    Continuing calibrations (CCAL) were performed on
a daily basis to check the response of the detector by
analyzing a mid-concentration standard and comparing
the calibration factor to that of the  mean  calibration
factor from the ICAL.

    Calibration factors were monitored hi accordance
with the SW-846 and OA-1 Methods. No CCAL was
performed for the 90-3 Walkley-Black Method.   Six
CCALs exceeded the 15 percent difference (%D) criteria
for various BTEX compounds. This resulted hi sample
results being qualified as estimated (J) and usable for
limited purposes.  Various  PAH compounds in six
SW-846 Method 8310 CCALs exceeded 15 %D for one
of the two detectors. Sample results for the compounds
exceeding 15 %D were qualified as  estimated (J) and
usable for limited purposes. SW-846 Method 8310 uses
two detectors, an ultraviolet detector and a fluorescence
detector.  Since one detector's CCAL response was
within QC guidelines, this data was considered useable.
    Retention  tunes  of the  single analytes  were
monitored through the amount of retention time shift
from the CCAL standard as  compared to the  ICAL
standard.   The retention tune windows  for SW-846
Method 8310 were set by taking three tunes the standard
deviation of the retention times that were calculated from
the ICAL and CCAL standards.  The retention tune
windows for SW-846 Method 8020 were set by PACE at
plus or minus 0.07 minutes for benzene, ethylbenzene,
m-xylene, and plus or minus 0.10 minutes for toluene.
 No CCAL retention times  for  the individual PAH
analytes were outside the  retention tune windows.
CCAL retention times for the individual BTEX analytes
were observed outside the retention time windows as set
by the ICAL. No samples were qualified based on this
QC  criteria because  the retention  time shifts were
adjusted appropriately by PACE for sample identification
and quantitation.

    Following the ICAL, a method blank was analyzed
to  verify  that  the  instrument  met   the  method
requirements.  Following this, sample  analysis may
continue for 24 hours. As stated hi SW-846 Method
8000, a CCAL must be analyzed and the calibration
factor verified on each working day.  Sample analysis
may continue as long as CCAL standards  meet the
method requirements.

Sample Analysis

    Specific PAH and BTEX compounds were identified
in a sample by matching retention times of peaks  found
after analyzing the sample with those compounds  found
hi PAH and BTEX standards. VPH was identified hi a
sample by matching peak patterns found after analyzing
the sample  with those  compounds  found  hi  VPH
standards.  Peak patterns may not always match exactly
because of the way the VPHs were  manufactured or
because of the effects of weathering.   When peak
patterns do not  match,  the analyst  must decide  the
validity of the identification of VPHs. For this reason,
peak pattern identification is  highly dependent on the
experience and interpretation of the analyst.

    Quantitation  of PAHs, BTEX compounds, TPHs,
and VPHs was performed by measuring the response of
the peaks hi the sample to those same peaks identified hi
the ICAL  standard.  The reported results of this
calculation were based  on wet weights (except  for
PAHs).  PAH data was reported on a dry-weight basis.
PRC converted this data to wet-weight based results.
Quantitation of TOC was performed by measuring the
volume  of  K2Cr2O7 titrated and  calculating  the
milliequivalents of  K2Cr2O7 titrated.   This value was
then multiplied by conversion factors  and subsequently
                                                  10

-------
 divided by the grams of sample.  TOC results were
 reported on a wet-weight basis.

    Sample  extracts  can  frequently  exceed  the
 calibration range determined during the ICAL.  When
 this occurred, the extracts were diluted to obtain peaks
 that fall within the linear range of the  detector.  For
 BTEX compounds and  VPHs, this linear  range  was
 defined as the highest standard concentration response of
 the analytes of interest analyzed during the ICAL.  The
 linear range for TPHs was defined as  an absorbance
 maximum of 0.8.  For  PAHs, as defined in SW-846
 Method 8310, the linear range was from 8 tunes the
 method detection limit (MDL) to 800 tunes the MDL
 with  the  following  exception:  benzo(ghi)perylene
 recovery at 80 tunes and 800 tunes the MDL are low.
 Once a sample was diluted to within the linear range, it
 was analyzed again.  Dilutions were performed when
 appropriate on the samples for this demonstration.

 Detection Limits

    The PACE reporting limit (PRL) for PAHs was
 calculated by  multiplying the calibration  correction
 factor based on dry weight,  tunes the MDL for each
 specific PAH.   PRLs  for  BTEX compounds were
 determined by the lowest concentration standard of the
 ICAL. The BTEX ICAL concentration range was from
 10 micrograms per liter (^g/L) to 100 jj,g/L.  The PRL
 for benzene, toluene, and ethyl benzene was 50 micro-
 grams per kilogram  C/g/kg) and 100 Mg/kg for total
 xylene.  The three levels of standard concentrations for
 the VPH ICAL ranged from 2 milligrams per milliliter
 (mg/mL) to 8 mg/mL.  The PRL for VPH was 5 mil-
 ligrams  per  kilograms  (mg/kg).    For TPH,  the
 calibration range  was calculated by calibrating  the
 infrared detector using a series of working standards. A
plot was then prepared of absorbance versus milligram
petroleum hydrocarbons per 100 milliliter (mL) solution.
The PRL for TPH was 10 mg/kg. The MDL for TOC
analysis was 10 mg/kg wet weight.

 Quality Control Procedures

    A number of QC measures were used by PACE as
required  by SW-846 Methods 8310 and 8020, EPA
Method 418.1,  Method  OA-1,  and the 90-3 Walk-
ley-Black Method.  These QC measures included the
analyses of method blanks, instrument blanks, laboratory
control samples (LCS), matrix spike (MS) and matrix
spike duplicates (MSD), and the use of sample surrogate
recoveries.

    All  method  and  instrument  blanks   met  the
appropriate QC criteria, except for two method blanks
 analyzed by Method  418.1.   TPH was  reported as
 slightly above the PRL of 10 mg/kg  in two  method
 blanks.  Due to  the low values reported in the two
 method blanks, the sample results were not qualified.

    Surrogate standards were added to all samples,
 method blanks, MSs, and LCSs for the SW-846 Methods
 8310 and 8020,  and  Method OA-1.   All  surrogate
 recoveries for SW-846 Method 8020 were within the QC
 acceptance criteria of 42 to 140 percent for soil.  Seven
 samples were qualified as estimated (J) and usable for
 limited purposes  based on surrogate  recoveries for
 Method OA-1. The QC acceptance criteria for surrogate
 recovery for Method  OA-1  was 67 to 127 percent.
 Thirty soil samples for SW-846 Method 8310 analysis
 were  qualified as estimated (J) and usable for  limited
 purposes based on surrogate recoveries observed outside
 the QC limits of 58 to 140 percent.  Two surrogates
 were  used for Method 8310.  Samples were qualified
 only when both surrogates were outside the QC limits
 and no dilution analysis was performed.  Numerous soil
 samples required dilution for the Method 8310 analysis
 because of petroleum  interference.  Dilution of these
 samples  resulted  in   corresponding  reductions  in
 surrogate  concentrations.  When this occurred, the
 resultant concentration of surrogate was below its MDL.
 In cases where dilution resulted in failure to detect the
 surrogate, no coding of the data was implemented.

    MS samples are samples to which a known amount
 of the target analytes are added.  There were 10 MSs
 performed during the analysis by Method 418.1.  Eight
 of the MS samples were affected by high concentrations
 of target analytes in the spiked samples.  No samples
 were qualified. Eleven MSs were performed during the
 analysis by Method 8310. All but three MSs and MSDs
 were  outside the QC limits for percent recovery and
 relative  percent  difference  (RPD).    These QC
 exceedences were due to petroleum matrix interference.
 The  data associated with the  QC  samples  was  not
 qualified because EPA guidelines state that samples
 cannot be qualified based on MS and MSD results alone
 (1991).  There were  seven MS and  MSD samples
 analyzed by Method  8020  and  five  by Method
 OA-1.  The MSs and MSDs analyzed by Method
 8020 did not meet QC  acceptance criteria for percent
 recoveries or RPDs. No samples were qualified based
on these MS  and MSD results due to the reasons stated
 above.   All MS and MSDs analyzed by Method
 OA-1.  and 90-3  Walkley-Black  Method met all QC
acceptance criteria and were considered acceptable.

    All LCSs  met QC acceptance criteria and were
considered acceptable for all soil samples analyzed by
SW-846 Method  8310,  Method  OA-1,  90-3  Walk-
ley-Black Method, and Method 418.1.  One soil LCS
                                                 11

-------
analyzed by Method 8020 was outside the QC control
limits. The soil LCS percent recovery for toluene was
below the QC limit.  Twenty soil samples were qualified
as estimated (J) and usable for limited purposes.

    Also, three equipment rinsate blanks and one trip
blank were  analyzed to assess the efficiency of field
decontamination and shipping methods,  respectively.
There was no contamination found above PRLs in any of
these type of  blanks indicating decontamination pro-
cedures were adequate.

Confirmation of Analytical Results

    Confirmation of positive results was not required by
any of the analytical methods performed except SW-846
Method 8310.  Confirmation of positive PAH results by
Method 8310 was performed by the use of two types of
detectors.    Both  an  ultraviolet  detector  and  a
fluorescence detector were used in the analysis of PAHs.
The only requirement for using either detector for
quantisation was that they meet the QC criteria for
linearity (ICAL) and %D  (CCAL).  If either detector
failed either of these criteria, it could not be used for
quantitation, but it could be used for confirmation of
positive results.

Data Reporting

    The results reported and qualified by the reference
method contained two types of qualifier codes.  Some
data was coded with a "J," which is defined by PACE as
detected but below the PRL; therefore, the result is an
estimated concentration. The second code, "MI," was
defined as matrix interference. Generally, the effect of
a matrix interference is to reduce or enhance sample
extraction efficiency.

Quality Assessment of
Reference Laboratory Data

    This section discusses the accuracy, precision, and
completeness of the reference method data.

Accuracy

    Accuracy   of   the  reference  method   was
independently assessed through the use of performance
evaluation (PE) samples purchased from Environmental
Resource Associates (ERA) containing a known quantity
of TPH.  In addition, LCSs and past PE audits of the
reference  laboratory were used to verify  analytical
accuracy.  Based on a review of this data, the accuracy
of the reference method was considered acceptable.
Precision

    Precision  for the reference  method results was
determined by evaluating field duplicates,  laboratory
duplicate, and MS and MSD sample results.  Precision
was evaluated by determining the  RPDs for  sample
results and their respective duplicate sample results.

    The  MS and  MSD  RPD results for  the PAH
compounds averaged 25 percent for all of the  11 MS and
MSD sample pairs.  However, there was one MS and
MSD  sample  pair that  had  a  RPD of  99.9  for
1-methyl-naphthalene.  If this point was removed, the
overall average would decrease to 20 percent.  The
average RPD for the seven BTEX MS and MSD sample
pairs was less than 25 percent. Only four BTEX RPDs
were  outside advisory QC guidelines  defined by the
PACE'S laboratory control charts.  All five VPH MS
and MSD sample RPDs met advisory QC guidelines set
by  the reference laboratory's control charts.  The
10 TPH MS and MSD sample pairs were  considered
acceptable.

    Laboratory duplicate samples  are two separate
analyses performed on the sample.  During the analysis
of demonstration samples, 10 TPH laboratory duplicate
samples  were  prepared  and  analyzed.    All TPH
laboratory duplicate RPD result values were less than
25 percent. This was considered to be acceptable.

Completeness

    Results were obtained for all of the soil samples.
PACE J-coded values that were detected below the PRL,
but above the MDL.  As discussed above, samples were
J-coded based on one or more of the advisory  QC
guidelines not being met (i.e., surrogate  and spike
recoveries)., Also, some samples were J-coded based on
BTEX CCALs not meeting QC guidelines.  PRC did not
consider this serious enough to preclude the  use of this
data because the %Ds for the CCALS did not exceed the
QC guidelines by more  than 10 percent of the acceptable
range. The analytes with %Ds outside the QC guidelines
were not detected in most of the samples associated with
the CCALs.  The J-coded data is  valid and usable for
statistical analysis because the QC guidelines were based
on advisory control limits set by either the Method or by
PACE and  this   data  set   should  be   considered
representative  of  data  produced   by conventional
technologies.  For this  reason, the actual completeness
of data used was 100 percent.

Use of Qualified Data for Statistical Analysis

    As noted  above,  100 percent of the  reference
laboratory results  were  reported  and validated by
                                                  12

-------
approved QC procedures.  The data review indicated
that  J-coded  data  was  acceptable  for  meeting the
demonstration objective of providing baseline data to
compare against die demonstrated technologies.

    None of the QA/QC problems were considered
serious enough to preclude the use of J-coded data for
this demonstration.  The surrogate and spike recovery
control limits were for advisory purposes only, and no
corrective  action  was   required  for  the  surrogate
recoveries that were outside of this range.  RPD results
for MSs and MSDs  that did not  meet advisory QC
control limits were common when the matrix contained
a high concentration of petroleum.   Again, these were
advisory limits and no corrective action was required.
These same  general  results  would be seen by any
laboratory using the reference analytical methods on
such highly contaminated samples.

    Also, rejection of a large  percentage of data would
increase  the apparent variation between the reference
laboratory data and the data from the technology.  This
apparent variation  would probably be of a similar
magnitude to that introduced by using the data.   For
these reasons, the J-coded data was used.

Chemical Cross Sections

    Chemical cross sections were created from the
reference analytical data produced  for the qualitative
data evaluation (see Appendix A).  These samples were
collected by a professional geologist on site during the
logging of boreholes.  The cross sections were  hand
contoured, and the  contour intervals were selected to
best represent the  range of contamination  detected.
These cross  sections were  intended to represent  a
conventional  approach to the  delineation of subsurface
contamination.  The  cross sections  are presented on
Figures 3-1 to 3-6.  A  written interpretation of  these
cross sections is presented below.

Atlantic Site

    The  five  sampling  nodes formed a northwest to
southeast trending transect across the site (Figure 3-1).
Node 1  on the far northwest edge of the cross section
represented an area  that was not  impacted by the
contamination from the Atlantic site. Just southeast of
this  location at Node  2,   two  distinct  layers of
contamination were identified.  The upper zone extended
from approximately 1 foot to 5  feet below ground
surface  (bgs).  This zone was characterized  by  TPH
contamination ranging from 100 to  10,000 ppm.   The
lower   zone   of  contamination   extended   from
approximately  22 feet  to  28 feet  bgs.   The  TPH
concentrations in flu's  zone ranged from 100 to greater
than  10,000  ppm.   These two zones expanded and
blended together as Node 3 was approached.  Around
Nodes 3 and 4 the thickness of the TPH plume remained
fairly constant, extending from approximately 1 foot to
31 feet bgs.  The central portion of this zone exhibited
TPH  contamination  greater than  1,000 ppm.    The
remainder of this zone exhibited TPH contamination in
the range of 100 to 1,000 ppm.  As the far southeastern
edge  of the transect was approached at Node 5, the
highest concentrations in the center of the plume pinched
out, leaving a contamination zone that extended from just
below the ground surface to approximately 27 feet bgs.
This zone exhibited contamination in the range of 100 to
1,000 ppm.

    The total  PAH cross section along this same transect
exhibited a slightly different distribution (Figure  3-2).
As with the  TPH cross section, the total PAH  cross
section began at Node 1 in an area exhibiting no signs of
contamination.    At Node 2, again two  zones of
contamination were detected.  The upper zone extended
from  the ground surface to approximately 7 feet bgs.
This 2:one deepened toward the east.  The concentrations
of total PAHs in this zone ranged from 10 to greater
than  100 ppm.   The lower zone  extended  from
approximately 14 to 30 feet bgs. The concentrations of
total PAHs in this zone ranged from 10 to greater than
100 ppm. Concentrations greater than 100 ppm were not
exhibited at this depth in the nodes occurring further
east.  The distribution of the  10 to  100 ppm dipped
below the ground surface at progressive depths farther
east of Node  2.  At Node 5, this upper zone began at
approximately 7  feet bgs. This zone  also reached its
maximum depth around Nodes 3 and 4, approximately
30 feet bgs.  Around Nodes 3 and 4 were two lenses of
total PAH contamination in excess of 100 ppm.   The
largest of these zones appeared to be thickest around
Node  3, extending from approximately 7 to 16 feet bgs.
This zone thinned to the east and pinched out between
Nodes; 4 and 5. A smaller lens of greater than 100 ppm
total PAH contamination was exhibited at Node 4.  This
zone extended between 7 to 9 feet bgs.  This zone was
not detected in Nodes 3 or 5.

York Site

    The five  sampling  nodes formed a north to south
trending transect. The TPH and total PAH distributions
appeared to be similar, except  at Node 5, at the  York
site (Figures 3-3 and  3-4).   At Node  5,  the  TPH
contamination was more extensive, extending from 1 to
25  feet  bgs.   At  this  same location,  the   PAH
contamination extended from 13 to 21 feet bgs.

    All of the nodes for this transect occurred in  areas
that were impacted by the contamination associated with
                                                   13

-------
FIGURE 3-1.  TPH REFERENCE METHOD CHEMICAL CROSS SECTION—ATLANTIC SITE
         SOUTHEAST
                 NODES
           0-
           -4-
           -3-
           -8-

           -8-
           -9-
          -10-
           -11-
          -14-
          -13-
          -18-
          -17-
          -18-
          -19-
          -20-
          -21-
          -22-
          -23-
-I?:
-28-
-29-
-30-
-31-
-32-
-33-
-34-
-35-
-38 —
-37-
                        NODE+
                                  NOOE3
                                                   NOOE2
                                                 DISTANCE (FEET)
                                                                                 NORTHWEST
                                                                               NOOE1
                                                                                       ::§

                                                                                        -5
                                                                                        -6
                                                                                        7
                                                                                        a
                                                                                        -»
                                                                                        -10
                                                                                        11
                                                                                       — 12
                                                                                       —13
                                                                                        -11
                                                                                        -IS
                                                                                        -10
                                                                                        17
                                                                                        -10
                                                                                        -IB
                                                                                        -22
                                                                                        23
                                                                                                 --27
                                                                                                 --21
                                                                                                 --29
                                                                                                 — 30
                                                                                                 — 31
                                                                                                 --32
                                                                                                 --33
                                                                                                 --34
                                                                                                 --35
                                                                                                 --35
                                                                                                 --37
          LEGEND
       75



1< 100 PPM


• 100 - 1,000 PPM
                                     125     1SO



                                   1.000 - 10.000 PPM


                                   > 10,000 PPM
                                                             175
                                                                           225
                                                             ND - . NOT DETECTED


                                                             PPM -  PARTS PER MILLION
                                                                                        27S
« - QUANTITATIVE
    REFERENCE DATA
          NOTE:  QUANTITATIVE REFERENCE DATA USED BECAUSE OF POOR SAMPLE RECOVERY
FIGURE 3-2. PAH REFERENCE METHOD CHEMICAL CROSS SECTION—ATLANTIC SITE
         SOUTHEAST
                 NODES
           0-
           -4
           -5
           -8
          -10
          -II
          -12
          -13
          -14
          -13
          -18
          -17
          -18
          -19
          -20
          -21
          IE
          -30-
          -31-
          -34-
          -37-
                        NODE4
                                  NOOE3
                                                    NODE2
          NORTHWEST
        NODE1
               -0

               —2
                                                                                       ::?
                                                                                       —8
                                                                                       •-•
                                                                                       ::j?

                                                                                       -14

                                                                                       —18
                                                                                       -17
                                                                                       -W
                                                 DISTANCE (FEET)
          LEGEND
                           a< 10 PPM
                                          l-_-_-J 10-100 PPM
                                                                 H> 100 PPM
                                                                                  230
                                                                                 T - < 1 PPM
                         PPU -  PARTS PER MILLION      . - QUANITATIVE REFERENCE DATA


          NOTE:  QUANITATIVE REFERENCE DATA USED BECAUSE OF POOR SAMPLE RECOVERY
                                                                                       :»
                                                                                       —29
                                                                                       —30
                                                                                       —31
                                                                                       —32
                                                                                       —33
                                                                                       —34
                                                                                       —33
                                                      14

-------
FIGURE 3-3. TPH REFERENCE METHOD CHEMICAL CROSS SECTION—YORK SITE
         NORTH
           0-
           -1-
          -2-
          -3-
          -4-
          -5-
          -8-
          -7-
          -8-
          -9-
          -10-
          -11-
          -12-
          -13-
          -14-
          -15-
          -18-
          -17-
          -18-
          -19-
          -20-
          -21-
          -22-
          -23-
          -24-
          -25-
          -26-
                 NODE1
                               NODE2
                                              NOOE3
                                                                  NODE4
                                                                               NODES

                                             377
                                              DISTANCE (FEET)
                                                   80
                                                                                          SOUTH
                                     0
                                      1
                                      2
                                      3
                                      4
                                      5
                                      8
                                      7
                                      8
                                     -»
                                      10
                                      11
                                      12
                                      13
                                      14
                                      15
                                      18
                                      17
                                      18
                                      It
                                      20
                                      21
                                      22
                                      23
                                      24
                                      25
                                      26
                                                                                          120
         LEGEND
                         < 10 PPM

                         > 10.OOO PPM
WpgiO - 100 PPM

 NO - NOT DETECTED
r-_-_-|ioo - i.ooo PPM    F
r%-— ~i              t
 PPM -  PARTS PER MILLION
11.000 - 10,000 PPM
FIGURE 3-4. PAH REFERENCE METHOD CHEMICAL CROSS SECTION—YORK SITE
         NORTH
           o-
           -1-
          -2-
          -3-
          -4-
          -5-
          -8-
          -7-
          -8-
          -e-
          -10-
          -11-
          -12-
          -13-
          -14-
          -15-
          -18-
          -17-
          -18-
          -18-
          -20-
          -21-
          -22-
          -23-
          -24-
          -25-
          -28-
                 NODE1
                                NODE2
                                              NOOE3
                                                                  NODE4
                                                                                  NODES
                                              DISTANCE (FEET)
                                                                                          SOUTH
                                                     -o
                                                     —1

                                                     —3
                                                     —4
                                                     --5
                                                     —8
                                                     —7
                                                     —8
                                                     —9
                                                     —10
                                                     —11
                                                     —12
                                                     —13
                                                     —14
                                                     —15
                                                     —18
                                                     —17
                                                     —18
                                                     —19
                                                     —20
                                                     —21
                                                     —22
                                                     —23
                                                     —24
                                                     —25
                                                     —28
                                                   80
                                                         70
                                                                8O
                                                                       90
                                                                             100
                                                                                   110
                                                                                          120
         LEGEND
                         • < 10 PPM
                         ;
                         •10-100 PPM
     1OO - I.OOO PPM

     > 1.000 PPM
    T -  < 1 PPM       PPM -  PARTS PER MILLION

    NO - NOT DETECTED
                                                  15

-------
FIGURE 3-5.  TPH REFERENCE METHOD CHEMICAL CROSS SECTION—FORT RILEY SITE
         SOUTH
           o-
           -1-
           -2-
           -3-
           -4-
           -5-
           -6-
           -7-
           -8-
           -9-
          -10-
           -11-
          -12-
          -13-
          -14-
          -15-
          -18-
          -17-
          -18-
          -19-
          -20-
           -21-
          -22-
          -23-
          -24-
          -25-
          -28-
          -27-
          -28-
          -29-
          -30-
           -31-
                              NODE2
                                                 NODES
                                                                    NODE3
                                                                                       NODE4
                                                                                              NORTH
                                                                        -o
                                                                        —i
                                                                        —2
                                                                        —3
                                                                        —4
                                                                        --5
                                                                        —6
                                                                        —7
                                                                        —8
                                                                        --9
                                                                        —10
                                                                        --11
                                                                        —12
                                                                        —13
                                                                        —14
                                                                        --15
                                                                        —18
                                                                        —17
                                                                        —18
                                                                        —19
                                                                        —20
                                                                        --21
                                                                        —22
                                                                        —23
                                                                        —24
                                                                        —25
                                                                        —28
                                                                        —27
                                                                        —28
                                                                        —29
                                                                        —30
                                                                        —31
                                                 DISTANCE (FEET)
                                                  ISO     180
          LEGEND
                          < 10 PPM

                          > 10.000 PPM
                gg$10 - 100 PPM

                 NO -  NOT DETECTED
q 1.000 - 10.000 PPM
                                                           PPM -  PARTS PER MILLION
FIGURE 3-6. PAH REFERENCE METHOD CHEMICAL CROSS SECTION—FORT RILEY SITE
          SOUTH
            0-
            -1-
           -2-
           -3-
           -4-
           -3-
           -«-
           -7-
           -8-
           -9-
          -10-
           -11-
          -12-
          -13-
          -14-
          -15-
          -18-
          -17-
          -18-
          -19-
          -20-
           -21-
          -22-
          -23-
          -24-
          -25-
          -28-
          -27-
          -28-
          -29-
          -30-
           -31-
                               NODE2
                                                  NODES
                                                                     NOOE3
                                                                                        NODE4
                                                                                               NORTH
                                                                         •0
                                                                         -1
                                                                         -2
                                                                         -3
                                                                         -4
                                                                         -5
                                                                         •-8
                                                                         -7
                                                                         -I
                                                                         -9
                                                                         -10
                                                                         -11
                                                                         -12
                                                                         -13
                                                                         -14
                                                                         -15
                                                                         -18
                                                                         —17
                                                                         —18
                                                                         •-19
                                                                         —20
                                                                         —21
                                                                         —22
                                                                         -23
                                                                         -24
                                                                         —25
                                                                         —28
                                                                         —27
                                                                         —28
                                                                         —29
                                                                         —30
                                                                         —31
                                                 ' DISTANCE (FEET)
                          110
                                120
                                      130
                                             140
                                                   150
                                                         180
                                                               •170
                                                                      180
          LEGEND
2»P4<:10PPM        r-_-_-lio - 100 PPM     m±tt|ioo - i.ooo PPM


 ND - NOT DETECTED     PPM - PARTS PER MILLION
                                                                                 T - < 1 PPM
                                                      16

-------
this site.  The contamination at this site appeared to
occur in a single band extending from approximately
10 to 22 feet bgs for total PAH and 2 to 25 feet bgs for
TPH contamination. This band of contamination thinned
from south to north across the transect. At the north end
of the transect, the TPH contamination thinned to a zone
extending from 12 to 19 feet bgs.  Concentrations of
TPH in this zone ranged from 10 to 10,000 ppm, and the
concentrations of total PAH contamination range  from
10 to  1,000 ppm.   TPH contamination exhibited  its
maximum concentrations in a lens around Nodes 3 and
4. This lens extended from approximately 12 to 16 feet
bgs,  and  exhibited TPH concentrations greater  than
1,000 ppm. This lens tended to thin and deepen  from
south to north.  Node 4 exhibited the greatest TPH and
total PAH contamination.  Two narrow lenses of total
PAH concentrations greater than 1,000 ppm existed at
approximately 14 to 16 feet bgs and 18 to 20 feet bgs.
A narrow  lens  of TPH contamination greater  than
10,000 ppm was detected at approximately 18 to 19 feet
bgs at Node 4.

Fort Riley Site

    The five sampling nodes formed a south to north
trending transect. The TPH and total PAH distributions
appeared to be similar at the Fort Riley site (Figures
3-5 and 3-6).  Node 4, situated at the far north end of
the transect, was not affected by contamination. All of
the remaining nodes for this transect occurred in areas
that were impacted by the contamination associated with
this site.  The contamination at this site appeared to
occur in a single zone extending from approximately
1 to 25 feet bgs for total PAH and 0 to 30 feet bgs for
TPH  contamination.   This  zone  of contamination
exhibited  relatively constant  thickness across Nodes
2, 3, and 5.  Concentrations of TPH in this zone ranged
from   100  to greater than  10,000  ppm,  and ithe
concentrations of total PAH contamination ranged  from
10 to 300 ppm. Total PAH contamination exhibited its
maximum  concentrations hi a  lens  around  Nodes
2, 3, and 5. This lens extended from  approximately 5 to
8 feet bgs at Node 3, becoming thicker and deeper at
Node 2 where it extended from 10 to 20 feet bgs.  The
TPH   concentrations   exhibited   two  lenses   of
concentration at greater than 10,000 ppm. These lenses
did not appear to be extensive and their occurrence was
limited to the areas around single nodes.  Node 5 in the
center  of the transect exhibited one  of these lenses of
highest TPH contamination extending  from 10 to 13 feet
bgs.   Node 2 has the other such lens which extended
from 17 to 19 feet bgs.
Quality Assessment of
Geotechnical Laboratory Data

    This  section discusses  the  data  quality  of the
geotechnical laboratory results, the data quality of the
borehole logging conducted by the on-site professional
geologist, and the soil sampling depth control.

Geotechnical Laboratory

    Soil samples submitted for textural determination
were analyzed by ASTM Method D-422 (1990).  ETS,
Petaluma, California, conducted these analyses. ASTM
Method D-422 does not define specific QAXQC criteria,
however,  it specifies the use of certified sieves, and
calibrated  thermometers  and  hydrometers.    ETS
followed the approved method and complied with all the
equipment certification and calibration requirements of
the method.  Based on this, the geotechnical data was
detennined acceptable.

Borehole Logging

    The  data  quality   of the  on-site  professional
geologist's borehole logs was checked through  on-site
audits  by a soil scientist,  and by  comparison  of the
geologist's descriptions for intervals corresponding to
samples  analyzed by ASTM Method D-422.    This
comparison is discussed later hi this report.

Sampling Depth Control

    At each site,  random checks  of the  reference
sampling intervals were made.  These checks consisted
of stopping drilling operations just before inserting the
split spoon sampler into the  hollow stem auger to  collect
samples.  At this tune, a weighted tape measure was
used to measure the top of the sampling interval.  The
measurement was checked against the intended sampling
depth.  If the difference between the intended  and actual
sampling  depth had varied by more than 1 inch, the
borehole would have been redrilled.  Depth checks were
made at a minimum of once per sampling day. None of
these depth checks resulted hi data exhibiting a greater
than 1  inch  difference between intended and  actual
sampling depth. Based on this, the reported sample
intervals were considered accurate.

Stratigraphic Cross Sections

    Stratigraphic cross sections were based on the data
produced by a professional  geologist during the logging
of boreholes during the demonstration.   The cross
                                                  17

-------
FIGURE 3-7.  REFERENCE METHOD STRATIGRAPHIC CROSS SECTION—ATLANTIC SITE
SOUTHEAST
       NODES
  0-
  -I-

  -4-
  -S-
  -0-
  -7-
  -n-
  -9-
 -10-
 -II-
 -14-
 -13-
 -IS-
 -17-
 -1B-
 -19-
 -20-
 -21-
 -21-
 -23-
 -24 —
 -25-
 -28-
 -27-
 -28-
 -29-
 -30-
 -31-
                                 NODE4
                                           NODE3
                                                            NODE2
                                                                                         NORTHWEST
                                                                                       NODE1
                                                                                               -1
                                                                                               -2
                                                                                               -3
                                                                                               -4
                                                                                               -6
                                                                                               -8
                                                                                               -7
                                                                                               -8
                                                                                               -9
                                                                                               .-10
                                                                                               -11
                                                                                               •-12
                                                                                               •-13
                                                                                               •-14
                                                                                               -15
                                                                                               -IB
                                                                                               -17
                                                                                               -18
                                                                                               •-19
                                                                                               -20
                                                                                               -21
                                                                                               -22
                                                                                               -23
                                                                                               —24
                                                                                               —25
                                                                                               —28
                                                                                               -27
                                                ;LOAM)
                                                ;LOAM)
                                                                                          (SAND)
                                                                                          (CLAY
                                                                                          .LOAM;
                                                DISTANCE (FEET)
                                                     --30
                                                     —31
                                                     —32
                                                     —33
                                                     —34
                                                     —35
                                                     --38
                                                     —37
                                              125
                                                     150
          LEGEND
; SILTY CLAY (CH>

 SILTY CLAY (CL)
                                                   SILT

                                                   SILTY SAND
   250     275


WELL GRADED SAND

POORLY GRADED SAND
                                       ( ) -  LABORATORY CLASSIFICATION (USDA)
sections were intended  to  represent  a conventional
approach to the delineation of subsurface stratigraphy.
The cross  sections are presented on  Figures  3-7 to
3-9.  A verbal interpretation of these cross sections is
presented  below  by site.   QA/QC consisted  of the
collection  of samples  for  textural  analysis  by  a
geotechnical  laboratory  (see  Appendix A).    These
samples are discussed at the end of each site-specific
discussion.

Atlantic Site

     The Atlantic site is located on the flood plain of the
Nishnabotna River, which is located about 0.7 mile west
of the site.  The flood plain is nearly level.  The surface
soil at the  site is a silty clay.  These  soils have most
likely  formed  from  alluvium.    Stratigraphic cross
sections produced during the  demonstration from soil
borings indicated that the subsurface  soil at the site
consisted of silts and  clays and silty clay interfingered
with each other to a depth of approximately 21 feet bgs
on the northwest end (Node 1)  and to 28 feet bgs on the
southeast (Node  5).  See Figure 3-7  for a  graphical
representation of the cross section.  A layer of sand was
present from 18 to 36 feet bgs at Node 2.  This zone
 remained relatively uniform from Node 2 to Node 5.
                  Seven soil samples  were collected  to  verify the
              geologist's borehole logging at the Atlantic site. The
              geologist's  classification   of  soils   matched  the
              geotechnical laboratory's classifications six out of seven
              tunes (Table 3-1). This one point of disagreement was
              sample DR16 from the 2- to 3-foot interval at Node 1.
              In this  sample,  the geologist  identified  silt  as the
              predominant size fraction of the  sample,  while the
              geotechnical  laboratory  identified  clay   as  the
              predominant size fraction.  This is a common point of
              variance between field soil classification and laboratory
              classification.  These common differences are magnified
              in grossly contaminated soils, and when the geologist is
              forced to wear plastic gloves during classification. This
              difference was noted,  and geologist's  Stratigraphic
              borehole  logs  meet  the  demonstration  DQOs for
              screening level data.

              York Site

                  The York site is located on the flood plain of Beaver
              Creek, which is located 0.1 mile southwest of the site.
              The site is situated on a nearly flat lying terrace above
              the river.  The surface soils is a silt loam. These soils
              most likely formed hi alluvium on stream terraces.  A
              Stratigraphic cross section based on soil borings during
              the demonstration was prepared (Figure 3-8). The top
                                                     18

-------
FIGURE 3-8. REFERENCE METHOD STRATIGRAPHIC CROSS SECTION—YORK SITE
         NORTH
           0
           -1
           -2
           -3
           -4
           -5
           -6
           -7
           -8
           -9
          -10
          -11
          -12
          -13
          -14
          -15
          -15
          -17
          -18
          -19
          -20
          -21
          -22-
          -23-
          -24-
          -25-
          -26-
                  NODE1
                                 NODE2
                                                NODE3
                                                                     NODE4
                                                                                      NODES
                                                DISTANCE (FEET)
           LEGEND
      50     80



CLAYEY SILT ft SILT (ML)

WELL GRADED SAND
                                                               SILTY CLAY (CL)

                                                               POORLY GRADED SAND
I
                                                                                             SOUTH
                                                       -0
                                                       —1
                                                       —2
                                                       --3
                                                       —4
                                                       --5
                                                       --8
                                                       —7
                                                       —8
                                                       —9
                                                       —10
                                                       —11
                                                       —12
                                                       —13
                                                       —14
                                                       —15
                                                       —18
                                                       —17
                                                       —18
                                                       —19
                                                       —20
                                                       --21
                                                       --22
                                                       —23
                                                       --24
                                                       —25
                                                       --26
                                                                                              120
SILT

SILTY CLAY (CH)
                       (  ) -  LABORATORY CLASSIFICATION (USOA)
FIGURE 3-9.  REFERENCE METHOD STRATIGRAPHIC CROSS SECTION—FORT RILEY SITE
         SOUTH
           0-
           -1-
           -2-
           -3-
           -4-
           -5-
           -6-
           -7-
           -8-
           -9-
          -10-
           -11-
          -12-
          -13-
          -14-
          -15-
          -16-
          -17-
          -18-
          -19-
          -20-
          -21-
          -22-
          -23-
          -24-
          -25-
          -28-
          -27-
          -28-
          -29-
          -3O-
          -31-
                  NODE1
                              NODE2
                                                 NODES
                                                                    NODE3
                                                                                      NODE4
                                                 (LOAM)
                                                 (SANDY
                                                  LOAM)
                                                                                              NORTH
                -I
                -2
                -3
                -4
                •-5
                •-6
                -7
                -8
                •-»
                -10
                -11
                -12
                -13
                -14
                -15
                -16
                —17
                —IB
                —19
                —20
                —21
                —22
                -23
                -24
                -25
                —26
                -27
                -28
                -29
               —30
                                                DISTANCE (FEET)
                   100    110    120    130     140    150    160    17O     180    190-   200    210     220
          LEGEND
                                           SILTY CLAY (CL)

                                           SILTY CLAY (CH)


                      (  ) - LABORATORY CLASSIFICATION (USDA)
                                        . WELL GRADED SAND

                                         POORLY GRADED SAND
                                                    19

-------
TABLE 3-1.  COMPARISON OF GEOLOGIST DATA AND GEOTECHNICAL LABORATORY DATA-
ALL SITES
Site
Geologist Classification
Geotechnical Laboratory Classification
Match
Atlantic     Silty Clay (ML)

            Clayey Silt (ML)

            Silt (ML)

            Well Graded Sand (SW)

            Clay (CL)

            Silty Clay (CL)

            Silty Clay (CL)

York        Clayey Silt (ML)

            Silty Clay (CL)

            Well Graded Sand (SW)

            Poorly Graded Sand (SP)

            Clayey Silt (ML)

            Sand (SW)

Fort Riley    Poorly Graded Sand (SP)

            Fill

            Poorly Graded Sand (SP)

            Clayey Silt (ML)

            Poorly Graded Sand (SP)

            Poorly Graded Sand (SP)

            Poorly Graded Sand (SP)

            Well Graded Sand (SW)
                            Sandy Lean Clay (CL)

                            Clay or Silt (CL or ML)

                            Silt or Clay (ML or CL)

                            Well or Poorly Graded Sand (SW or SP)

                            Sandy Lean Clay or Sandy Silt (CL or ML)

                            Sand Lean Clay or Sand Silt (CL or ML)

                            Silt or Clay (ML or CL)

                            Silt or Clay (ML or CL)

                            Silt or Clay (ML or CL)

                            Silty to Clayey Sand (SM or SC)

                            Poorly Graded Sand with Silt or Clay (SW-SC or

                            Silt or Lean Clay (CL or ML)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand with Gravel (SC or SM)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand (SM or SC)

                            Silty or Clayey Sand (SM or SC)

                            Silty or clayey sand (SM or SM)
                                                    Noa

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Yes

                                                    Noa

                                                    Yes

                                                    Yes

                                                    Noa

                                                    No

                                                    Yes

                                                    Noa

                                                    No

                                                    Noa

                                                    Noa

                                                    Noa

                                                    Noa
Notes:
        These failures to match were due to the geologist underestimating the percentage of fines in the sample.
        Unified Soil Classification System two-letter code.
1 to 2 feet of the cross section was fill.  From 2 to 14
feet bgs, the cross section consisted of clayey silt with
some lenses of silty clay and silt. At approximately 14
feet bgs, there were thick lenses of silt, sandy silt, and
sand.  These lenses were approximately 7 feet thick and
were interfingered with each other. At approximately 21
feet bgs, the material became primarily well graded sand
to the bottom of the section at 25 feet bgs.
                                            Six soil  samples  were  collected  to  verify the
                                        geologist's  borehole logging at the York  site.   The
                                        geologist's   classification   of  soils   matched  the
                                        geotechnical laboratory's classifications  four out of six
                                        times (Table 3-1).  The two points of disagreement were
                                        samples DR27 (Node 1, 15 to 15.5  feet bgs) and DR 29
                                        (Node 3, 12 to 13 feet bgs).  In both cases, the geologist
                                        underestimated the percentage  of silt  and clay size
                                                 20

-------
particles hi the samples.  This is a common point of
variance between field soil classification and laboratory
classification.   These differences are  magnified in
grossly contaminated soils, and when the geologist is
forced  to  wear plastic  gloves  during classification
activities.   The variances  described above are not
uncommon hi environmental studies,  and  thus, the
geologist's stratigraphic borehole logs, while exhibiting
some  disagreement  with  the laboratory  data,  are
considered  to  meet  the  demonstration  DQOs for
screening level data.

Fort Riley Site

    The Fort Riley site is located on the flood plain of
the Kansas River, which is located 0.1  mile southeast of
the site. The site is situated on a nearly flat lying terrace
above the river.  The surface soil is a silt loam.  This
soil  is  most likely  formed from  deep alluvium.  A
stratigraphic  cross  section  based  on  soil borings
conducted  during the demonstration is  presented on
Figure 3-9.  This cross section showed typical deposition
hi an alluvial setting with interfingered beds of clay, silt,
silty clay, clayey silt and sand.  In the center of the cross
section, the  top  8 feet  was fill.   The  northern and
southern edges of the cross section were silt or silty clay
at the surface. In the northern half of the cross section,
poorly graded sand was present from 5 to 17 feet bgs.
In the southern half of the cross  section from 8 to
approximately 18 feet bgs, the cross section consisted of
interfingered lenses of clay, silty sand, and sand. Below
20 feet bgs, the cross  section became primarily sand
with silt and clay lenses of various thickness intermixed
to the terminal depth of the cross section.

    Eight  soil samples were  collected to verify the
geologist's borehole logging at the Fort Riley site. The
geologist's   classification  of  soils   matched  the
geotechnical laboratory's classifications one out of eight
times (Table  3-1).  Both classifications did correctly
identify the dominant particle size fraction.  In all of the
cases of disagreement, the geologist underestimated the
percentage of silt and clay size particles.  Small shifts hi
the estimation of these particles can alter the descriptive
modifier used  hi classification. The variances described
above affect the accuracy of the reference stratigraphic
cross sections as  far  as the secondary classification
modifiers are concerned.  The baseline classification as
to the dominant particle size is accurate. This data met
the demonstration's  DQOs, however, decisions based
solely on differences in classification modifiers should be
qualified as semiqualitative.
                                                     21

-------
                                             Section 4
                Site Characterization and Analysis Penetrometer System
    This section describes the SCAPS sensors that were
evaluated during this demonstration.  The description
provided is based  on information  provided  by the
developer,  on information PRC obtained from reports
and journal articles written about the technology, and on
observations  made  during the demonstration.   The
description includes background information  on the
technology and its components,  general  operating
procedures, training and maintenance requirements, and
the cost of the technology discussed.

Background Information

    The SCAPS LIF sensor was developed by the Army
(U.S. Army Corps of Engineers, Waterways Experiment
Station  [WES]  and the Army Environmental Center
[AEC]), Navy (Naval Command, Control and Ocean
Surveillance Center),  and the  Air Force (Armstrong
Laboratory).   This system uses laser light to cause
compounds in  soils to fluoresce and  measures the
resulting fluorescence.  Currently, this technology is
most  commonly  used  to detect  PAH compounds
associated with petroleum fuels. The U.S. Army holds
a patent for this combination of a sapphire window and
cone penetrometry. The LIF sensor was modified from
a design developed  by the Navy for use in detecting
petroleum,  oils, and lubricants in seawater.

    The SCAPS CP  sensor  is  a  standard sensor
commercially available.

Components

    This section describes the components of the SCAPS
LIF  and CP  system,  which consists of a  cone
penetrometer truck, modified  CP,  sampling tools,  a
nitrogen (N^ laser, and a fluorescence detection system.

Cone Penetrometer Sensor

    A complete CP system consists of a truck, hydraulic
rams and associated controllers, push rods,  samplers,
and the CP sensor itself.   The weight of the truck
provides a static reaction force, typically 20 tons, against
which  the  hydraulic  system  works   to   advance
1-meter-long  segments  of  3.57-centimeter-diameter
threaded push rod into the ground.  The CP, which is
mounted on the end of the series of push rods,  contains
sensors that continuously  log  tip  stress  and sleeve
friction.  The data from these sensors is used to map
subsurface stratigraphy. Conductivity or pore pressure
sensors can be driven into the ground simultaneously
with the tip resistance and sleeve friction sensors. The
conductivity and pore pressure sensors  are  used to
further define subsurface stratigraphy.

    Soil, groundwater, and soil gas sampling tools can
also be used with the CP system.  These capabilities are
discussed  in  greater  detail hi the  general  ITER.
Generally, sampling tools and sensors cannot  be used
concurrently.

    In favorable stratigraphies, push depths of 50 meters
or greater have been achieved. The CP can be pushed
through asphalt, but concrete must be cored  prior to
advancing the  CP.  Advancing sensors and sampling
tools with the cone penetrometer truck may be  difficult
in the following subsurface environments:
        Gravel units
        Cemented sands and clays
        Buried debris
        Boulders
        Bedrock
    The cone penetrometer truck used with the SCAPS
sensors is fitted with a steam cleaner to decontaminate
the push rods as they are withdrawn from the ground.
The  decontamination  water  is  contained  in  the
decontamination apparatus  and  it  can  be  directly
discharged into a storage container. In addition, the
combination CP and LIF sensors used in the SCAPS is
modified to provide automatic grouting of die CP hole
during the retraction of the push rods. The decontami-
                                                  22

-------
nation water, pressure sprayer, and grouting pump are
mounted in a trailer that can be towed behind the cone
penetrometer truck.

    The SCAPS system is mounted on  a  specially
engineered 20-ton truck designed with protected work
spaces  which provide  additional  health  and safety
protection to SCAPS workers at hazardous waste sites.

LIF Sensor

    The SCAPS  LIF sensor's main components are the
N2  laser, fiber  optic  cable, and the fluorescence
detection system, and the computer system.   The N2
laser creates  laser light of a known wavelength.  The
laser light passes along a fiber optic cable and into the
soil through a sapphire window, 2 millimeter (mm) thick
and 6.35 mm in diameter, mounted 65 centimeters (cm)
above the terminal end of the CP probe hi which it is
mounted.  Induced fluorescence from the soil is returned
to the fluorescence detector along a second fiber optic
cable.  The fiber optic cables are all silica fiber optic
cables,  365  micrometers  0/m)  hi  diameter.    A
photodiode  array  (PDA)  and optical multichannel
analyzer (OMA) is used as the fluorescence detector,
and the data is processed by a computer system.  The
return fluorescence data and soil stratigraphy data (from
the  CP) are collected and  interpreted by the same
computer system.   A diagram of the SCAPS sensor
configuration is shown on Figure 4-1.

    To   operate  the  SCAPS  sensors,  the  cone
penetrometer truck must be positioned over a designated
penetration point.  At this tune, the LIF sensor's
response is checked using a standard rhodamine solution
held against the  sapphire window.  This procedure is
carried out before and after each push. The CP and LIF
sensor are then advanced into the soil at a rate  of
2 centimeters per second (cm/s) or approximately 4 feet
per minute.

    The LIF sensor is operated  with a N2 laser that
provides excitation pulses at a rate of 10 pulses per
second (Hz).  The PDA accumulates the fluorescence
emission response over 10  laser shots, and  then  an
emission spectrum of the soil fluorescence is retrieved
from  the  PDA by the  OMA and computer  system.
Therefore, at the data acquisition rate of 10 Hz and a
penetration rate of 2 cm/s, the spectral resolution of the
LIF detection system under these operating conditions is
2 cm.  The fluorescence intensity at  peak  emission
wavelength for each stored spectrum is displayed hi real
tune on a panel plot,  which also includes the soil
classification data from the  CP sensor  (Figure 4-2).
This sensor is described hi detail in the general ITER.
LIF Sensor Components

The mam SCAPS LIF sensor components are:

    «   N2 laser

    «   Fiber optic cable (365  pm diameter) and
        modified CP fitted with a sapphire window

    •   Fluorescence detection system

    «   Computer system

    Each SCAPS LIF sensor component is discussed hi
more detail below.

N2 Laser

    Laser radiation excitation is produced by a pulsed
nitrogen laser made by Photon Technology, Inc. (PTI).
The laser produces light at a wavelength of 337 nano-
meters (nm) with an intensity of approximately 1 mega-
joule  (Mj).   The emitted laser radiation is  focused
through a lens and directed into the excitation fiber.

Fiber Optic Cable

    Each laser pulse is  focused  through  a lens and
directed  into an Ensign-Bickford  hard coat, all-silica
optical fiber  with a core  diameter of 365  jum.  The
core/cladding diameter is approximately 400 /um. The
optical  fiber along  with  a  return  fiber   (same
specifications), instrumentation cables, and a grout line
are all protected  by a neoprene shrink tubing jacket
forming the sensor umbilical, which is passed through
the center of each push rod.   The transmit  fiber is
terminated at a 2-mm-thick, 6.3-mm-diameter sapphire
window, which is coated with an anti-reflective material
to reduce 337 nm light backscatter into the return fiber.
This sapphire window is removable to facilitate periodic
replacement as necessary. The sapphire window passes
the laser light onto the  soil surface adjacent to  the
window.   The fluorescence signature  of the  soil is
returned  by  another optical  fiber  with  the  same
specifications of the transmit fiber.   The return fiber
passes the returned light into  the monochromator
(EG&G Princeton Applied Research Company [PARC],
Modd 1229 Monochromator).

Fluorescence Detection System

    When return fluorescence travels up through the
return fiber,  it first enters the monochromator.  The
monochromator contains mirrors and a grating so that
                                                  23

-------
r
              FIGURE 4-1. TRI-SERVICES SCAPS
                  Space Far Extra
                    Proboa
              FIGURE 4-2. SCAPS PANEL PLOT - NODE 4 ATLANTIC SITE
                            Cone Resistance
                            qc (tons/ft')
                            1     100
                                     1000
 SlMv* Friction
 f, (tons/ft1)
0 Z 4  6  B 10
                                                      CPT based SOIL
                                                      CLASSIFICATION
  Ss.S  2
  *> 4J  » 4J a a o
e •-• c C c
                                                                    O  300 1000 1300 MOO
                        «r«it*nc* Inttntuv    Hcvtltngttl
                         =•«"» - SMB1» •««   at Piak 
-------
the returned light is  diffracted into its  component
wavelengths. The light then exits the monochromator
and  enters  an  EG&G  PARC Model 1421B-1024-G
intensified silicon PDA detector,  which is attached
directly to the monochromator.  The detector is capable
of being gated and provides a blue to mid-spectrum
response using a 1024 element array.  The intensity of
the returned light causes the internal diodes to produce
an electrical signal directly proportional to the intensity
of the incident light for each of the 1024 elements of the
PDA.    Each  element corresponds  to  a  particular
wavelength.  The PDA detector is  controlled by a
EG&G PARC Model 1460 OMA. The OMA receives
the data and  displays  the  spectral signature  of  the
returned signal.  The OMA can be used as a stand-alone
processor using its display and touch screen technology
to control the detector and communicate  with external
devices.  However, the system also can be controlled by
an external computer via a GPID interface.

Computer System

    The computer system is comprised of two Hewlett
Packard 486DX33 Vectra computers. One computer is
used as the data acquisition computer and  the  second
computer is used for post-acquisition processing.  The
data acquisition computer is used to communicate and
transfer data from the OMA and record measurements of
soil stratigraphy.   Both  the soil classification and LIF
sensor response are displayed in real time  during  the
advancement of the CP. Once the push is completed,
the data is transferred (through a local area network) to
the  post-processing computer  where   the  data  is
manipulated and plotted.   It  should be  noted that
although normal sensor data consists of the fluorescence
intensity response at peak emission wavelength,  SCAPS
LIF sensor is configured to collect and store the entire
fluorescence spectrum  from approximately  300  to
800 nm.

General Operating  Procedures

    Four people are needed to operate the  SCAPS as
currently deployed.  The  crew chief, the LIF sensor
operator,  the  hydraulic ram operator,  and the rod
handler.  The crew chief is an experienced engineer that
plans and manages the total deployment of the SCAPS.
This  involves  predeployment and post-deployment
logistics, push rod decontamination,  and grouting.  The
actual  collection of data on site is handled by the three
other crew members.   The hydraulic  ram operator
operates the hydraulics of the cone penetrometer truck
and monitors CP depth and soil stratigraphy data.  The
rod handler screws the push rods into place as the CP is
advanced.  The LIF sensor operator monitors both the
LIF sensor response and the soil classification data as the
push is executed.  The LIF sensor operator also handles
the post-acquisition data processing between penetration
events,  and produces the final chemical and physical
characterization reports.

Cost

    The SCAPS LIF and CP sensors are commercially
available.  However, there are a number of SCAPS units
currently available to various  Federal agencies under
cooperative work agreements with the U.S. Army Corps
of Engineers.   WES has produced five SCAPS units:
one for research, three for the U.S. Army Corps of
Engineers, and one for the Department of Energy.  The
Navy has produced two units for deployment and one for
research use.    Similar  LIF  and CP technology  is
available from either Hogentogler or Applied Research
Associates,  Inc., both of  which have  non-exclusive
licenses from WES to use LIF technology with cone
penetrometry.

    WES has produced an operations manual for the
SCAPS and has limited training for U.S. Army Corps of
Engineers SCAPS operators.

    Currently,  WES estimates the daily rate for use of
the SCAPS LIF and CP sensors would be $3,500. This
cost represents operating costs.   The cost does not
include  normal resources associated with commercial
application, such as  marketing,  research, and profit.
Mobilization and operator per diem costs are included in
the daily rate. Based on the daily use charge of the LIF
and CP sensors, a total cost of approximately $20,000
was realized for the three site characterization activities.
This cost  includes the  initial mobilization  and the
subsequent inter-site mobilization required for two days
of travel. The data was generally generated in two days
at each site and a total of two days of travel between all
3 sites was used. For comparison, the predemonstration
activities used conventional field screening  and produced
similar data at the three sites; however, it required more
personnel and  on-site analytical  capabilities.   The
approximate  three   site  characterization  cost  was
$43,000.   This effort resulted hi fewer data points,
relative to the  continuous data output of the  SCAPS
sensors. In addition, the predemonstration activity only
produced one borehole log at each site.   Another cost
comparison can be made relative to  the costs  accrued
producing  the  reference  cross  sections  for  this
demonstration.    The reference cross  sections  cost
approximately   $55,000,  including  approximately
$30,000 for drilling services, approximately $8,000 for
an  on-site  geologist and  a  sample, approximately
$12.,000  for   off-site   analytical   services,   and
approximately   $5,000 for  handling and disposal of
investigation derived waste.
                                                   25

-------
Observations

    Observations recorded during the demonstration of
the SCAPS LIF and CP sensors are briefly summarized
below.

    The SCAPS conducted a total of 30 grid pushes and
6 non-grid pushes during the demonstrations at the three
demonstration sites.  The following discussion reflects
observations made by Dr. Harry Ellis of PRC during the
demonstration of the SCAPS. Dr. Ellis did not operate
the SCAPS equipment and required no training prior to
the demonstration.  Because WES was responsible for
operations, Dr. Ellis was an observer only.

    The crew  operating  the SCAPS  unit  was a
developmental group, rather than a general operational
crew. The only difference between these crews involved
the number of personnel.  The developmental group
crew  included a dedicated person to post-process  the
data.

    In some cases, the size of the SCAPS truck made
on-site maneuvering in confined spaces difficult.

    Once at a demonstration site, it took 3 to 4 hours to
convert the SCAPS unit from a road travel mode to
operating  mode.   This  included  unpacking   the
computers, LIF sensor, and other sensitive components,
connecting and testing these components, connecting the
trailer to the truck, and so on. Moving about the site
from  one  push  location to  another   required  no
adjustments except lifting up the  access ladder.  The
decontamination  and grouting trailer can be moved
separately from the truck hi close quarters.  If this is
done, the connecting lines will  usually have to be
disconnected and reconnected which  takes  a  few
minutes.   Demobilization  in preparation  for road
movement to a new site takes about 2 hours.

    Based  on   the  progress  of work during   the
demonstration,  it is possible to estimate the speed of
operations.  The average push was  35 feet below grade.
These estimates are as follows:

    •   About 1 hour per day for watering, fueling, and
        minor maintenance

    •   Approximately 1 to 1.5 hours per 35 foot push.
        This includes all operations from placement on
        location to placement on the location for  the
        next push.  This includes  truck movement
        between  points,   rod   advancement  and
        withdrawal, grouting,  and decontamination.
    If the deepest pushes of the demonstration are
considered, 75 feet below grade, it would add 10 to
15 minutes per push and require more frequent filling of
the clean  water storage  tank.   This  suggests  that
additional depths can be achieved with minimal impact
on throughput. The automatic decontamination and hole
grouting while the push rod is  being withdrawn are
advantages  of  this technology  and  greatly  increase
sample throughput.

    Rock, debris, and similar items downhole may stop
advancement of the push rods  and  sensors.   It is
necessary to have a good idea  of the actual subsoil
conditions  before estimating production rates  for the
SCAPS at a given site.  The terrain can limit the use of
the SCAPS.   It  needs  about 20  feet of overhead
clearance.  Side slopes and rough terrain can limit its
use. The leveling jacks can compensate within limits.

    The N2 laser used  by the SCAPS  LIF sensor
consumes nitrogen gas.   Currently,  the nitrogen gas
cylinders are mounted on the  decontamination  and
grouting trailer. Whenever the clean water tanks on the
trailer need refilling, operation  of the LIF sensor is
stopped because the nitrogen source for the N2 laser is
attached to the decontamination and grouting trailer.
This potential downtime could be decreased by mounting
the nitrogen cylinders on the cone penetrometer truck
itself rather than the trailer.  However, this would create
an ergonomic problem of lifting these heavy items into
position. If this were done, the trailer could be taken for
water refill during a calibration and push, and part of the
1 hour per day for replenishment  would be eliminated.

    Normal wear and tear does slow down operations.
Two such items were noted during this demonstration.
The fiber optic cable hi use during initial  pushes was
nearing the end of its useful life (about 150 pushes is
estimated).  This made it difficult to achieve optimal
operating conditions during the daily ICALs.  Also, the
jaws which grip the probe rod were worn and caused
intermittent problems with probe withdrawal due to
internal slippage.

    Of  the 36 pushes done during this demonstration,
three resulted in catastrophic unit failure.  This equates
to an 8  percent catastrophic failure rate. Catastrophic
failure either resulted in the physical loss of the CP and
LIF sensor, LIF sensor down-tune in excess of 8 hours,
or the disabling of one  of the  SCAPS components.
Specifically:

    •   While at Grid 5  at the  York site, the grout
        pump seized due to concrete clotting in the
                                                   26

-------
        helical  pump.    This pump was  original
        equipment (about 5 years old).  Hand grouting
        was used until the pump could be temporarily
        repaired.    This  caused  a  work delay  of
        approximately 4 hours and an  added  1  to
        1.5  hours to the completion tune for  each
        subsequent push due to the tune associated with
        hand grouting.

    •   During the first nongrid push at the York site,
        the probe stopped producing a response.  The
        probe was pulled, and neither a post calibration
        or a flashlight shining directly into the sapphire
        window would elicit any response from the
        fluorescence detector.  The OMA appeared to
        be functioning normally, so it was concluded
        that  the  fiber  optic  cable  had   broken.
        Therefore, the crew rigged a new probe and
        umbilical and  resumed the push.   The old
        umbilical was returned to WES for repair.  The
        repair was estimated to cost $2,000 and was
        expected  to require  a day's labor  from two
        instrumentation technicians. This cable break
        resulted  hi  a work delay of approximately
        2 hours.  This delay was minimized by the fact
        that the  SCAPS is  deployed with a second
        sensor,  and umbilical cord which is already
        threaded through a second set of push rods.

    •   During the last nongrid push at the Fort Riley
        site, the sensor array was lost downhole due to
        push rod breakage during retrieval. The broken
        end  of the  push rods that  were retrieved
        exhibited a fracture along the male threads.

    The primary maintenance practice with the SCAPS
LIF and CP sensors is to inspect and repair as necessary.
It will be useful to accumulate the experience necessary
to predict the useful life of various  SCAPS components
and set up a more detailed schedule for overhaul  or
replacement of components.

Data Presentation

    To qualitatively assess the abilities of the SCAPS
CP  sensor  in  identifying  the  subsurface  textural
properties of a site, it was required to collect soil texture
data during its advancement at each of the five sample
nodes  at each site.   The  nodes were  arranged  hi a
transect line across  a known area of subsurface soil
contamination  identified  during  predemonstration
sampling and previous investigations conducted at  each
site.   Sampling at a node was continuous from the
surface of the soil a depth of 50 feet.
    The soil texture data generated by the technology
was used to produce soil texture cross  sections along
each transect line.  A comparison of its data to that of
the reference methods is discussed in Section 5.  The
following sections present chemical and physical data for
the SCAPS site.

Chemical Data

    Two types of SCAPS  data are presented hi this
section.  The data used in the qualitative evaluation is
presented and discussed as cross sections. The SCAPS
data used in the quantitative  evaluation are  listed  hi
Tables 4-1, 4-2, and 4-3.   This data is discussed  hi
Section 5.

    The SCAPS LIF logs  are used to describe the
relative distribution of subsurface  contaminants and
produce  contaminant cross sections.   Although the
SCAPS  produced   its  own   cross  sections,  PRC
transferred the data and plotted it on a scale that matched
the ones used  for  the reference  method.    The
transformed  SCAPS  chemical  cross   sections are
presented on Figures 4-3, 4-4, and 4-5.  An example of
the standard SCAPS LIF graphic is shown on Figure
4-6.

    The LIF sensor data was reported as intensity at the
peak wave length.  Theoretically, changes  hi intensity
relative to background can be  used to assess relative
changes hi the concentration of subsurface  fluorescing
materials.   In theory, as  the LIF  sensor  intensity
increases,   the   concentration   of    contaminants
(fluorescing)  also may increase. One objective of this
demonstration was to evaluate  this  relationship.  The
following data presentation was produced by the SCAPS
operator  and  represents  a  typical  narrative data
evaluation provided by SCAPS. These narratives can
often be generated hi the field within 24 hours of data
acquisition.

Atlantic Site

    The standard operating procedures for the SCAPS
LIF sensor data interpretation include review of panel
plots.  These plots include soil stratigraphy, fluorescence
peak intensity,  and wavelength at peak intensity. An
example panel plot is seen on Figure 4-2.  Based on the
fluorescence response at various depths, the fluorescence
emission spectra for a particular depth was inspected to
determine if different contaminants were present.

    At  the Atlantic  site,  the LIF sensor  response
indicated  that sampling  Node 1 was at background
fluorescence  levels.   Sampling Node 2 showed the
presence of a fluorescing contaminant at 20.5 to 24 feet
                                                   27

-------
TABLE 4-1. QUANTITATIVE EVALUATION DATA FOR THE ATLANTIC SITE
Node
2
2
3
4
4
4
5
5
Depth (feet)
21-22
24-25
16-17
6.5 - 7.5
10-11
27.5 - 28.5
16-17
23.5 - 24.5
Number of
Readings
5
6
5
5
5
5
5
6
Maximum
Fluorescence
Readinq
9,078.00
10.33
1,638.0
425.2
1,195.0
13,623.0
21,225.0
36.25
TABLE 4-2. QUANTITATIVE EVALUATION DATA
Node
2
2
3
4
4
4
5
Depth (feet)
15-16
13.5 - 14.5
17-18
17-18
14-15
18-19
1.5-2.5
Number of
Readinqs
5
5
5
5
5
6
6
Maximum
Fluorescence
Readinq
331.9
1,154.0
923.2
1,527.0
1,526.0
992.8
313.9
TABLE 4-3. QUANTITATIVE EVALUATION DATA
Node
1
1
2
2
5
5
Depth (feet)
2-3
13-14
6-7
17-18
10.5-11.5
16-17
Number of
Readings
5
6
5
6
5
6
Maximum
Fluorescence
Readinq
1,979.0
453.7
1,787.0
12,561.0
4,289.0
3,684.0
Minimum
Fluorescence
Reading
4,058.0
1.50
682.71
35.33
543.7
1,127.0
13,557.0
25.20
FOR THE YORK
Minimum
Fluorescence
Reading
118.1
493.9
33.83
518.1
219.9
166.4
275.5
FOR THE FORT
Minimum
Fluorescence
Readinq
306.1
265.7
54.82
1,245.0
1,731.0
1,620.0
Mean
5,809.0
5.37
1,323.0
213.4
837.1
6,310.0
19,574.0
29.88
SITE
Mean
221.3
723.8
268.5
908.4
775.7
412.5
297.1
RILEY SITE
Mean
1,143.0
366.6
1,136.0
4,853.0
2,923.0
3,036.0
Standard
Deviation
2,007.0
3.38
391.5
154.2
326.6
5,765.0
3,420.0
3.73

Standard
Deviation
93.15
263.6
370.1
388.9
642.6
311.1
12.91

Standard
Deviation
761.6
76.21
648.7
4,147.0
990.0
1,402.0
                                    28

-------
FIGURE 4-3.  SCAPS CHEMICAL CROSS SECTION—ATLANTIC SITE
         SOUTHEAST
                 NODES

           0-
           -1-
           -2-
           -3-
           -4-
           -5-
           -6-
           -7-
           -8-
           -9-
          -10-
           -11-
          -12-
          -13-
          -14-
          -15-
          -16-
          -17-
          -18-
          -18-
          -20-
          -23-
          -24-
          -25-
          -28-
          -27-
          -28-
          -29-
          -30-
          -31-
          -32-
          -33-
          -34-
          -35-
          -35-
          -37-
    NODE4
              NODE3
                               NODE2
                                                            NORTHWEST
                                                          NODE1
                                                                  -0
                                                                  —1
                                                                  —2

                                                                  —4
                                                                  --5
                                                                  —«
                                                                  —7

                                                                  --»
                                                                  --10
                                                                  —11

                                                                  ::!§
                                                                  —14
                                                                  --15
                                                                  —16
                                                                  —17
                                                                  —18
                                                                  —19
                                                                  —20
                                                                  --21
                                                                  —22
                                                                  --23
                                                                  —24
                                                                  —25
                                                                  --28
                                                                  "27
                                                                  —28
                                                                  —2«
                                                                  —30
                                                                  --31
                                                                  —32
                                                                  —33
                                                                  —34
                                                                  --35
                                                                  —38
                                                                  —37
                                                 DISTANCE (FEET)
                                                      150
                                                                                 250     275    300
          LEGEND
30 - 100 (COUNTS)
.-_-_-] 100 - 1.OOO
] 1.000 - 10.000
 FIGURE 4-4. SCAPS CHEMICAL CROSS SECTION—YORK SITE
          NORTH
            0-
            -1-
            -2-
            -3-
            -4-
            -5-
            -8-
            -7-
            -8-
            -9-
           -10-
            -11-
           -12-
           -13-
           -14-
           -15-
           -18-
           -17-
           -18-
           -19-
           -20-
           -21-
           -22-
           -23-
           -Z4-
           -25-
           -28-
                   NODE1
                                  NODE2
                                                  NODE3
                                                                      NODE4
                                                                                       NODES
                                                                                               SOUTH
                                                                   -0.
                                                                   —1
                                                                   —2
                                                                   —3
                                                                   —4
                                                                   —5
                                                                   --6
                                                                   —7
                                                                   —10
                                                                   —11
                                                                   —12
                                                                   —13
                                                                   --14
                                                                   —15
                                                                   —IS
                                                                   —17
                                                                   —IB
                                                                   —19
                                                                   —20
                                                                   —21
                                                                   —22
                                                                   —23
                                                                   —24
                                                                   —25
                                                                   —28
                                                  DISTANCE (FEET)
                                                       60
              LEGEND
  10 - 100 (COUNTS)
                                                        - 1,000
                                                                         31.000 - 10,000
                                                     29

-------
FIGURE 4-5. SCAPS CHEMICAL CROSS SECTION—FORT RILEY SITE
         SOUTH
           0-
           -1-
           -2-
           -3-
           -4-
           -5-
           -8-
           -7-
           -8-
           -9-
          -10-
          -11-
          -12-
          -13-
          -14-
          -15-
          -18-
          -17-
          -18-
          -19-
          -20-
          -21-
          -22-
          -23-
          -24-
          -25-
          -28-
          -27-
          -28-
          -29-
          -30-
          -31-
                  NODE1
                              NODE2
                                              NODES
                                                             NODE3
                                                                                     NODE4
                                                                                             NORTH
                                                                                    -7
                                                                                    •-8
                                                                                    -9
                                                                                    •-10
                                                                                    -11
                                                                                    -12
                                                                                    -13
                                                                                    -14
                                                                                    -15
                                                                                    -18
                                                                                    •-17
                                                                                    -18
                                                                                    -19
                                                                                    -20
                                                                                    -21
                                                                                    —22
                                                                                    •-23
                                                                                    —24
                                                                                    -25
                                                                                    —28
                                                                                    •-27
                                                                                    -28
                                                                                    -29
                                                                                    -30
                                                                                    -31
                                                DISTANCE (FEET)
                   100  — 110
                               120
                                     130
                                                  ISO
                                                        180
                                                              170
                                                                    180
                                                                          190
                                                                                200
                                                                                       210
                                                                                             220
          LEGEND
                              - 100 (COUNTS)
                                                    - i.ooo
                                                           11.000 - 10.00O
FIGURE 4-6.  SCAPS STRATIGRAPHIC CROSS SECTION—ATLANTIC SITE
          SOUTHEAST
                 NODES
            0-
           -5-

           -7-
           -8-
           -9-
           -10-
           -11-
           -12-
           -13-
           -14-
           -15-
           -18-
           -17-
           -23-

           -%'-
           -a-
IE
-34-
-35-
                      NODE4
                                 NOOE3
                                                  NODE2
  NORTHWEST
NODE1
        -0
        —1
        — 2
        — 3
        — 4
        — 5
        —8
        ~7
        — 8
        —9
        —10
        — 11
                                                                                                17
                                                                                                IB
                                                 DISTANCE (FEET)
                                                                                               -20
                                                                                                21
                                                                                                22
                                                                                                23
                                                                                                28
                                                                                                27
                                                                                                -28
                                                                                                29
                                                                                                30
                                                                                                31
                                                                                                32
                                                                                                33
                                                                                                34
                                                                                                35
                                                                                                38
                                                                                                37
                    25     50     75     100     125    150     175     200


            LEGEND       »CAY    F^jsAND    Mis™ MIX  M\™
                                                                                250
                                                                                             300
                                                    30

-------
bgs  that  had  a   fluorescence  emission  peak  at
approximately 480 nm. The same pattern was observed
at both pushes in this sampling node.  Sampling Node
3 indicated fluorescence at 10 feet which continued until
approximately 25.5  feet.  Inspection of the emission
spectra for this contaminant zone indicated contaminants
with different spectral features from those observed at
Node 2 and, therefore,  the possible presence of two
different products.  The fluorescence response for Node
4  indicated significant  contamination areas  at  8  to
30 feet. The emission spectra changed with depth.  Near
the  top  of the zone,  the  spectra  indicated  peak
fluorescence at approximately 425 nm.  The spectra at
28.7 feet indicated a possible  mixture. The spectra at
30 feet had an emission maximum at 480 nm.  This was
similar to  the spectra from Node 2 at 22 feet.  The
significant fluorescence response for Node 5 began at
approximately 8 feet and continued to 22 feet.  The
emission   spectra  indicated   a  product   that  was
significantly different from the contaminant at the other
nodes.  This product had  an emission maximum at
400 nm.

    The above descriptions are based on the real-tune
outputs from the LIF sensor.   Thus, the operator can
identify different wastes as pushes  are  made.   It is
important to note that this data is only used to identify
differences in wastes and not identify specific wastes or
classes of contaminants. Figure 4-2 shows the panel plot
for Node 4 at the Atlantic site.  Shifts in "Wavelength at
Peak (nm)" seen in the far right of the panel plot clearly
shows  the emission wavelength shifts discussed above,
and  used  to  identify  changes  hi  waste  type
characteristics.  Figure 4-9 shows three individual plots
of fluorescence  intensity versus wavelength for select
depths from the Node 4 push.  These plots  can  be
produced  after  a  push  and represent waveforms  at
distinct depths  during a push.  This  data is used to
confirm the conclusions regarding waste type differences
based on the panel plots.

York Site

    The  initial  review of the panel plots during and
immediately after pushes indicated the potential presence
of three distinct contaminant types based on the different
wavelengths observed for the peak fluorescence response
at various  depths.   The spacial distribution of these
different contaminants on site was relatively consistent
across the five sampling nodes. However, the intensity
of the  fluorescence response for each contaminant did
vary significantly between  a number of the  sampling
nodes.  The contaminant near the ground surface was
consistently found to have  an emission spectrum that
peaked at  approximately 400 nm.  The contaminants
detected at greater depths always  yielded  emission
spectra with peak wavelength that increased with depth.
Review of individual spectra at various depths indicated
a contaminant with an emission spectrum peaking at a
lower wavelength (450 nm) overlaying a contaminant
with  an  emission spectrum  peaking  at  a  longer
wavelength (480 to 500 nm). These observations will be
discussed hi detail below for the individual pushes.

    The two Node 1 pushes yielded panel plots that were
similar.   The  stratigraphy was similar, as was the
fluorescence response.  This sampling node yielded low
fluorescence  response (300 to  500 counts)  at 14 to
18  feet.   The emission spectra of this  fluorescence
response indicated a contaminant with peak fluorescence
at approximately 490 to 500 nm.

    The two Node 2 pushes yielded panel plots that were
similar except for two observations. First, the second
push 'indicated the presence of a contaminant near the
surface (4 to  7 feet) that had an emission maximum at
410 nm that was not observed in the first push.  This
was the first observation of this fluorescence response on
this site.  The second difference between these two panel
plots was that the fluorescence response observed at
greater depths in the second push was resolved into two
bands (12 to 16 feet and 17 to 20 feet), while no similar
spacial resolution was observed for the first push.
However, the spectra for the two contaminant regions hi
the second push were very similar to those  observed
from the top to the bottom of the band observed between
11  to 18 feet for the  first push.  The  fluorescence
response of the upper  zone hi this band indicated a
contaminant with a spectrum peaking between 410 and
430 nm.  The middle of the band had a spectrum that
peaked at 450 nm, and the bottom of the band had a
spectrum that peaked at 480 nm.  It should be noted that
the change hi the emission spectra could be inferred
during the  push  by  observing the change  hi the
wavelength  at  peak  fluorescence  panel  that  was
generated in real tune during a push event.  As discussed
earlier, the pattern observed at this sampling node was
generally repeated at the other  sampling nodes.

    The spacial distribution and spectral characteristics
of the fluorescence response observed hi the two pushes
at Node 3 were very similar to those observed hi Node
2.  However, the intensity of the fluorescence response
hi the second push at Node  3 was  lower than that
observed for the first push at Node 3.  The wavelength
of  peak  fluorescence  intensity for the  contaminants
detected at 11 to 15 feet and  16 to 20 feet were very
similar to the spectra obtained hi the Node 2 pushes at
similar depths.  The two different fluorescence spectral
responses were well resolved spatially hi both  the Node
3 pushes.
                                                    31

-------
    The two pushes  at Node 4  again indicated  a
fluorescence response near the surface (0 to 1.5 feet).
The spectrum of this contaminant was the same as that
observed near the surface hi the second push at Node
2  (maximum fluorescence at 410 nm).  The pattern
observed for Node 2 and Node 3; increasing wavelength
of peak fluorescence response with increasing depth was
observed for the two pushes at Node 4. The intensity of
the fluorescence response  at this  sampling node was
similar to those observed at Nodes 2 and 3.

    The fluorescence response observed  for the Node
5  pushes indicated the low wavelength fluorescence
(410  nm)  response near the surface and  the  longer
wavelength fluorescence (400 to 500 nm) response at
greater depths. The intensity of the fluorescence at this
node was significantly less than that observed at Nodes
2, 3, and 4.

    The extra push at Node 6 (not part of the formal
demonstration) was very similar to the general pattern
observed  on  this site.   The  fluorescence response
indicated three different contaminants at different depths.
The low wavelength (410 nm) contaminant was observed
near the surface and the longer wavelength response was
observed at depth (450 nm and 480 to 500 nm).  Again,
the  differences  in the  spectra for  the  different
fluorescence responses observed at various depths can be
inferred from the wavelength at fluorescence peak panel
of the standard panel plot.

Fort Riley Site

    The results obtained at the Fort Riley site indicated
a fairly homogeneous distribution of a single contaminant
(emission  spectra with  peak  fluorescence at  about
410 nm). Node 1 indicated low level fluorescence near
the surface and from approximately 10 to 20 feet. The
first push at sampling Node 1 indicated a higher level of
contaminant in a narrow band at about 21 feet.  In the
area around sampling Nodes 2, 3, and 5, the panel plots
indicated low to high level contamination beginning at
approximately 7 to 10 feet and continuing to about 20 to
22  feet.   Sampling Node 4 indicated essentially no
fluorescence contamination.

Textural Data

    The SCAPS CP uses ASTM methods to generate the
subsurface  textural data.  The individual CP logs were
used to construct stratigraphic cross sections for each of
the demonstration sites. Although the SCAPS produced
its own cross sections, PRC transferred the data and
plotted it on a scale that matched the one used for the
reference method.   The  transformed  SCAPS  strati-
graphic  cross  sections  are  presented  on  Figures
4-6, 4-7, and 4-8.  The SCAPS data package did not
include a narrative of the stratigraphic cross sections,
therefore, a PRC geologist provided the descriptions
presented below.  The SCAPS CP software  uses the
term "mixed" to modify the soil classification.  When
the "mixed" modifier is used, the dominant particle size
is named and the term mixed  is added to indicate a
significant percentage  of different  size particles are
present.

Atlantic Site

    The SCAPS CP sensor identified an unclassified unit
hi the  surface 1 foot across the cross section.  Figure
4-2 is the SCAPS stratigraphic cross section for the site.
The SCAPS CP sensor identified a thin layer of mixed
silt from 1 to 2 feet bgs across the cross section.  From
2  feet  bgs to approximately 21 feet bgs the SCAPS
sensor identified primarily clay.  The  SCAPS  sensor
identified a silt  mix lens in the northern three nodes of
the cross section at 8 to 15 feet bgs.  This lens thinned
from 7 feet thick at Node 3 to 2 feet thick at Node 1.  In
the southern  two nodes  clay was present from 3 to
28 feet bgs. A 2-foot-thick silt mix layer was below the
clay in the southern two nodes. This is followed by sand
to the bottom of the section. In the northern three nodes
from 21 feet bgs to the bottom of the cross section, the
SCAPS sensor identified primarily sand with several thin
clay, silt mix, and sand mix lenses throughout. The
SCAPS sensor also identified a 2-foot-thick peat layer at
19.5 feet bgs hi Node 5.

York Site

    The SCAPS CP  sensor identified sand, sand mix,
and silty mix hi the upper  2 feet of the cross section.
From 2 feet to 17 feet bgs, the CP sensor logged thick
lenses  of clays  and silt mix at the York site.  Figure
4-4 is a SCAPS stratigraphic cross section for the site.
From 17 to 25 feet bgs (the bottom of the section), the
SCAPS sensor logged many thin beds of silt, clay, sandy
silt, silt mix, and sand.

Fort Riley Site

    The SCAPS CP sensor identified extensive layers of
clays,  silts, sands, and mixtures throughout this cross
section.  Figure 4-5  is a SCAPS sensor stratigraphic
cross section for the site. From the surface to a depth of
10 feet bgs, the SCAPS sensor  identified silt mix and
clay hi the south three quarters of the cross section.  In
Node 4, the SCAPS sensor identified sand and sand mix
                                                   32

-------
FIGURE 4-7. SCAPS STRATIGRAPHIC CROSS SECTION—YORK SITE
         NORTH
           0-
           -1-
           -2-
           -3-
           -4-
           -5-
           -6-
           -7-
           -8-
           -9-
          -10-
          -11-
          -12-
          -13-
          -14-
          -15-
          -16-
          -17-
          -18-
          -19-
          -20-
          -21-
          -22-
          -23-
          -24-
          -25-
          -28-
                  NODE1
                                NODE2
                                                NODE3
                                                                    NODE4
                                                                                     NODE5
                                                DISTANCE (FEET)
                                                     80
                                                                                            SOUTH
                                                                  -0
                                                                  —1
                                                                  —2
                                                                  —3
                                                                  —4
                                                                  --5
                                                                  —8
                                                                  —7
                                                                  --»
                                                                  —9
                                                                  "10
                                                                  —11
                                                                  —12
                                                                  --13
                                                                  —14
                                                                  —15
                                                                  —18
                                                                  —17
                                                                  —18
                                                                  —19
                                                                  —20
                                                                  --21
                                                                  —22
                                                                  —23
                                                                  —24
                                                                  —25
                                                                  —28
                                                                                             120
          LEGEND
           [,'•:.; .[SAND

CLAY TO SANDY SILT
                                                                          1 SAND MIX TO SANDY SILT
FIGURE 4-8.  SCAPS STRATIGRAPHIC CROSS SECTION—FORT RILEY SITE
         SOUTH
           0-
           -1-
           -2-
           -3-
           -4-
           -5-
           -8-
           -7-
           -8-
           -B-
          -10-
           -11-
          -12-
          -13-
          -14-
          -19-
          -16-
          -17
          -18-
          -19-
          -20-
           -21-
          -22-
          -23-
          -24-
          -25 H
          -28-
          -27-
          -28-
          -29-
          -30-
           -31-
             90
                  NODE1
                              NODE2
                                                NODES
                                                                   NODE3
                                                                                     NODE4
                                                                                             NORTH
                                                                   -0
                                                                   -1
                                                                   -2
                                                                   -3
                                                                   -4
                                                                    5
                                                                   •-8
                                                                   -7
                                                                   -8
                                                                   -9
                                                                   -10
                                                                   -11
                                                                   -12
                                                                   -13
                                                                   -14
                                                                   -15
                                                                   -18
                                                                   -17
                                                                   -18
                                                                   -19
                                                                   -20
                                                                   -21
                                                                   -22
                                                                   -23
                                                                   -24
                                                                   -25
                                                                   —28
                                                                   -27
                                                                   -28
                                                                   —29
                                                                   —30
                                                                   —31
                                                DISTANCE (FEET)
                   100    110    120     130     140    ISO    160    170    180     190    200    210
                                                                                             220
          LEGEND
                                         JStLTY CLAY
                                                       •I SILT MIX
                                                                      I SAND MX
                                                      l.V.-SAND
                                                     33

-------
FIGURE 4-9. FLUORESCENCE INTENSITY VS. WAVELENGTH—NODE 4 ATLANTIC SITE
   01
   4-1
   C
   
-------
                                               Section 5
                                         Data Comparison
    The data produced by SCAPS were evaluated using
the criteria described hi Section 2.  The qualitative and
quantitative data evaluations are discussed separately..
The qualitative evaluation compares the chemical and
stratigraphic cross sections produced by SCAPS relative
to cross sections from the reference  methods.   The
quantitative evaluation statistically compares the SCAPS
data with analytical data produced  by the reference
methods.

Qualitative Assessment

    The qualitative assessment presents the evaluation of
both the stratigraphic and chemical mapping capabilities
of the SCAPS sensors relative to the reference methods.
In addition, the potential affects of TOC on the system's
measurements are examined.  Both the reference and
technology cross sections were produced from collocated
sampling areas as discussed in Section 2.  Since these
methods were  sampling spatially  different locations,
matrix heterogeneity will impact the comparisons of both
the physical and chemical cross sections.  Based on a
review of the demonstration data, this impact appears to
have  had a minimal impact on the qualitative data
evaluation.

Stratigraphic Cross Sections

    The following sections present descriptions of the
similarities  and  differences observed between the
stratigraphic cross sections produced by the SCAPS CP
sensor and the reference methods. For this comparison,
PRC used the SCAPS cross sections shown in Section
4. These cross sections were produced directly from the
technology's raw data, however,  they are scaled to
match the reference method  cross sections shown in
Section  3.  These comparisons are qualitative and, as
such,  are  subjective  in nature.    However,  these
comparisons  were  made  by a  certified professional
geologist (American Institute of Professional Geologists)
with over 17 years of experience in this field.
Atlantic Site

     The SCAPS  sensor and the reference method's
stratigraphic cross  sections exhibited good correlation.
The surface materials identified as silts and silty clays by
SCAPS were  identified as fill and silty clay by the
reference methods.   Fill  is  defined  as  a man-made
deposit of rock and/or soil. Sand was identified hi the
northern (Node 1)  portion of  the cross section by both
the SCAPS and the reference methods at approximately
21 feet bgs.  A 7-foot-thick silt mix lens was identified
by both the  SCAPS  and the reference methods hi the
center  of the cross section extending from 8 to 15 feet
bgs.    The  SCAPS  and reference methods showed
relatively good correlation at Nodes 1 and 2, except that
the different strata were mapped at slightly shallower
depths  by the SCAPS relative to the reference methods.
The SCAPS also identified a  2-foot-thick peat layer at
19.5 feet bgs in Node 5, while  the field geologist saw no
evidence of peat hi the soil core.

    One notable variation between the  SCAPS sensor's
and reference methods cross sections was observed.  The
reference method had trouble collecting samples for
logging purposes in  the running sands that generally
occurred from 20  feet bgs to the termination of the
reference  borehole.   This  lack of complete  sample
recover/ is common for this method of borehole logging,
and caused the geologist to use circumstantial evidence
to fill in the resultant gaps in the borehole logs at depth.
The circumstantial evidence used was direct feedback
from the driller on changes in drilling characteristics,
cuttings, and interpolation based on what was recovered.
The SCAPS  did not need to physically collect  soil
samples to produce borehole  logs, and thus, is not as
affected by running  sands.  This explains the greater
detail  shown  hi  the SCAPS cross section below
approximately 20 feet bgs.

    Seven samples were collected at the Atlantic site for
geotechnical analysis.   The results of these analyses were
                                                   35

-------
compared to the corresponding SCAPS stratigraphic
data.  Four out of the seven samples showed intermethod
agreement.  The remaining samples were not matched
due  to the SCAPS  lack of reporting or detecting
increases in sand content. This resulted in the SCAPS
identifying intervals as clays when they were identified
by the reference method as sandy clays or  silts. This
indicates that the SCAPS may not be sensitive to small
shifts in particle size distribution.

York Site

    The SCAPS CP sensor and the reference method's
cross sections exhibited fairly good  correlation.  The
SCAPS identified sand, sand mix, and silty mix in the
top 2  feet of the  cross  section,  while the reference
method identified the same interval as fill. The SCAPS
identified components of fill and, therefore,  for this
zone, the SCAPS  and the reference method are most
likely  identifying  the same material.   From 2  to
17  feet bgs, the  SCAPS identified thick lenses  of
mixtures of clays and silt, while the reference method
identified thick layers of clayey silt with lenses of silt
and silty clay. From 17 to 25 feet bgs (the bottom of the
cross section), the SCAPS identified thin beds of silt,
clay,  sandy silt, silt mix, and sand.   The reference
method identified  this interval  as being composed of
primarily lenses of sand with sandy  silt and silt.  The
small lenses of silty clay were not identified in  the
reference method's  logs.    The  lack  of  correlation
relative to the thin sand,  silt, and clay  lenses  may be
more representative of the reference method's inability
to resolve thin strata. The detail of the reference method
can be increased by spending more time examining
sample cores, however,  time and cost factors often
prohibit fine detailed examination of sample cores.  The
SCAPS produces the same level of detail whenever it is
used. Running sands were not a problem at this site.

    Six samples were collected at the York  site  for
geotechnical analysis.  The results of these analyses were
compared to the corresponding SCAPS stratigraphic
data.  Four out of the six samples showed intermethod
agreement.  The remaining two samples did not show
good agreement. This was due to the SCAPS apparent
inability to detect small increases of decreases in coarse
or fine particle sizes.  This indicates that the SCAPS
sensor may not be sensitive to small shifts hi abundance
in secondary particles sizes.

Fort Riley Site

    The SCAPS CP sensor and the reference method
cross sections are generally well correlated when the
cross sections are  considered as  a  whole, however,
minor differences occurred when individual layers were
examined.  The SCAPS identified many more variable
clays, silts, sands, and mixture layers than the reference
method. Nodes 1 and 4 were quite similar in both cross
sections with the exception mat the lower  sand  was
logged at  different depths in each.   (At Node  1, the
SCAPS located the beginning of the sand at 20.5  feet
bgs, while the reference method located its upper limit
as 19 feet bgs).   In  Node 4, the SCAPS logged sand
from 1 feet bgs to the terminal depth of the push, while
the reference method logged sand from 10 feet bgs to the
termination of the borehole.  Below 19 feet bgs,  across
the cross section, the SCAPS identified numerous  thin
lenses  of silt, silt mix, and  sand, while the reference
method identified primarily sand.  This may be  due to
the occurrence of running sands below 19 feet bgs. This
is similar to the differences observed at the Atlantic site.

    Eight samples were collected at the Fort Riley site
for geotechnical analysis. The results of these analyses
were  compared  to   the   corresponding  SCAPS
stratigraphic data.  Only two of the samples showed
intermethod matches.  However, both methods identified
the dominant particle  size  as  sand.  This lack of
intermethod agreement was due to the SCAPS lack of
sensitivity to small changes hi particle size distributions
for minority constituents in a given strata.  In all cases
of poor intermethod matching, the SCAPS identified the
sample as a sand when the reference method laboratory
identified the sample as a silty or clayey sand.  This type
of disagreement was  also seen between the geologist's
classifications  and the  reference  method  laboratory
classifications (see Section 3).

Summary

    The SCAPS CP sensor  and the reference method
produced similar geologic cross sections; however, the
SCAPS data showed more detailed spatial resolution. In
addition, limited QC checks of the SCAPS stratigraphic
data showed good correlation with the reference method.
The SCAPS  was not  as  sensitive to  small  shifts hi
particle size distribution relative to the reference method.
The SCAPS provided a finer  resolution of thin strata by
identifying more  thin  stratigraphic  units  than  the
reference method. This difference was magnified when
running sands were encountered at the Atlantic and Fort
Riley sites. This may be due to the CP sensor's  ability
to continuously acquire soil textural data during  a push
and the common limitations of a geologist's logs where
strata are less than several niches thick. It is possible
that the SCAPS cross sections are more representative of
the actual site stratigraphy below 19 feet bgs at the Fort
Riley and Atlantic sites.  An additional difficulty with
the reference method was its inability to retrieve samples
from running sands. This caused significant data gaps at
depth.  The SCAPS does not require active soil sampling
                                                   36

-------
 to  log a hole, and therefore, it is not as affected by
 running  sands, and  may be more representative of
 subsurface  stratigraphy  than the reference method in
 running sands.

 Chemical Cross Sections

     The following sections present descriptions of the
 similarities   and  differences  observed  between the
 chemical cross sections  produced by the SCAPS LIF
 sensor and  the reference method.  Unless otherwise
 specified the comparisons are made in consideration of
 both reference cross sections for TPH and total PAH.
 PRC used the SCAPS LIF sensor's cross sections shown
 in  Section  4.   These cross sections were produced
 directly from the SCAPS raw data, however,  they are
 scaled to match the  reference  method cross  sections
 shown in Section 3. These comparisons are qualitative,
 and as such are subjective hi nature.   The effects of
 heterogeneity  may  influence  this data  comparison,
 however, the qualitative nature of this comparison should
 greatly reduce the potential impact of heterogeneity in
 contaminant distribution.  These comparisons were made
 by a soil scientist with over 9 years of experience hi site
 characterization activities.

 Atlantic Site

    Both the SCAPS LIF sensor and the reference
 method  showed  good  correlation  for  background
 characterization. This is exhibited by the data from both
 SCAPS and  the reference method showing Node 1 to be
 outside the area of contamination. Both reference cross
 sections  detected the zone of contamination at Node
 2, which extended from approximately 20 to 28 feet bgs
 for TPH and from 16 to 31 feet bgs for total PAH. The
 SCAPS identified this zone being from 2 to 9 feet
 thinner than the reference method for TPH and total
 PAH, respectively. The SCAPS identified the zone as
beginning almost  2 feet shallower and being 2 feet
thinner than the reference method, relative to the TPH
cross section. This difference is acceptable and can be
explained as an artifact of data interpolation, which was
used for the reference method  to create the reference
method cross section. This is common when relatively
few samples are used to define zones of contamination.
The major difference between SCAPS and the reference
method in Node 2 dealt with the failure of the SCAPS to
detect the zone of elevated contamination 1.5 feet bgs
identified by the reference method.  This difference may
have been due to spatial variability exhibited across the
node. The size of the shallow contaminated zone may be
an artifact of data interpolation.  Overall, the zones of
elevated SCAPS LIF data corresponds well with general
zones of contamination shown in both reference method
cross sections.   The shape of  each cross section is
 heavily influenced by the contour intervals used, and
 therefore, it is not possible to say which reference cross
 section shows  the  closest match to  the  SCAPS  cross
 section.  The quantitative data evaluation will answer
 this question.  Interpolation can often lead to the  over-
 estimation of layer thicknesses.

     Another way to examine the relationship between
 the  LIF  sensor's data and the qualitative  reference
 method data is to superimpose the two data types on a
 single plot  of fluorescence intensity  and  reference
 method concentration against depth.  To make the plot
 scales meaningful,  the  SCAPS  LIF  data   and the
 reference method data had to  be normalized.  The
 reference method data was normalized to  the highest
 TPH and total PAH concentrations measured during the
 qualitative sampling.  The LIF data was  normalized to
 the average high reading measured  over a qualitative
 method reference sampling point at this site.   This
 normalization allows a general comparative evaluation of
 the data.

     Figures 5-1, 5-2, and 5-3 show the normalized data
 plots. A review of this data shows that the qualitative
 reference  method  data and  the LIF sensor's  data
 generally agree in their identification of zones of high,
 medium, and low contamination.  The major exception
 to this is exhibited in Node 2 (1 to 1.5 feet bgs). In this
 zone, the reference method exhibited both TPH and total
 PAH contamination hi the range of  50 percent of the
 high reading for the site. This is opposite of the SCAPS
 LIF data which exhibited contamination hi the range of
 1 percent of the high LIF reading. This difference may
 have been  an artifact of  the  heterogeneity  of the
 contaminant  distribution,   the   relative  constituent
 distribution of the  waste,  or  it could reflect a  false
 negative reading.  Minor differences were seen in the
 relative readings produced by both  data sets for the
 zones of lowest contamination.  In these cases the LIF
 data was generally lower. This is probably an artifact of
 the normalization of the LIF data.  In these cases the LIF
 sensor did  detect  increased fluorescence, however,
 relative to the high, this fluorescence was generally less
 than 1 percent.

 York Site

    The SCAPS LIF sensor's cross section showed little
 correlation to the reference method  cross sections at
 Node 5.  Both the TPH and total PAH reference cross
 sections  exhibited  zones  of  elevated  contaminant
concentrations at Node 5.  The zone of elevated  total
PAH contamination extended from approximately 13 to
22 feet bgs, and the TPH contamination extended from
approximately 1 to 24 feet bgs.  The LIF sensor  only
                                                   37

-------
r
            FIGURE 5-1. NORMALIZED LIF AND QUALITATIVE REFERENCE DATA—ATLANTIC SITE
Node -2 Node -3 Node -4 Node -5
Percentage of High Reading Percentage of High Reading Percentage of High Reading Percentage of High Reading
0 50 100150200250 0 100 200 300 400 0 100 200 300 400 100 300 500
n OR
1.96
3.92
5.78
7.74
9.61
—11.34
£.13.33
t 15.08
18.94
20.94

2469
26.69
28,58

i i i i
—•.







1 — « — _
*"



0.07
2.1
4.1
5.85
7.83
9.58
— 11.3
£13.31
£ 15.13
§"17 1"5
Q1'-10
18.92
20.94
2284
24.77
26.81
28.61
1 t 1




	
T

j"
L
e- == 	




007
2.12
4.11
6.04
7.9
9.76
—11.52
£13.49
£15.35
 1733
Qi9:o9
21.05
22.91
24.65
26.62
28.52
i i i



E=s:
fe=
£^

J

c_
£
—

/Total PAH oTPH
-LIP Data

U ZUU 4UU
0.06
1.9
3.9
5.65
7.65
— 11.25
£13.26
flB.12





*
•fefc^

^L
r
y





0.06
1.81
3.78
5.61
7.6
8.47
11.48
g13.47
|.16.48
* 17.6
18.28
21.17
23.19
2527
2724
29
1 1 1
~^— P
f






K
IS.





















1
                                                 38

-------
 FIGURE 5-3.  NORMALIZED LIF AND QUALITATIVE REFERENCE DATA—FORT RILEY SITE
Node-1 Node -2 Node -3 Node -5
Percentage of High Reading Percentage of High Reading Percentage of High Reading Percentaqe of High Readinu
0 50 100 150 200 250 20 60 100 140
nnft
U.UD
1.83
3.83
5.86
7.82
9.58
11.43
§13.41
5^15.19
,§"17.18
19.03
21.04
22.9
24.59
26.58
28.46

i I i i

r2"




>
?•

*~ 	




0 100 200 300 400 <
0.06
1.78
3.74
5.48
7.46
9.34
11.35
§13.13
t~~ 14.9
16.89
18.65
20.63
22.5
24.5
26.38
28.25
i i i



7
^
f~
'fr
^r-S"
p>





0.06
1.86
3.85
5.68
7.67
9.56
^11.41
gl3.41
§15.29
17.27
19.01
21
22.83
24.68
26.71
28.61
) 40 80 120

I
jl
= \
v~
?~
i
~ ">•
g=-
	 "
t-f



0 100 200 300 400
33.95
1.46
3.28
5.14
7.11
8.86
10.87
g.12.76
£ 14.7
g 16.58
18.31
20.31
22.17
24.18
26.07
27.91

nTotal PAH nTPH .LIF Data \

' '



i

^•'^
^__
^^ir
.. 	




















identified limited fluorescence from 1 to  3 feet bgs.
These  differences  may have  been  due  to  spatial
variability  in contaminant distribution, however, the
vertical extent of this contamination probably indicates
more than isolated spots of contamination.   Nodes
1, 2, 3, and 4 showed better correlation between the
reference method cross sections and the SCAPS cross
section.  Overall,  the SCAPS LIF cross section was
relatively similar to the two reference cross sections.
Since the contour intervals strongly influenced the shape
of the contours, it is not possible to identify which
reference cross section most closely match the SCAPS
LIF cross section. The quantitative data evaluation will
answer this question.   The differences between the
reference method cross sections and the SCAPS cross
section could be the combination of an artifact of data
interpolation for the reference method cross sections,
and the finer definition provided by the SCAPS, which
produces continuous profiles with a 2 cm resolution.
The zone of low SCAPS readings shown around Node
3 may be a reflection of the spatial variability of the
contamination or the small elementary sample volume
used by the technology.

    Figure 5-2 shows the normalized line graphs of the
five SCAPS LIF sensor pushes at the  York site.  The
qualitative reference data for TPH and total PAH is
superimposed on these line graphs, at the sample depths
they represent.  The reference data has been normalized
to the high  average LIF  reading measured at the
qualitative  reference method sampling depths.   This
normalization makes the data comparable on a relative
scale.

    A review of this data  shows that  generally the
relative magnitudes between the two types of data were
hi agreement.   Zones of  high reference readings
corresponded to zones of high  LIF readings.   This
relationship appears to hold for medium and low zones
of contiimination. At the low end of this comparison it
appears as if the LIF data is much less than the reference
data.  This is an artifact of the normalization procedures.
In several cases, the LIF data produce relatively much
higher readings.  This can be seen in Node 3 (17 to
18.5 feet bgs) and hi Node 4 (17 to  18  feet bgs).  In
these cases, the LIF data was  100 to 50 percent of the
high reading, while the reference method data was at
approximately 1 percent of the high reading.  These are
examples of false positive readings for the LIF data.
However, heterogeneity of contaminant distribution, or
the constituent composition, could have influenced this
data.

Fort Riley Site

    The SCAPS LIF sensor's cross section showed little
correlation  to the reference method's cross sections at
Node 4,, The reference method cross sections exhibited
                                                  39

-------
an isolated zone of elevated contaminant concentrations
at Node 4.  This isolated detect may be an artifact of
limited reference sampling at this node.  Examination of
the drilling logs for this node indicate that this was the
only depth interval at Node 4 to exhibit elevated (above
background) readings on the portable photoionization
detector (PID).  The SCAPS detected three zones of
elevated fluorescence readings along Node 4.   These
relatively   small  zones  of  fluorescence  may  be
representative of the spatial variability of contamination
at Node 4  and the small representative elementary
sampling volume for the technology. Aside from Node
4 on the northernmost end of the transect, the remaining
nodes  produced  cross  section  data  that  showed  a
relatively  good match between Jhe technology and the
reference  methods.  The greater definition of potential
contaminant lenses in the SCAPS cross sections is most
probably due to the 2 cm sampling resolution provided
by the technology. The need to interpolate data for the
reference  method reduces the potential for identifying
distinct smaller lenses of contamination.  Overall, the
SCAPS cross section exhibited a good  match with the
reference method cross sections.  Since the shape of the
cross sections  is heavily influenced by  the  selected
contour intervals, it is not possible to  identify  which
reference cross section exhibited the closest match to the
SCAPS LIF cross sections.   The  quantitative  data
evaluation will answer this question.

    Figure 5-3 shows the normalized line graphs of the
five SCAPS LIF sensor pushes at the Fort Riley  site.
The qualitative reference data for TPH and total PAH is
superimposed on these line graphs, at the sample depths
they represent.  The reference data has been normalized
to the  highest TPH and  total PAH  concentrations
detected. The SCAPS LIF data has been normalized to
the high average LIF reading measured at the qualitative
reference method sampling depths. This normalization
makes the  data comparable on a relative scale.  A review
of this  data shows that for  all pushes the general
contamination trends identified by the technology match
the trends detected by the  qualitative reference data.
Similar zones of low, medium, and high contamination
were identified by the  technology and the reference
method.

Summary

    Generally,  the SCAPS  LIF sensor  showed a good
relative correlation with the reference method's cross
sections.   The closest  match  was exhibited  when
technology's cross section was compared to the total
PAH reference method's cross sections.  The TPH
reference  method cross  sections generally appeared to
show more resolution than either the technology's or
total PAH  reference method cross sections. In addition,
the SCAPS data and qualitative reference method data
were well correlated hi their identification of zones of
low, medium, and high contamination.

    The observed differences between the cross sections
for the SCAPS LIF sensor and reference method could
have been caused by several factors.  The  SCAPS
sampling volume covered a circle less than 0.5 cm hi
diameter,  and  approximately one  micrometer thick
(approximately 0.2 cubic centimeters). This makes the
SCAPS hyper-sensitive to the  natural spatial variability
of contaminant distribution. The reference method's use
subsample  from a homogenized  12-inch sampling
interval, approximately 1,000  grams of soil.  This base
sample volume is several thousand times larger  than the
sample volume  used by SCAPS.  This larger sample
volume may average  out  the smaller heterogeneities
detected by the  SCAPS sensor. Some of this relative
sample volume effect is canceled out by the fact that the
technology collects much more data.  In the case of this
demonstration, the reference  method  used  a  total of
76 samples, compared to the over  1,300 sample points
the SCAPS produced.

Total Organic Carbon

    PRC  compared the  SCAPS sensor's intensity
measurements for areas free from  contamination to the
corresponding TOC concentrations.   This  evaluation
examined the potential for gross humics to  affect LIF
sensor intensity measurements.  SCAPS data from the
York,  Atlantic,  and Fort  Riley sites  were  reviewed.
This evaluation focused on contamination-free  zones to
eliminate the carbon from the site contaminants from
biasing the results.   The samples  collected  for this
evaluation exhibited TOC concentrations ranging from
not detected to over 3,000 ppm.  Based on the limited
data base (11 samples), there appears to be no affect of
TOC concentrations on LIF data at any of the  three sites.
This is based  on  the fact  that  although  the TOC
concentrations varied over three orders of magnitude,
the LIF  intensity  measurements  remained relatively
constant. However, it is possible that TOC  becomes a
potential interferant in the presence of organic  solvents
or petroleum products.  This interference may be created
by the contaminants' activation of fluorescent properties
hi  the  TOC,   specifically humics.   Isolation and
examination of  the potential for this  activation  of
fluorescence in humics was beyond the scope of this
demonstration.

Quantitative Assessment

    This section presents the comparative evaluation of
the SCAPS LIF sensor's data and the reference method's
analytical  data,  and an evaluation of the technology's
                                                   40

-------
precision and resolution.  The precision and resolution
discussion will be presented after the regression analysis
discussion.

    The  reference  method  sampling  and  analysis
identified considerable heterogeneity in the distribution
of contaminants in the soil matrix.  The experimental
design of this demonstration expected heterogeneity and
intended  to  define  it through collocated  replicate
sampling.  This sampling did define the heterogeneity.
However, in many cases the heterogeneity was greater
than expected.   In almost 50 percent of the 21 quanti-
tative sample intervals the heterogeneity produced ranges
between maximum  and  minimum  concentrations in
excess of one order of magnitude.  This heterogeneity
coupled  with the developers  inability to  specifically
identify the compounds they are measuring, the lack of
a reference analytical  method that monitors  the exact
suite of the compounds measured by the technology,  the
mixed distribution of constituents hi the contamination,
and the varied age of the contaminants cause uncertainty
to be introduced into the point by point comparison of
data  in  the  quantitative evaluation.   Therefore,  any
conclusions stated in this section should be considered as
trend indicators  and not  definitive statements  on
technology performance.  However, the conclusions  are
likely to be duplicated if similar field in situ verification
is attempted.

    The quantitative assessment evaluated SCAPS LIF
sensor's  data   at  distinct   intervals   relative  to
corresponding data from the reference method.  This
evaluation is intended to quantify relationships between
the technology's data and compound or class-specific
analytical data produced by the reference methods. The
target compounds for this evaluation were TPH, VPH,
BTEX, total BTEX, naphthalene, 1-methylnaphthalene,
2-methylnaphthalene,   acenaphthene,   fluoranthene,
phenanthrene, pyrene, benzo-a-pyrene, total PAH, and
total naphthalene. The TPH,  VPH, total naphthalene,
total PAH, and total BTEX groupings were made in an
effort to more closely match the technology's data. The
developers felt that classes of compounds would show
the closest match to the technology's data.

    This data evaluation involved regression analysis of
the SCAPS LIF data against the corresponding reference
method data. As defined in the approved demonstration
plan, a  correlation coefficient (r2)  of 0.80  or better
defines a useable predictive model.

    The SCAPS LIF sensor made two collocated pushes
at each node. The first push was intended to produce the
primary data for both the qualitative and  quantitative
evaluations.  The second push was intended to examine
the teclinology's precision.   The second  push also
produced continuous LIF data to depth.  The primary
data evaluation focused on the data from the first push at
each node,  however, PRC did examine the possible
impact of averaging the two pushes for the regression
evaluation.  This averaging had very limited impact on
the outcome of the regression analysis and will only be
discussed where its findings differ from the first push
data.

    The data sets were initially examined as a whole and
then post-hoc techniques were used to eliminate data
outliers.  The total data set for this evaluation consisted
of 21 sampling intervals, 8 at the Atlantic site, 7 at the
York site,  and 6 at the Fort Riley site.  Each one of
these intervals produced one data point for the regression
analysis,, however, each of these data points represented
the mean concentration from five collocated samples.
Therefore, this evaluation was  based on the analytical
results of 105 individual samples and analyses. The data
presented is based on  non-transformed data.   PRC
mirrored   the  analyses   discussed   below   with
log-transformed  data, however,  in no  case did the
correlations improve.  This suggests that the high and
low concentration points did not disproportionally bias
the regression.

    PRC also examined the data hi its raw form, prior
to averaging the  reference method data.  This approach
did not improve  the correlation of the data.

    The initial regression analysis examined the data set
of  mean concentrations  as a whole.    From  this
evaluation, no i^s  of greater than 0.20 were observed
(Table 5-1).  An examination of the maximum and
minimum concentrations  for each  set  of  collocated
samples  indicated  that  several locations at  each site
exhibited considerable heterogeneity. This was expected
and is normal for environmental sampling.  Using data
pouits from reference sampling depths that exhibited
wide ranges hi contamination  introduced  additional
uncertainty into  the data evaluation.  In these cases, it
was hard to define  representative mean concentration.
Concentrations were highly location dependant.  In an
effort to reduce  the impact of this heterogeneity on the
data evaluation, all data pouits exhibiting a greater than
1 order of magnitude range between the maximum and
minimum  were eliminated.   Ranges, which  are
nonparaimetric statistics, were selected for this post-hoc
data reduction since  they are not dependent on data
distribution.  In most cases, this resulted hi at least a
50 percent reduction hi useable data.  For this reason,
the  subsequent  data  analyses  should be  considered
indicators of trends  hi correlation, and not well-defined
predictive models.
                                                    41

-------
TABLE 5-1. REGRESSION ANALYSIS RESULTS FOR SCAPS AND THE REFERENCE METHODS-
ALL SITES
Initial Regression
Compound
TPH
VPH
Benzene
Toluene
Ethylbenzene
Xylene
Naphthalene
1 -Methylnaphthalene
2-Methylnaphthalene
Alenaphthene
Fluoranthene
Phenanthrene
Pyrene
Benzo-a-Pyrene
Total Naphthalene
Total PAH
Total BTEX
Notes:
n
21
20
16
19
20
20
21
18
20
12
19
21
17
19
21
21
20

r2
0.01
0.02
0.02
0.03
0.00
0.01
0.03
0.07
0.01
0.00
0.12
0.02
0.01
0.03
0.01
0.01
0.02

n Number of sample points, each
r2 Coefficient of determination.
slope
0.12
0.01
0.20
0.85
0.19
0.67
-0.18
0.01
-0.00
-0.00
0.00
-0.00
-0.00
0.00
-0.01
-0.00
2.20

sample
y-intercept
(ppm)
3,340
274
4,943
13,023
11,391
30,510
46.4
47.0
34.2
45.2
2.19
18.2
8.03
1.25
202
117
56,539
Final Post-Hoc Data Reduction
n
7
9
8
10
10
11
9
9
9
4
12
10
7
8
8
9
10

r2
0.89
0.94
0.34
0.41
0.88
0.94
0.01
0.29
0.43
0.31
0.10
0.02
0.76
0.50
0.50
0.06
0.94

slope
2.2
0.16
1.9
9.6
7.0
17.4
0.00
0.02
0.01
0.09
-0.00
-0.00
0.01
0.00
0.07
0.02
38.7

x-intercept
y-intercept (fluorescence
(ppm) intensity)
-346
-53.8
3,028
4,376
607
-3,377
14.6
38.4
10.7
0.90
1.80
15.2
-0.65
0.69
50.0
101
-11,370

point is the mean concentration from five collocated
-157
336
-1,594
-518
-87
194
No Data
-1,920
-1,070
-10
No Data
No Data
0.02
No Data
-714
-5,050
294

samples.
ppm Parts per million.
    After these data points were removed, the regression
analysis was run again.  No significant changes hi the
regression  parameters were observed  (Table  5-1).
However,  a  post-hoc examination of the  residuals
identified several outliers for each regression.

    A final regression analysis was conducted on the
data  sets after  the  outliers were removed.    This
regression showed considerable  improvements in the
data correlation (Table 5-1). TPH, VPH, ethylbenzene,
xylene, and total BTEX all exhibited i^s  above the
0.80  criteria  for acceptance.   Pyrene had  an r2 of
0.76,  almost  meeting  the acceptance  criteria  for
correlation. The slope data cannot be used to assess data
quality since the LIF data was not in the same units as
the reference method data. However, the slope data can
indicate trends hi relative fluorescence.  The slopes of
the ethyl benzene, xylene, and total BTEX regressions
were all much greater than  1.0.  This indicates that
relatively large changes hi contaminant concentration are
required to cause  changes in LIF data.  Conversely, the
VPH  regression had  a  slope much less than 1.0,
indicating that small changes in VPH can cause relatively
larger changes in LIF data. This can be  translated into
a  general  conclusion  regarding  the  LIF  sensor's
sensitivity.   Based on the slope data, the LIF sensor
appears to be most sensitive to the compounds measured
hi the VPH analysis relative to the TPH, ethylbenzene,
xylene, and total  BTEX analyses.
                                                  42

-------
    Although the r2 data at this point indicates that the
concentrations of the above  compounds appear to  be
correlated, the small size of the data set limits the
usefulness  of any predictive models based on these
regression parameters.  The regression parameters for
TPH and VPH could best be used to produce general
predictive models for concentration based on LIF data.
Due to the negative y-intercepts, these models could not
be applied to LIF data for intensities below 157 for TPH
and 336 for VPH. These intensities correspond to the x-
intercepts for the respective regression models when the
concentration of contaminants is 0 mg/kg.

    The number of compounds that exhibited acceptable
correlations suggests that these relationships are real.
However, the lack of correlation observed for many of
the compounds  may not  be  wholly attributable  to
technology performance. Rather, poor correlations are
likely due to a combination of effects such as matrix
heterogeneity, the lack  of a definitive match between
reference analytical methods and the suite of compounds
measured by the technology, the variable distribution of
contaminant constituents, and the variable  ages of the
contaminants.

    Similar conclusions are drawn if the data from the
two SCAPS LIF sensor pushes are used in the regression
analysis. The only exceptions are for ethylbenzene and
pyrene. In this data set,  ethylbenzene no longer exhibits
an acceptable r2, but pyrene does (Table 5-2). The same
trends  in slopes are observed for this data set, the LIF
sensor seems to be more sensitive to the VPH, and hi
this data set the PAH compound pyrene.   The TPH and
VPH  shows the most conducive  data for creating  a
predictive model, just as above.

    The quantitative determination of a  detection limit
for the SCAPS LIF  sensor was not possible given the
data produced from this demonstration.

    Qualitative observations regarding  the detection
limits of this technology  can be made with the data
produced  from  this   demonstration.    Measurable
fluorescence was reported for TPH concentrations as low
as  60   mg/kg and  VPH  concentrations  as  low  as
19 mg/kg.  At no point during the demonstration did the
SCAPS  LIF  sensor report no   fluorescence above
background for soils exhibiting contamination detectable
by the  reference methods.  Another qualitative method
for assigning a detection threshold is to determine the x-
intercept for the TPH and  VPH regression models
discussed above.  The x-intercept  for  these models
represents  the  point   at   which  TPH   or  VPH
concentrations are 0 mg/kg.  For TPH the fluorescence
intensity at the x-intercept is 157 and  for VPH it is
336. Cross checking these pseudo thresholds against the
information hi Table 5-3, for data remaining after the
initial removal of outliers based on heterogeneity, shows
that hi most cases SCAPS  LIF  readings  in  these
threshold ranges corresponded to the lowest contaminant
concentrations.

    To examine the potential for site-induced effects on
the data evaluation, the data was divided by site  and
regression analyses were run on the resultant three data
sets.   This regression  analysis  began with data  sets
whose gross outliers had been removed.  These outliers
were defined as data points where the maximum  and
minimum values varied by over one order of magnitude.

    This  site-specific  regression showed  that only
naphthalene  and  fluoranthene  exhibited  acceptable
correlations at the Atlantic site; no compounds showed
acceptable correlations (r2 greater than 0.80) at the York
site; and acceptable correlations for toluene, VPH,  and
total BTEX were found at the Fort Riley site.   The
number  of  samples resulting  hi  these  acceptable
correlations ranged from 3 to 4, out of a maximum of
6 to 8.  This small sample set greatly limits the use of
this data to form predictive models.  However,  these
regressions exhibited the same trends in their slopes, as
exhibited hi the data set as a whole.  The slopes for the
VPH arid PAHs were all less than  1.0, and the BTEX
compounds produced regression equations with slopes
greater than 1.0.

    Inherent instrument precision for the SCAPS LIF
sensor measurements at the York and Atlantic sites was
evaluated by calculating  the percent RSD of each set of
10 replicate measurements, taken at single depths (Table
4-1).  The SCAPS  took no precision measurements at
the Fort Riley site. These precision measurements were
taken from intervals where  peak wavelengths ranged
from 350 to 650 nm. The percent RSD was calculated
by  dividing the standard deviation by the mean, then
multiplying the result by  100.  The range of RSDs at the
Atlantic site were  1.1 to 4.1.  The range of RSDs for the
York site were  1.0 to  1.6.  Based on this data,  the
standard deviations noted on Table 4-1 are most likely
due to heterogeneity hi contaminant distribution.   The
maximum inherent instrument precision of the SCAPS
LIF sensor observed during this demonstration  was
4.1 and 1.7 percent for the  Atlantic and York sites.
With this high degree of inherent instrument precision,
it is possible to identify  the cause of the wide range of
measurement standard deviations exhibited  hi Tables
4-1, 4-2, and 4-3. All but 1 to 5 percent of this variance
can be attributed to matrix heterogeneity hi the vertical
direction.  The small area and volume of the SCAPS LIF
measurements tend to accentuate matrix heterogeneity hi
the soil matrix.
                                                   43

-------
TABLE 5-2. REGRESSION ANALYSIS RESULTS FOR THE AVERAGE OF BOTH SCAPS PUSHES
AND THE REFERENCE METHODS—ALL SITES
Initial Regression
Compound
TPH
VPH
Benzene
Toluene
Ethylbenzene
Xylene
Naphthalene
1-Methylnaphthalene
2-Methylnaphthalene
Alenaphthene
Fluoranthene
Phenanthrene
Pyrene
Benzo-a-Pyrene
Total Naphthalene
Total PAH
Total BTEX
Notes:
n
21
20
16
19
20
20
21
18
20
12
19
21
17
19
21
21
20

r2
0.09
0.11
0.14
0.11
0.05
0.05
0.00
0.16
0.01
0.00
0.19
0.00
0.00
0.09
0.00
0.01
0.09

n Number of sample points, each
r2 Coefficient of determination.
ppm Parts per million.
slope
0.63
0.05
0.86
3.0
1.1
3.5
0.001
0.01
0.00
0.00
0.00
-0.00
0.00
0.00
0.00
0.01
9.1

y-intercept
(ppm)
2,485
205
3,619
9,251
9,725
25,403
37.6
41.6
30.0
44.0
1.76
17.1
7.27
1.05
183
103
44,820
Final Post-Hoc Data Reduction
n
7
10
8
10
10
11
11
11
9
4
12
9
7
7
10
10
10

r2
0.84
0.95
0.50
0.42
0.69
0.88
0.04
0.09
0.72
0.29
0.08
0.01
0.92
0.32
0.32
0.08
0.89

slope
2.1
0.22
2.6
9.7
10.7
17.0
0.00
0.01
0.02
0.12
-0.00
-0.00
0.02
0.00
0.05
0.02
37.7

y-intercept
(ppm)
375
-72.9
1,874
5,248
1,891
1,120
14.3
37.2
2.6
-15.8
1.76
16.1
-1.38
0.37
39.8
96.9
392

x-intercept
(fluorescence
intensity)
-179
331
-721
-541
-177
-66
No Data
-3,720
-130
132
No Data
No Data
69
-185
-796
-4,845
-10

sample point is the mean concentration from five collocated samples.
    The wavelength resolution of the SCAPS LIF sensor
was also examined during this demonstration.  During
the precision measurements at the  Atlantic site, the
deviation reported peak wavelengths for the York site
ranged from 0.6 to 1.7 percent.  Based on an exam-
ination of the spectral wave forms produced by SCAPS
during this demonstration, PRC determined that the
reported peak wavelength could vary by approximately
plus or minus 5 percent before significantly affecting the
reported intensity.   The  inherent  instrument peak
wavelength resolution is less than 5 percent and, thus, it
should not affect instrument performance.
                                               44

-------
TABLE 5-3
Site
Atlantic







York






Fort Riley





Notes:
a r>at=
. DATA
Node
2
2
3
4
4
4
5
5
1
2
2
3
4
4
5
1
1
2
2
5
5

a no into re
FOR MEAN SCAPS,
Depth
(feet)
21-22
24-25
16-17
6.5 - 7.5
10-11
27.5 - 28.5
16-17
23.5 - 24.5
15-16
13.5 - 14.5
17-18
17-18
14-15
18-19
1.5-2.5
2-3
13-14
6-7
17-18
10.5-11.5
16-17

imoininri aftar lha initial
TPH, AND VPH— ALL
SITES
SCAPS
Fluorescence
Intensity TPH
(mean) (mean mq/kq)
5,809
5.37
1,323
213.4
837.1
6,310
19,674
29.88
221.3
723.8
268.5
908.4
775.7
412.5
297.1
1,143
366.6
1,136
4,853
2,923
3,036

rammf^l f\f *"\ii+li»r^ I^Qi^Afl
11,090a'b
4,044
425
255
2,436a'b
1,094
201 a
239a
773
1,539
497
778a,b
2,281 a'b
1,878
60a'b
5,728
1,416
2,169
13,150a'b
22,480a
3,926a'b

^"if» rv^^vimi trvi *

VPH
(mean mq/kq)
1,402a
538
112a,b
43
1,320a
452
96a
77a,b
No Data
25a,b
20a'b
19a.b
64a,b
175
ND
48
184
42
790a,b
334a'b
442a,b

•i»l*"l rv-itriirvti irv% s*s%t>v

Total PAH
(mq/kq)
673
291
5.8
8.3
78
121
2.4
3.0
260
246
160
230
515
799
0.45
89
31
11
154
246
65


        Data point used in the final regression analysis.
ND     Not detected.
mg/kg   Milligram per kilogram.
                                                   45

-------
                                               Section 6
                                    Applications Assessment
    The SCAPS technology is designed to be operated
by trained technicians from the AEC, Army Corps of
Engineers, U.S. Navy, WES, and other licensees. The
SCAPS  technology is  available for use  by private
citizens or corporations, although it is available to state
and  federal  agencies.    Hogentogler  and  Applied
Research Associates, Inc., have nonexclusive licenses
from  WES  to  use  the  LIF  sensor  with   cone
penetrometry. A similar technology is operated by Loral
Corporation, under secondary license from Hogentogler.
The SCAPS technology's current usage has been focused
on contamination detection and delineation at  military
installations.  The  target contaminants  are primarily
PAHs,  and most often this technology is  applied at
petroleum fuel release sites.   As demonstrated, this
technology can rapidly acquire and plot data  defining
zones of general contamination if the contamination has
a fluorescent signature. This data can greatly facilitate
site characterization activities.

    The  qualitative  assessment   portion   of this
demonstration showed that this technology is comparable
to reference methods in its ability to map subsurface
contaminant  plumes at petroleum  fuel and  coal tar
contamination sites.  This demonstration showed that
both the SCAPS LIF sensor and the reference  methods
identified similar zones of subsurface petroleum and coal
tar contamination at each of the three sites. Many of the
differences  between the  SCAPS and  the reference
methods can be explained by their respective methods of
data collection.  The technology produces a continuous
profile, while the reference methods take a few selective
samples  and  target  boundaries   and   zones  of
contamination. In addition, the reference methods had
difficulty retrieving samples in running sands, adding
potential  data  gaps.    The   technology produced
continuous data without the need to physically retrieve
samples.  The SCAPS technology can produce relatively
continuous data on petroleum or coal tar contaminant
distribution over a 35-foot depth in approximately 1 to
1.5 hours.  The reference methods would be able to
collect samples over this  interval, however, definitive
analytical services would require, at best, several days,
and  the  costs  associated with analyzing  continuous
samples collected every 2 niches would be prohibitive.
Even if the reference methods used on-site analysis and
produced only screening level data, it would take several
hours to provide data on the samples.  Therefore, on-
time critical projects that can use screening level data,  or
on projects where it is more critical to cover large areas
in greater detail, the SCAPS technology seems to have
distinct advantages.  The  cost  of this  technology  is
comparable to conventional approaches, except that this
technology produces greater resolution for similar cost.
However, this resolution is at a lower data quality level
than the reference methods.

    Another powerful aspect of this technology is that it
provides continuous descriptions of the subsurface soil
concurrently with the chemical data. This demonstration
found  that the  subsurface  logging capabilities  of the
SCAPS CP sensor was of comparable accuracy to the
reference  methods,  however, it  appeared to exhibit
greater resolution.  Site-specific calibration borings were
not used for this demonstration, and the technology still
produced    acceptable   accuracy   for   subsurface
stratigraphic logging.

    The quantitative data assessment for this technology
indicated that the resultant LIF data may be correlated to
VPH,  TPH, ethylbenzene, xylene,  and total  BTEX
concentrations.   In addition, this data suggests that a
detection  threshold  for the SCAPS  may  be around
157  fluorescence units for  TPH and 336 fluorescence
units for  VPH.  These values generally matched the
lowest TPH and VPH concentrations measured.  The
lowest TPH and VPH concentrations measured by the
reference methods were 60 and 19 mg/kg, respectively.
Both of these  samples exhibited fluorescence  above
background.  Due to the data set sizes, the predictive
models based on this data should only be used for the
most general estimates. The original reference data sets
were reduced by as much as 50 percent when data points
exhibiting excessive heterogeneity were eliminated. The
                                                    46

-------
qualitatively identify changes in waste characteristics and
possibly types.  The regression analysis  showed some
correlation between the technology's results  and individual
compounds, however, this may have been an artifact of
then- relatively systematic distribution within a larger class
of  compounds, TPH  or VPH,  most  closely  being
monitored. Based on the results of this demonstration,  the
use of site-specific calibration samples for the application
of the SCAPS LIF sensor may increase its performance in
a qualitative node, however, it seems unlikely that they
would improve its quantitative performance due to matrix
and contaminant interferences. Site-specific calibration
samples were not used during this demonstration and  the
technology still produced similar contaminant distributions
to  the  reference  methods.    Even  with site-specific
calibration,   in the  configuration  deployed  hi this
demonstration, it  is not likely that the technology can
produce definitive data, however, site-specific calibration
may  allow  an  estimation  of  relative contaminant
concentrations.   This would be true if  the observed
correlations were real.

    Based on this demonstration, this technology appears
to produce  screening level  data for both physical and
chemical characterization sensors.  The failure to achieve
better quantitative correlations for the chemical data may
not be wholly attributable to the technology performance.
This may  have been due to the relatively small reference
method data set size, the lack of a reference method that
measures the same suite of  compounds as the SCAPS LIF
sensor monitors, the  complex interactions between  the
fluorescing compounds and the soil matrix which resulted
in the observed heterogeneity. The first two factors can
be  addressed with changes  hi experimental design and
innovations  in  analytical  methods, however, the  final
factor will require more research to isolate specific matrix
interactions,  and the heterogeneity  issue may not  be
solvable given current technology.

    If the SCAPS LIF  sensor performance is  to  be
evaluated in the field, this demonstration has  shown that
on a point-by-point quantitative basis, it is possible that
little to no correlation to reference data will be observed.
This  is due to a combination of heterogeneity effects,
limitations hi conventional  sampling and analysis, and  the
complex  interaction  of waste  aging and   constituent
distribution of relative fluorescence.  Therefore, based on
the  results of this demonstration, field evaluations of this
technology should be restricted to qualitative evaluations
consisting of cross section comparisons and comparisons
of normalized LJF and to verify that LIF highs correspond
to higher levels  of contamination. This latter comparison
will also be affected by effects listed above.

    In the configuration used during this demonstration,
the  SCAPS LIF and CP sensors provided screening level
chemical and stratigraphic data in real time, at a rate faster
than conventional approaches, and with apparently greater
resolutions.  The LIF data was relatively correlated to the
reference  chemical data hi that both data sets  tended to
identify the same zones  of high,  medium,  and  low
contamination. The added benefit of sensors mat function
without physical sampling allows them to produce data hi
subsurface  environments  that prohibit   conventional
sampling. An example of such an environment are the
running sands encountered at the Atlantic and Fort Riley
sites.  The cost of this  technology is comparable to
reference  methods, hi fact, on a per-data point basis, this
technology is much less expensive than reference methods.

    Although  there are  many  advantages  to   this
technology, a potential user should be aware of several
disadvantages. This technology has a sampling volume
several thousand times smaller than conventional sampling
analysis.  This makes the technology more sensitive to
matrix heterogeneity. Some of this sensitivity is reduced
(vertically)      by       the       averaging      of
10  data, points every 2 cm.   This  effect can also be
minimized by the sampling of more push locations to
reduce ithe sensitivity hi a horizontal orientation.  At a
developear-claimed data collection rate up to 400 linear feet
per day  (6,096 data points), additional pushes can be
conducted without greatly increasing  project  duration.
The LIF  results  can  be  influenced  by  the  age  and
constituent distribution  of wastes.   This  coupled  with
heterogeneity effects, and a lack of instrument calibration,
makes quantitation or field verification of LIF results
difficult.  The use of the  LIF and CP sensors is  restricted
to the maximum push depth of the cone  penetrometer
truck. This depth can be as much as 150 feet, or in the
case of this demonstration, 30 to 70 feet. These shallow
depths  were  realized  when  deeper  strata  exhibited
increased cone tip resistance and sleeve friction, and at
locations  where  strata  at  shallower depths would not
provide adequate lateral  support for the push rod. These
condition greatly increase the chance  for  push  rod
breakage and sensor loss.

    This  technology   can currently  provide  rapid
assessment of the distribution of fluorescent material hi the
subsurface. When these materials are PAHs or petroleum
fuels, the technology can be used to map the general extent
of subsurface contamination. This  data can be used to
guide  critical conventional soil sampling,  and  the
placement of groundwater monitoring wells. All of this
data can be produced and interpreted hi the field.  This
real-time  sampling  and  analysis allows  the use  of
contingency-based sampling which assists  hi character-
izing a site with a single mobilization. These aspects
coupled with  its  low volume  waste production during
decontamination make this technology a powerful  and
effective, site characterization tool.
                                                     47

-------
can be used to guide critical conventional soil sampling,
and the placement of groundwater monitoring wells. All
of this data can be produced and interpreted in the field.
This real-tune sampling and analysis allows the use of
contingency-based sampling which assists  in character-
izing a site with a single mobilization.  These aspects
coupled with its low volume waste production during
decontamination make  this technology a powerful and
effective site characterization tool.
                                                    48

-------
                                              Section 7
                   Developer Comments and Technology Status Update
    The developer of SCAPS submitted both editorial
and technical  comments on the draft ITER.  Where
appropriate, the editorial comments were addressed.
The  developer's  technical comments  are  presented
verbatim below in italics.   PRC's response to the
comments is presented below each developer comment
in plain type.

1.  The graphical representations, produced by PRC, of
    the physical and chemical  cross sections may be
    sufficient to represent "tradition data," but it is a
   poor representation of what was produced by the
    our system while it was in the field.

    Panel plots from the SCAPS LIF and CP sensors
    have been included in the ITER.  The data  from
    both of these sensors is often plotted in color cross
    sections to assist in the interpretation of the  data.
    Color plots for this  demonstration were  submitted
    by  the SCAPS operator.   These plots  generally
    show  greater resolution than the ones used in the
    ITER.  The developer's color plots are hi the TER;
    they were not added to  the ITER due to the
    complexities and costs associated with reproducing
    color graphics.

2.  There is a general editorial comment concerning the
    "negative" tone to  the discussions.  There are
    numerous examples of paragraphs starting with a
    negative sentence and then followed with several
   positive comments.  The report could just as easily
    be  written to highlight the positive aspects of the
    technology.

    The ITER was reviewed regarding its tone.  Where
    the tone  disproportionately stressed either the
    negative or positive, the text was altered to present
    a more uniform presentation of the data.

3.  Considering the lack of precision and accuracy in
    the reference "quantitative" methods, it does not
    seem appropriate to judge SCAPS correlation with
    those methods.  We have never claimed to be more
    than a screening tool, and therefore should not be
    judged by a tougher standard.

    The ITER has been clarified.  It now indicates that
    the developer claimed the technology demonstrated
    was designed to produce screening level data.  In
    addition, the inclusion of the quantitative evaluation
    was; explained as an attempt to develop baseline data
    on  the   current quantitative capabilities  of the
    technology.

    The developer's comments regarding the precision
    and accuracy of the  reference methods is noted.
    The; ITER has been modified to explain and consider
    the impact of heterogeneity in the soil matrix, and
    the problems observed with the reference methods,
    primarily sample collection methods.

4.  The site descriptions do not adequately address the
    heterogeneous contaminant distributions that were
    observed. This can be illustrated, by the variation
    observed in some of the replicates of the reference
    samples.  This variance represents a horizontal
    heterogeneity at these sights.   In addition,  the
    vertical  heterogeneity observed over the one foot
    averaged area in the  SCAPS data, indicates that a
    nonhomogeneous distribution of the stratigraphy and
    contamination exists.

    The ITER has been rewritten to address the issue of
    heterogeneity at all levels of data comparison.

5.  Precision data indicated a high level of precision for
    the SCAPS  technology, while statements in the
    report imply  that fluorescence intensity variations
    were  due  to  the technology  rather than  the
    heterogeneous distribution of the contaminant.

    The ITER has been rewritten to consider the effects
    of heterogeneity on all levels of data comparison.
                                                   49

-------
                                             Section 8
                                            References
American Society for Testing and Materials (ASTM). 1990. "Standard Test Method for Particle-Size Analysis of
        Soils."

Environmental Protection Agency (EPA). 1986. "Laboratory Manual Physical/Chemical Methods." la. Test Methods
        for Evaluating Solid Waste.  Volume IB.

EPA. 1991. National Functional Guidelines for Organic Data Review.  Contract Laboratory Program. June.

EPA. 1993. Data Quality Objectives Process for Superfund - Interim Final Guidance.

Page, A. L., ed.  1982.  "Chemical and Microbiological Properties." In. Methods of Soil Analysis. 2nd Edition.
        Number 9.  Part 2.

PRC Environmental Management, Inc.  1994.  "Final Demonstration Plan for the Evaluation of Cone Penetrometer-
        and Geoprobe*-Mounted Sensors." August.

University Hygienic Laboratory. 1991.  "Method OA-1 for Determination of Volatile Petroleum Hydrocarbons
        (Gasoline)." University of Iowa. Iowa City, Iowa.
                                                 50

-------
                                       APPENDIX A
                Qualitative, Quantitative, Geotechnical, and TOC Data
A-1.    Qualitative Reference Laboratory Data for TPH and PAH - Atlantic Site  	  A-1
A-2.    Qualitative Reference Laboratory Data for TPH and PAH - York Site	  A-2
A-3.    Qualitative Reference Laboratory Data for TPH and PAH - Fort Riley Site  	  A-3
A-4.    Quantitative Reference Laboratory Data - Atlantic Site	  A-4
A-5.    Quantitative Reference Laboratory Data - York Site	  A-5
A-6.    Quantitative Reference Laboratory Data - Fort Riley Site	  A-6
A-7.    Geotechnical and TOC Data - Atlantic Site	  A-7
A-8.    Geotechnical and TOC Data - York Site .. >	  A-8
A-9.    Geotechnical and TOC Data - Fort Riley Site	  A-8
                                            51

-------
TABLE A-1. QUALITATIVE REFERENCE LABORATORY
DATA FOR TPH AND PAH—ATLANTIC SITE
Node
Number
2
2
2
2
2
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
Depth
(feet)
1-1.5
8-9
13-14
25-26
35-36
1-2
10-11
20.5-21.5
33.5 - 34.5
2-2.5
6-6.5
6.5-7
7-7.5
7.5-8
8-8.5
8.5-9
9-9.5
9-10
9.5-10
10-10.5
10.5-11
15-16
27-28
1-2
5-6
7-8
27-28
33.5 - 34.5
TPH
(ppm)
1,680
15
24.7
NS
19.3
55.4
1,130
222
54.2
149
330
614
1,650
4,170
541
73.7
1,680
897
2,880
2,960
3,820
1,170
118
399
ND
275
146
ND
PAH
(ppm)
88.19
3.59
0
0
.06
11.40
158.37
6.06
.99
13.39
.258
5.267
21.494
44.205
128.811
71.760
55.482
40.644
80.999
104.487
107.437
62.091
48.879
7.007
0.020
18.496
3.481
0.030
Notes:

ppm   Part per million.
NS    Not sampled.
ND    Not detected.
                     52

-------
TABLE A-2. QUALITATIVE REFERENCE LABORATORY
DATA FOR TPH AND PAH—YORK SITE
Node
Number
1
1
1
1
2
2
2
2
3
3
3
3
3
3
3
3
4
4
4
4
4
4
5
5
5
5
Notes:
ppm
ND
Depth
(feet)
12-13
14-15
17-18
22 - 22.5
8.5-9
10.5-11
14-14.5
20-21
10-11
12-13
16-16.5
16.5-17
17-17.5
17.5-18
18-18.5
21.5-22.5
8-9
11-12
14-15
17-18
18-18.5
21.5-22
10-11
12-13
17-18
21-22

Part per million.
Not detected.
TPH
(ppm)
26.1
345
13.7
ND
ND
417
855
10.2
10
259
2,570
3,650
57.5
12.7
27.8
ND
115
174
8,150
137
13,100
74.2
23.7
66
377
ND


PAH
(ppm)
1.09
48.81
.88
.01
0
7.72
127.62
.060
0
37.62
134.67
313.97
1.90
0.23
0.20
0.01
0.66
182.09
1,412.16
10.33
1,130.18
14.35
0
9.31
128.09
0.165


                   53

-------
TABLE A-3. QUALITATIVE
DATA FOR TPH AND PAH-
REFERENCE LABORATORY
-FORT RILEY SITE
Node
Number
1
1
1
2
2
2
2
2
3
3
3
3
4
5
5
5
5
5
5
5
5
5
5
5
Notes:
ppm
ND
Depth
(feet)
1-1.5
18-19
28.5 - 30
5-6
6-7
15-16
23.5 - 25
28.5-30
1.5-2.5
5.5 - 6.5
15-16
23.5 - 24
15-16
2.5 - 3.5
5-6
10-10.5
10.5-11
11-11.5
11.5-12
12-12.5
12.5-13
13-13.5
24-25
29-30

Part per million.
Not detected.
TPH
(ppm)
47.1
482
ND
37.4
233
6,720
89.3
96.9
112
2,670
1,850
ND
37
ND
1,280
6,730
32,800
19,300
9,360
12,700
2,830
2,550
9.94
ND


PAH
(ppm)
0.667
12.964
0.036
0.108
0.142
137.885
1.707
2.713
1.358
118.471
18.730
0
0
0
1.655
143.622
338.566
344.984
206.888
190.693
87.639
64.711
0
0


                    54

-------
TABLE A-4.  QUANTITATIVE REFERENCE LABORATORY DATA—ATLANTIC SITE

Chemical Minimum
Node 2 (21 to 22 feet)
Ethyl-
benzene 25,000.00
TPH 8,850.00
VPH 910.00
Total
PAH 70.42
Total
BTEX 15,400.00

Chemical Minimum
Node 3 (16 to 17 feet)
Ethyl-
benzene 3,600.00
TPH 104.00
VPH 88.00
Total
PAH 4.43
Total
BTEX 25,590.00

Chemical Minimum
Node 4 (10 to 11 feet)
Ethyl-
benzene 29,000.00
TPH 959.00 '
VPH 1,200.00
Total
PAH 12.26
Total
BTEX 218,700.00

Chemical Minimum
Node 5 (16 to 17 feet)
Ethyl-
benzene 390.00
TPH 87.80
VPH 36.00
Total
PAH 0.48
Total
BTEX 5,490.00

Maximum


42,000.00
15,400.00
2,000.00

918.32

293,000.00

Maximum


4,600.00
1,290.00
130.00

6.84

33,550.00

Maximum


35,000.00
3,780.00
1,400.00

148.10

307,000.00

Maximum


2,100.00
516.00
160.00

4.72

28,700.00
Standard
Mean Deviation


ISliilBfi 7,035.62
lilgOJJI 2,689.89
427.22

672.69 354.08

221,200.00 53,049.03
Standard
Mean Deviation


IllPJJQjpJI 469.04
425.25 577.00
21.35

HI 1-05

mafflg 3,728.07
Standard
Mean Deviation


S1}600.g§ 2,408.32
1,129.72
83.67

77.91 60.07

fHI^7,4,QTOIJ 34,028.49
Standard
Mean Deviation


811.80
177.36
59.36

1.79

11,794.98

Minimum Maximum
Node 2 (24 to 25 feet)

250.00 39,000.00
36.00 9,880.00
7.90 1,400.00

3.99 691.99

1,250.00 260,000.00

Minimum Maximum
Node 4 (6.5 to 7.5 feet)

250.00 3,600.00
20.70 412.00
10.00 110.00

3.91 13.73

320.00 37,300.00

Minimum Maximum
Node 4 (27.5 to 28.5 feet)

1,300.00 23,000.00
117.00 3,030.00
42.00 970.00

29.51 270.81

13,700.00 193,000.00

Minimum Maximum
Node 5 (23.5 to 24.5 feet)

940.00 1,700.00
48.20 893.00
52.00 160.00

1.96 5.11

14,330.00 22,200.00

Mean


14,690.00
4,004.40
537.58

290.96

93,950.00

Mean


1,410.00
254.93
43.33

HB

10,435.00




10,520.00
1,093.60
452.40



96,740.00

Mean



238.50





Standard
Deviation


16,663.45
4,592.73
579.81

324.69

109,210.58
Standard



1,897.71
166.39
57.74

4.63

17,931.46
Standard
Deviation


10,146.77
1,156.83
461.91

98.83

85,721.28
Standard
Deviation


301.30
366.76
46.38

1.26

3,340.21
Notes:
       Values used in the final regression equations.
                                          55

-------
TABLE A-5. QUANTITATIVE REFERENCE LABORATORY DATA—YORK SITE

Chemical

Minimum

Maximum

Mean
Standard
Deviation
Node 1(1 5 to 16 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX

Chemical

200.00
53.00
a

64.65

580.00

Minimum

1,900.00
2,270.00
a

755.70

4,700.00

Maximum

smm
773.20
a

260.31



Mean
*
948.75
905.13
a

284.20

2,317.79
Standard
Deviation
Node 2 (17 to 18 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX

Chemical

1,600.00
88.70
5.40

40.41

1,900.00

Minimum

7,200.00
1,380.00
33.00

252.47

14,100.00

Maximum

1SS1S2S
496.94




8,946.0Q

Mean

2,846.58
535.84
12.51

98.64

5,131.92
Standard
Deviation
Node 4 (14 to 15 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX

Chemical

2,200.00
647.00
37.00

166.06

4,540.00

Minimum

19,000.00
6,450.00
97.00

1,048.82

36,300.00

Maximum

ilfiSSfl



£!l»§i

19,542.00

Mean

6,412.49
2,361.49
24.87

342.68

11,891.11
Standard
Deviation
Node 5 (1.5 to 2.5 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX
Notes:
a
ND

ND
15.20
ND

0.01

ND

No data.
Not detected.

ND
138.00
ND

0.95

ND




ND

ND

0.45

ND




ND
67.83
ND

0.40

ND

Minimum

Maximum Mean
Standard
Deviation
Node 2 (13.5 to 14.5 feet)

1,800.00
156.00
14.00

159.85

1,800.00

Minimum
Node 3 (17 to 18 feet)

290.00
261.00
7.50

97.45

1,090.00

Minimum
Node 4 (18 to 19 feet)

92.00
18.20
6.00

1.40


13,000.00 IflgJlgfl
2,710.00 1,539.20
45.00

466.41 Msum

23,280.00 7,556.00

Maximum Mean


2,700.00 W§i$i§M
1 ,450.00 SSBUfl
30.00 Jg5J

359.67

4,900.00 msam

Maximum Mean


57,000.00 21,810.50
4,000.00 1,878.15
280.00: 175.33

2,332.21 798.91

274.00 128,800.00 42,528.80





























4,714.69
1,035.82
11.90

129.85

8,853.57
Standard
Deviation


874.25
513.04
9.30

116.76

1,641.76
Standard
Deviation


27,363.23
2,155.00
148.01

1,116.53

59,970.23














      Values used in the final regression equations.
                                         56

-------
TABLE A-6. QUANTITATIVE REFERENCE LABORATORY DATA—FORT RILEY SITE
Chemical
Minimum
Maximum
Mean
Standard
Deviation
Node 1 (2 to 3 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX
Chemical
79.00
27.50
6.00
0.97
339.00
Minimum
3,700.00
15,800.00
110.00
260.60
20,610.00
Maximum
1,075.80
5,727.98
47.50
89.15
6,412.40
Mean
1,502.51
6,698.52
44.90
105.57
8,295.04
Standard
Deviation
Node 2 (6 to 7 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX
Chemical
107.50
48.60
9.00
0.05
89.00
Minimum
2,000.00
7,720.00
98.00
42.05
10,070.00
Maximum
762.50
2,169.32
41.83
10.98
2,956.63
Mean
1,072.32
3,186.75
48.87
18.15
4,756.48
Standard
Deviation
Node 5 (10.5 to 11. 5 feet)
Ethyl-
benzene
TPH
VPH
Total
PAH
Total
BTEX
1,400.00,
17,700.00
250.00
157.14
23,270.00
29,000.00
32,800.00
430.00
340.26
96,300.00
13,680.00
waflfwm
SfeSbEHS

24M8

11,238.86
5,934.81
80.81
67.41
27,580.73
Minimum
Maximum Mean
Standard
Deviation
Node 1(1 3 to 14 feet)
100.00
32.90
9.20
0.02
230.00
Minimum
20,000.00 8,595.00
3,110.00 1,416.00
320.00 183.55
96.62 31.44
70,000.00 26,497.60
Maximum Mean
10,009.87
1,607.95
152.78
44.47
35,204.28
Standard
Deviation
Node 2 (17 to 18 feet)
28,000.00
7,050.00
530.00
60.66
147,000.00
Minimum
60,000.00
16,900.00 SBIIiiPJl
1,200.00 SM
224.76
254.ooo.oo wmmm
Maximum Mean
13,464.77
4,182.11
259.71
72.14
47,704.30
Standard
Deviation
Node 5 (16 to 17 feet)
20,000.00
1,090.00
170.00
18.23
63,100.00
55,000.00 ESiHi
9,630.00 £9JffipJ
930.00
162.28 gg^l
219,700.00
13,612.49
3,470.96
289.34
58.45
62,975.05
Notes:
      Values used in the final regression equations.
                                         57

-------
TABLE A-7.  GEOTECHNICAL AND TOC DATA—ATLANTIC SITE
Node/
Grid
1/F
1/F
1/F
1/F
1/F
4/C
4/C
Notes:
mg/kg
mm
Depth
(feet)
2-3
10-11
20.5-21
30.5 - 31
35 - 35.5
9-10
15-16

TOC
(mq/kq) %>2 mm
4.000 .03
ND 0
600 1
200 28.38
400 0
3,800 0
3,200 0

% Sand
(0.5-2 mm)
12.43
36
50.84
62.78
24.72
19.34
44.79

% Silt
(2-50 «m)
58.33
43.78
34.17
4.72
44.73
51.24
30.57

% Clay
(<2 urn)
29.21
20.22
13.99
4.12
30.55
29.42
24.64

USDA
Classification
Silty clay loam
Loam
Loam
Sand
Clay loam
Silty clay loam
Loam

uses
Classification
Sandy lean clay (CL)
Silt or clay (CL or ML)
Silt or clay (CL or ML)
Well to poorly graded
sand (SW or SP)
Sandy lean clay or
sandy silt (CL or ML)
Sandy lean silt or
sandy lean clay (CL or
ML)
Silt or clay (CL or ML

Milligram per kilogram.
Millimeter.
urn      Micrometer.
USDA   United States Department of Agriculture.
USCS   Unified Soil Classification System, ( ) two-letter classification code.
ND      Not detected.
                                               58

-------
       TABLE A-8. GEOTECHNICAL AND TOC DATA—YORK SITE
Node/
Grid
1/G
1/G
1/G
1/G
3/C
3/C
Depth
(feet)
5-6
7-8
15-15.5
18.5-19
12-13
16.5-17
TOC
(ma/ka)
ND
2,800
1,400
490
3,200
2,600
%>2 mm
0.00
0.05
0.23
30.54
0.00
6.69
% Sand
(0.5-2 mm)
13.66
26.08
60.24
46.38
8.90
52.48
% Silt
(2-50 Mm)
58.94
51.05
20.93
12.09
60.43
17.92
% Clay
(<2 itm)
27.40
22.82
18.60
10.99
30.67
22.91
USDA
Classification
Silly clay loam
Silt loam
Sandy loam
Sandy loam
Silty clay loam
Sandy clay loam
uses
Classification
Clay or silt with sand
(CL or ML)
Clay or silt with sand
(CL or ML)
Silty to Clayey sand
(SM or SC)
Poorly graded sand
with silt or clay
(SW-SC or SP-SC)
Silt or lean clay with
sand (CL or ML)
Clayey or silly sand
(SM or SC)
       Notes:

       mg/kg   Milligram per kilogram.
       mm     Millimeter.
       MID     Micrometer.
       USDA   United States Department of Agriculture.
       USCS   Unified Soil Classification System, ( ) two-letter classification code.
       ND     Not detected.
       TABLE A-9. GEOTECHNICAL AND TOC DATA—FORT RILEY SITE
Node/
Grid
4/H
4/H
4/H
4/H
2/E
3/G
Notes:
mg/kg
mm
Mm
USDA
USCS
Depth TOC
(feet) (ma/kg)
2 - 3 3,400
7.5 - 8.5 600
15-16 800
29 - 30 300
15-16 4,600
5.5 - 6.5 9,000

%>2 mm
0.00
.16
0.00
20.36
.11
.10

% Sand
(0.5-2 mm)
31.32
60.76
62.44
57.48
55.13
47.61

% Silt
(2-50 urn)
43.48
22.08
19.16
10.46
25.87
36.02

% Clay
25.20
17.00
18.40
11.70
18.89
16.27

USDA
Classification
Loam
Sandy loam
Sandy loam
Sandy loam
Sandy loam
Loam

Milligram per kilogram.
Millimeter.
Micrometer.
United States Department of Agriculture.
Unified Soil Classification System, ( ) two-letter classification code.
*U.S. GOVERNMENT PRINTING OFFICE:  1995-653-424
                                                       59

-------

-------